y |
‘ is Bl CA ae Ly |
Wille Ca licidl Acganold ¥ |

AM bf bs & ZL, |

MEL’ ,

FALCONER’S

PALHZONTOLOGICAL MEMOIRS
&e.

VOLUME I.

_ Mare tone Mle lon Off. _Vlmcree

2 arrenen es ©

: > LONDON
PRINTED BY SPOTTISWOODE 4ND CO” -
NEW-STREET SQUARE

PALAONTOLOGICAL MEMOIRS AND NOTES

OF THE LATE

HUGH FALCONER, A.M, M.D.

VICE-PRESIDENT OF THE ROYAL SOCIETY;

FOREIGN SECRETARY OF THE GEOLOGICAL SOCIETY OF LONDON ;
AND FOR MANY YEARS

SUPERINTENDENT OF THE H. E. I. COMPANY’S BOTANICAL GARDENS
AT SUHARUNPOOR AND CALCUTTA.

WITH A

Hiographical Shetch of the Author.

COMPILED AND EDITED BY

CHARLES MURCHISON, M.D., F-.R.S.

FELLOW OF THE ROYAL COLLEGE OF PHYSICIANS OF LONDON.

PREPAID ELI

VOL. I.

FAUNA ANTIQUA SIVALENSIS.

a
a NAT. HIST. @&
1 9 SEP 1968

LONDON :
ROBERT HARDWICKE, 192 PICCADILLY.
1868.

TO

COLONEL SIR PROBY T. CAUTLEY, K.C.B,

MEMBER OF THE COUNCIL OF INDIA,

This Volume,

EMBRACING THE RESULTS OF HIS OWN RESEARCHES
AND THOSE OF HIS DEPARTED FRIEND AND COLLEAGUE
ON THE FOSSIL FAUNA OF THE SEWALIK HILLS,
WHICH OBTAINED FOR THEM
A WORLD-WIDE REPUTATION IN SCIENCE,

ds Dedicated

BY

THE EDITOR.

PREFACE.

Qs the announcement of the death of Dr, Huan Fat-
CONER, on January 31,1865, there was but one feeling
among men of science—that a master mind had passed
away, and left little behind of the vast amount of pa-
lxontological knowledge acquired during a period of
thirty years. Gifted with a memory rarely equalled,
to which he too often confided the results of his re-
‘searches, and having a natural aversion to publish his
views without thoroughly sifting every fact which could
be brought forward to corroborate or refute them,
there can be no doubt that the loss to science conse-
quent on his death is in a great measure irreparable.
For many years, however, Falconer had been in the
habit of noting down careful descriptions, with measure-
ments, of many specimens which he believed to elucidate
the subject of his investigations. The present work is
a collection of memoirs, some of which were published
during his lifetime, but many were not, together with
such passages from his Note-books as have appeared to
the Editor most important and complete.

The work is divided into two volumes, of which the
first gives the results of the author’s investigations on the
Fossil Zoology of the Sewalik Hills, and the second is

Vill PREFACE.

composed of memoirs and observations, for the most
part written subsequently to his return to Europe.
Many of the memoirs in the first volume have been
already published, but in journals which are accessible
to few scientific men in England or on the Continent.
Others, though written many years ago, are now pub-
lished for the first time. The researches on the Fossil
Fauna of the Sewalik Hills were conducted by Dr. Fal-
coner jointly with his friend Captain (now Sir Proby
T.) Cautley, and all the new species were announced
in their joint names. Some of the published memoirs
appeared in the names of both, and others in the name of
one only of these observers; but all have been here col-
lected, so as to bring together as complete an account of
the Sewalik Fossils as is now possible. With the same
object, several short notes by Messrs. Baker and Durand*
have likewise been inserted. Most of the papers have
also been enriched by appendices, in the form of extracts
from Dr. Falconer’s Note-books and correspondence, or
from his published descriptions of the Sewalik Fossils in
the Catalogue of the Asiatic Society’s Museum in Cal-
cutta. The first volume likewise comprises an Index to
all the published plates of the ‘ Fauna Antiqua Siva-
lensis,’ compiled by the Editor from entries in the
author’s Note-books, from references to the figures in
his published memoirs, and from labels in his hand-
writing on the figured specimens now in the Palzonto-
logical Gallery of the British Museum. For assistance
in identifying these specimens the Editor is under obli-
gations to Mr. W. Davies of the British Museum. In
the first volume will also be found a Report by Dr.

* Now General Baker, and General Sir Henry M. Durand, K.S.I.

PREFACE. 1X

Falconer of his expedition to Cashmeer and Little
Tibet, in 1837 -38, with notes and appendices.

Of the second volume certain of the memoirs have
been already published, but many now appear for the
first time, and particularly the important memoir and
memoranda on the fossil species of Rhinoceros ; while
all have received valuable additions from the author’s
Note-books and letters.

It is necessary to state that many of the memoirs
and notes now published were written long before the
author’s death, and that possibly he may have seen rea-
son subsequently to modify or change certain of the
views expressed. Any one who knew Falconer’s ex-
treme caution, and the frequency with which he revised
and re-studied all his views and observations before com-
mitting them to the press, will probably feel that he
would have deprecated the exposure which has now been
made of what in some instances may have been only
first impressions of the specimens which he described.
But unfortunately, the powerful and discriminating in-
tellect which would have erased the error of first im-
pressions and moulded scattered observations into a har-
monious whole was gone; and the question to be decided
was, whether or not the observations carefully made,
aud in many instances minutely and accurately recorded
by a shrewd, experienced, and conscientious searcher
after truth, were to be for ever lost? As an executor of
my departed friend, I have not hesitated in adopting the
alternative which I believe will conduce most to the
perpetuation of his name in Paleontological Science;
while I am not the less assured that the publication of
these volumes will meet with the approval of his former
friends who still labour. in the same field of research.

x PREFACE,

At the same time it ought to be distinctly understood,
that many of the memoranda and notes which are now
made public are not only fragmentary, but were simply
the expressions of the author’s mind at the dates which
they bear. Throughout the volumes these notes are
printed in smaller type than that of the finished memoirs.

Since Dr. Falconer’s death the Editor has deposited
in the Palwontological Department of the British
Museum a large number of the specimens, casts, and
drawings which are now for the first time described, and
which will thus be accessible to those who may be in-
terested in their examination. ‘These include all the
original drawings illustrating the important descriptions
of Fossil Rhinoceros and Llephas Melitensis, copies of
seventeen unpublished plates of the ‘Fauna Antiqua
Sivalensis,’ as well as outline tracings for the drawings
necessary to complete that work.

Falconer was not only a paleontologist, but a botanist
of repute. His Indian career was spent as Superin-
tendent of the Botanic Gardens of Suharunpoor and
Calcutta, and in Calcutta he was also Professor of
Botany in the Medical College. Of late years, however,
he paid comparatively little attention to botany. But,
although it is as a paleontologist that he will hereafter
be remembered, several of his botanical memoirs are
of great interest and value. A complete list of them is
given in the first volume, but none of them have been
reproduced.

The biography which is given in the first volume will
not only interest Falconer’s former friends, but will ex-
plain the circumstances under which some of his re-
searches were undertaken, and others suspended. It has
been constructed from brief memoranda written by him-

PREFACE. xl

self, from letters addressed by him to his former fellow-
students, the Rev. Dr. Gordon of Birnie, and the Rev.
Duncan Campbell of Pentridge Rectory, to the late
Professor Jameson of Edinburgh, and to his intimate
friends Sir Proby Cautley, Mr. Arthur Grote, Colonel
Wood, and M. Ed. Lartet, which have been kindly en-
trusted to me for the purpose, as well as from other
sources. It might have been possible to extend it; but
I have preferred incorporating the scientific information,
which I have been enabled to extract from Dr. Fal-
coner’s correspondence, in the form of foot-notes, with
the memoir to which in each case it refers. Throughout
the work, the orthography of Indian names, on which
scarcely two writers agree, is what was adopted by Dr.
Falconer himself; while that of the specific names of
fossils, as was Falconer’s habit, has been rendered in ac-
cordance with Bronn’s ‘ Nomenclator Paleeontologicus.’

The admirable lithographs which illustrate the work
have been executed by Mr. Joseph Dinkel, and have
been partly copied from figures in the ‘ Fauna Antiqua
Sivalensis,’ or from original drawings belonging to Dr.
Falconer, and partly drawn from the original specimens
in the British Museum or in other collections. The
Rey. John Gunn of Irstead, with his wonted readiness
to render his valuable collection available for the advance-
ment of science, has forwarded several of his specimens
to be figured.

To the Editor of these volumes it is a matter of regret
that the work which he has now completed has not been
executed by some one more competent to deal with the
subjects of which they treat, and less liable to pro-
fessional interruptions. But as the object has been to
place the results of Dr. Falconer’s labours before the

x1 PREFACE.

scientific public without any addition or comment, an
expert in the science did not seem necessary; while the
time required for collecting and arranging his widely-
scattered observations would, probably, have precluded
any of the small number of British paleontologists from
undertaking the task. Lastly, although further delay
might have rendered the work more complete, this
advantage would have been more than counterbalanced
by the loss to paleontologists, in the meantime, of the
results of some of Falconer’s researches.

‘Nec dubitamus multa esse que nos preeterierunt: homines enim
sumus et occupati officiis, subsecivisque temporibus ista cura-
: %3
vimus.

1 Pliny, Preef. in ‘ Nat. Hist.’

CONTENTS

OF

TORE LR SL | VOL Lo. ULMe Be

BioGRAPHICAL SKETCH. . ‘ ; : P

List oF Botanican Mrmorrs anD Reports BY Dr. FALCONER

I. Introductory Observations on the Geography, Geological
Structure, and Fossil Remains of the Sewalik Hills .

II. On the Structure of the Sewalik Hills, and the Organic
Remains found in them. By Proby T. Cautley, Esq., Capt.
Bengal Artillery : : : : : ;

III. On the Fossil Species of Sr aan and Mastodon found in the
Sewalik Hills

APPENDIX.—1. peat of Fossil Remains of Ele-
phants in the Museum of the Asiatic Society of
Bengal

2. Descriptions of Fossil Remains of Mastodon in the
Museum of the Asiatic Society of Bengal

3. Description of Fossil Molars of Mastodon Pandionis
from the Deccan

4, Note on the Teeth of the Mastodon a dents étroites of
the Sewalik Hills. By Capt. P. T. Cautley

5. Note on the Mastodons of the Sewaliks. By Capt.
P. T. Cautley . . : : ;

IV. On the Fossil Hippopotamus of the Sewalik Hills. By Hugh
Falconer, M.D., and Capt. P. T. Cautley

APPENDIX. 2g) Description of specimens of Fossil Bho.
potamus in the Museum of the Asiatic Society of
Bengal

142

XIV

Wie

Vil.

VII.

IX.

XII.

CONTENTS OF

2. Note upon a young Skull of Merycopotamus dissimilis
from Burmah . : . : : ’

. Description of a Fragment of a Jaw of an unknown extinct

Pachydermatous Animal, from the Valley of the Murkunda.
Tetraconodon magnum vel Cherotherium 5 > ;

On the species of Fossil Rhinoceros found in the Sewalik
Mills . 31) ee at Minin Jy ,

AppEnDIx.—l. Description by Messrs. Baie and Du-
rand of the Fossil Rhinoceros of the Sewalik Hills.

2. Description of Fossil Remains of Rhinoceros in the
Museum of the Asiatic Society of Bengal < ;

On the Fossil Rhinoceros of Central Tibet, and its relation to
the recent upheaval of the Himalayahs_. ‘ : .

On the Fossil Equide of the Sewalik Hills

AppENDIX.—Descriptions of Fossil Equide in the Mu-
seum of the Asiatic Society of Bengal

On some Fossil Remains of Anoplotherium and Giraffe from
the Sewalik Hills. By H. Falconer, M.D,, and Capt. P. T.
Cautley :

AppENDIx.—l. Description of Fossil Remains of ' Giraffe
in the Museum of the Asiatic Society of Bengal

2. On the number of Existing Species of Giraffe .

. On Chalicotheritum Sivalense . ‘ , é F -

AppENDIXx.—1. Description of Upper Jaw of Chalicothe-
rium Goldfussi in the Museum of Darmstadt, and of
the remains of the ‘ Nestoritherium’ of Kaup in the
Museum of Munich . . . . :

2. On the inconvenience of retaining the Dichobunes in
Anoplotherium

. On the Fossil Camel of the Sewalik Hills. By H. es

M.D., and Capt. P. T. Cautley . : : :
AppENDIx.—1. On the Fossil Camelide of a Sewaliks.
By Capt. P. T. Cautley . ‘ . : .

2. Description of Fossil Remains of Camel in the Mu-
seum of the Asiatic Society of Bengal . -

Sivatherium giganteum; a new Fossil Ruminant Genus from
the Valley of the Murkunda. By Hugh Falconer, M.D.,
and Capt. P. T. Cautley . 3 t : - : <

Apprnpix.—1. Note on a Specimen of Sivatherium dis-
covered by Col. Colvin . : : : ‘

PAGE

247

266

THE FIRST VOLUME.

2. Notes on Bones of Sivatherium not described in
original memoir

3. Description of Fossil Remains of Sivatherium in the
Museum of the Asiatic Society of Bengal

XIII. The Fossil Bovide, Cervide, and Antilopide of India

AppENDIx.—1l. Description of Remains of large Bovine
Ruminants from the Sewalik Hills in the Museum
of the Asiatic Society of Bengal

2, Description of Fossil Remains of Bos, from the Ner-
budda, in the Museum of the Asiatic gel of
Bengal

3. Description of Fossil Remains of ee from the
Sewalik Hills, in the Museum of the Asiatic aa
of Bengal (Antilope Paleindica)

4, Description, by Capt. Baker, of the Fossil cree of
the Sewalik Hills : :

XIV. On the Fossil Quadrumana of the Sewalik Hills .

1. Notice on the Remains of a Fossil Monkey from the
Tertiary Strata of the Sewalik Hills. By ree pau
Cautley and H. Falconer, M.D. :

Sub-Himalayan Fossil Remains of the Didiencos Col-
lection. By Lieuts. W. E, Baker and H. M. Durand

. An additional Fossil Species of the Order Quadrumana
from the Sewalik Hills. By H. Falconer, M.D., and
Capt. P. T. Cautley c : : :

4, An additional Quadrumanous Remains from the Ter-
tiary Deposits of the Sewalik Hills. By H. ea
M.D., and Capt. P. T. Cautley .

Note on a Correction of Published Statements re-
specting Fossil Quadrumana. By H. Falconer, M.D.

Lo

is)

cx

XV. On Felis cristata, anew Fossil Tiger from the Sewalik Hills.
By H. Falconer, M.D., and Capt. P. T. Cautley .

XVI. On Ursus (Hyenarctos) Sivalensis, a new Fossil Species from
the Sewalik Hills. By Capt. P. T. Cautley and H., Fal-
coner, M.D. : , : ; : ‘ : :

APPENDIX.—1. Reply to M. de Blainville’s observations
on Ursus Sivalensis .

2. Comparison of Ursus (Hyenarctos) Stivalensis with
Brayard’s Ursine Head from Buenos Ayres

XVI. On Enhydriodon (Amyxodon), a Fossil Genus allied to
Lutra, from the Tertiary Strata of the Sewalik Hills .

284.

300

309

315

Xvi CONTENTS OF

Apprenpitx.—l. Comparison of Head of Indian Otter
with Skulls of Otters in the College of Surgeons’
Museum . : : . . : :

2. Description of Fragment of Lower Jaw of Enhydriodon

XVIII. On the Fossil Carnivora of the Sewalik Hills

APpPENDIX.—1. Description, by Messrs. Baker and Du-
rand, of Fossil Carnivora from the Sub-Himalayahs

2. Description of Fossil Remains of Cainivora from the
Sewalik Hills in the Museum of the Asiatic Society
of Bengal F , :

3. Description of Fossil ike

XIX. On the Fossil Crocodiles of the Sewalik Hills. By Capt.
P. T. Cautley .

Apprnpix.—l., On the Crocodiles of India, recent and
fossil

Memoranda on Crocodiles in the Belfast Museum

Memorandum on Crocodiius Schlegelit .

rece iho

Description of Fossil Remains of Crocodiles from Ava,
Perim Island, and the Nerbudda, in the Museum of
the Asiatic Society of Bengal .

XX. On the Colossochelys Atlas, a gigantic Fossil Tortoise from
the Sewalik Hills
1. On some Remains of the Megulochelys Sivalensis

2. On the Osteological Characters and Palzeontological
History of the Colossochelys Atlas

Further Observations on the Colossochelys Atlas

gs

4, Anatomical Description of the Colossochelys Atlas

ce

Description of Fossil Remains of Colossochelys and
Testudo in the Museum of the Asiatic Society of
Bengal

XXI. On a Fossil Species of Emys from the Sewalik Hills, refer-
rable to the existing Emys tecta (Bell) F

AppENDIXxX.—Further observations on the Fossil Lmys
tecta of the Sewalik Hills

XXII. Note on the Fossil Freshwater Shells from the Sewalik
Hills. By the late Professor Edward Forbes. - é

XXIII. Description of some Fossil Remains of Dinotherium, Giraffe,
Bramatherium, and other Mammalia, from Perim Island,
Gulf of Cambay, Western Coast of India

PAGE

THE FIRST VOLUME.

ApprenpIx.—1. Description of a Fragment of Fossil Bone
from Perim Island, in the Collection of Mr. Henry
Duckworth . - ¢ : : : c .
2. Microscopic Characters of Mr. Tait’s supposed Vertebra
of Dinotherium from Perim Island : : .
XXIV. Note on certain Specimens of Fossil Animal Remains from
Ava . . ‘ : 5 ; . : : :
XXYV. Note on Fossil Remains found in the Valley of the ee
below Attock, and at Jubbulpoor . c - :
XXVI. Note on an extraordinary variety of Elastic Sandstone é
XXVII. A description of the Plates in the ‘Fauna Antiqua Sivalensis’
XXVIII. Official Report of oe to Cashmeer and Little Tibet,
in 1837-88 . : - C
APpPENDIx.—1. Note on the Hivos or. Cervus Cashmee-
rianus d > : a : : : .
2. Note on a variety of Musk-Deer, or Kustora :
3. The Gcats of Cashmeer . .
4, Note on the Mule Breeds ee the Yak and the
Cow in Tibet . : : : . Meyers .
5. The Cashmeer Marten : : : . - :
6. The Tibetian Hare. : - : c
7. The Tibetian Marmot, Arctomys Tibetana .
XXIX. Miscellaneous Notes on Indian Zoology . : : :
1. Note on Cervus Duvaucellit
2. Note on an Indian Species of Esox

3. Memorandum regarding the Predaceous Habits of
Indian Frogs . : ° hits

VOL. I. | a

XVill

PLATE

Iil.

. Skull and lower jaw of Merycopotamus dissimilis

LIST OF ILLUSTRATIONS

LIST OF ILLUSTRATIONS

THE FIRST VOLUME.

A. PLATES.

PAGE

. Portrait of Dr. Falconer, drawn on stone, by Mr. Joseph

Dinkel, from a photcgraph by Mr. Ernest Edwards. _ Frontispiece

; Map illustrating Dr. Falconer’s observations on the Geology

of India, . - : - : : : : 6
Map of Country between the Sutlej and the Ramgunga . 380

. Sections of teeth of ae insignis, £. lanes e" and £.

Africanus , ‘ - 74
. Sections of teeth of Biopha Ey pone tcus, E. Iie and £.
primigenius - - , Sle
. Sections of teeth of Elephas ¢ Ganesa, Mastodon nies Dino-
thertum Indicum, and D. giyanteum . ° ; 82
. Sections of teeth of Mastodon Sivalensis, M. Ohiatious, and
Dinotherium giganteum ; 6 5 ; : apres:
. Lower jaw and molars of Mastodon Andium . . 100
. Molars of Mastodon Sivalensis and M. Perimensis . > 1s
. Skull of Mastodon Sivalensis . . 128

. Skull of Aippopotamus (Hexapr Pee Sivalensis . : . 1380

. Upper and lower jaws with molars and incisors of Hippopotamus

(Hexaprotodon) Sivalensis . : : . . . . 134

Skull and molars of Rhinoceros Sivalensis, Skull and lower
jaw of R. platyrhinus . ; : . 158

. Skull of Merycopotamus dissimilis from oe Fragments

of bones of fossil Rhinoceros from the Niti Pass in Tibet . 178

Vertebra and teeth of Camelopardalis Sivalensis. Teeth of C,
affinis ‘ ‘ . : . ‘ a

IN THE FIRST VOLUME.

XVII. Chalicotherium Sivalense. Skull, lower jaw, teeth, &c. :
XVUL. Camelus Sivalensis, Skull, lower jaw, and teeth. : :
XIX. Skull of Sivatherium giganteum . : 5 . : .
XX. Skull of Stvatheriwm giganteum . . . . : A

XXI. Sivatherium giganteum. var restoration of skull, lower
jaw, and sternum . : : : .

XXII. Skulls of Bos Paleindicus and Bos Namatdicus . . :
XXIII. Skull of Antilope Paleindica . 0 . . . :
XXIV. Fossil Quadrumana of the Sewalik Hills. i F ;

XXV. Skull of Felis cristata. Fragment of skull and of lower jaw
of Depranodon (Machatrodus) Spat and canines of

Drepanodon latidens, : : é :

XXVI. Skull and lower jaw of Ursus Uh yenar a Sivalonsis.
Molars of Ursus Namadicus . : . : . :

XXVIL. Skulls of Enhydriodon Sivalensis. Skull and lower jaw of
Lutra Paleindica . - . : : : 5

XXVIII. Fragments of fossil skulls of Crocodilus bombifrons and Lep-
torhynchus Gangeticus . : : : . . .

XXIX. Fragments of fossil skulls and lower jaws of Crocodilus
crassidens, Crocodilus ee and Leptorhynchus Gange-

ticus . . . . . . .
XXX. Colossochelys ies Bisel Syphosema and Ungual
bones. . . . .

XXXI. Colossochelys Atlas. Skull, acer and Paes Pe wh ae

XXXIL Shell of fossil Hmys tecta. Lower end of humerus of Va-
ranus Sivalensis, or fossil Monitor of the Sewalik Hills .

XXXII. Molars of Bramatherium Perimense and of Dinotherium
Indicum from Perim Island . , A 3 - 3

XXXIV. Molars of Mastodon Pandionis and of Antoletherium. Fossil
Vertebra from Jubbulpoor . . . . : .

B. WOOD ENGRAVINGS.

1. Section through Himalayahs and Sewalik Hills . pee ;
2, Section at Deoni, on the Murkunda River . ; ; ; :
8. Section at Tuloagpoor . c - c . - * ; é
4, Section showing dislocation of strata below the village of Tuloag-

poor : ; ea ec eM paca he Ta aaeieots aie

5. Teeth of Tetraconodon, or Cheerothertum magnum .

6, Astragalus of Sewalik fossil Monkey, upper surface. : :
a2

x1x

PAGE

210
232
248
250

266
280
290
300
316

324

334

xx LIST OF ILLUSTRATIONS IN THE FIRST VOLUME.

PAGE
7. Astragalus of existing Semnopithecus entellus, upper surface . « 294
8. Astragalus of Sewalik fossil Monkey, under surface . - . 294
9. Astragalus of existing Semnopithecus entellus, under surface « . 294

10. Sketch by the late Professor Edward Forbes, illustrating the
Hindoo mythological conception of an Elephant bearing the
world, and supported on the back of a Tortoise . : . . 297

11. Canine tooth of fossil Monkey, with lower jaw of the Orang-
Outany of Sumatra . 3 c : - . : : . 304

12. Diagrammatic restoration of Colossochelys Atlas. . 5 5 362

13. Sketch by the late Professor Edward Forbes, showing the Ble-
phant victorious over the Tortoise, supporting the ‘world, and
unfolding the mysteries of the ‘Fauna Sivalensis’ . - 2 Old:

14, Map of the district near Perim Island. . : : : iim. OO

15. Head of Cervus aia ga from a drawing hes Gade Vigne,

Esq. : é . 578

16. Cervus Cashmeerianus, oe a drawing G. aly Viene, ag. . 578

Page 22 line

» 23 5

» 23 4, 7
i ko 30.
Paol. 3; 5
pe OLS sy 7
» 388 ,,

“5 53 Cs, 1
59) gee lGy Gamer 35
a 62) -,, 21
Oa 2
pe 2S) ss 1
oy elite ae 8
ny 6218) 55 13
ear oO 5;
weOs 4, 2
pe. 339

» 440 ,, 2
» 441 , 6&7.
» 467

me 408) yy
O02

» 902 and 503.
» 513line 10.
» 024 ,, 4
» 546 ,, 6
» 554

» 982 ,,18&22.
» 587

CORRIGENDA AND ADDENDA.

15,

For Netee Pars, read Niti Pass.

1. For Brahmatherium, read Bramatherium.

. For Antelope Paleindicus, read Antilope Paleindica.

For Successively, read successively.
of foot-note.
of do.
In description of Fig. 4, for ‘a. Clay,’ read ‘a. Upper soil.’
For de read di.

For planes, read plains.

Add inyerted comma. after mountains.

of foot-note 3.

. For tichorinus, read tichorhinus.

. For Maginnud, read Moginund.

For Maginnud, read Moginund

. For conicle, read conical.
. For anteror, read anterior.

. For vallies, read valleys.

8rd column of Table. For Pethecus, read Pithecus.

. For channeled, read channelled.

Since this page was printed, the Editor has had an oppor-
tunity of perusing a letter from Dr. Falconer to M. de
Blainyille, dated October 4, 1847, in which reference is
made to two species of fossil Hyena, and to two species of
fossil Felis (Ff. cristata and F. paleotigris), from the
Sewalik Hills. Casts of these fossils were transmitted to
the Jardin des Plantes.

. Omit Reproduced in Plate IX. of Vol. II.

For first, or antepenultimate, read pre-antepenultimate (a),
antepenultimate (0).

Under fig. 5 of Plate XXXVI., for last? read penultimate.

7. For Portwick, read Postwick.

5. Omit The drawing shows the descending process of the jaw.

Substitute upper jaw for lower jaw, in descriptions of fiqures
1, 3, 11, and 15 of Plate LXTI,

For tichorinus, read tichorhinus.
of description of Fig. 4. For premolars, read premolar.
of description of Fig. 7. For p. 354, read p. 554,

In a letter to M. de Blainville, dated October 4, 1847, Dr.
Falconer designated certain remains of the Sewalik fossil
birds, ‘ Struthio paleindica,’

For Martin, read Marten.

For Cervus Duvancelli, read Cervus Duyaucellii.

BIOGRAPHICAL SKETCH.

N the 29th of February, 1808, Hucu FaLconer was born

at Forres, in the North of Scotland—a town beautifully
situated on the banks of the Findhorn, but best known from
its traditional connection with the ‘blasted heath’ of Macbeth.
He was the youngest of a family of five sons and two
daughters, his father, Mr. David Falconer, being a des-
cendant of an ancient family, the Falconers of Lethen and
Halkerton. He received his early education at the Gram-
mar School of Forres, where he attracted the notice of his
teachers from his wonderful memory and facility for ac-
quiring languages; and then, aided by the resources of
an elder brother in India, he entered the University of
King’s College, Aberdeen, to pass through the established
Scotch Academical curriculum of classical literature and
science, extending over a period of four years. A fellow
student, who was very intimate with him at that time, re-
members well his diligence and steady conduct, and adds
that even then, ‘he was an immense favourite with his class-
fellows, and was remarked for his playful genial humour, and
frank, generous, winning disposition,’ and that ‘they who
watched him closely could detect in the young student the
penetrating intellect and shrewdness in observing which
distinguished the future paleontologist.’ Reading was
always his favourite amusement, and although the subjects
which he studied were very varied, he was especially fond of
everything relating to Natural History. A book entitled
‘The Three Hundred Animals’ was an especial favourite,

1 Letter from the Rey. Duncan Campbell,

XXIV BIOGRAPHICAL SKETCH.

and the intervals of his sessions at King’s College, when he
was a pupil of Drs. Smith and Adams of Forres, were chiefly
spent in studying the botany of the neighbourhood and in
watching the habits of the many animals which he kept as
pets. After receiving the degree of Master of Arts at
Aberdeen, in 1826, he proceeded to Edinburgh to enter on the
study of Medicine. Here he eagerly followed up his early
tastes for Natural History, under the systematic tuition of
the late Professor Graham in Botany, and Professor Jameson
in Geology and the other branches of Natural History. <Ac-
cording to the testimony of one who knew him well, ‘ he
laboured with untiring energy. Shut up in his apartments
for days together, he allowed himself little relaxation except
that of accompanying the great Wernerian Professor in his
Geological excursions with wallet and hammer, which he
never failed to do.’! His range of study, however, at this time
was perhaps too excursive for solid attainment in any one
walk; for, besides attendance on the numerous classes in Medi-
cine and Natural History, he matriculated as a student in
Divinity, in order to benefit by the renowned teaching of Dr.
Chalmers, then Professor of Divinity in the University. In
1829 he received the degree of M.D. from the Edinburgh
University, his graduation thesis being entitled ‘ De Chorea.’

In the same year he was nominated to the appointment of
Assistant-Surgeon on the Bengal establishment of the Hon.
East India Company, but not having attained the required
age of 22 years, he proceeded to London, where he devoted
the necessary interval to assisting the late Dr. Nathaniel
Wallich in the distribution of his great Indian Herbarium ;
and, under the generously bestowed instruction of Mr. Lons-
dale, to the further study of Geology and Paleontology. The
Museum of the Geological Society of London, under the
charge of Mr. Lonsdale, gave him access to the collection of
Indian fossil mammalia from the banks of the Irrawaddi,
formed by Mr. John Crawfurd during his mission to Ava.
The description of these remains by Mr. Clift had excited
much interest in the scientific world, as the first instance in
which ground had been broken in the paleontology of tro-
-pical regions.

2 The Rey. D. Campbell.

BIOGRAPHICAL SKETCH. XXV

In 1830 Dr. Falconer proceeded to India as an Assistant-
Surgeon in the Hon. Hast India Company’s service, and
arrived in Calcutta in September of the same year. Here he
at once undertook an examination of fossil bones from Ava,
in the possession of the Asiatic Society of Bengal, and
published a description of them in the third volume of the
‘Gleanings in Science,’ an Indian journal then conducted by
Mr. James Prinsep. This notice! was slight and modest in
its scope ; but the cultivators of Science in India were then
few in number, and its appearance at once gave Falconer a
recognized position in their roll.

Early in 1831 Dr. Falconer was ordered to the army
station at Meerut, in the North-Western provinces. His first
and last military duty, during twenty-six years of service, was
to take charge of a detachment of invalids proceeding from
Meerut to the sanitarium of Landour, in the Himalayahs.
This led him to pass through Suharunpoor in April 1831,
where the late Dr. Royle was then superintendent of the
Botanic Garden. Kindred tastes and common pursuits soon
knit Falconer and Royle together ; and, at the instance of his
friend, Falconer was speedily appointed to officiate for him
during leave of absence, and, on Royle’s departure for
Europe, in 1832, to succeed him in charge of the Botanic
Garden. Thus, at the early age of twenty-three, did he find
himself advanced to a responsible and independent public
post, offering to a naturalist the most enviable opportunities
for research; so fertile was the Indian service then in
chances to rise for any young officer who chose to make the
exertion.

Suharunpoor is situated in lat. 29° 58’ N. and long. 77° 30’
E., between the Jumna and Ganges rivers, outside the belt of
Terai forest, which lies between the mountains and the
plains, and is distant about twenty-five miles from the
Sewalik Hills, beyond which rise the Himalayahs. It is thus
most favourably situated as a central station for Natural
History investigations; the rivers, plains, forests, and hills
teeming with life in every shape, and the range of elevation
combining, within a short distance, the features and produc-
tions of tropical temperature and alpine regions insensibly

1 See p. 412 of this yolume.

XXV1 BIOGRAPHICAL SKETCH.

blended. Being a remote provincial station, with at that
time only half-a-dozen European families, the white man
had to draw on local means in all emergencies where the
appliances of civilized life were required; but the intelligence,
docility, and exquisite manual dexterity of the natives, backed
by their faith in the guiding head of the European, furnished
an inexhaustible fund of resource. To construct, for example,
a barometer for mountain explorations, broken tumblers were
melted and blown into a tube; mercury was distilled from
cinnabar purchased in the bazaar; a reservoir was turned out
of box-wood felled in the mountains ; and finally a brass scale
was cast, shaped, and even graduated, by a native blacksmith,
under the superintending eye of the amateur. Or again, he
might be seen superintending the expression of some indi-
genous oil as a substitute for salad oil, when the European
supply had been exhausted. Such discipline was of infinite
value in training the young officer to habits of self-reliance

and to kindly relations with those among whom his lot was -

cast, and no doubt contributed to that great fund of uni-
versal information for which Falconer was afterwards so
remarkable.

In 1831 Dr. Falconer commenced his field explorations, by
investigating the geological formation of the Sewalik Hills.
Captain Herbert, in his mineralogical survey of the N. W.
Himalayahs, had referred the Sub-Himalayahs to the age of
the ‘ New Red Sandstone;’ but Dr. Falconer, on his firsti
visit, from finding beds of incoherent gravel covering the
northern slope of the range, from the occurrence of seams of
lignite and dicotyledonous woods discovered by Lieutenant
Cautley in 1827, and from the mineral characters of the
different strata, inferred that they were of a tertiary age,
and analogous to the Molasse of Switzerland.'! Thirty years
of subsequent research by other geologists have not altered
that determination, although our exact knowledge of the
formation has been greatly extended. Early in 1834 Dr.
Falconer gave a brief account of the Sewalik Hills, describing
their physical features and geological structure, with the
first published section showing their relation to the Hima-
layahs.? The name ‘ Sewalik’ had been vaguely applied

1 Journal Asiatie Society of Bengal, | Edinburgh University, February 8,

March 1832, vol. i. p. 96; and letter | 1836.
to the late Professor Jameson of; 7? Ib., yol. iii. p. 182.

Ee a,

BIOGRAPHICAL SKETCH. XXVIil

before then by Rennell and others to the outer ridges of the
true Himalayahs, and the lower elevations towards the
plains. Dr. Falconer restricted the term definitely to the
flanking tertiary range, which is commonly separated from the
Himalayahs by valleys or Dhoons. The proposed name was
not favourably received at the time by geographical authori-
ties in India; but it is now universally adopted in geography
and geology, as a convenient and well-founded designation.
When, in 1831, Dr. Falconer determined the tertiary age
of the Sewalik Hills, the confirmatory evidence of animal
remains was wanting; but he was led to the conclusion ‘ that
the remains of Mastodon and other large extinct mammalia
would be found either in the gravel or in other deposits
occupying the same position in some part of the range, and
the notice in Ferishta’s Indian History of the bones of giants
being found in the hills in which the Sutlej took its origin
made this opinion the more probable.’’ Still other geolo-
gists, including Govan, Herbert, and a sharp-eyed observer
like Jacquemont, had previously gone over the ground, but
had failed to detect any fossil remains. Towards the end
of 1831, Dr. Falconer, from the indication of a ‘ black cylin-
drical fossil,’ found some years before by his friend and
colleague Capt. (now Sir Proby T.) Cautley, but the real
nature of which had been previously overlooked, was led to
discover bones of crocodiles, tortoises, and other fossil re-
mains in the tertiary strata of the Sewalik Hills. A brief
notice of this important discovery, extracted from a letter by
Dr. Falconer, appeared in the ‘ Journal of the Asiatic Society ’
for March 1832.2 In April 1834, Dr. Falconer discovered
the shell of a fossil Tortoise in the Timli Pass, and imme-
diately after the search was followed up with characteristic
energy by Capt. Cautley in the Kalowala Pass by means of
blasting, and resulted in the discovery of more perfect re-
mains, including Miocene mammalian genera. The finding,
therefore, of the fossil fauna of the Sewalik Hills was not
fortuitous, but a result led up to by researches suggested by
previous special study, and followed out with a definite aim.
The researches thus begun were followed, about the end of
1834, by the discovery by Lieutenants Baker and Durand of

1 Letter to Professor Jameson. letter to Professor Jameson, already re-
? Op. cit. vol. i. pp. 96 and 249, and | ferred to,

XXVill BIOGRAPHICAL SKETCH.

the great ossiferous deposit of the Sewaliks, near the Valley
of Murkunda, and below Nahun. Lieut. Baker’s attention
had been directed to this by his having been presented by
the Nahun Rajah with a fragment of tusk and the fossil
tooth of an Elephant,! which had been picked up at Sumroti,
near the Valley of Pinjore, and which the Rajah regarded as
the remains of giants destroyed by the redoubtable Ram-
chundra. Capt. Cautley and Dr. Falconer were immediately
in the new field. Falconer’s enthusiasm may be judged of
from the following extract of a letter to Mr. Prinsep :—

‘You have heard from Capt. Cautley and Lieut. Baker about the late
fossil discovery up here. I have come in for a lion’s share of them.
In one of my tours I had to return by Nahun, and having heard of the
tooth presented by the Rajah, in October, to Lieut. Baker, I made
inquiry, and had a fragment of tooth presented to me also. I gota
hint of where they came from, and on going to the ground, I reaped a
splendid harvest. Only conceive my good fortune. Within six hours I
got upwards of 300 specimens of fossil bones! This was on November
20th, a couple of days after Lieuts. Baker and Durand had got their first
specimens through their native collectors.’?

Similar remains were subsequently discovered by Dr. Fal-
coner in the Sewalik range eastward of the Ganges, near
Hurdwar.* By the joint labours of Cautley, Falconer, Baker,
and Durand, a sub-tropical mammalian fossil fauna was
brought to light, unexampled for richness and extent in any
other region then known. It included the earliest discovered
fossil Quadrumana, an extraordinary number of Proboscidea
belonging to Mastodon, Stegodon, Loxodon, and Huelephas;
several extinct species of Rhinoceros; Chalicotherium ; two
new subgenera of Hippopotamus, viz. Hexaprotodon and Mery-
copotamus; several species of Sus and Hippohyus, and of

Equus and Hippotherium ; the colossal ruminant Sivatherium, .

together with fossil species of Camel, Giraffe, Cervus, Antilope,

Capra, and new types of Bovide; Carnivora belonging to the

new genera Hyenarctos and Enhydriodon, and also to Drepano-

don, Felis, Hyena, Canis, Gulo, Lutra, &e; among the Aves,

species of Ostrich, Cranes, &c. ; among the Reptilia, Monitors,
1 This tooth was described and figured 2 Journ. As. Soe., vol iv. p. 57.

by Lieut. Baker in the Journ. As. Soc. 3 ‘Note on the Occurrence of Fossil
of Bengal for December, 1862, and was | Bones in the Sewalik Range, eastward of

subsequently determined by Dr. Falconer | Hurdwar.’—Journ. As. Soc. of Bengal,

to belong to Elephas Ganesa. vol. vi. p. 233, 1837.

i i

BIOGRAPHICAL SKETCH. Xxix

and Crocodiles, of living and extinct species, the enormous
Tortoise, Colossochelys Atlas, with numerous species of Hmys
and Trionyxz; and among fossil Fish, Cyprinide and Siluride.
The general facies of the extinct fauna exhibited a congre-
gation of forms participating in European, African, and
Asiatic types. Of the mammalian remains all belonged to
extinct species, but of the Reptilia and Freshwater Shells,
some of the fossil species were identical with species now in
existence on the continent of India; and from this fact,
more than thirty years ago, Dr. Falconer was led to draw
important inferences as to the antiquity of the human race.

Thrown suddenly upon such rich materials, the ordinary
means resorted to by men of science for determining them
by comparison were wanting. Of paleontological works or
osteological collections in that remote quarter of India there
were none. But Falconer was not the man to be baffled by
such discouragements. He appealed to the living forms
abounding in the surrounding forests, rivers, and swamps to
supply the want. Skeletons of all kinds were prepared; the
extinct forms were compared with their nearest living ana-
logues, and a series of memoirs by Dr. Falconer and Captain
Cautley, descriptive of the most remarkable of the newly
discovered forms, appeared in the ‘ Asiatic Researches,’ the
‘Journal of the Asiatic Society of Bengal,’ and in the ‘ Geo-
logical Transactions.’ The Sewalik explorations soon at-
tracted notice in Europe, and in 1837 the Wollaston Medal,
in duplicate, was awarded for their discoveries to Dr. Falconer
and Capt. Cautley by the Geological Society, the fountain of
geological honours in England; while the value of the dis-
tinction was enhanced by the terms in which the president,
Mr. (now Sir Charles) Lyell was pleased to announce the
award.

‘When,’ remarked the President, ‘Capt. Cautley and Dr. Falconer
first discovered these remarkable remains, their curiosity was awakened,

_ and they felt convinced of their great scientific value; but they were
not versed in fossil osteology, and, being stationed on the remote con-
fines of our Indian possessions, they were far distant from any living
authorities or books on Comparative Anatomy to which they could
refer. The manner in which they overcame these disadvantages, and
the enthusiasm with which they continued for years to prosecute their

, researches, when thus isolated from the scientific world, are truly

XXX BIOGRAPHICAL SKETCH.

admirable. Dr. Royle has permitted me to read a part of their corre-
spondence with him, when they were exploring the Sewalik mountains,
and I can bear witness to their extraordinary energy and perseverance.
From time to time they earnestly requested that Cuvier’s works on
Osteology might be sent out to them, and expressed their disappoint-
ment when, from various accidents, these volumes failed to arrive. The
delay, perhaps, was fortunate; for, beg thrown entirely upon their
own resources, they soon found a Museum of Comparative Anatomy in
the surrounding plains, hills, and jungles, where they slew the wild
tigers, buffaloes, antelopes, and other Indian quadrupeds, of which they
preserved the skeletons, besides obtaining specimens of all the reptiles
which inhabited that region. They were compelled to see and think

for themselves, while comparing and discriminating the different recent”

and fossil bones and reasoning on the laws of comparative osteology,
till at length they were fully prepared to appreciate the lessons which
they were taught by the works of Cuvier.’ !

These Sewalik researches, interrupted for a time by distant
employment on other duties, were subsequently resumed by
Dr. Falconer in England.

Concurrently with these investigations, Dr. Falconer’s
official duties, as Superintendent of the Suharunpoor Botanic
Garden, led him to explorations in the snowy range of the
neighbouring Himalayahs.

In 1834 a Commission was appointed by the Bengal Go-
vernment to inquire into and report on the fitness of India
for the growth of the tea-plant of China. Acting on the
information and advice supplied by Dr. Falconer,? the Com-
mission recommended a trial. The Government adopted the
recommendation; the plants were imported from China, and
the experimental researches were placed under Falconer’s
superintendence in sites selected by him. Tea culture has
since then greatly extended in India, and the tea of Bengal
bids fair to become an important commercial export from
India, as Falconer long ago predicted.

In one of his expeditions, in July and August 1834, in
search of sites for tea-plantations, Dr. Falconer succeeded in

1 Address at anniversary meeting of
Geol. Soe. February 17, 1837; ib. vol.
vi. p. 890.

? On the Aptitude of the Himalayahs
for the Culture of the Tea Plant.—
Journ. As. Soc. of Bengal, 1834, vol. iii.
p- 182. The following is an extract from
a letter from Dr. Falconer to Sir Proby

Cautley, dated May 2, 1844: ‘My tea
services are undeniable. The experi-
ment was attempted on my recommend-
ation, and conducted under me; the first

tea was manufactured under me, and the

produce declared by three sets of brokers
to be equal to the best China tea,’

a

BIOGRAPHICAL SKETCH. XXX1

ascending the Jumnootree as far as the hot springs at the
sources of the Jumna. His diary kept during this journey
is full of interest. He dwelt with much force on the singular
contrast to the Alps of Switzerland, or the Highlands of
Scotland, presented by the Himalayahs in the absence of
lakes,!—a fact which many years afterwards he handled with
his wonted vigour in the discussion upon the Glacier-Erosion
theory of lake-basins. Between Mussooree and Deobun he
discovered greenstone trap. He also described a section of
the bed of the Jumna at Burkot, which he believed to
possess great interest in connection with the Sewalik Hill
formation. It consisted of schistose primitive rocks with a
superimposed layer of unstratified gravel, forming a cliff
upwards of 50 feet above the level of the river, and exactly
like that found in the Kheeri and Kalowala passes. It
appeared to Dr. Falconer that this was due to-‘a diluvial
action higher up the valley, perhaps the sudden disruption
of a large volume of water, forming a mighty wave which
swept along suspended in it an enormous quantity of gravel
and left it deposited as at Burkot, at intervals in its course,
where the velocity of the stream might have diminished, and
which rushed on to the plains, where, disemboguing from its
contracted channel, it spread out and left the gravel deposit
now crowning the Sewalik Hill formation, previous to their
upheaval.’

A letter written to the Rev. Dr. Gordon of Birnie, N.B.,
soon after his return from the Jumnootree, contains the fol-
lowing paragraph :—

_ €The rock formations of the Himalayahs are all primary: the Sub-
Himalayan is very recent. In the outer ridges you get limestone and
the newer primary rocks (transition). As you go on, gneiss, mica-
slate, &c. succeed. In the outer ridges the volcanic rocks are greenstone
traps (I believe I was the first to make this out), often with porphy-
ritic crystals, and here and there unstratified quartz rock. As you go
inwards you get granite and syenite. On the southern side of the snow
peaks there are more recent formations, and I should not have said that
the Himalayahs are entirely primary. You there get limestone with
Ammonites, Orthocerates, Trilobites, and Terebratule, as in the moun-
tain limestone of England. The snowy range or central ridge has an

1 ‘Viakes are scarcely anywhere seen | Highlands of Scotland or travelled inthe
in the Himalayahs. This singular fea- | Alps,’ &c.
ture strikes any one who has seen the

XXXli BIOGRAPHICAL SKETCH.

elevation varying from 15,000 to 26,000 feet. Perhaps the mean
height may be from 18,500 to 19,000. The snowy mountains are not,
as in the Andes, interrupted peaks here and there of porphyries and
other traps, but a continuous line of ridges, and the highest of them are
certainly primary schists, such as gneiss, &c. You may remember,
perhaps, Jameson’s doubts about this point. But I am convinced that
they are only huge masses of the same formation as the lower ridges,
upheaved to a greater elevation. The scenery is magnificent; like
Byron’s ocean, “ Endless, boundless, and sublime”; huge, vast, and
awe-striking. To give you an idea of some of the views: I got up on
the top of a high mountain called Choor, half way between the snowy
range and the plains, with an elevation of about 13,000 feet. In front,
looking to the north, the eye took in a continuous line of snowy ridges,
varying from 15,000 to 24,000 feet, or no less than 90° on a quadrant
of the horizon. This is no exaggeration. Between me and them
stretched an ocean of mountain waves, I overtopping all. In the rear,
or south, stretched another sea of mountain-ridges, with the plains of
India in the distance, level as a lake, traversed here and there by a
streak of silver marking the tiny show made by the mighty rivers
Jumna and Ganges, and then turning to right and left was a stretch of
ridge upon ridge and of mountain upon mountain, bounded only by the
limits of vision. I stood upon pinnacled masses of granite, which made
a noble and harmonious offset to the whole. Follow me on another
occasion to the source of the river Jumna, at the foot of the mountain
Jumnootree, 21,000 feet high, I walking in the bed of the river, ina
narrow winding channel, cutting off the view in every direction, with a
lofty wall of rock on either hand. Imagine now a sudden bend of the
channel opening a vista in front, and the mountain bursting on the
view rising nearly two miles in height right over me, its black front
patched over and its summit crested with snow, looking like an enormous
wave curling with foam and rolling on to overwhelm us. So vivid was
this impression, that astounded awe was the first feeling, and it required
an exertion of reason to get over it.’

In this and other expeditions, as well as by means of
trained native collectors dispatched in different directions
as far as Cashmeer and the borders of Chinese Tartary, Dr.
Falconer made large additions to Indian Botany, which have
been freely acknowledged by Dr. Royle and other botanical
writers. Dr. Royle! states that Falconer’s ‘untiring zeal
induced him to travel much in the midst of the rainy season,

1 Tilustrations of Botany of the Hima- | in honour of his friend Dr. Falconer,
layahs, vol. i. 1839, pp. 362, 367. <A | ‘whois as zealous and able a botanist

new genus of the family of Antidesmee | as he has shown himself to be a dis-
was designated by Dr. Royle, Falconeria, | tinguished zoologist.’—Ib., p. 354,

4

BIOGRAPHICAL SKETCH. XXXill

to the great risk not only of his health, but of his life,’ and
from Falconer’s diary it appears that during his trip to
Jumnootree he was so seriously ill that on two occasions he
thought it necessary to bleed himself to a large amount.

Not content with investigating the Natural History pro-
ductions of the country surrounding Suharunpoor, Dr. Fal-
coner had repeatedly expressed a desire to visit Cashmeer
for the same object. Accordingly in June 1837, on the
occasion of Burnes’s second mission to Caubul, which pre-
ceded the Afghan war, Dr. Falconer, along with Lieutenant
Mackeson, was ordered by Lord Auckland, on the recom-
mendation of Captain Wade and Dr. N. Wallich, to join the
party, and then proceed into Cashmeer and the countries north
of that valley. United at Peshawur, the party consisted of
Burnes, Mackeson, Leech, Lord, Wood, and Falconer. Of
these six officers, Wood, the explorer of the Oxus, alone
survives. In his journey from Loodianah to Peshawur, Dr.
Falconer found the Sewalik fossils all along the Sub-Hima-
layan range from Jhelum on to Rawul Pindee. After ex-
ploring the neighbourhood of Peshawur he detached himself
from the rest of the party and proceeded westward to Kohat
and the lower part of the Valley of Bunguish, in order to
examine the Trans-Indus portion of the Salt-Range, and
then, in company with Lieutenant Mackeson, who many
years later was cruelly assassinated by a fanatic at Peshawur,
he made for Cashmeer, reaching the town of Cashmeer at the
latter end of September. Soon after their arrival Lieutenant
Mackeson received instructions to return at once to Peshawur,
but Dr. Falconer remained at Cashmeer, where he passed
the winter and spring examining the natural history of the
valley and making extensive collections. He lived in the
same house which had formerly been occupied by the cele-
brated traveller, Moorcroft, and here, for many weeks in
December and January, he suffered from an alarming illness,
which reduced him to a state of extreme prostration. The
following summer (1838) he crossed the mountains to
Iskardoh in Bulkistan, and by the aid of Rajah Ahmud Shah
traced the Shiggur branch of the Indus to its source in the
glacier on the southern flank of the Mooztagh range, now
ascertained to be 28,200 feet above the level of the sea.
Having examined the great glaciers of Arindoh and of the
VOL. I.

XXXIV BIOGRAPHICAL SKETCH.

Braldoh valley, he returned to Cashmeer by the Valley of
Astore, where he discovered the Assafcetida plant of commerce,
which he was the first to describe. On October 10th he
reached Cashmeer, from which he set out again on the 22nd,
on his return journey through the Punjab to Suharunpoor.
During this second visit to Cashmeer and part of the journey
through the Punjab he was again severely ill. It was during
this return journey that he discoverd the Sewalik formation
with the remains of Mastodon, &c., in the hills between the
Punjab and Cashmeer. During his stay in Cashmeer, Dr.
Falconer transmitted to the Botanic Gardens at Suharunpoor
650 grafted plants, comprising all the more valuable fruit-
trees of Cashmeer, with plants of the Prangos Grass ; and on
his arrival at Suharunpoor, early in December 1838, his col-
lections in Botany, Zoology, and Geology amounted to nine
cart-loads. Besides the plants yielding Assafcetida and koost,
articles of considerable commercial value, they included nu-
merous new species of plants, 587 sorts of seeds, 234 skins of
birds, and 30 specimens of Mammalia, including one new
species of Cervus, two of Capra, and one of Moschus—the
details of which will be found in Dr. Falconer’s official account
of the expedition, and in the notes and appendices.!

In 1841, Dr. Falconer addressed a letter to the Secretary
of the Asiatic Society on the then recent Cataclysm of the
Indus, and while advocating a careful Government investiga-
tion of its causes, suggested as an explanation a temporary
obstruction of the river with snow or ice above Iskardoh.
This he supposed had dammed up the water and caused the
river to be so low, that at Attock, in place of being as usually,
many fathoms, it was fordable. All at once the obstacle had
given way, and a mighty flood coming down had swept every-
thing before it.?

In 1840, Dr. Falconer’s health, shattered by previous
attacks of severe tropical diseases—the results of incessant
exposure—gave way; and alarming symptoms setting in, he
was compelled in 1842 to seek for recovery by returning to
Hurope on sick leave. He brought with him the Natural
History collections amassed during ten years of exploration.
They amounted to seventy large chests of dried plants from

1 See vol. i. p. 557.
? Journ. As. Bengal, vol. x. p. 615. July 1841.

BIOGRAPHICAL SKETCH. XXXV

Cashmeer, Afghanistan, Tibet, the Punjab, the Himalayahs,
the plains of the N.W. provinces, and from the neighbourhood
of Darjeeling, Assam, and Sylhet; and forty-eight cases con-
taining five tons of fossils bones, together with geological
specimens, illustrative of the Himalayan formations from the
Indus to the Gogra, and from the plains of the Punjab across
the mountains north to the Mooztagh range.

From 1843 to 1847 Falconer remained in England. He
occupied this time in publishing numerous memoirs on the
geology and fossil remains of the Sewalik Hills, which
appeared in the ‘Transactions’ of the Geological Society,
and in the ‘ Proceedings’ of the Zoological Society, of the
British Association, and of the Royal Asiatic Society, and
which have been reproduced in these his collected works.
He had now an opportunity of comparing the Indian fossils
with the metropolitan collections of Paleontology and Com-
parative Anatomy. On his arrival he was at first so weak that
for several weeks he was unable to walk, but his first visits were
to the Royal College of Surgeons and the British Museum ;
and he at once wrote off a glowing account of the treat which
he had received to Captain Cautley, in India. He might well
contrast the advantages enjoyed by the paleontological
student in London with the difficulties which he and his
colleague had so ably surmounted in India. He also communi-
cated several important papers on Botanical subjects to the
Linnean Society ;! of which may be specially mentioned that
on Aucklandia Costus, the Cashmeer plant which yields the
Kostos of the ancients; and that on Narthex Assafetide,
which was the first determination of the plant, long con-
tested among botanists, which yields the Assafcetida of com-
merce, and which he had found growing wild in the Valley of
Astore, one of the affluents of the Indus. His extensive
botanical collections, on which he had bestowed so much
labour, were unfortunate. Having partially suffered from
damp on the voyage to England, they were left deposited in
the Hast India House during Falconer’s second absence in .
India, and the specimens underwent a ruinous process of
decay. ‘Those which escaped were obtained in 1857 from the
Court of Directors, by Dr. J. D. Hooker, for the Museum at

1 See List of Botanical Memoirs at p. ly.
b 2

XXXV1 BIOGRAPHICAL SKETCH.

Kew. Respecting this collection, Dr. Hooker and Dr. Thomson
thus wrote in the ‘ Flora Indica,’ published in 1855 :—

‘We cannot conclude this comprehensive catalogue with-
out an allusion to the labours of Dr. Falconer, one of the
most estimable, able, and accomplished of Indian botanists ;
to whose liberality and good offices we were in many ways
indebted as travellers in India, and are still as workers at
home. Dr. Falconer was one of the first botanists who
visited Cashmeer and Little Tibet, where he formed magni-
ficent collections, as he also did in Kumaon and the Punjab,
illustrating his specimens with voluminous notes and details
of their structure. His collections are, we believe, still in the
India House, where they have been for many years. They
constitute the only herbarium of importance to which we
have failed to procure access, and we are hence unable to do
our friend that justice in the body of this work to which, as
the discoverer of many of the plants described, he is pre-
eminently entitled.’! It may be added, that since his death
Dr. Falconer’s voluminous botanical notes, with 450 coloured
drawings of Cashmeer and Indian plants, have been deposited
in the Library at Kew.

But his main work during his residence at this time in
England was the determination and illustration of the
Indian fossils in the British Museum and in the East India
House. Captain Cautley, in 1840, had presented his vast
collection, the result of ten years’ unremitting labour and
great personal outlay, to the British Museum, the Geological
Society having declined to accept it, as it was beyond their
means of accommodation. Its extent and value may be
estimated from the fact that it filled 214 large chests, the
average weight of each of which amounted to 4 cwt., and
that the charges on its transmission to England alone, which
were defrayed by the Government of India, amounted to 6021.
Dr. Falconer’s selected collection was divided between the
India House and the British Museum; the great mass was
presented to the former, but a large number of unique or
choice specimens required to fill blanks or improve series
were presented to the latter.2 Other collections of the

1 <Plora Indica.” By J. D. Hooker, |. collection which were figured in the
M.D., and Thos. Thomson, M.D. Lond. | ‘Fauna Antiqua Sivalensis’ were subse-

1855, pp. 67-8. | quently removed to the British Museum.
2 All the specimens in the India House

BIOGRAPHICAL SKETCH. XXXVI

Sewalik fossils had been presented to the Museum of the
Edinburgh University by Colonel Colvin, and to the Oxford
University by Mr. Walter Ewer. The bulk of the specimens
in the British Museum were still unarranged and embedded
in matrix. In July 1844, a memorial, signed by the Presidents
of the various Scientific Societies,! was presented to the
Court of Directors of the Hon. Hast India Company, point-
ing out the desirability of having the specimens in the
National Collection prepared, arranged, and displayed, and
also of publishing an illustrated work, which would ‘ convey
to men of science in both hemispheres a knowledge of the
contents of the Sewalik Hills,’ and suggesting Dr. Falconer
as the person most fitted to superintend the preparation
and arrangement of the specimens and to edit the work.
At the meeting of the British Association held at York
in the following October, a committee, consisting of the
President of the Association with the President of the
Royal and Geological Societies, &c., was appointed to me-
morialize Her Majesty’s Government with the same object.
Sir Robert Peel, then at the head of the Government, res-
ponded to this appeal by making a grant of 1,000/. to
prepare the materials in the British Museum for exhibition
in the Paleontological Gallery. Falconer was in December,
1844, entrusted with the superintendence of the work, and
rooms were temporarily assigned to him by the Trustees of
the British Museum. The Court of Directors also of the East
India Company liberally employed him on duty, on the footing
of service in India; and at his instance they caused to be
prepared a series of coloured casts of the most remarkable of
the Sewalik fossil forms, sets of which were presented to the
principal Museums of Great Britain and Hurope.? Under
the patronage of the Government and of the Hast India
Company, each of which subscribed for forty copies, an illus-
trated work was also brought out, entitled ‘Fauna Antiqua
Sivalensis.’. The work was to have appeared in twelve parts,
and six years were calculated as the time necessary for its

1 The Marquis of Northampton, Pre- | Universities of Oxford, Cambridge, and
sident of the Royal Society ; Lord Auck- | Edinburgh; Trinity College, Dublin; the
land, President of the Asiatic Society; | Royal College of Surgeons; Paris, St.
Henry Warburton, Esq., M.P., President | Petersburgh, Copenhagen, Stockholm,
of the Geological Society ; Mr. (now Sir | Vienna, Berlin, Bonn, Munich, Florence,
Roderick) Murchison, President of the | Rome, Leyden, Brussels, Calcutta, and
Geographical Society ; Dr. Buckland, &e. | Bombay.

* The Museums were those of the |

XXXVIil BIOGRAPHICAL SKETCH.

completion; but within three years there appeared nine
parts of the work, each containing twelve folio plates, ex-
ecuted in a style of lithography rarely, if ever, equalled. Mr.
Ford, the artist to whom the work was so much indebted,
bestowed in several instances as much as 180 hours upon a
single plate. No fewer than 1,123 specimens are figured in
these plates; and of many specimens three, four, or five
different views are given. Besides the Sewalik fossils proper,
the ‘Fauna Antiqua’ includes illustrations of a very valuable
and important series of mammalian remains from the Pliocene
deposits of the Valley of the Nerbudda, together with illus-
trations of the Miocene fauna of the Irrawaddi and of Perim
Island in the Gulf of Cambay. The descriptive letter-press
unfortunately did not keep pace with the plates. After a little
progress, Dr. Falconer ‘found that the labour in comparing
and identifying the enormous mass of materials was so great
and the references to be consulted so numerous, that, if he
had given up his time to the letter-press, he would have
been unable to finish the preparation and arrangement of
the collection in the British Museum during the period
within which his stay in England was, by the rules of the
Indian service, peremptorily limited.’ In December, 1847, he
was compelled to return to India, where he found it impos-
sible to continue the work, as he had hoped, at a distance
from the specimens. On his return to England in 1855,
many of the unpublished plates! had been erased from the
stones on which they had been drawn, and many of the
original subscribers were dead, so that the work could
only have been continued under extraordinary difficulties.
It was Dr. Falconer’s. intention, nevertheless, to have com-
pleted it, and in October 1856, he applied to the Trus-
tees of the British Museum for accommodation and access
to the specimens, to enable him to carry out this object.
Bad health, however, which compelled him to seek a warmer
climate, and his ambition to master every detail con-
nected with the fossil mammalian fauna of Europe, before
proceeding to generalize on that of the Sewalik Hills, caused
him to postpone its execution until it was too late. To collect

1 Proof copies of seventeen of these | been deposited in the Library of the
plates, together with outline tracings for | British Museum. (See yol.i. pp. 538,
the remaining plates of the work, have | 554.)

a ee

= ee

BIOGRAPHICAL SKETCH. XXKLX

and arrange the notes which he left behind has been the en-
deavour of the Editor of these volumes. The great Indian
fossil collection, mainly the gift of Sir Proby Cautley, but
the specimens of which, unique in their richness, stupendous
size, and fine preservation, were prepared, identified, and
arranged by Falconer, has long constituted one of the chief
ornaments of the Paleontological Gallery of the British
Museum. There it may be well said of Falconer and of
Cautley : ‘Si monumentum queris, circumspice.’

In June 1847, on the retirement of the late Dr. Wallich,
Dr. Falconer was appointed his successor as Superintendent
of the Calcutta Botanic Garden, and Professor of Botany in
the Medical College; but for six months, at a considerable
pecuniary sacrifice, he continued to prosecute his work in
connection with the ‘Fauna Antiqua Sivalensis ;’ and it was
not until December 20, when a longer delay would have
‘involved the forfeiture of his commission and his right to a
pension,’ that he left England. In February, 1848, he entered
upon his new duties in Calcutta, and became at once the
referee and adviser of the Indian Government and of the
Agricultural and Horticultural Society on all matiters per-
taining to the vegetable products of India. In 1850 he was
deputed to the Tenasserim Provinces, to examine the Teak
forests, which were threatened with exhaustion from reckless
felling and neglected conservation. His report, suggesting
remedial measures, was published in 1850 in the ‘ Selections
from the Records of the Bengal Government,’ and is a model
of clearness and preciseness on the subject of which it treats.’
In 1852 he published, in the ‘Journal of the Agricultural
and Horticultural Society of India,’ a paper on the quinine-
yielding Cinchonas and on their introduction into India; and
in the following year the writer of this sketch saw in the
Calcutta Botanic Garden a Wardian case containing speci-
mens of Cinchona Calisaya, in which Falconer took great
interest. Dr. Falconer was not at the time cognizant of
Weddell’s accurate determination of the species of Cinchona ;
but he recommended a trial of them in India, and indicated
the hilly regions in Bengal and the Neilgherries in Southern
India as the most promising situations for experimental

1 See List of Botanical Memoirs and Reports, at page lvi, No. 17,

xl BIOGRAPHICAL SKETCH.

nurseries. The subject was taken up some years afterwards ;
the bark-yielding Cinchonas were then introduced from
South America, and they are now thriving ard promise to
be a new source of wealth to India. From what has been
stated, it will be seen that Falconer was not only instru-
mental in rescuing from destruction the Teak forests of
Tenasserim, but in introducing the cultivation of Tea and
Cinchona Bark into our Indian Empire. During his resi-
dence in Calcutta, Dr. Falconer communicated to the Agri-
cultural and Horticultural Society many botanical and other
reports of great economic value, among which may be
mentioned those on ‘The Woods for Railway Sleepers in
India,’ on ‘The Timber Trees used for Fuel,’ on ‘The best
means of Tapping the Caoutchouc Tree of Assam,’' and on
‘The Wild Cochineal insect of Assam.’? He was likewise
employed in the selection and arrangement of the illustra-
tions of the Botanical products of Bengal forwarded to the
Great Exhibition of 1851; and he was the author of the
Report on Cashmeer Shawls which appeared in the Official
Catalogue. In 1854, assisted by his friend the late Mr.
Henry Walker, Professor of Physiology in the. Medical
College of Calcutta, he undertook a descriptive catalogue of
the fossil collections in the Museum of the Asiatic Society of
Bengal, which was published as a distinct work in 1859.4
The labour which this involved was immense. No separate
record had been kept by the Society of the numerous pre-
sentations of fossils which had been made to it from time to
time, and the specimens from different localities were mixed
up in the most hopeless confusion. Fossil bones from the
Lias of England, from the Cape of Good Hope, Ava, Perim
Island, the Valley of the Nerbudda, and the Sewalik Hills
were huddled together in heaps in various rooms, and in
ninety-nine cases out of a hundred without a label or mark
of any kind to indicate whence they came. Dr. Falconer’s
familiarity with the characters of the fossils from different
sources in India enabled him to convert what was little more

1 See List of Botanical Memoirs and 4 Descriptive Catalogue of the Fossil
Reports at page lvi. Remains of Vertebrata from the Sewalik
2 Journ. Agr. Hort. Soe. of India, | Hills, &c., in the Museum of the Asiatic
vol, vii. p. 33. Society of Bengal. Calcutta 1859. 8yo.
* Offic. Descript. and Ilust. Cat., vol. | p. 261,
lL. p. 934,

BIOGRAPHICAL SKETCH. xi
than a chaotic heap of rubbish into a collection of fossils
accurately identified and worthy of the Capital of India.
In this work he pointed out the distinguishing characters
of the matrix of the fossils from different parts of India, as
follows :—

1. Those from Ava are black and heavy, and often strongly
impregnated with hydrate of iron.

2. Those from Perim Island are usually embedded in a
yellow marly conglomerate.

3. The matrix of the Nerbudda specimens is white, soft,
friable, and adhering to the tongue, without any ferruginous
or calcareous infiltration.

4. The majority of the Sewalik fossils are embedded in a
hard sandstone matrix, but others are black, heavy, impreg-
nated with iron, and scarcely distinguishable from the Ava
specimens.!

As a teacher of botany in the Medical College, Falconer
was eminently successful and always a great favourite with
his pupils. His house in Garden Reach will long be remem-
bered for the hospitality dispensed to the many who were
reckoned among his friends, and particularly to young officers
on their first arrival in India. In the spring of 1855 he
retired from the Indian service, and on his return home he
visited the Holy Land, whence he proceeded along the
Syrian coast to Smyrna, Constantinople, and the Crimea,
during the siege of Sebastopol.

On his arrival in England he at once resumed his paleon-
tological researches, and in 1856 he published an essay to
vindicate the principle propounded by Cuvier, that the laws
of correlation which preside over the organization of animals
is the guide to the reconstruction of extinct forms.” His
time was now mainly occupied in studying the fossil species
of Mastodon, Elephant, Rhinoceros, and other Mammalia,

' The writer has received the follow-
ing note from Sir Proby Cautley on a
fossiliferous stratum lying below the
great sandstone and shingle deposits of
the Sewaliks :—‘ The Kalowala Pass de-
posit of clay-marl in which the small
black (hydrate of iron) fossils were
found in such numbers by me, and which
Durand also found north (or on the
Himalayah side) of Nahun is a totally

different one from that in which the
larger fossils were found in the upper
sandstone strata, and lying on the sur-
face amongst the detritus of the sand
rocks.’

? On Professor Huxley's attempted
refutation of Cuvier’s Laws of Correla-
tion in the reconstruction of Extinct
Vertebrate forms.—‘ Annals and Maga-
zine of Nat. Hist.,’ June 1856,

xlii BIOGRAPHICAL SKETCH.

and the Cave fauna of England and of the Continent. With
this object he visited and examined for himself almost every
museum and private collection of any note not only in Eng-
land, but in France, Italy, and Germany, and took careful
and detailed notes upon the spot of all the more important
specimens. The winters and spring of 1858, 1859, and
1860-61, which he was obliged to spend in the south of
Europe on account of his health, were devoted to a study of
the valuable collections in the Museums of Lyons, Mont-
pellier, Italy and Sicily; and in the autumn of 1863, in
company with his friend M. Lartet, he visited the various
collections of fossil remains of Rhinoceros and Cervus in
Chartres and Puy-en-Velay. In 1857 he communicated to
the Geological Society two memoirs ‘ On the Species of Mas-
todon and Elephant occurring in the fossil state in England.”!
In these essays he attempted to discriminate with precision
the Mastodon of the Crag (M. Arvernensis) from M. longirostris
and M. angustidens; and to prove that three British fossil
Hlephants, E. primigenius, EH. antiquus, and EF. meridionalis,
had till then been confounded under the name of ZL. primi-
genius. So far as materials were available he showed the
range of existence geographically and in time of each of
these species, and the mammalian fauna with which each
was associated. He likewise produced a synoptical table
showing the serial affinities of all the species of the Probos-
cidea, fossil and living, then known, of the former of which a
large number had been either discovered or determined by
himself. In 1845, at the meeting of the British Association
at Cambridge, Dr. Falconer had endeavoured to prove by
specimens of crania and teeth, that there was a continuous
passage between the Mastodons and Elephants, the forms
included by Clift under M. Elephantoides constituting the
connecting links. This view was further developed in the
published plates of the ‘Fauna Antiqua Sivalensis’ and in
the two memoirs just referred to. He was the first to
establish the constancy of the Ternary and Quaternary ridge-
formule in the Mastodons, as a means of ranging all the
known species under the two natural groups of Trilophodon
and Tetralophodon; and he extended the same principle of

1 See vol. ii. pp. 1, 76.

BIOGRAPHICAL SKETCH. xhii

the ridge-formula to the arrangement of the rest of the
Proboscidean forms, or Elephants, under the divisions of
Stegodon, Loxodon, and Huelephas. In 1858 he urged upon the
Paleontographical Society the propriety of ‘bringing out a
series of figures of the natural size and with descriptions of
all the teeth of each species of fossil Hlephant and Mastodon
found in the British strata, so that, wherever a specimen
might be discovered, there might be a standard figure by
which any competent observer might be able at once to
identify it.2! In 1862 Dr. Falconer communicated to the
British Association at Cambridge an account of Hlephas Me-
litensis, the pigmy fossil Elephant of Malta, discovered with
other extinct animals by his friend Capt. Spratt, C.B., in the
ossiferous cave of Zebbug. This unexpected form presented
the Proboscidea in a new form to naturalists.2 Farther
researches on general questions concerning the same family
appeared in a memoir on Klephas Columbi, published in
the ‘ Natural History Review’ for 1863, the title of which
but inadequately indicated the range of the subjects dis-
cussed.?

His researches on the fossil species of Rhinoceros were
scarcely less important. Among many notes and papers
which never appeared during his lifetime may be mentioned
a most important memoir ‘On the European Pliocene and
Post-Pliocene Species of Rhinoceros.’?* In this memoir it
is attempted to be shown that there are four distinct Plio-
cene and Post-Pliocene species of Rhinoceros, three of which
have long been confounded by Cuvier and other Palzontolo-
gists under the name of R. leptorhinus. One of these, the
original R. leptorhinus of Cuvier, founded upon the Cortesi
cranium (R. megarhinus, Christol), has no bony nasal septum ;
two, R. Htruscus, Falc., and R. hemitechus, Fale. (R. lepto-
rhinus, Owen), have a partial bony septum ; while the fourth,
R. antiquitatis, Blamb., or R. tichorhinus, Fisch. and Cuv.,
has the bony septum complete.

Soon after his return to England Dr. Falconer devoted
much study to the new Purbeck mammalian genera disco-
vered by Mr. Beckles, near Swanage. The specimens were

1 Letter to J. Bowerbank, Esq., Sec. $ See vol. ii. p. 212.
Pal. Soc., March 28, 1858. 4 See vol. ii. p. 309.
? See vol. il. p. 292.

xliv BIOGRAPHICAL SKETCH.

subsequently transferred to Mr. Owen for description; but,
in 1857, Dr. Falconer published an account of one of these
remarkable Purbeck genera, Plagiaulax, and this was fol-
lowed, in 1862, by a second paper on the disputed affinities
of the genus."

Having occupied himself during several years with the
special investigation of ihe Mammalian fauna of the Pliocene,
as distinguished from that of the Quaternary period of
Europe, he was conducted to the examination of the Cave
fauna of England. In 1860 he communicated a memoir to the
Geological Society on the Ossiferous Caves of Gower explored
or discovered by his friend Colonel Wood, of Stout Hall.?
The existence of Hlephas antiquus and Rhinoceros hemitcechus
as members of the Cave fauna was then for the first time esta-
blished, and the age of that fauna precisely defined as pos-’
terior to the Boulder-clay, or period of the Glacial submergence
in England. At the time of his death he was busily engaged
in collecting materials for memoirs on the fossil remains of
Cervus, Hyena, Spermophilus, and other genera.?

In 1860, while on a visit to Torquay, he was induced to
examine the vegetable fossils of the Bovey Tracey Coal, for
which he was prepared by previous researches on the vege-
table fossils of the Burdwan Coal in India,‘ and he was led to
the conclusion that the Bovey Coal, which for twenty years
had vibrated in the minds of geologists between Hocene and
Post-Pliocene, belonged really to the Miocene and was in
correlation with the Coal formations of Germany and Swit-
zerland.® This opinion was confirmed by the more detailed
investigations of Professor Heer, the results of which were
embodied in a memoir presented to the Royal Society in
1862.

While exploring the Himalayahs in his early days, Fal-
coner’s attention had been closely directed to the physical
features which distinguished them from mountain ranges in
temperate regions, and more especially to the general absence
from their southern valleys of the great lakes so common in
corresponding situations in the Alps. When the hypothesis

1 See vol. ii. pp. 408, 430. Reports at page lv, No. 2.

2 See vol. ii. p. 498. ° Letter to His Grace the Duke of
8 See vol. ii. pp 452 to 481. Argyll, April, 1860.

‘ See List of Botanical Memoirs and

ve

BIOGRAPHICAL SKETCH. xlv

of the excavations of lake-basins by glacial action was brought
forward, he took a share in the discussion, and combated the
view by an appeal to the contradictory evidence furnished by
the Himalayahs, the lakes of Lombardy, and the Dead Sea.
In connection with this subject it may be mentioned that his
last public act—the last occasion in fact in which he left his
house—was to attend the Council meeting of the Royal
Society, with the object of advocating the grant of 100I. to
Sir Henry James, for accurately determining by levelling the
amount of depression of the Dead Sea below the level of the
Mediterranean, an object which since Dr. Falconer’s death
has been accomplished.’

In 1861 Dr. Falconer gave important evidence before the
Royal Commission appointed to inquire into the sanitary
condition of India. He distinguished between the removable
and the irremovable causes of disease, and under the latter
he ranked excessive heat and excessive moisture as telling
most on the health. He expressed the opinion that fever
often resulted from malaria produced by vegetable decompo-
- sition.

For nearly thirty years Dr. Falconer had been engaged
more or less with the investigation of a subject which has
lately occupied much of the attention both of men of science
and of the educated classes generally, viz. the proofs of the
remote antiquity of the human race. In 1833, fossil bones
procured from a great depth in the ancient alluvium of the
Valley of the Ganges in Hindostan were erroneously figured
and published as human. The subject attracted much atten-
tion at the time in India. It was in 1835, while the interest
was still fresh, that Dr. Falconer and Captain Cautley dis-
covered the remains cf the gigantic Miocene fossil Tortoise of
India, which by its colossal size realized the mythological
conception of the Tortoise which sustained the Hlephant and
the World together on its back.? In the same formations as
the Colossochelys the remains were discovered of a smaller Tor-
toise, identical with the existing Hmys tecta. About the same
time also several species of fossil Quadrwmana were disco-
vered in the Sewalik Hills, one of which was thought to
have exceeded the Orang-Outang, while another was hardly

1 See vol. ii. p. 65. 2 See vol. i. p. 367.

xlvi BIOGRAPHICAL SKETCH.

distinguishable from the living ‘ Hoonuman’ Monkey of the
Hindoos. Coupling these facts with the occurrence of the
camel, giraffe, horse, crocodiles, &c., in the Sewalik fauna, and
with the further important fact that the plains of the Valley of
the Ganges had undergone no late submergence, and passed
through no stage of glacial refrigeration, to interrupt the
previous tranquil order of physical conditions, Dr. Falconer
was so impressed with the conviction that the human race
might have been early inhabitants of India, that he was con-
stantly on the look out for the upturning of the relics of man
or of his works, from the Miocene strata of the Sewalik Hills.
In April 1844 he wrote thus to his friend Captain Cautley :—
‘ Joining the indication given by the Hindoo mythology with
the determined fact of the little Hmys tecta having survived
from the Fossil period down to the present day, I have put
forward the opinion that the large Tortoise may have survived
also, and only become extinct within the human period. This
as a most important matter in reference to the history of man,
The same view was publicly announced in 1844 at the Zoolo-
gical Society.'! In the account of the gigantic fossil Tortoise,
which appeared in the joint names of Dr. Falconer and Cap-
tain Cautley, the paleontological and mythological bearings
of the case are summed up as follows: ‘ The result at which
we have arrived is, that there are fair grounds for entertain-
ing the belief that the Colossochelys Atlas may have lived down
to an early epoch of the human period and become extinct
since.’ Ten years later, while investigating the fossil remains
of the Jumna, he pointed out that they were ‘most promising
of results bearing upon the human period.’ ?

In May, 1858, having the same inquiry in view, while occu-
pied with his Cave researches, he communicated a letter to the
Council of the Geological Society, which suggested and led to
the exploration of the Brixham Cave, and the discovery in it of
flint-implements of great antiquity associated with the bones
of extinct animals. In conjunction with Professor Ramsay
and Mr. Pengelly he drew up a report on the subject, which,
communicated in September of the same year to the Councils
of the Royal and Geological Societies, excited the interest of

1 See vol. i. p. 866. in the Museum of the Asiatic Society of
* Preface to descriptive Cat. of Fossils | Bengal, 1855, p. 7.

BIOGRAPHICAL SKETCH. xlvii

men of science in the case. Although, ‘loath in the last de-
gree to leave England, having so much unfinished work on
hand,’ his health compelled him to seek a warmer climate for
the winter. He followed up, however, the same object by
proceeding to Sicily to examine the ossiferous caves of that
island, and he there discovered the ‘ Grotta di Maccagnone,’
in which flint-implements of great antiquity were found ad-
hering to the roof-matrix, mingled with remains of hyenas
now extinct in Europe. An account of this important cave
was communicated to the Geological Society.?, Having ex-
amined the collection of M. Boucher de Perthes, on his route
to Sicily, he was impressed with the authenticity of some of
the flint-implements discovered in the Valley of the Somme,
and he urged his friend Mr. Prestwich, than whom there is no
higher authority in this branch of Geology, to proceed there
and investigate the conditions of the case. This led to Mr.
Prestwich’s celebrated memoir on the flint-yielding Quater-
nary deposits of the Somme. Thus, in 1859, the subject of
the antiquity of the human race, which had previously been
generally discredited by men of science, was launched upon
fresh evidence. Since then it has been actively followed up
by numerous inquirers, and Dr. Falconer himself was con-
templating, and had indeed actually commenced, a work ‘On
Primeval Man.’* In 1863 he took an active share in the
singularly perplexed discussion concerning the human jaw
of Moulin-Quignon; and in the Conference of English
and French men of science held in France he expressed
doubts as to the authenticity of the specimen, but in that
guarded and cautious manner which was characteristic of
him. Dr. Falconer’s honesty of purpose and love of truth
were well exemplified in this controversy. Before proceeding
to the Conference he had publicly expressed an opinion un-
favourable to the authenticity of the jaw. The results of the
first few days of the Conference seemed to be in his favour,
and he wrote thus to London: ‘I have every confidence from
the present aspect of the matter that we shall establish our
case. But I am open to conviction, and will give a true and
honest verdict to the best of my convictions.’ At the close
of the Conference, he wrote again in these words: ‘In the

1 See vol. il. p. 491. 3 Th. p. 570.
2 Tb. p. 548.

xlvill BIOGRAPHICAL SKETCH.

long run matters went against us; and I am very glad that
they did, as truth alone was our object.’ The doubts which
he still entertained were set forth in a memoir which he sub-
sequently wrote, but which never appeared in his life-time.’

In the spring of 1864 he published a notice on the remark-
able works of art by ‘ primeval man,’ discovered by Messrs.
Lartet and Henry Christy in the ossiferous caves of the Dor-
dogne,? which he had visited himself in conjunction with M.
Lartet; during the subsequent summer he was occupied in
preparing a memoir ‘ On the Asserted Occurrence of Human
Bones in the Ancient Fluviatile Deposits of the Nile and
Ganges, with Comparative Remarks on the Alluvial Formation
of the two Valleys’;? and in September he accompanied his
friend Mr. Busk to Gibraltar, to examine caves in which
marvellously well-preserved remains of man and mammals of
great antiquity had been discovered. Before starting, he
drew up, in conjunction with Mr. Busk, a preliminary report
on the specimens brought from Gibraltar to this country,
which was presented to the meeting of the British Associa-
tion at Bath. He attached great importance to the results
of this expedition; and on his return home he continued to
prosecute the examination of the fossil remains of Gibraltar,
the results of which he contemplated elaborating, in conjunc-
tion with those of his explorations in Sicily, into a separate
memoir on the Mediterranean Cave Fauna.*

But his labours were at an end. From Gibraltar he
hastened home to support at the Council of the Royal Society
the claims of Charles Darwin for the Copley Medal. He
suffered much from exposure and fatigue, consequent on the
breaking down of the diligence on the Sierra Morena, on his
return journey through Spain from Gibraltar, so that the in-
clement winter told with additional force upon a constitution
naturally susceptible of cold and weakened by long residence
and serious diseases in India. On January 19th, on his
return from a meeting of the Council of the Royal Society,
he felt depressed and feverish. The attack speedily became

1 See vol. 11. p. 601. as a letter to Sir W. Codrington, the
2 Tb. p. 626. Governor of Gibraltar, has been pub-
3 Tb. p. 632. lished in the Quart. Journ. Geol. Soc.,

4 Since Dr. Falconer’s death, a pre- | (see vol. ii. p. 554.) A more detailed
liminary report on the Gibraltar Caves, | report by Mr. Busk may shortly be ex-
drawn up by Dr. Falconer and Mr. Busk, | pected.

BIOGRAPHICAL SKETCH. xlix

developed into acute rheumatism, complicated with disease
of the heart and lungs, which proved fatal on the morning
of January 31,1865. On February 4 his remains were com-
mitted to their last resting-place, at Kensal Green, in the
presence of a large number of his sorrowing friends and
fellow-labourers.

At the time of his death Dr. Falconer was a Vice-President
of the Royal Society and Foreign Secretary of the Geological
Society. He had been elected a Fellow of the Royal Society
in 1845, and in the same year he had been offered the Secre-
taryship of the Geological Society, which he had been obliged
to decline, as his time was fully occupied with the Sewalik
collection in the British Museum. Foreign countries had
not failed to acknowledge his transcendent merits as a Na-
turalist. Besides being a Fellow of the Royal, Linnean,
Geological, Zoological and Horticultural Societies at home,
and a Member of the Asiatic Society of Bengal, he had been
elected a Corresponding Member of the Academy of Natural
Sciences of Philadelphia (1836), Foreign Member of the
Imperial Austrian Society of Agriculture (1840), Correspond-
ing Member of the National Institute of Washington (1840),
Corresponding Member of the Royal Academy of Sciences of
Turin (1844), Hon. Member of the Natural History Society
of Hesse Darmstadt (1846), Hon. Member of the Academia
Valdarnese del Poggio (1859), Corresponding Member of the
Imperial Society of Emulation of Abbeville (1863), Corres-
ponding Member of the Imperial and Royal Geological Society
of Vienna (1863), and Corresponding Member of the Academy
of Natural Sciences of Italy (1863).

As an additional proof of the high esteem in which he was
held by men of Science, it may be mentioned that, at a
meeting held in London on February 25th, 1865, Sir Proby T.
Cautley, K.C.B., in the Chair, it was resolved ‘ to record the
great loss sustained by Science in the early death of the late
Dr. Falconer, and to perpetuate his name as a naturalist and
a scholar by a suitable memorial.’

It was also resolved that this memorial should include
a marble bust, to be placed in the rooms of one of the
Scientific Societies, or elsewhere, in London, as might be
determined.

One of the objects in which Dr. Falconer had taken deep
VOL. I. c

] BIOGRAPHICAL SKETCH.

interest, up to the latest hour of his life, having been the
foundation of Fellowships or Scholarships in the University
of Edinburgh, to enable deserving students to prolong their
studies beyond the usual academical period, it was further
resolved to collect funds for the purpose of founding, in that
University, a Fellowship or Scholarship in Natural Science,
tenable for a limited term of years, and to be called ‘ The
Falconer Fellowship’ or ‘ Scholarship.’

A Committee, including the Presidents of the Royal,
Linnean, Geological, Geographical and Ethnological Socie-
ties, was at once formed to promote the objects of the
‘Falconer Memorial,’ by the exertions of which a sum of
nearly 2,000. was collected. The marble bust, by Mr.
Timothy Butler, has been accepted by the Royal Society, and
placed in their rooms; and by a separate subscription another
marble bust has been placed at Calcutta in the Museum of
the Asiatic Society of Bengal, to the early reputation of
which Falconer so materially contributed. Before long the
‘Falconer Fellowship’ will be founded in the University of
Edinburgh, and will be the real monument of the genius of
the man whose name it bears.

From what has been said, it is obvious that Falconer did
enough during his lifetime to render his name as a paleon-
tologist immortal in Science; but the work which he published
was only a fraction of what he accomplished. The amount
of scientific knowledge that perished with him was very
ereat; for not only did he die in the prime of life and in the
fulness of his power, but he was cautious to a fault; he
always feared to commit himself to an opinion until he was
sure that he was right; and thus, as too often happens under
such circumstances, he constantly deferred publishing his
views, and others reaped the credit of observations originally
made by him. No scientific man was ever more deeply
imbued with the sentiment, so eloquently expressed by
Cicero in the following passage, than he.

‘ Quid est enim temeritate turpius? aut quid tam temerarium tamque
indignum sapientis gravitate atque constantia, quam aut falsum sentire,

aut, quod non satis explorate perceptum sit et cognitum, sine ulla
dubitatione defendere?’ (Cic. de Nat. Deor. lib. i.)

These volumes, however, will suffice to rescue his name

BIOGRAPHICAL SKETCH. li

from any charge of idleness which some, in ignorance, may
have attributed to him.

The rapturous enthusiasm with which Falconer prosecuted
his favourite researches, as well as the inferences he drew
from the teachings of geological science, are made evident by
the following extracts from his Note-books and corres-
pondence.

Writing in 1840 of his Sewalik discoveries, he says :—

‘What a glorious privilege it would be, could we live back—were it
but for an instant—into those ancient times when these extinct animals
peopled the earth! To see them all congregated together in one grand
natural menagerie—these Mastodons and Elephants, so numerous in
species, toiling their ponderous forms and trumpeting their march in
countless herds through the swamps and reedy forests: to view the
giant Sivatherium, armed in front with four horns, spurning the timidity
of his race, and, ruminant though he be, proud in his strength and
bellowing his sturdy career in defiance of all aggression. And then the
graceful Giraffes, flitting their shadowy forms like spectres through the
trees, mixed with troops of large as well as pigmy horses, and camels,
antelopes, and deer. And then last of all, by way of contrast, to con-
template this colossus of the Tortoise race, heaving his unwieldly frame
and stamping his toilsome march along the plains which hardly look
over strong to sustain him.

‘ Assuredly, it would be a heart-stirring sight to behold! But
although we may not actually enjoy the effect of the living pageant, a
still higher order of privilege is vouchsafed to us. We have only to
light the torch of philosophy, to seize the clue of induction, and like
the prophet Ezekiel in the vision, to proceed into the valley of death,
when the graves open before us and render forth their contents; the dry
and fragmented bones run together, each bone to his bone; the sinews
are laid over, the flesh is brought on, the skin covers all, and the past
existence—to the mind’s eye—starts again into being, decked out in all
the lineaments of life. “ He who calls that which hath vanished back
again into being enjoys a bliss like that of creating.” Such were the
words of the philosophical Niebuhr, when attempting to fill up the
blanks in the fragmentary records of the ancient Romans, whose
period in relation to past time dates but as of yesterday. How much
more highly privileged then are we, who can recall as it were the
beings of countless remote ages, when man was not yet dreamt of; not
only this, but if we use discreetly the lights which have been given to
us, we may invoke the spirit of the winds, and learn how they were
tempered to suit the natures of these extinct beings. We may con-
template the soil on which they were afterwards to move and breathe, at
first reposing under the depths of the ocean, and then raised tranquilly

hi BIOGRAPHICAL SKETCH.

into the air, or disrupted by measureless forces, and projected in
mountain-ranges high into the heavens. All this may we see, and even
date the various events with nearly as much certainty in regard to past
time, as we now do human occurrences which refer to the period of the
Olympiads. For the Almighty Creator, infinitely beneficent in regard
to the wants of his creatures, and thrifty in the use of means, has left
imperishable monuments and inscriptions of the past operation of his
laws, more durable than the pyramids, and more legible than the
hieroglyphics of the Egyptian porphyries. He has engraved on our
natures, as well as in the record of revelation, “‘ Seek, and ye shall find ;
knock, and it shall be opened unto you.” In the exercise of this
high endowment of our nature, we glorify in the highest degree the
attributes of our Creator; and who is he that shall dare to say such
pursuits are unholy or opposed to the ends of man’s being? That
person, if such there be, is an anachronism in time, and a traitor to his
being. Better that he should never have been, or that he should have
lived amidst the darkness of the middle ages, than that he should deck
himself out in the pride of ignorance, as in a marriage garment, and
mislead the helpless prejudices of the unlearned. Truly that man
arraigns the wisdom and beneficence of God, and vilifies the dignity of
his own nature.’

Again, in 1854, he wrote thus to a relative :—

‘Tt has never yet been pretended that there has been a divine revela-
tion expounding the knowledge of the natural world. The Almighty
has given us reason, and left us, by the adequate exercise of that power,
to investigate the laws and order of Creation. Take Astronomy, and
see what has been done in it. Is there any educated person now living
that believes “that the sun was made to rule by day, and the moon by
night,” as servile attendants on the earth ?—No; not one. Does any
one now believe that the sun rolls round the earth?—No. Yet in
former times the universal belief of mankind at the present day was
denounced as a damning heresy, opposed to the Bible. Geology is now
passing through the ordeal that Astronomy did in the days of Galileo.
When the ignorant and bigoted fail in reason and argument, they raise
the yell of intolerance, and charge the science with infidelity. The
odium of the term serves their end for a time, but what follows? This
denounced infidel doctrine, after the lapse of a few years, becomes the
accepted faith of all mankind, philosophical and religious. When,
therefore, in a good cause the imputation of infidelity is raised, one
need not be ashamed of it. There can be no two truths in Nature
opposed to each other. True religion and true science can never be
irreconcileable. As regards the creation of the world, the evidence is
as clear that millions and millions of years must have elapsed between
the first appearance of life on the earth and the present day, as that

BIOGRAPHICAL SKETCH. li

youvand I possess eyes and ears, and have a living existence. The dif-
ference merely is, that the evidence is not of the same nature. The
one is- complete; the other fragmentary, but equally significant and
strong. For instance, a tooth, or the end of a joint found in a rock, is
as conclusive evidence of the former existence of an animal, as if all the
structure—skin, flesh, blood, and living limbs—were before us. ‘The
only difference is, that in the one case the evidence is cumulative and
complete in every detail; while in the other it is fragmentary and in-
ductive; but it is equally clear and conclusive in both. For the
Almighty has so ordained that reason can safely reproduce all that has
been lost, and restore to the tooth all that was correlative to it in life.
But, remember that what I have said here bears solely upon our know-
ledge of the physical world, and not to doctrines of faith for our moral
and religious guidance.’

Lovers of science and they who knew Falconer well can
best appreciate his penetrating and discriminating judgment,
his originality of observation and depth of thought, his
extraordinary memory, his fearlessness of opposition when
truth was to be evolved, the scrupulous care with which he
awarded to every man his due, and his honest and powerful
advocacy of that cause which his strong intellect led him to
adopt. They have also occasion to deplore the death of a
staid adviser, a genial companion, and a hearty friend.

LIST

OF

BOTANICAL MEMOIRS AND REPORTS,

By Hucu Fatconer, M.D., F.R.S.

1. On the Aptitude of the Himalayan Range for the Culture of the
Tea Plant. Journal of the Asiatic Society of Bengal, vol. iii. p. 178,
April 1834.

2. Plates prepared for the illustration of an unpublished Memoir on
the Fossil Plants of the Burdwan Coal-field, intended to have appeared
in the ‘ Asiatic Researches,’ 1833-6. [These consist of nine plates
engraved on stone, and comprising thirty-three figures. They have
been deposited in the Library at Kew. ]

3. Description of the Prangos Grass, Prangos pabularia, Lind., of
Tibet and Cashmeer. [See vol. i. p. 568. ]

4. Report to the Bengal Government on the Cultivation of Hemp in
India. Suharunpoor, about 1839.

5. Some account of Aucklandia, a new genus of Composite, believed
to produce the Costus of Dioscorides. Transactions of the Linnean
Society, vol. xix. p. 23. Read November 17, 1840.

6. On a reformed character of the genus Cryptolepis, Brown. Trans-
actions, Linnean Society, vol. xix. p. 53. Read June 15, 1841.

7. On Edgeworthia, a new genus of plants of the order Myrsinee.
Transactions, Linnean Society, vol. xix. p. 99. Read March 15, 1842.

8. Notes on several new genera of Indian Plants, including a new
genus called Enantia. Hooker's Journal of Botany, vol. iv. p. 173.
1842.

9. Description of the Assafwtida Plant of Central Asia. T'rans-
actions, Linnean Society, vol. xx. p. 285. Read November 2, 1846.

10. Account of Gamoplewis, an undescribed genus of Orchideous
Plants. Transactions, Linnean Society, vol. xx. p. 293. Read Feb-
ruary 2, 1847.

lvi LIST OF BOTANICAL MEMOIRS AND REPORTS.

11. On Athalamia, a new genus of Marchantiew. Transactions,
Linnean Society, vol. xx. p. 897. Read June 1, 1847.

12. Memorandum respecting Timber Trees and materials used for
Fuel in India. Journ. of Agric. and Hortic. Society of India, vol. vi.
p- 165. 1848.

13. Report on the ‘Pooah’ of Nepaul and Sikkim, ‘Poee’ of Gurhwal
and Kumaon, the fibre of Behmeria frutescens. Don. Urtica frutescens.
Journ. Agric. and Hort. Soc. of India, vol. vi. p. 240. 1849.

14. Report on the ‘ Chu-Ma,’ or Grass-Cloth Plant of China, and its
identity with the ‘Kunchoora’ of Bengal, and ‘ Rhiea’ of Assam.
Urtica tenacissima, Roxb. Journ. Agric. and Hort. Soc. of India, vol.
vil. p. 18. 1850.

15. Report regarding Fruit Trees imported into India from North
America in Ice. Journ. Agric. and Hortic. Soc. of India, vol. vii.
p- 172. 1850.

16. Report on some Soap-yielding Pods of a leguminous plant from
China, allied to Cesalpinea. Journ. Agric. and Hortic. Soc. of India,
vol. vil. p. 280. 1850.

17. Report on the Teak Forests of the Tenasserim Provinces, being
No. IX. of ‘Selections from the Records of the Bengal Government.’
Calcutta 1852. 8vo. p. 286, with coloured plates and maps.

18. On the Quinine-yielding Cinchonas and their introduction into
India. Journ. Agric. and Hortic. Soc. of India, vol. viii. p. 13. 1852.

19. Memorandum on the Victoria Regia Water Lily in the H. C.
Botanic Garden. By Mr. Robert Scott and Dr. Falconer. Journ. Agric.
and Hortie. Soc. of India, vol. viii. p. 17. 1852.

20. Remarks on Scevola Taccada, quoad Rice-paper. Journ. Agric.
and Hortic. Soc. of India, vol. viii. p. 66. 1852.

21. Remarks on the best mode of tapping the Caoutchouc Tree of

Assam. Journ. Agric. and Hortic. Soc. of India, vol. viii. p. 102.
1852.

22. Report to the Indian Government on the best Woods for ele
Sleepers in India.

23. Coloured Drawings of nearly 500 Plants from Cashmeer, Tibet,
and Upper India, which, with descriptions of many hundreds of Indian

Plants—mainly Or ee, manuscript, have been deposited in the
Library of Kew.

ee ? |

FAUNA ANTIQUA SIVALENSIS.

I. INTRODUCTION.

BY H, FALCONER, M.D.

A.

INTRODUCTORY OBSERYATIONS—ANTIQUITY OF HUMAN RACE IN INDIA
—ITS MYTHOLOGY— ANTIQUARIAN CONDITION OF INDIA BEFORE MAN—
EVIDENCE AFFORDED BY ORGANIC REMAINS—HISTORY OF DISCOVERY OF
FOSSILS IN JNDIA.

THE antiquities and literature of the Hast have been, from its
commencement, the special field of investigation to this So-
ciety, and to the parent institution in Calcutta. A rich vein
has been opened, branching in a thousand ramifications, and
fertile in results of the deepest interest. The human race
has been traced farther back into time in the Hast than in
any other quarter of the globe; and the tendency of all in-
quiries has been to show that the civilization of at least a
large section of mankind first dawned in the valley of the
Ganges. The language or the mythology, the arts and the
sciences of India, have all been found more or less engrafted
on surrounding nations, and even that civilization which we
now boast of—which has shot so far ahead of the parent stock
—may be followed back to a spring-head in India, whence it
travelled westward through Egypt, and spread over Greece
and Italy.

Nor is it remarkable that it should be so. Man, ceteris
paribus, must have progressed most rapidly where most
favourably placed in regard to the external conditions which
regulate the increase of his race and the development of his
social relations. Neither the valleys of the Nile, nor of
the Huphrates, Tigris, or Oxus, in extent and fertility to-
gether, or in the richness and variety of their productions,

1 This introduction is mainly based | India some years before; and Part C. has
on the manuscript of two discourses de- | been, in a great measure, constructed by
livered by Dr. Falconer before the Royal | the Editor from private letters written by
Asiatic Society of Great Britain, on June | Dr. Falconer, between the years 1844
1 and 8, 1844, a very brief abstract of | and 1847. This will account for the de-
which only appeared in the Journal of | signations of several of the fossils being

the Society (No. xv. Pt. i. p. 107). Part | differentfrom those given in the published
B., however, was probably written in ! abstract.—[Ep.]

VOL. I. B

2 FAUNA ANTIQUA SIVALENSIS.

can admit of comparison with the valley of the Ganges.
Sugar, rice, cotton, and the golden fleece of the silkworm,
with numerous other industrial products, have from the
remotest antiquity been the common staples of the country,
and have been brought forth in such surprising abundance,
that other less-favoured nations could only embody the idea
by conceiving that the sands of India were strewed with gold.
So far down even as the days of the Roman empire, we find
Pliny describing sugar as a kind of honey which exuded natu-
rally from reeds in India, like a gum, but assuming the form
of acrystal. To benefit suitably from such favoured cireum-
stances, we find that the Indian variety of the Caucasian
branch of the human family has the most perfect develop-
ment of that physical conformation which is observed to be
associated withthe highest capability for mental improvement.

It is beside my object, on the present occasion, to do more
than barely allude to this line of research and its bearings, as
introductory to the subject I am desirous to bring before
you this evening, which is to take up the antiquarian history
of the animal races of India, back from the epoch where we
lose all indications of mankind.

There is a point up to which we can follow man back
‘through the records of language and art and the shadowy
indications of mythology and tradition, but beyond which we
cannot go. Every trace of the human race then fails us; and
if we wish to dive further into remote antiquity we have to
fall back on another order of antiquarian research, of the
highest interest, resting on monuments and inscriptions con-
structed by nature, more enduring than the colossal sculpture
of Elephanta, as legible as the scroll on the Bactrian coins,
and infinitely more certain in their indications than language,
tradition, or mythology.

The Colossochelys Atlas, or gigantic fossil tortoise of India,
discovered by Captain Cautley and myself, supplies a fit repre-
sentative of the tortoise, which sustained the elephant and the
infant world in the fables of the Pythagorean and Hindu cos-
mogonies. It isa point of great interest to trace back to a pro-
bable source a matter of belief like this, so widely connected
with the speculations of an early period of the human race.

You are aware that at the present day every climate and
every great division of the globe are characterized by their
peculiar race of animals. So constant are the laws which
regulate this distribution that, if told of the existence of
certain species, we can predicate with confidence regarding
the temperature of a country, and the general character of
the vegetation with which they are associated. Nature, uni-
form in the operation of her laws, has followed the same

INTRODUCTION. 3

order in past time as we observe at present, and thrifty in
her handiwork, has not merely created and extinguished
numerous successions of beings, but at the same time left
indelible records of each in the strata which constitute the
surface of the globe. We can trace by their fossil remains
the different faunas, from the oldest up to the existing crea-
tures. The further we go back into antiquity, we find the
animated races to differ more and more from what they are
at present; and as we descend toward the human period, we
detect a progressive approach to the existing kinds of beings.
The order of succession has been followed out in Hurope by
the concurrent labours of a vast number of observers, with
wonderful precision; and the changes in the animals have
been shown to have been accompanied by corresponding al-
terations of climate, or of other physical conditions. We
have the most certain proofs that England had at one time
the heat of the tropics with a similar vegetation and tropical
animals; and there are the strongest grounds to believe that
at a later period it was either covered with glaciers or
sheeted over with the drift-ice of an Arctic ocean.

The smaller number of observers in tropical countries has
necessarily led to our knowledge in regard to them being in-
finitely less advanced than as regards Europe. But much
has been done lately in America and in India. Passing over
more remote periods, we shall now proceed to consider what
was the condition of the animated creation in the latter
country during the period which preceded the appearance of
the human race upon it. When the plains of Europe fed the
Mammoth, the EHlasmotherium, and the Rhinoceros, and
America the Mastodon and the Megatherium, what were the
kinds of animals that then peopled India ?

The evidences employed in inquiries of this nature, in lieu
of the monuments and inscriptions used by the ordinary an-
tiquary, are the fossil remains of extinct animals found in the
newer strata of the earth. But before entering upon them it
may be well to say a few words regarding the method upon
which the evidence is worked out. Every organised being is
made of a number of parts, which have a definite and con-
stant relation to each other and to the common functions of
the aggregate form. For instance, a predaceous animal like
the tiger can only live on flesh; its alimentary apparatus is
constructed to digest this kind of food; its jaws and teeth
are formed to act like a scissors in cutting it up; its claws to
seize its prey and tear it to pieces; its extremities are built
to enable it to spring; and in connection with the rest, it has
an instinct which leads it to lie in wait, or come stealthily
on its victim, and so on throughout its system. On the

B 2

4 FAUNA ANTIQUA SIVALENSIS.

other hand, an herbivorous animal has its teeth and jaws con-
structed to act as a grinding apparatus; its extremities are
not required to seize prey, and are, therefore, formed merely
to sustain the body, and for speed to enable it to escape from
danger ; and so on with corresponding modifications through-
out its organisation. In short, every part of an animal bears
an invariable relation to every other part, and is of itself an
index to the general form. The bony skeleton constitutes
the basis of the construction. When, therefore, fossil bones
are detected in the strata of the earth, each fragment is a
monument of the existence of a former race of animals, and
every tooth or articular surface a distinct inscription, as it
were, in regard to the special character of the animal from
which the bones were derived.

The first notice of the occurrence of fossil bones in India
is contained in that excellent compendium of the history of
the Moghul and Pathan emperors by Ferishta. We are there
told, that under the reign of Feroz Shah the Third, in the year
1360, ‘The emperor on his return to Delhi, in the month of
Rujub (May 1360), was informed that near Hirdar! there was a
hill out of which issued a stream of water called Sursooti,
running into the Sutlej, and beyond a watercourse called
Selima. It was stated that if a great mound between these
two streams were dug through, the water of the Sursooti
would flow into the Selima, and thence through Sirhind and
Munsoorpoor on to Sunnam, and that the supply of water
would be perennial; whereupon the emperor proceeded in
that direction, and having ordered 50,000 labourers to be as-
sembled, he caused the eminence to be dug through, so as to
form a junction of the two streams. In course of the opera-
tions bones of elephants and men were discovered in the
unbedded mound. Those of the human forearm measured
three yards. Some of the bones were petrified, while others
were still in the condition of bone.'

The bearing of this passage upon Indian paleontology was
first observed by General Briggs. The locality Hirdar refers
to a place in the Sewalik hills, where abundant fossil remains
have been found since.

Captain Webb, in his admirable survey of heights of the
Himalayah mountains, was the first to prove the existence of
fossil bones in that chain. They are called ‘ Bijli ki har,’ or
‘lightning bones’ from their supposed origin, and are found
in the elevated plain of Tibet behind the sources of the

Ganges. They are collected by the inhabitants, and exported

to the plains as charms. In this way they were brought to
the notice of Captain Webb, who communicated them to Mr.
1 Perwar, according to Prinsep.—[ Ep. ]

INTRODUCTION. 3

Henry Colebrooke, by whom they were sent to Europe. They
are referred to in Dr. Buckland’s ‘ Reliquiz Diluviane,’ but
no detailed account of them has yet been published. The
very important inferences connected with these remains in
regard to the elevation of the Himalayahs will be noticed
hereafter.

The next notice of fossil remains in India was by Mr.
Crawfurd, who during his embassy to Ava, in 1826, dis-
covered a deposit of silicified bones of large animals along
the banks of the Irrawaddi, consisting of remains of Masto-
~don, Rhinoceros, Hippopotamus, Crocodiles, Deer, and other
animals. Such of the species as admitted of identification
were described by Mr. Clift in the ‘ Geological Transactions,’
and attracted great interest at the time. Mr. Clift established
the important fact of the former existence of two species of
Mastodon peculiar to India.

The next discovery of fossil bones was made by Captain
Cautley and myself, in 1831, in that range of tertiary hills
skirting the foot of the Himalayahs, to which we have applied
the name of Sewalik hills.!' They were at first found spar-
ingly, but in increasing numbers up to 1834, when through
information supplied by a native rajah, Lieutenants Baker
and Durand were guided to a tract where they were found in
the utmost abundance near the sources of the Sursooti river.
Karly in the investigation, in 1835, an account was published
of a very remarkable animal, called the Sivatherium, which
awakened attention to the subject in India generally ; and
similar remains were found in the valley of the Nerbudda by
Dr. Spilsbury, and in Perim Island in the Gulf of Cambay by
Dr. Lush and Lieut. Fulljjames. It is interesting to keep these
facts in mind, as Perim Island, the Irrawaddi, and the western
part of the Sewalik hills, form as it were three points in a
ereat triangle spread over the whole width of India, showing
that the same ancient race of animals formerly extended over
the continent.

1 MSS. note written by Dr. Falconer,
in 1832. ‘Fossil bones were first dis-
covered by Lt. Cautley. He met with
a single piece which was so imperfect
that he did not imagine it to be a por-
tion of animal remains. On going to
the locality some years afterwards, I
met with a few fragments which satis-
fied me of the existence of fossil bones
in the lower hill formation. They con-
sisted of portions of a testudinous shell,
a vertebra of one of the reptilia, &e. I
met also with some laminar pieces of
crystalline carbonate of lime, simulating
greatly the appearance of a compressed

bivalve shell, which subsequent examina-
tion proved them not to be. A slight
notice of the circumstance was given by
Dr. Royle in the Journal of the Asiatie
Society (vol. 1. p. 96). The fragments
up to this date were so imperfect, that
little beyond conjecture could be made
out of them, but during last cold weather,
on a visit to the Timli pass, I found a
fragment of a well-marked testudinous
remain, and since then Lt. Cautley has
been so fortunate as to discover several
other portions of bone, which set the
matter at rest.’—[Ep].

FAUNA ANTIQUA SIVALENSIS.

Before entering on the particulars of the fauna, it may be
well to refer briefly to the geographical relations and the geo-
logical structure of the Sewalik hills.

iB:

ON THE GEOGRAPHICAL POSITION, PHYSICAL CHARACTERS, AND GEOLOGICAL
STRUCTURE OF THE HYSUDRO-GANGETIC PORTION OF THE SEWALIK HILLS
IN NORTHERN HINDOSTAN.

By ‘Sewalik hills, it is here meant to designate that
range of lower elevations which stretches along the SW.
foot of the Himalayah mountains, for the greatest portion of
their extent from the Indus to the Brahmapootra, where those
rivers respectively debouche from the hills into the plains of
India. The name Sewalik,! or Scevalik, has hitherto had no
definite application, nor has it been universally adopted by
the geographers of India. By some it has been restricted to
the tract between the rivers Jumna and Ganges; by others to
that between the Ganges and Gogra forming the plain-ward
bonndary of the province of Kumaon; and by others to that
between the Gogra and Gunduck forming the northern boun-
dary of the province of Oude; while in Arrowsmith’s Map of
1816, and in the ‘Grand Trigonometrical Survey Map,’ the
designation is nowhere applied. We have resorted to the
term Sewalik as geognostically preferable to Sub-Himalayahs,
which were equally applicable to a range at the northern as
at the southern foot of the great chain: the ambiguity and
inconvenience of a term of this sort having been felt in the
restricted use of ‘Sub-apennine.’ Sewalik is also convenient
as a geographical designation; and its having been applied
by various authorities to different portions of one range, other-
wise without a name, is a sufficient reason for its adoption for
the whole of that range.

Special object of the Memoir.—The object of this memoir is

1 Sewalik, or Sceyalik, derived from
Siva, or Mahadeo the Hindoo God:
these hills, as well as the Himalayahs
being connected in Hindoo mythology, in
various ways with the history of Siva.
Major Rennell (Memoir, p. 72) applies
the name to the tract from the Sutlej to
Hurdwar (p. 233).

Rennell, describing the Ganges (3rd
edit. 1793, p. 313), says, ‘At Hurdwar it
(the Ganges) opens a passage through
Mount Sewalick, which is the chain of
mountains that borders on the level

country on the north of the province of
Delhi. Even Sewalick would be deemed
a lofty range, but for the presence of
Mount Himmaleh or Imaus, which rises
above it, when viewed from the plains of
Hindostan.’

Idem, p. 368. ‘Sirinaguris situated on
an exceedingly deep, and yery narrow
valley, formed by Mount Sewalick* the
northern boundary of Hindoostan on the
one side, and the vast range of Himmaleh
or Jmaus on the other, &e.

* Sewalik is the term according to the
common acceptation; but Captain Kirk-
patrick proves, from the evident etymo-

logy of it, that it should be Sewa-luck.—
RENNELL, p. 368.

DESCRIPTION OF PLATE II.

Map illustrating Dr. Falconer’s Observations on the Geo-
logy of India (p. 28). This map is reduced from a larger one
-which Dr. Falconer had coloured, and to which he had affixed
the following explanatory note. The different shadings cor-
respond to the colours in the original.

‘The great mass of light shading represents the supposed insular form
of the continent of India at an early period of the Tertiary epoch, the
island forming a sort of triangle, of which the eastern and western Ghats
formed the sides and the great Vindhya range the base, with an irregular
patch of mountainous country stretching north forming the Aravalli
range.

‘The dark shading represents the plains of India, forming the valley
systems of the Ganges and Indus drainage, which were formerly narrow
ocean straits. These straits were the recipients of the silt and alluvium
washed out of the Himalayahs, and were at length elevated above the
sea, so as to form the existing continent. The Sewalik Fauna then
spread over the continent, from the mouth of the Irrawaddi to the Gulf
of Cambay 2,000 miles, and north to the Jhelum 1,500 miles. After
the long establishment of the Sewalik Fauna, a great upheavement took
place along the line of the Himalayahs, elevating a narrow belt of the
plains into the Selawik Hills, and adding many thousand feet to the
height of the Himalayahs. The red stripe represents the Sewalik hills,
stretching from the Hydaspes to the Gunduck River, 800 miles. The
small red patch behind the Himalayahs represents the ossiferous plain of
Tibet about 16,000 feet above the sea. The other red patches repre-
sent the Nerbudda and the Gulf of Cambay fossil tracts.’

VOL. I. «

OL

Voll Plas ie

Ci) 8S. G0

Tindee \Aoosh

De CON Hls kepream |

Lake

|
Z\h
; Wount gins

hin,
a tg ae aT

aN pee
~ \BooTAN

m
AN

mii
Pa ggg ON ee ester

Pe

85. go
Man Hanhart Lith

7o IS

Map illustrating D* Faleoners observations on the Geology of India

INTRODUCTION. 7

to illustrate that portion of the Sewalik hills extending from
the Hysudrus or Sutlej, on to the Ganges, which has lately
yielded an immense quantity of the fossil remains of a variety
of new species of mammalia. This is the only part of the
range which we have been able to examine, but for an expo-
sition of its geological relations it will be necessary to notice
theextent of the tract north-west and south-eastof this portion.
Range and Extent.—The Sewalik hills appear to commence
near the western bank of the Jhelum or Hydaspes, about E.
long. 74°, and to run down to the eastward of the Gunduck
in E. long. 85°, along an extent of at least 11°. It is pro-
bable that they are prolonged to the eastward in the Lohara
Dunga range, till they are lost in the marine formations of
_ Assam. From long. 74° to 82° the direction is about NW.
and SH. To the eastward of 82° they bend so as to run
almost due HE. and W. Their axis is parallel to that of the
great chain of the Himalayahs, and in the examined portion
the strata have the same direction and dip. They are inter-
sected along their whole line, at short intervals, by the
numerous streams which unite to form the two great extra-
montane river systems of drainage for the Himalayahs, that
of the Indus to the west and of the Ganges to the east, the
Brahmapootra being intra-montane along its entire course.
Hxtent of the known Tract.—The tract along which we have
examined these hills from the Sutlej at Roopur, long. 76° 30’,
to the Ganges at Hurdwar, 78° 10’, although considerable in
itself, is small for the extent assigned to them; and some
doubt might be entertained of the identity of the range along
so protracted a line. But their physical and geographical
characters are so marked, that a glance at the map would
alone be convincing. Besides, from the hills at the eastern
bank of the Sutlej, near Roopur, we have seen them stretch-
ing far offinto the Punjab with the same north-westerly direc-
tion, and with a continuity interrupted only by the bed of the
river ; and in the same way they are seen from the heights at
Hurdwar, running out of sight to the eastward as they skirt
the hill province of Kumaon. Mr. McClelland! also gives a
section and description of the Sub-Himalayan heights near the
Gogra, long. 80° 20’, which show that Sewalik hills there are
formed of the same beds, and attain similar heights as in the
tract between the Jumna and Ganges. We are therefore cer-
tain of their geological identity along a line of 4° of longitude,
or of about 272 miles. Captain Herbert? has also described
them at the outlets of the Ramgunga and Cossillah rivers, as
showing the same characters as at Hurdwar, and imagines the

1 Geology of Kumaon. 2 Herbert, Mineralogical Report, MSS. p 316.

8 FAUNA ANTIQUA SIVALENSIS.

same formation to extend from beyond Cashmeer to Patna,
with valleys along the whole line.!

From our researches, we have been led to the opinion that
the whole line of the Sewalik hills extending from the Indus
nearly to the bay of Bengal, and formed by the débris of the
Himalayahs, transported by the same causes acting under the
same circumstances as at present, has been broken up from
the plains of Hindostan and assumed its present Alpine
characters within a very late geological epoch, posterior to
the long establishment upon the north of India of animals
so highly organised as the Quadrumana, of species of Camel,
Ox, and Antelope, and of Crocodiles now existing in India.
The importance and imposing character of the deductions
demand a rigid investigation; and with the risk of being
considered tedious and discursive, before entering on the
Geological details of the formation, we shall prefix a sketch
of the Physical characters of the neighbouring plains of
Hindostan, and of the conterminous Himalayah chain.

The course of the Nerbudda, in lat. 23° to 24° N. naturally
and geognostically divides the continent of India into two
great portions: the southern or hilly, and the northern or
‘plains of Hindostan.’ The former is surrounded on all
sides by distinct Alpine ranges, disposed pretty nearly in
the form of a triangle; with the Great Vindhya range run-
ning EH. & W. for its base, and the eastern and western
Ghats for the sides. The two latter meet in the apex at the
southern extremity of the continent near Cape Comorin, and
at their northern limits they jom on respectively with the
eastern and western extremities of the Vindhya chain. From
the western and central portion of the latter, subordinate
ranges such as the Aravalli are sent off to the north, jutting
into the division of the plains. The northern portion, or
plains of Hindostan, is composed of the two great extra
mountain basins of the river systems of the Himalayahs, the
Ganges and Indus. These form two great alluvial plains.
The one commencing in the delta of the Ganges and Megna
stretches NW. from the Bay of Bengal in 22°, to the
Guggar in lat. 31°, in one unbroken flat included between
the Himalayahs on the NE. and the northern extremity of
the eastern Ghat on the west, beyond which it expands to
the west behind the Vindhya chain. The other commences
in the delta of the Indus, and stretches north and east as
marked by the five rivers of the Punjab, to the foot of the
Himalayahs. The two basins are conterminous about half- ©
way between the Jumna and Sutlej in lat. 30°, where they

? Herbert, p. 69, para. 57. Para. 284 and 285.

INTRODUCTION. 9

intermingle in one continuous flat, and with no hilly ridge
intervening ; and they stretch united across the continent
through the plains of Hurrianah and the sandy desert beyond.

From the mouths of the Ganges up to the northern
extremity of the Doab,' the Gangetic valley gradually rises,
till at Suharunpoor it attains the height of 1,000 feet above
the level of the sea, after a stretch of about 1,200 miles. The
eulf of Scinde is about equi-distant from the same point, and
the slope of the plains in that direction across the continent
may be considered the same. In the north-western exten-
sion of the plains which form the basin of the Indus there
appears to be no considerable elevation above the northern
limit of the Gangetic valley, as without any mountain barrier
to alter their direction the rivers converge to the west to flow
into the Gulf of Cutch; and when they are conterminous,
the Soamb, a branch of the Indus system, crosses the line of
a canal, one branch of which runs towards the Indus and the
other to the Bay of Bengal.

Himalayah Chain.—The Himalayah mountains bound the
whole extent of this immense plain on its north-eastern side.
This mighty chain, in all its features, is the grandest accu-
mulation of mountain masses on the surface of the globe.
Its lofty pinnacles, so long the subject of controversy, not
only surpass all others in individual peaks, but maintain
their overtowering elevation along lines of hundreds of miles.
They are covered with vegetation where analogy would
mantle them with snow. They embosom within their belt
extensive plateaux or valleys abounding in numerous races,
and covered with the works of man, as high as the loftiest
peaks of Alps. Their rivers water the most fertile regions
of the earth; their accumulated débris has formed a conti-
nent which supports a population equal to half of that of
Hurope. They separate two of the most densely peopled and
distinct sections? of the human race, each of -which claims
for itself and the mountains above them a remoteness of
antiquity reckoned only by millions of years; and so effec-
tual a barrier do they oppose, that these races but a hundred
miles apart are less known to each other than they are to
the nations of Hurope, divided from thence by thousands of
miles of ocean.

Explored portion—What of them is at all well known
is but a limited tract included between the rivers Sutlej and
Gogra, a line of about 270 miles; and here the scientific
labours of Hodgson, Herbert, Webb, and the Gerards, have
been so successful that the physical outlines of the mountains

' Mesopotamia of Hindostan. 2 The Mongolian and Caucasian races.

10 FAUNA ANTIQUA SIVALENSIS.
have been as well laid down as any equal extent of Alpine tract
in Hurope.!

All westward from the Sutlej to the Indus is unexplored,
and the same may be said of the tract from the Gogra to the
Bay of Bengal.

One or other of us, Captain Cautley or myself, has had
opportunities in repeated journeys of examining that portion
of the mountains between the Ganges and the Sutlej. These
journeys have extended as far as the sources of the Ganges
and Jumna, and have intersected the lower tracts of mountain
in various directions. As the chain possesses a great deal of
uniformity of outline and physical characters generally, for
nearly 1,000 miles, it may safely be presumed that the geo-
enostical relations of the mountain masses, as exhibited in
the tract we have examined, may be taken as a type of the
whole ; excepting the western prolongation along the valley
of Cashmeer, where fossiliferous limestones are found.
Further we have not had an opportunity of examining the
fossiliferous limestones on the northern or Tartary slope.

As we attribute the formation of the whole line of the
Sewalik hills to the alluvial degradation of the Himalayahs,
we shall make no apology for entering at some length on their
physical outline, river systems, and geological structure.

The Himalayah mountains are generally described as com-
mencing in long. 75°, lat. 35°, where they join on with the
Hindoo Koosh, and run down to Bootan, long. 90°, a course
of 1,500 miles, skirting the plain of Hindostan. Their line
of direction from the Indus is about from NW. to SE.
South of the Gogra they get more easterly.

They form one of the mountain boundaries of an elevated

He has described as

1 We especially particularise the late
Captain Herbert, who, besides an impor-
tant share in the Trigonometrical labours
of the Survey, investigated with great zeal
the Geological and Mineralogical charac-
ters of the whole tract, and furnished a
voluminous report to the Indian Govern-
ment on the subject. Unfortunately,
Captain Herbert was a self-taught and
book geologist, and he was called upon
to describe the geology of an unknown
field—a subject new to him, at the very
time when he was acquiring his first
knowledge of geological science. The
consequence is, that his labours have
been less valuable than they otherwise
would have been from his talents and
general scientific acquirements with
longer study. He has fallen into several

important errors.
gneiss an enormous protrusion of granite
which forms the axis of the snowy range
—shown in a colossal section across
many miles, near the sources of the
Ganges; and he has fallen into the same
error with regard to a porphyritic trap,
which forms a most important member of
the Himalayan rocks. He has restricted
the rock formations to granite and gneiss,
and attached minor importance to an
enormous formation of primary sand-
stone. We have been favoured lately
with a perusal of his manuscript report
unpublished, and in consequence, in writ-
ing on the same subject, we deem it ne-
cessary to notice the above, while our
numerous references to his report tell
how largely we have drawn from it.*

* Captain Herbert’s Report was pub-

lished as an appendix to vol. xi. of the

Journal of the Asiatic Society.—[Ep. ]

INTRODUCTION. ll

central tract from which all the great rivers of Asia radiate ;
on the Indian side their drainage being effected by the three
ereat systems of the Indus, Ganges, and Brahmapootra. The
most remarkable feature about the Himalayahs is a line of
snowy peaks which may be considered as forming the axis
of the chain, a plain supported on which, from Cashmeer to
the Delta of the Ganges, would be elevated upwards of 20,000
feet above the level of the sea. Between this central range
and the plains of Hindostan there is a belt of mountains of
minor altitude, with an average elevation of about 7,000 or
8,000 feet. These intermingle with the Sewalik range, which
rises abruptly from the plains.'
* * % * *

The great chain of the Himalayahs rises in a ridge with an
abrupt steep face against the plains of about 6,000 feet in
height; there is then a slope from the crest of the ridge
towards the north. This is the general character of the
Himalayahs. The mountains on the side of the snowy range
consist of a series of nearly parallel ridges, with intermediate
valleys or hollows. They throw off spurs in all directions
into the hollows, forming subordinate valleys. There is
nothing like table-land (perhaps in the whole of the moun-
tains, with the exception of Nepaul), and the valleys are rather
broad, wedge-shaped chasms, contracted at the bottom to a
mere water-course, than anything else ; in fact, the ridges and
intermediate valleys, as a general law, form a series of salient
and re-entering angles, as seen in the sketch (fig. 1, p. 19).
In consequence, the quantity of level or nearly level ground to
be met with is most inconsiderable. From the dip or slope
being towards the north, and the abutment to the south being
steep, the great mass of vegetation has a northern exposure,
and the southern faces of the mountains are generally naked.
The formations are primary; the first towards the plains
consist of vast strata of limestone, lying on clay-slate, crowned
by slate, greywacke, or sandstone. Beyond the limestone
tract, gneiss, clay-slate, and other schistose rocks, occur.
Granite, so far as I know, is not found in the outer ridges.
It occurs in the mountains nearer the snowy range. I have
not gone so far, and have not yet seen granite in situ.
The igneous rocks, which have been concerned in the up-
heavement of the outer tracts, are of the green-stone trap
series, and are very generally met with in dykes intersecting
and rising through the regular strata. The formations

1 The manuscript here breaks off ab- | layan range for the Culture of the Tea
ruptly. Thethree paragraphs which fol- | plant’ (Journal of the Asiatic Society,
low are extracted from an essay by Dr. | April, 1834).—[Ep.]

Falconer, ‘On the Aptitude of the Hima-

2 FAUNA ANTIQUA SIVALENSIS.

have a remarkable feature. The strata are in all directions
fractured or comminuted; the slaty rocks are broken into
small fragments, as if they had been crushed ; and the lime-
stone rocks are vesicular or cavernous, and broken up into
masses.

The arrangement and nature of the soil take their character
from the rocks. From the high angle at which the latter are
inclined, and the northern direction of the slope, the soil is
chiefly accumulated on the northern sides, where is also the
vegetation. From the presence of schistose strata and lime-
stone, the soil underlying the vegetable mould is clayey and
calcareous, or limestone gravel. There is little sandy soil or
sandy gravel. From the extreme richness of the vegetation,
undisturbed for ages, and the moisture of the climate, there is
usually a great accumulation on the northern slopes of vege-
table mould; on the southern faces, the great steepness leaves
little room for the accumulation of soil ; where it occurs it is in
patches, and consists of clays or limestone gravel, mixed up
with vegetable mould. There is here also little sandy soil.
Towards the crest of the slopes the soil is usually dry, from the
moisture running speedily off; but lower down, and wherever
the ground is tolerably level, the soil is quite damp, and
perhaps it is rarely dry in the most parching seasons."

* * % * *

The Himalayah mountains are skirted on the SW. by a
range of lower hills which separate them from the plains of
India. These commence at Roopur on the Sutlej (see Plate
III.), in sandy elevations of inconsiderable height, and run
down a long way to the south, following the direction of the
great chain. In some places they run up to the Himalayahs,
and in others an intermediate valley lies between the two
ranges, as that of the Dhoon. On their SW. side, which
looks towards the plains, they are bounded by a broad belt of
luxuriant Terai jungle. The following observations refer
to that portion of the Sewalik hills which lies between the
Jumna and Ganges. I have not had an opportunity of
seeing them, where they extend to the east and west of the
Ganges and Jumna respectively, but there is little doubt that
the type of all that extends to the west of the Jumna, and
of several hundred miles of the tract of hills to the east of the
Ganges, is to be found in what lies between these two rivers.

The geological characters of the Sewalik hill formation
possess great interest. They appear to consist of an upheaved
portion of the plains of India lying at the foot of the Hima-
layah mountains. The nature of their mineral contents and
depth of the strata give evidence of their having been formed

1 What follows is another portion of MSS. on the same subject.—[Ep.]

ee

INTRODUCTION. 13
during a comparatively recent geological era, and of a vast
series of ages having transpired during their deposition up to
the period of their upheavement. Their coincidence with the
ereat chain of the Himalayahs in parallelism, line of direction,
similarity of dip, in contrast with the opposite nature of their
rocks, connect them closely with the ingenious speculations
of Elie de Beaumont, regarding the eras of upheavement of
parallel mountain chains.

The Jumna-Gangetic portion has already attracted the at-
tention of Indian geologists. The first published account of
them, so far as I am aware, is by Lieut. Cautley, Superintendent
of the Doab Canal, who in vol. xvi. of the ‘ Asiatic Researches,’
has given a very accurate description of the mineral characters
of the strata, in connection with the occurrence of coal and lig-
nite, which first attracted his attention. Itis to his zeal that
we are indebted for the discovery of animal organic remains in
the Sewalik hills. Captain Herbert, in the same volume of the
‘Researches,’ has described the tract from the Sutlejto the Kali
rivers, and enters fully on the nature of the coal or lignite met
with there. Heseems to have formed the opinion that the for-
mation is that of the rock marl or new red sandstone of Eng-
land. Dr. Govan, in a paper on the Physical Geography of the
Himalayahs, has given a brief sketch of the lower hills. He
considers them as belonging to the oldest of Buckland’s alluvial
deposits.!_ The lower hills were examined by Mons. Jacque-
mont in 1831, but I am not aware that any account of them
by him has been published.

The Himalayah mountains north of the valley of Deyra
consist of primary stratified rocks, dipping towards the east
of north ; their abutment istothe south. Their line of direc-
tion is from NW. to SE. or thereabouts. The rocks consist
chiefly of argillaceous schists and vast beds of limestone.
The strata are inclined at a high angle. The mountains
here at once rise with an abrupt mural front to about 7,000 ft.
above the level of the sea, and 6,000 above the plains in the
neighbourhood. No organic remains have as yet been met
with in the outer ridges in the tract between the Sutlej and
the Ganges, or on this side of the snowy range.?_ Of the un-
stratified rocks, greenstone traps, which were first, so far as I
know, observed by Lt. Cautley, occur in considerable abund-

1 Brewster's Journal of Science for
1825, vol. ii. p. 32.
' 2 In my possession there is a small
spiral-chambered univalye, closely re-
sembling a Turritella, which I received
from my friend Lt. Vicary, amost zealous
and indefatigable traveller in the Hima-
layahs. Mr. Vicary met with it near
Mussovres. The specimen was put by,

without the fossil being observed till
several days after, Mr. Vicary could not
subsequently hit upon the spot where it
was found, and as it might have been
broken from a boulder, I have, pending
the uncertainty about the matrix in
which it occurred, attached no import-
ance to the fossil till it should have been
met with again 7 sitz,

14 FAUNA ANTIQUA SIVALENSIS.

ance, rising in dykes through the stratified rocks, also
diallage rock. I am not aware that granite is found in the
outer ridges. It occurs abundantly in the Choor mountain.
Its place in the outer ranges is occupied by the greenstone
traps and diallage.

At the foot of the Himalayahs lies the valley of Deyra,
stretching from the Jumna to the Ganges, and bounded on
the south by the lower or Sewalik hills. It is situated about
1,400 ft. above the level of the sea. The plains in front of
the Sewalik hills are about 1,000 feet above the sea, and
stretch up to the foot of the hills with so slight an inclination
as to be imperceptible to the eye. The hills rise abruptly
from the plains. Their direction is from NW. to SH. par-
allel with the great chain of the Himalayahs. Their dip is
mainly to the north-east. The strata are inclined at an angle
of about 30° more or less, at different points. Within the
British territories, they commence at Roopur on the Sutlej
in inconsiderable heights. Stretching on towards the south
they include the valley of Pinjore, and are here of but low
elevation, and continue on to the Jumna but a few hundred
feet in height. On the eastern bank of the Jumna they at-
tain a greater height, till about half way on to the Ganges,
where in the Kheeri pass they rise to an elevation of 2,000 ft.
above the plains, or 3,000 above the sea. They fall off to-
wards the Ganges, where their height is perhaps about 1,000
feet above the plains. On the east bank of the Ganges they
are of nearly the same height, and run down a long way to
the southward, skirting the flank of the Himalayahs.'

Between the Jumna and Ganges, the Sewalik hills are
about 8 miles in breadth across their direction. They are
intersected by numerous gorges or passes, connecting the
plains with the Dhoon valley. The gorges where the Jumna

and Ganges emerge into the plains are broad. The other .

passes which form the beds of torrents during the rains are
wide at their mouths towards the plains, and gradually con-
tract towards the Dhoon.

Approaching from the plains, the Sewalik hills are seen to
rise with an abrupt, irregular,and deeply indented front.
There are no round-backed hills. The crests are sharp ridges
descending on the one side in a steep precipitous cliff, and on
the other in a smooth and highly inclined slope. Ridge suc-
ceeds ridge in this manner, so as to form, across the line of
direction, a series of close-packed serrated peaks, or a succes-
sion of pretty regular salient and re-entrant angles. So
marked is this feature that in the approach one imagines he

} Captain Herbert, vol. xvi. Asiatic Researches.

ee’

INTRODUCTION. 15

is coming upon rocks of an older date and more durable
structure. The strata may be divided into two classes: 1st,
and lowermost, sandstone and conglomerate, containing sub-
ordinate beds of clay ; 2nd, and uppermost, gravel.

The sandstone is a whitish grey arenaceous rock, varying a
good deal in characters. It is generally of a whitish grey
fine quartz basis, containing scales of white and some dark
mica. It is incoherent and crumbles easily from pressure of
the fingers. In some places the texture is so loose that it
looks, where the surface of a cliff has been acted on by the
weather, like well-packed sand. In others, although still
loose in texture, it resists a sturdy blow from the hammer.
The cementing basis is carbonate of lime, which is upon the
whole abundant in the rocks. In the friable descriptions it
is less so; the harder varieties effervesce strongly with acids.
In some places the sandstone is extensively coloured by de-
composed iron pyrites,! and this especially occurs where it
contains lignite, as in the Kalowala pass.. Where it comes
in contact with the beds of clay, or the strata of conglomerate,
the cementing basis is largely composed of argillaceous
matter ; and the rock possesses great hardness and tenacity.
The sandstone, as above described, forms by far the greatest
portion of the rocky mass of the lower hill strata. In some
places fine sections are exhibited of great extent. About
the middle of the Kheeri pass, a perpendicular cliff rises
from the bed of the pass to a height of nearly 2,000 ft. It
is of course inaccessible, except at the bottom; but so far as
the eye can detect, it consists nearly uniformly of deep strata
of whitish grey sandstone, without conglomerate or beds of clay.

The conglomerate consists of a clayey and arenaceous
basis, in most instances highly impregnated with carbonate of
lime, and cementing waterworn fragments of the older rocks
of the Himalayahs. Captain Herbert has enumerated quartz,
greywacke, granite, hornblende, and limestone: the fragments
are usually of no great size. The strata vary in depth from
a few inches to many feet, and alternate with the strata of
sandstone. They are conformable, and regularly inclined with
the sandstone strata.

The beds of clay are found in the sandstone and conglome-
rate, and modify the character of the rocks where they occur.
The clay possesses different characters. In some places it is
blue and tenacious, mixed up with sand ; in others, yellow or
light flesh-coloured, and greasy to the feel. It is generally
lumpy and unequal, and rarely shows any appearance of
foliation. The beds are of unequal depth, and rarely of any

1 Cautley, loc. cit.

16 FAUNA ANTIQUA SIVALENSIS.

great extent. The depth, from a few inches to a foot and a
half, thinning off and disappearing within a line of a few
yards. It is very often accompanied by thin seams of coal or
lignite, varying from a few lines to four inches—rarely more.
The beds are very unequally distributed in the different por-
tions of the range. In the Kheeri pass, clay is nearly ab-
sent, or seen only in very small quantities. In the Kalowala
and Timli passes it is seen in more abundance. It is de-
scribed as showing itself in a thick bed at Silani, by Lieut.
Cautley.!

The sandstone and conglomerate occupy about three-fourths
of the sections across the range, as exhibited in the different
passes.

Lying uppermost, and in strata in every respect con-
formable with the sandstone, we meet with the deposits
of gravel. The hills of which it is formed exhibit the
same appearance as the other parts of the range. The same
precipitous cliffs are seen towards the plains, and the same
slope towards the Dhoon; but the strata never attain
the height to which the sandstone reaches. Where the
gravel is in contact with the uppermost bed of sandstone
there is a gradual transition from the one rock to the other,
or no very marked contrast in texture between them. The
gravel here contains a great abundance of sand, the pebbles
are small, seldom above an inch or two in size. The texture
of the rock is very loose, and it crumbles under the fingers—
lines of stratification are distinctly marked. Progressively as
we get on towards the upper beds, the size of the pebbles in-
creases, and the quantity of sand decreases, till in the upper-
most beds, which are still clearly stratified, and inclined at
the same angle asthe rest of the range—about 30° to 35°, the
eravel consists chiefly of large water-worn boulders, about
half a foot or more in diameter. The deposit here has ex-
actly the characters of the rolled boulders and gravel which
form the bottom of the pass, which is itself in the rains the
bed of a rapid stream, entirely dried up during the hot
weather. The extent of the gravel is about two miles, and
the cliffs in some places attain a height of 800 to 1,000 ft.
above the beds of the passes.

Besides the beds of clay, the sandstone and conglomerate
contain tabular masses of sandstone of a more compact tex-
ture, with the plain of gravitation parallel with that of stra-
tification. The lower beds of gravel contain similar masses,
but appearing to belong to the sandstone of the range.
Nodules of clay are contained in the sandstone and conglo-

1 Described as foliated orshaly at Silani. Cautley, Joc. cit.

INTRODUCTION. 17

merate, and in some places the surface of exposed conglome-
rate is seen to be patched over with portions of a thin very
crystalline carbonate of lime, simulating very much the
appearance of portions of compressed marine bivalve shells,
which I at first suspected them to be. They have a round,
conchoidal surface, and a suitable thickness; but I have seen
no specimen possessing unequivocal characters by which it
could be referred to a shell.

Organic Remains.—The existence of coal or lignite in the
Sewalik hills has been known since the Goorka war, and has
led at different times to the idea that it might be found in
sufficient quantities for mining speculation. Cautley and
Herbert have described the circumstances under which it is
found, and the latter has given a detailed account of its dif-
ferent varieties. The probability of any extensive coal or
lignite deposit occurring in the formation will be afterwards
considered. The carbonaceous matter is found in the sand-
stone and conglomerate, or in the beds of clay. I have
specimens of it passing through every variety from a brown
lignite, with the scarcely altered characters of a dicotyle-
donous wood, to a scarcely bituminised coal, of rich black
with a high lustre and with no trace of woody structure.
In its most perfect state it is but slightly bituminised, and it
has more the characters of jet than of coal. It is found
deposited in two ways; first, in thin wavy lamine, from a
few lines to a few inches deep, along the flexuous surface of
the thin beds of clay. It is likely that here it was formed of
vegetable matter left in small patches of limited lacustrine
basins, such as the small jhils of Hindostan, or in the bed of
a small sluggish stream. Secondly, in solitary masses, tra-
versing the sandstone strata, and forming the ends of a log.
Tt is often in such situations accompanied with a discolora-
tion of the sandstone about it, owing to impregnation with
iron from decomposed iron pyrites. It here has been formed
of the imbedded trunk of a tree deposited in the sandstone.
When of this description, it often retains a highly ligneous
character. Of a number of specimens which I have ex-
amined from different localities, the lignite has always been
of a dicotyledonous wood. I have seen no trace of any
monocotyledonous woody remains. I have not been able to
refer the woods to any class among the dicotyledones. No
specimen has shown any of the characters of the wood of
the Coniferz, although the fibres were examined under a
powerful microscope. One very perfect specimen in my pos-
session, consisting of a large portion of the transverse
diameter of the trunk of a large tree, has the bark converted
into a substance nearly as hard and lustrous as jet; the weody

VOL. I. c

18 FAUNA ANTIQUA SIVALENSIS.

fibres are carbonised, and the interstices between them im-
pregnated partly with siliceous matter, and partly with hy-
drated (carbonate of?) iron. No other vegetable remain has
been seen, and so far as I know no portion of a leaf or other
vegetable structure of any kind. In the Timli pass, I came
upon a fine section of sandstone, in which from eight to ten
feet are exposed longitudinally of the trunk of a large dico-
tyledonous tree, one foot nine inches in diameter, and upwards
of five feet in cireumference. The wood is silicified, part of the
bark converted into lignite, and impregnated with crystals of
sulphate of lime.

From the above description it will be seen that the Sewalik
hills consist of a succession of beds, irregularly distributed, of
sandstone, shingle, and conglomerate, surmounted by gravel.
The section will give a general idea of the arrangement of
the rocks ; but it by no means professes to be exact in the
details (see fig. 1).

Very opposite opinions have been advanced regarding the
age of the Sewalik hill formation. Captain Herbert, from
the occurrence of lignite and from the stretch of the range
on to the Sutlej, was led to think that it was connected with
the saliferous beds of Lahore, and might be identical with
the new red sandstone of England. Captain Cautley ad-
vanced a similar opinion, more with the laudable object of
having the matter agitated than determining it himself;
but he has long since dropped it. Dr Govan designated the
formation as belonging to the older alluvial deposits of Buck-
land. One can scarcely imagine more opposite geological
conceptions of a formation. Mons. Jacquemont examined
the lower hills in 1831; but Iam not aware of any opinion
he may have formed regarding them.

[The manuscript here ends abruptly, but the following
passage is extracted from the Essay on Tea Cultivation above
referred to, page 11.—Ep. |

‘I regard these hills as an upheaved portion of the plains
at the foot of the Himalayahs, and that they are formed of
the débris of the mountains washed down by streams and
other natural causes. They are covered with vast forests of
saul, toon, and fir, and are uninhabited.

‘The soil of the Sewalik hills and of the valley of Deyra
takes the character of the rocks. It is dry, sandy, or gravelly,
with a considerable quantity of calcareous matter, and it ap-
pears to me to possess the character indicated for the tea
districts in China.

‘The Himalayahs have a direction running from NW. to SE.
They consist, on this side of the snowy range, chiefly of
primary rocks inclined at a considerable angle. The dip of

INTRODUCTION. 19

the strata is to the H. of N.and their abutment to the W.
of 8. On the flank of the great range there is a line of low
hilis, the Sewalik, which commences at Roopur on the Sutlej,
and run down a long way to the south, skirting the great
chain. In some places they run up to and rise upon the Him-
alayahs; in others, as in this neighbourhood (Suharunpoor),
they are separated by an intermediate valley. Between the
Jumna and Ganges they attain their greatest height, which
Captain Herbert estimates at 2,000 feet above the plains
at their foot; or 3,000 above the sea. Suharunpoor is about

rgd

a, Level of the sea at Caleutta; 6, Level of Scvharunpoor, 1,000 feet above the
sea; cc, The Sewalik hills; cc, The strata of sandstone and conglomerate ; ¢’c”,
Strata of gravel; dd, The valley of Deyra; ec, Strata of the Sewalik hills, in
some places rising on the Himalayahs; ff, Outer ridges of the Himalayahs; gg,
Primary strata; 4, The valleys or hollows between the ridges.

1,000 feet above the sea. About twenty-five miles north are
the Sewalik hills. They are here about six or seven miles
wide. ‘To the east of the Ganges and west of the Jumna
they gradually fall off. They have the same direction with
the great chain, and agree generally in dip; their slope being
towards the north, and abutment to the south. They rise at
once against the plains, with an abrupt mural front. They
are serrated across their direction, forming a succession of
“searcely parallel ridges, with a steep face on one side, and
slope on the other. The strata are inclined at an angle of
25° to 30°. They are of recent tertiary or alluvial formation,
and consist of friable sandstone or gravelly conglomerate,
agglutinated by a calcareous cement, containing subordinate
beds of clay; the upper strata are entirely gravel. Beyond
these hills lies the valley of Deyra, 1,200 or 1,400 feet above
the sea, and then the great chain of the Himalayahs. The
foregoing rude sketch will perhaps give an idea of the
_ whole better than description; the distances are not in pro-

portion in the section.’

C.
THE SEWALIK FAUNA—ITS EXTENT AND PECULIARITIES.!

The fossils were either collected at the foot of the cliffs, or
blasted out of rock, or excavated. They were of two sorts,
those formed in the sandstone being hard, while those in the

’ This section is mainly constructed from letters written by Dr. Falconer.—[ Ep. ]

Ge2

20 FAUNA ANTIQUA SIVALENSIS.

clay were soft. The fossil Proboscidea are the most striking
among the remains. The ordinary Proboscidea (Dinotheriwm
being regarded as an aberrant form of the same family)
may for convenience be divided into two genera, Mastodon
and Elephas; although the Sewalik fossils show that these
so-called genera are indistinguishable through any characters
derived from the form and structure of the teeth. There is,
indeed, a gradual and continuous passage in the structure of
the teeth between the mastodon and elephant, the forms which
have been included under the name of Mastodon Elephantoides
by Clift and one of the Sewalik species constituting the
intermediate links. These links are so complete in regard
to the development of enamel, ivory, and cement, and the
number and form of the dental ridges in the molars, as to
break down the technical distinctions founded upon the
structure of the teeth which it has been attempted to make
between mastodon and elephant. The species of mastodon
may be ranged under two sections, viz. :—

I. Trilophodon, or the mastodon having a ternary formula in
the ridges of the intermediate molar teeth.

Il. Tetralophodon, including the species with a quaternary
formula.

Of Trilophodon there are no Indian species yet discovered,
the group being at present limited to Hurope and America,
such as M. giganteus, M. angustidens, and M. Tapiroides.
Of Tetralophodon there are three Indian species, viz. M.
Sivalensis, M. Perimensis, and M. latidens (of Clift ex parte).

Elephas we divide into three sectional groups, viz. :—

I. Stegodon, being the species which Owen calls Transi-

tional mastodons—the M. Hlephantoides of Clift, and of

which there are three, and perhaps four, Indian fossil species,
viz. Hlephas Cliftii, HE. insignis, H. bombifrons, and E. Ganesa.

Il. Lowxodon, the species allied to the African elephant,
of which we have but one Indian fossil species, viz. H.
planifrons.

III. Hlasmodon,' or the thin-plated Elephants (quoad mo-
lars), being the species allied to the mammoth and existing
Indian elephant, of which two fossil species are found in
India, viz. EH. Hysudricus, a huge species with a concave
forehead, and H. Namadicus, the species from the Nerbudda
valley, with the remarkable bulge around the forehead. In
all, ten fossil species of mastodon and elephant, as well as a
new fossil species of Dinotheriwm, have been found in India.
Six of these species have been found in the Sewalik hills.
Mastodon Pervmensis and Dinotheriwm Indicum come from

1 Afterwards called Huelephas (See description of plates in Fauna Ant. Siy.
Plates 42 to 44),.—[Ep. ]

eee in eee

INTRODUCTION. 21
Perim Island, Mastodon latidens and Hlephas Cliftii from
Ava, and Hlephas Namadicus from the valley of the Ner-
budda.! It follows, that in the Sewalik hills alone there
were nearly as many fossil species of mastodon and elephant
as there are now species of the whole order of Pachydermata
upon the continent of India.?

Next come the Hippopotami, of which five fossil species
have been found in India. One of these, H. Paleindicus, is a
true hippopotamus with four incisors allied to the H. major,
which we have included under the subgenus Tetraprotodon.
Three, H. Sivalensis, H. Iravaticus, and H. Namadicus, differ
from all species hitherto described in having the same number
of incisors as the hog, viz. six in both jaws, and conse-
quently we have constituted them into a new subgenus,
Hexaprotodon. The fifth is so distinct, and in its teeth it so
nearly approaches the ruminants as to form a new genus,
Merycopotamus. The Tetraprotodon and Hexaprotodon Nama-
dicus are from the valley of the Nerbudda; the Hex. Irava-
ticus from Ava; the Hex. Sivalensis, and Merycopotamus dissi-
milis from the Sewalik hills.®

Two fossil species of rhinoceros have been discovered in
the Sewalik hills, Rhinoceros Sivalensis and R. platyrhinus ;
the cranium of the former being remarkably concave on its
upper surface, that of the latter being flat. Rhinoceros
Sivalensis was evidently unicorned, and it was also remarkable
in having six instead of four incisors in both jaws. In this
respect it resembled the contemporary hippopotamus of the
same formations. <A third species of fossil rhinoceros comes
from Perim Island, R. Perimensis; and a fourth from the
valley of the Nerbudda, R. Namadicus.

1 For an account of the discovery of
fossils in the valley of the Nerbudda,
‘in consequence of a hint from Dr. Hugh
Falconer,’ see Journ. Asiat. Soc. October
1832, vol. i. p. 456.

Extract of letter to Sir Charles Lyell,
1855 :-—

‘T havealready made out two perfectly
distinct Tertiary Faunas im India, the one
in the Nerbudda (i.e. Central India), com-
paratively late, and characterised by
Llephas Namadicus, Hippopotamus pale-
indicus, and a large species of Bubalus,
with other Bovide, all of them perfectly
different from those found in the Sewalik
hills. The elephant is closely allied to
the species Lhavenamed Elephas antiquus
of Astiana in Piedmont, the Cromer (in
part) and other fresh-water beds of Nor-
folk, and also of the caves in England
and Fouvent in France. Strange it is
that E/ephas Namadicus is nearer to the

English species than to the existing In-
dian species, although in time only a
little ahead of the latter. This is a fact
for Darwin.

‘The hippopotamus of the Nerbudda
is like the existing African and Val
d’Arno fossil species, a Tetraprotodon,
while the Sewalik and Ava species are
Hexaprotodons. We have never yet found
Mastodon in the Nerbudda Fauna; but
we have got sparing remains of one of
the Sewahk Stegodons, i.e. intermediate
elephants, from it.

‘The large ruminants of the Sewalik
Fauna are none of them of the modern
types ofBovide.’—| Ep. ]

2 The above account is derived from a
letter to Mr. Broderip, dated April 6,
1845.—[ Ep].

3 Mainly from a letter to M. de Blain-
yille in 1846.—[ Ep. |

22 FAUNA ANTIQUA SIVALENSIS.

The Sewalik fossils include three species of Swide: Sus
Hysudricus, Sus giganteus, and a new genus, Hippohyus
Sivalensis, with teeth exhibiting a strong tendency towards
the hippopotamus ;! also the Chalicotheriwm Sivalense, one
of the most remarkably aberrant pachyderms that have yet
been met with, either in the fossil or recent state, closely
allied to Anoplotherium, but showing a return from the ru-
minant tendencies of the Cuvierian species back to a more
pachydermatous type, and a closer affinity with rhinoceros.

The family of Hquidae was represented by three Sewalik
species, Hqwus Sivalensis, Equus Paleonus,? and Hippotherium
Antelopinum, the last exhibiting the characters of a small
horse, drawn out into the attenuated proportions of an
antelope. The remains of a third species has been found in
the Netee Pars, and another fossil species, Hqwus Namadicus,
has been obtained from the valley of the Nerbudda.

The fossil ruminants of the Sewaliks are surprisingly
rich, and include almost every type, fossil or recent, known
in the order. In the first place, there are two species of
giraffe, Camelopardalis Sivalensis and C. affinis, and a species
of camel, Camelus Sivalensis, neither of which genera have
before been found in the fossil state. Then there is the
new genus, the gigantic Sivatheriwm, bearing four horns,
nearly approaching the elephant in size, and considerably
exceeding the rhinoceros. Among its chief peculiarities
are the immense width of cranium for muscular attach-
ments, the fore and hind horns, the arched form of the nasal
bones, as in Tapir; the massiveness, width, and shortness
of the face, and the curving upwards of the grinding plane
of the teeth, as in Darwin’s American ox. As to its affini-
ties, the teeth are those of a giraffe; in its four horns,
it resembles the Antilope quadricornis, while the near horns
resemble those of a dicranocerine antelope from North
America. In the head, the nearest affinities are to the ox,
in regard to the plane of the frontai and occipital, the parie-
tals being joined on to the occipital. The upper lip was
prolonged into a trunk. The front horns were remarkable
for their flat form, and in the absence of a bur, notwithstand-
ing the guttering. The neck vertebre are short, and the
legs of huge dimensions. Altogether, the Sivatheriwm was
a remarkable form of animal, unlike anything living. Closely

1 As no description of these fossil | p. 661; also vol. iv. p. 568.—[Ep.]
Suide was ever published, the reader is 2 M. Lartet, in a letter to Dr. Fal-
referred to the description of the plates | coner, dated August 1855, was inclined
in the Fauna Antiqua Sivalensis (Ixix. | to regard Equus Pa/@onus as a young
to lxxi.), and also to a memoir by Messrs. | individual of either E. Sivalensis or E.
Baker and Durand in the Journal of the | Namadicws.—[ Ep. |
Asiatic Society for October, 1836. Vol. y.

INTRODUCTION. 23

allied to it is another new genus, the Brahmatheriwm, from
Perim Island. Among the Cervide are two species of Se-
walik Cervus, and the Dorcatheriwm moschinum ; two other
species of fossil cervus, Cervus Namadicus, and Cervus Pale-
indicus, have been obtained from the valley of the Nerbudda.
The Sewalik fossils also include at least two species of an-
telope, Antelope Paleindicus, and A. gyricornis; and nume-
rous species of Bovide, viz. Bison Sivalensis, Bos occipitalis,
Amphibos acuticornis, Amphibos elatus, Amphibos antelopinus,
Hemibos triquitriceras; two additional species, Bos pale-
indicus, and Bos Namadicus,! have been found in the valley
of the Nerbudda.

The Sewalik Carnivora comprehend fossil species of Felis,
Hyena, Canis, Fox, Mustelide, Machairodus, and the new
forms of Hyenarctos and Enhydriodon, or fossil otter.

The Quadrumana are represented by four fossil species,
and there are also several forms of Rodentia and Insectivora,
including Hystriz, Mus,? and a new undescribed genus,
Typhlodon.

The Sewalik Reptilia are exceedingly rich in forms, par-
ticularly of the Crocodiles and Chelonians (Emys, Trionyx
and Testudo), some of which, such as Leptorhynchus Gangeti-
cus and Emys tectum, are indistinguishable from existing
species; while the Oolossochelys Atlas is a prodigy of size in
the order. It is in every part of its organisation a true
land tortoise, estimated from numerous remains to have had
a shell twelve feet long and six feet high. The possible
connection of this fossil with the fossil which figures so pro-
minently in the Pythagorean and Hindu cosmogonies is a
subject of interesting speculation.

Among the Sewalik fossils there are also the remains of
several species of Birds, including Grallae, greatly surpassing
in size the gigantic crane of Bengal (Ciconia Argala); and of
Fishes, Crustacea, and Mollusca.

x % % * *

At the previous meeting,® when I had the honour of address-
ing you, [ had to wade through such a number of facts, in
explaining, although in the briefest way, the various animals
which entered into the Sewalik fauna, that the time to which
I was limited was more than over before I was done with
them. I had, in consequence, to terminate abruptly, before
T could indicate to what general consequences the mass of
facts tended. For you are not to suppose that the subject is

1 Few of these species have as yet 2 Journ. Asiatic Soc. December 1839.
been described, but the specimens named | Vol. iv. p. 706, and vol. v. p. 296.
by Dr. Faleoner are in the British 3 See note, p. 1.—[ Ep. ]
Museum.—[Ep. |

24 FAUNA ANTIQUA SIVALENSIS.

one merely of curious interest, resting on the numerous
animals and remarkable forms with which we are occupied.
It involves important considerations regarding the former
condition of India, in respect of climate, geographical cha-
racter, and the alterations of surface which it has undergone
in the upheavement of the Himalayahs and of the Sewalik
hills. To these points I shall now direct your attention ;
but in order to bring the subject clearly before you, it will be
necessary in the first instance to review some of the leading
features of the Sewalik fauna.

The first character which strikes us is the wonderful rich-
ness and exuberant variety of forms. In the Pachyder-
mata, which are now restricted in continental India to four
genera and four or five species, there were then twice the
number of genera, and about five times the number of species.
Of the Proboscidean Pachydermatua alone, including the ele-
phant and mastodon, there were as many Sewalik species as
are now comprised in the whole order in India, and the same
holds good as tothe Ruminantia. Besides a large number of
species representing those which now inhabit the continent,
such as the ox, buffalo, bison, deer, antelope, musk-deer, and
others, there were more than one species of giraffe and of
camel, together with the Sivatherium—in fact, representatives
of every type known in the order, fossil or recent, either in
India or elsewhere. And so on through the Carnivora, which
were singularly rich and varied in forms, and through the
Quadrumana, Rodentia, Insectivora—Birds, Reptiles, Fish,
Crustacea, and Mollusca. Nor is there the slightest proof
that this mass of species were not alive at the same time,
but interpolated Successively at different periods. The re-
mains of all were found reposing together, in the same beds
of the same strata.

The next striking feature is the close analogy between the
existing fauna of India, so far as it goes, and the extinct
fauna. What we have at present is but a reduced, or as it
were remnant, representation of what existed before. India
has no marsupial animals now; and up to the present time no
remains referable to that class have been detected among the
Sewalik fossils. The same holds in regard to the Edentata.
India contains at present but a single species, the Manis
crassicaudata ; and we have not found one in the fossil state,
although at least one may yet be expected. In like manner,
the cetacea which are represented by the solitary Platanista
Gangetica, or dolphin of the Ganges, have as yet yielded no
fossil representative, although, as in the case of the manis,
one or more may be expected. It is interesting to keep this
fact of parallel representation in mind, as the same law has

ia gia i See he eevee 8

a ee ee. a! Se

INTRODUCTION. 25

been found to hold generally good in two other remarkable
instances, Australia and America. The former country is at
present the head-quarters of the marsupial tribes, and the
same is the leading character of what is known of its fossil
fauna. In like manner America, which is now the great
home of the Edentata, has yielded almost all the gigantic
forms of that order known to us—such as the Megatheriunr
and others. But in marking the general analogy which runs
between the ancient and the existing fauna of India, we are
struck, considering the number of forms which have become
extinct, with the extreme reduction of species. Not only
have the Proboscidean pachyderms declined from five to one,
and the Equidee from three to one, but numerous genera have
died off entirely: we have nothing remaining of the Hexa-
protodon Hippopotami, the Merycopotamus, Anoplotherium,
Hippohyus, Enhydriodon, Hyzenarctos, Sivatherium, Camel,
Giraffe, and other forms which I could enumerate. The
conclusion is irresistible, that the era of the great force and
development of the vertebrated animals in India has gone
by, and that what we now see as our contemporaries are, as
it were, but a ragged remnant representation of the rich
garment of life with which the continent was formerly clothed.

The next remarkable character is the singular mixture of
representatives of old and new, past and existing forms,
which are grouped together in the Sewalik fauna. I allude
especially to the Anoplotherium Sivalense, the bones of which
were discovered crossed in the same clay matrix with those of
camel, antelope, and giraffe. The species comes nearest to
that which has been described by Kaup as the Chalicotherium
Goldfussi, from the miocene beds of Eppelsheim, the generic
distinction having been founded apparently from mistaking
the false molars for incisors.' But the great development of
the genus is in the Eocene tertiaries of Europe; while in the
Sewalik hills the species is associated with several quadru-
mana closely allied to existing forms, and even with fossil
reptilia, now known to us as existing species. In addition
to the Anoplotherium, excellent observers like Messrs. Baker
and Durand, and Dr. McClelland, have mentioned the Paleo-
therium as a Sewalik fossil ; but no remains referable to that
genus have yet come under our observation. Some of the
other Sewalik animals, such as the Machairodus or Ursus cul-
tridens, and the Merycopotamus, which is closely allied to the
Anthracotherium, indicate a similar tendency towards the
faunas of the older tertiaries of Europe, in a portion of the
Sewalik fauna. At the present day we only know the giraffe

1 Dr. F. afterwards regarded Anoplotherium Sivalense as belonging to Kaup’s
genus, Chalicotherium.—l[Ep. ]

26 FAUNA ANTIQUA SIVALENSIS,

and the hippopotamus as strictly confined to the African
continent. But the head-quarters and force of these genera
appear to have been formerly in India. In short, it would
seem as if all the geographical divisions of the old continent,
and all the epochs, from the Eocene downwards, had contri-
buted their representatives to constitute one vast and com-
prehensive fauna in ancient India.

The next striking point is the peculiarity of type and
number of transitionary forms which run as a general feature
through the Sewalik fauna. The mastodons and elephants
pass into each other through intermediate species. The
hippopotami have the full complement of incisive teeth, and
the same is indicated in regard to forms allied to rhinoceros.
The Hycenarctos is the most abnormal in its dentition of any
known ursine form. The Enhydriodon was a Lutrine animal,
the size of a panther: while the Colossochelys tortoise was
a prodigy of size in its order.

In regard to the nature of the species, in so far as the
evidence has yet been worked out among the mammifers, all
the ascertained species have turned out to be extinct, and in
almost every instance different from those known elsewhere
in the fossil state. But I put forward this statement with
the reservation that the evidence has not in every case been
complete enough to be decisive, and that in several instances
the fossil forms make the closest approach to species now
living in India. This is the case with several of the carni-

vora; while the teeth of one of the species. of giraffe comes °

so near those of the existing African species in size and
form as to be indistinguishable. The Sewalik reptilia, on.the
other hand, exhibit a mixture of recent and extinct species.
The same appears to be the case with the fish; but in this
order the evidence has not yet been gone into sufficiently to
justify pronouncing with any confidence. In regard to the
mollusca, which are regarded as the main evidence for deter-
mining the age of geological formations, the species belong
to land and freshwater genera now common in India. Mr.
Benson, our best authority on Indian shells, considered
the most, if not the whole, of them as identical with existing
species. They are now in the hands of my distinguished
friend Professor Forbes, by whom they will soon be carefully
worked out. Iam permitted to say that he has already been
able to identify some of the species with existing forms.
Nothing approaching human remains or industrial monu-
ments has ever been met with among the Sewalik fossils col-
lected along a line of 360 miles, thus confirming what the
evidence derived from all sources goes to show, the late origin
and very modern advent of man into the system. Yet when

ee

INTRODUCTION. 27

we found the remains of such forms as the camel, giraffe, and
quadrumana with existing reptilia pouring in upon us, each
successive ascertained form appeared to indicate a nearer
and nearer approach to the human period ; and when we had
exhausted the list, the question used to arise, what shall we
find next ?—but man and his works were to the last wanting.
The Sewalik fauna was not merely surprisingly rich in species,
but equally so in the vast number of individuals which the
plains of ancient India subsisted. The collection of fossil
bones which Captain Cautley presented to the British Museum
amounted to 200 chests, averaging about a hundred-weight
each of contents. Another collection formed by myself was
nearly as extensive. Captains Baker and Durand, in April
1856, at an early stage of their collection, took the trouble of
tabulating the number of heads and jaws with teeth contained

among their fossils ; and the following are extracted from their
list.

(fs Fragments of upper Lower jaws. Mutilated frag-
ELEPHANT AND jaws and heads. ments of jaws,
MASTODON. Ra
110 101 56
Crania and Lower jaws.
HIPPOPOTAMUS. CEES:
46 63
PACHYDERMATA J Epper lair:
GENERALLY. | 222 225
RUMINANTS. 143 230

Now when it is remembered that the aggregate collections
have been more than tenfold increased since, and that the
remains were either excavated or found in débris of cliffs, and
that the explored surface bears a very small proportion to that
which has not yet been investigated, one may form an esti-
mate of the prodigious number of animals which must have
lived together in the former plains of India, making every
allowance for the bones having accumulated during many
successive generations.

Viewed as a whole, what designation are we to assign to
the Sewalik fauna? The shell evidence is still to be worked
out; but it has already been shown either that the majority

‘are identical with existing forms, or that there is a mixture
of recent species with a series of extinct forms closely re-
presenting existing ones. The evidence from the vertebrate
animals is of a double character; half of them are so like the
fauna which we now have in India, that they might pass for
the creatures of yesterday, whilethe other half represents the

28 FAUNA ANTIQUA SIVALENSIS.

characters of the middle and older tertiaries of Europe. That
they belong tothe vertebrate series which immediately preceded
the existing race of animals is hardlysusceptible of doubt from
the admixture of existing reptiles.! * * * And as changes of the
fossil fauna of Europe, which mark the different subdivisions of
the tertiary series, have been shown to have been coincident
with changes of climate, and repeated elevations and depres-
sions of temperature—if we can only show that the climate of
India has been less subject to great oscillations during the
tertiary period, and that the surface of the land enjoyed longer
periods of repose, it would be, perhaps, not unphilosophical
to conceive that the epoch of the Sewalik fauna may have
lasted through a period corresponding to more than one of
the tertiary periods of Europe.

iD;
GEOLOGICAL AND CLIMATAL BEARINGS OF THE SEWALIK FAUNA.

Besides the mere zoological interest of the subject, the
Sewalik inquiries involve these conclusions.

1. The upheavement of a narrow belt of the plains of India
at the foot of the Himalayahs into hills 3,500 ft. high along 11°
of longitude, or about 800 miles, after the long establishment
on the continent of such modern forms as quadrumana, camel,
giraffe, and existing species of reptilia.

2. A great upheavement of the Himalayahs, extending to
many thousand feet, and equal to the elevation of a tract
which formerly bore a tropical fauna, up to a height which
now causes a climate of nearly arctic severity. Remains of
rhinoceros, antelope, hyzena, horse, large ruminants, &c., found
at 16,000 feet above the sea.?

3. Conditions in India during the tertiary period precisely
the reverse of what have held in Europe. Instead of a suc-
cession of periods with successive decrease of temperature,
India has now as high a temperature, if not higher, than it
ever had during the tertiary period. The upheavements have
operated to increase the heat. In lat. 30°, at 7,000 feet above
the sea, the mean temperature, making the compensation for
the elevation, and reducing it to the level of the sea, is 81-2°
Fahr., or equal to that of the equator. The same excess of
temperature holds generally over the continent, as contrasted
with the eastern side of the continent of Asia.

4, Instead of numerous subdivisions of the tertiary period
with successive faunas, facts tend to the conclusion that India
had one long term, and one protracted fauna, which lived

1 Manuscript defective.—[Ep. ]

? On this matter, see further the memoir on The Fossil Rhinoceros of Tibet.—
| Ep].

INTRODUCTION. 29

through a period corresponding to several terms of the ter-
tiary period in Europe.

The continent of India, at an early period of the tertiary
epoch, appears to have been a large island, situated ina bight
formed by the Himalayahs and Hindoo Koosh ranges. The
valleys of the Ganges and Indus formed a long estuary, into
which the drainage of the Himalayahs poured its silt and
alluvium. An upheavement took place, which converted these
straits into the plains of India, connecting them with the
ancient island, and forming the existing continent. The
Sewalik fauna then spread over the continent from the Irra-
waddi to the mouths of the Indus, two thousand miles; and
northwest to the Jhelum, fifteen hundred miles. After a long
interval of repose, another great upheavement followed, which
threw up a strip of the plains of India, forming the Sewalik
hills, and increased the elevation of the Himalayahs by many
thousand feet. This event, and the climatal changes which
it involved, caused the extinction of the Tibetian and Sewalik
faunas. Asa result of the climatal changes implicated in
these upheavements, it may be inferred that India is now
enjoying ‘the summer of the great cycle ;’ and that, in contrast
with what has taken place in Europe, there has been no de-
crease of temperature in that country, which has now as warm
a climate, if not warmer, than it ever had during any part of
the tertiary period. (See Plate IT.)

cO FAUNA ANTIQUA SIVALENSIS.

Il. ON THE STRUCTURE OF THE SEWALIK
HILLS, AND THE ORGANIC REMAINS
FOUND IN THEM)

BY PROBY T, CAUTLEY, ESQ., CAPT. BENGAL ARTILLERY, F.G.S.

(Read March 9, 1836.)

THE mountains, a part of which I am about to describe, range
(with the exception of the debouchures of rivers) almost un-
interruptedly from the Sutlej, which separates the territory
of the British Government from that of the Sikh chieftain,
Runjeet Sing, to the Brahmapootra river, and the district of
Cooch Behar. Their general bearing, in the portion near
the Sutlej, is NW. and SE.; but in that approaching the
Brahmapootra it is many points nearer direct east and west.
They lie at the foot of the great Himalayan chain, with which
they are in some parts connected by a succession of low moun-
tains; but in others, as in the districts under review, they
are separated by valleys from three to ten miles in width, and
called by the natives Deyra Dhoon, Kearda Dhoon, &c., or
the Valley of Deyra, the Valley of Kearda, &c. The highest
peaks do not far exceed 3,000 feet, and the generality of the
summits vary from 2,000 to 2,500 above the level of the sea.
It is necessary, however, to premise, that my observations
have not extended further eastward than the Ganges; and
that whatever I may record in this paper beyond that river,
is given on the authority of others.

Between the Ganges and Jumna rivers, this line of moun-
tains is separated from the great chain of the Himalayahs by
the Dhoon, or valley of Deyra, the average width of which
may be under ten miles. At the Ganges, or south-eastern
extremity, a road passes between the chain and the river;
and no difficulty whatever is offered to the approach to the
- valley. At the opposite or Jumna extremity, on the contrary,
the mountains are in many places scarped into the river ;
and the footpath used by the boat and raft people is an irre-
gular track, over ridge and through hollow, and impinges

‘ This memoir is reprinted from the Geological Transactions, vol. v., Second

Series, p. 267.—[ Ep. ]

DESCRIPTION OF PLATE III.

Map of the country between the Sutlej and the Ramgunga,
to illustrate Capt. Cautley’s Memoir on the Structure of the
Sewalik Hills (p. 30). Copied from the original in the

Geol. Trans. Sec. Ser., vol. v.

VOL. I.

Vol |}. Plate 3

x
GandgesR.
&
%
%
\,
~Badshak mirhkul
. %
o Suharunpoor
Q
Loal Canal
&
s
; Jumna R
Kurni
SY Se
Seale of English Miles .
5 i) Yi Cae 50
Latitude of _. 29° 57' North
Longitude ......-..-..--- _.78. 2 East

M6N Hankart Lith

Map of Country between the Sule] and the Ramounda.

STRUCTURE OF SEWALIK HILLS. 31
upon the river at those points only where the stream, in its
meandering course, strikes the opposite side of the channel.
The intermediate roads, or ghats, follow the rivers or hill-
streams; for in such an utter confusion of mountains any
other passage is impossible. A few of these roads are pass-
able for wheeled carriages, but the generality are bad foot-
paths. Between the Jumna and the Sutlej are two other val-
leys, the Kearda and the Pinjore, separated from each other by
one of those complex masses of mountains which connect
the lower with the upper Himalayah range. The hills are
generally less precipitous in this neighbourhood; and .the
higher points and ridges are separated by valleys, affording
a drainage to the country, and displaying on each side high
and beautiful sections of the stratification. These valleys, or
plateaux—for, with reference to the rivers running below
them, they may be considered as such—vary in width, but ex-
tend along the whole course of the mountain-streams.

As much confusion may be avoided, in the absence of a
native name for the whole range of these mountains between
the Sutlej and the Brahmapootra, by establishing one which
may be considered legitimate, I wish to propose that of
Sewalik,! formerly applied to the portion between the Ganges
and the Jumna; a name which appears to me better than
that of the ‘Lower Hills’ or the ‘Sub-Himalayan,’ terms,
equally applicable to the subordinate ranges northwards of
the Himalayahs.

These preliminary observations, with the appended sketch of
the country (Plate ITT.), will make the reader acquainted with
the field of our fossil discoveries; and I proceed to the de-
scription of the geological features of the tract. The forma-
tions composing the hills consist of beds of boulders or shingle,

' In Smith’s Exotic Botany, vol.i.p.9., | residing at Deyra. ‘The derivation sup-

is the description of the Rhododendron
arboreum. He refers it ‘to the moun-
tainous tract called the Sewalik chain,
which separates the planes of Hindostan,
between 75° and 85° HE. long. from the
Himalayah mountains. I make the quo-
tation for the value of the name, though
the statement is evidently wrong; the
rhododendron in question growing in
the Himalayahs themselves at a high ele-
yation, and in company with oaks. The
chain separating the plains of Hindostan
from the Himalayahs, which is the one
now under review, is subject to a mean
temperature perfectly inimical to oaks
and the Rhododendron arboreum. The
name is quoted also in Dow’s History,
and in some traditional writings in the
possession of the high priest or Mahant

plied by the high priest is as follows:—

‘ SEWALIK, a corruption of Shibwal/a,
a name given to the tract of mountains
between the Jumna and the Ganges,
from having been the residence of
Ayshoor Shib, a name of Mahadeo and
his son Gun, who, under the form of an
elephant, had charge of the westerly
portion, from the village of Doodhli, to
the Jumna; which portion is also called
Gungujur (gujur, elephant): the portion
eastward from Doodhli, ‘or-between that
village and Hurdwar, is called Deodhar,
from its being the especial resjdence of
Deota, or Ayshoor Shib. The whole
tract, however, between the Jumna and
the Ganges, is called Shibwalla, or the
habitation of Shib.’

32 FAUNA ANTIQUA SIVALENSIS.

either loosely mixed or agglutinated by clay and carbonate of
lime; of sands of various degrees of consistency; of marl or
clay conglomerate ;! and of an infinite variety of clays; the
two latter being most extensively developed to the westward
of the Jumna. The strata dip from 15 to 35 deg., generally
towards the north; and the breadth of the inclined beds is
from six to eight miles.

The succession of the strata is irregular, with the excep-
tion of the shingle in the more northern tracts, where it ap-
pears to overlay as well as to alternate with the sandstones ;
whereas, although the débris is in considerable abundance in
the water-courses, and on the flanks of the hills, I have never
met with these shingle-beds south of the most northern half
of the hills. A similar remark applies to the marl, but in a
contrary direction, as it occurs only to the south. The want
of the marl in the northern sections may be accounted for by
supposing it to have been either a local deposition or a gene-
ral one, which is not exposed throughout its entire range.

Although I consider the whole of this tract of mountain
as composed of one formation, it may be of use to divide the
portion between the Jumna and Ganges, from that west-
ward of the former river, as it appears to me that there is a
marked difference (as before stated) in the shingle strata, as
well as in the position and number of the clay beds ; although
perhaps nothing further than would be exhibited on any ex-
tended line of country, formed of débris from different tracts
of mountains. The causes which upheaved this district ap-
parently acted more energetically in the contracted surface
eastward of the Jumna than in that to the westward;
though, in the latter, the surface upheaved is much more
extended. JI shall therefore divide these notes into two sec-
tions ; the first consisting of the tract between the Jumna
and Ganges, the second, that westward of the Jumna. To
commence, therefore, with the range between the two rivers.

HILLS BETWEEN THE JUMNA AND GANGES.

Shingle and Sandstone.—The beds of shingle are of enor-
mous thickness, and alternate with the sandstone. The
former precisely resemble the shingle in the beds of the
existing great rivers of the country, and consist of boulders
of granite, gneiss, mica slate, quartz, hornblende schist, and

1 T call this stratum clay-conglomerate, | acids. The rock is, in general, exceed-
the beds being composed of fragments of | ingly hard and tough, as is shown in
an indurated clay, cemented together by | those parts where it is in contact with
clay, sand, and carbonate of lime; the | the stream of the rivers, which appear to
clay itself effervescing strongly with |-have had little power upon it.

STRUCTURE OF SEWALIK HILLS. 33

traps; and every other rock, through which these rivers hold
their course. If the beds of the Jumna and Ganges were to
be upheaved, in the same way as those of former rivers, the
appearance of the strata would be exactly similar. The
sandstone consists either of grains of pure quartz, with
different proportions of mica, or of an admixture of the other
ingredients so common in all river sands. The presence of
oxide of iron causes a great variety in colour, from red to gray,
whilst the induration of the rock appears to depend on the
proportion of carbonate of lime. In the more easterly limit
opposite Hurdwar, as well as other places, the stone is
quarried for building; and in many localities where it is
erystalline it is highly valuable in architecture. On the
Jumna are the remains of an ancient hunting palace, built
by the emperor Shah Juhan at the end of the seventeenth
century; and although now in utter ruin, amongst its frag-
ments are capitals of columns, and scolloped archwork, &c.,
cut in this sandstone, which prove it to be well adapted for
architectural purposes. It is easily worked, and the mixture
of mica gives two splitting surfaces. On exposure to weather,
however, it exfoliates and crumbles, as is conspicuously shown
in some lintels, still in position at Badshahmuhul. The colours
of the clays are endless; and a light blue variety, which is
found under the marl, as well as higher up in the series, where
it contains fresh-water shells, is exceedingly pure. The strata
of this blue clay are thin, but those of the other varieties are
of all dimensions.

Inguite.—Carbonaceous matter occurs throughout the
sandstones, either in detached fragments exhibiting vege-
table origin, or in strata or seams composed of sandstones
and lignite in equal proportions. Lignite is also found in
the marl, but generally in the form of black dust; leaving,
on its removal, an indistinct, vegetable impression. In one
instance I also met with it in the shingle strata. It has
never been found in sufficient abundance to excite an inquiry
as to its adaptation for economical purposes; though it is
common to the whole of this tract of mountains. Hastward
of the Ganges it has been found, to the north of the Morada-
bad district ; and it agrees in every respect with that found
elsewhere. It has also been met with at points still more
easterly ; but the examination of that tract has been very
slight. The most interesting point at which I have found
lignite is at the mouth of the Kalowala pass, one of the en-
trances into the Deyra Dhoon, from the plains. <A stratum
or bed of yellow and red sand, about eighteen inches thick,
is there completely charged with lignite, either in long flat-
tened masses, or in transverse sections of trunks of trees,

VOL. I. D

34 FAUNA ANTIQUA SIVALENSIS.

which show, by their elliptical form, effects of vertical pres-
sure. The lignite, in these cases, constitutes merely the
outer covering ; the interior of the trunk being composed of
the sandstone matrix. The bed is enveloped in strata of the
marl, abounding in the remains of animals; and both the
lignite bed and the marl appear at this point most distinctly
to have been deposited in a hollow; the beds at their lateral
extremities gradually attenuating, until they meet the sand-
stone rock; and the lower surface being concave. Although
the marl is evidently limited at this spot, it seems to exist so
generally in extended strata, that, supposing it to have
formed tracts of marsh land, those tracts must have been
very extensive. I have before remarked, that I have seen
the marl only on the southern limits of these hills, between
the Jumna and Ganges. In crossing the former river, how-
ever, the same stratum, with the same organic remains, is
met with on the north of the mountain ridge, on which the
town of Nahun stands. Here the upheavement has been
more violent than elsewhere; and the point of junction of
this tract with the higher Himalayahs is consequently much
dislocated. It is necessary to state, that there are appear-
ances of trap! in the neighbourhood of these disturbances ;
a fact which may lead hereafter to an interesting geological
disquisition on the point of junction. Fossil wood, appa-
rently dicotyledonous, abounds in the sandstone. The woody
fibre is generally perfect, but impregnated with the sandstone,
and frequently mixed with carbonaceous matter; in which
case the fossil is black, and on fracture has the appearance of
an intimate mixture of coal and sand, or of an imperfect
coal. The only remains of animals yet found are fragments
of tortoises.

Marl, with Organic Remains.—As the marl is in greatest
abundance in this section, I will now give an account of its
organic remains, confining myself to classes and genera.
These fossils are in appearance perfect, and the deep black
colour which they have derived from hydrate of iron renders
them ornamental. The medullary cavities are, in many cases,
filled with pyrites, and in others with pure white crystallised
carbonate of lime. The greater part of the fossils already
procured is from the deposit in the Kalowala Pass; and-as
my collection is not only large, but abundant in different
genera, the number already found in such a limited space
indicates the great variety of animal remains which a dili-
gent and extensive search may produce. Teeth, and the
more solid part of the bones, are found in the greatest

1 Dr, Falconer has made the same | vol. iv. p. 50; date of communication
observation, Journ. Asiatic Soc. Bengal, | January 3, 1835.

™

156 teetaatins

STRUCTURE OF SEWALIK HILLS. 35

quantity. Most of the former are perfect, and as sharp in
their outline as when the animals existed ;’ and even in the
fragments, the sharpness of the fracture proves that they
were quietly deposited in the sediment in which they are
found. The following is a list of the fossils already disco-
vered in the marl :—

MAMMALIA.

Pachydermata . . Teeth, and the remains of a species of Anthracothe-
rum.

Carnivora. . . . Genera doubtful; but some of the teeth correspond
with the third incisor of the hear.

Rodentia . . . . Rat, anda small variety of castor.

Ruminantia . . . Deer , several varieties, and one molar of a very
small species.

Solipeda . . . . Horse, one incisor, and one molar from the right

side of the upper jaw: they exhibit a peculiarity
in the form of the enamel flexures,

REPTILIA.

Crocodilia. . . . Gharidl. (Gavial of naturalists). Teeth and bones
in abundance.

Crocodile. Teeth, in great abundance, fragments
of the osseous plates, vertebree, and other bones.

Chelonia . . . . Emys. Fragments of the plates of the back very
perfect; also ribs, with the attached osseous part
of the buckler.

Trionyx. Ditto, ditto. Some marked differences
in the form and position of the rugous surface of
the fragments of the buckler, may “hereafter point
out a variety of species.

PIscEs.
Vertebre and Scales—The latter doubtful.

TESTACEA.
Bivalve . . . . Imperfect, but resembling Unio.
Univalve. . . . A cast resembling that of the Paludina of the present

fresh-water.

Besides these more easily determinable remains, there have
been found a number of incisor teeth, which I cannot assign
to any genus; also quantities of bones, in fragments, and
portions of ribs, one of which must have belonged to a large
animal; also vertebre, metatarsal or carpal bones, &c. With
the exception of the teeth, the remains are generally in frag-
ments, and separated; nor does it appear probable that we
shall meet with anything approaching to a complete skeleton.

1 The only univalve which has been | Nahun. Ihave entered it here, consi-

yet found in the marl is the cast above- | dering, as I do, that the deposits are
mentioned, from the rock north of! identical.

D2

36 FAUNA ANTIQUA SIVALENSIS.

In the tract then between the Jumna and Ganges, the
fossil remains as yet discovered are thus disposed :-—

Shingle, or Gravel Beds Lignite, scarce.

Sandstone. . . . Trunks of dicotyledonous trees in great abundance ;
lignite, and remains of reptiles.

Marl . . . . . Remains of mammifers, reptiles, fishes, shells, and
lignite.

The mineral products are, carbonate of lime, the general
cement of the whole formation, also in stalactites and stalag-
mites ; selenite, in small tabular crystals; and pyrites, but
apparently in the neighbourhood of organic remains only.
Soda abounds throughout the mountains, efflorescing on the
shingle and sandstone rocks; and the presence of this alkali
may explain the partial disintegration of the boulders of
which the shingle is composed: for I believe I am right in
asserting, that every variety of boulder, from granite to quartz
and sandstones, has been acted on ; and when it happens that
the boulders can be removed entire from the bed, they fall to
pieces, either after exposure to the air for a time, or by pres-
sure of the hand immediately. With regard to fractures in
some of the boulders, and their consolidation on different
planes, as in a slip of stratification, no remark whatever
appears requisite, as these are the necessary consequences of
that movement which raised the beds from their horizontal
position. There does not appear to be any further remark
necessary on this tract, more than can be reserved for the
general summary; I shall therefore proceed to describe the
hills westward of the Jumna.

HILLS WEST OF THE JUMNA.

These, as I before stated, consist of the same series of
shingle, sand, clays, and marls; but they differ, in the beds
of shingle being less abundant, although equally inclined,
and in containing a different description of boulder. The
rolled fragments eastward of the Jumna are all of primary
or lower rocks, whilst those to the westward are confined to
varieties of clay slate and quartz. The marl, which, between
the Jumna and Ganges shows itself in the southern limits, is
here exposed at a point north of Nahun, cropping out on the
northern slope of the mountain; and the fossil remains re-
semble those found in the marl eastward of the Jumna, con-
sisting of mammifers, crocodiles, tortoises, fishes, and shells.
From Nahun to the plains there is a succession of sandstones
and clays, without any abundance of shingle (fig. 2). The
sandstone, which, in the vicinity of Nahun, is much indu-
rated, and used for building, becomes softer on approaching

STRUCTURE OF SEWALIK HILLS. 37

the plains; long before reaching which, the whole formation
consists of an interminable succession of sandstones and clays,
the latter being in the greatest abundance, of every variety
of colour, and dipping, on an average, 20° to the north. The
topographical outline of these mountains shows a consider-
able southing of upheavement, in the hills westward of the

Fre. 2.

PLATEAU UPON WHICH THE VILLAGE OF DEONI IS SITUATED, ON THE MURKUNDA RIVER.
a. White Sandstone. 0. Red and Brick-coloured Clay. ¢. Débris.

Jumna, as will be observed in the map;! and the circum-
stance of the fossil remains abounding in the sandstones and
clays in this tract, and not in that between the Jumna and
the Ganges, may probably be due to the non-upheavement of
the line on its prolongation eastward of the Jumna. The
drawings which accompany this paper (figs. 2, 3, and 4) show
that the action, in all probability, was exerted irregularly ;

Bigs a3:

ae

gs
wi

SLIP UNDER TULOAGPOOR.

and although in the large scale we may lay down the dip and
direction with accuracy, the former as varying from 15° to 35°,
and the latter from NH. to SW., local details give very dif-
ferent results.

1 Plate" III.

38 FAUNA ANTIQUA SIVALENSIS.

The sandstone rock, from which the fossil remains sent to
the Society’s Museum were extracted, reposes at the above
angle, over numberless beds of clays, more or less rich in tes-
taceous remains. The fossil bones lie in great abundance on
the surface of the slopes in the neighbourhood of the sand-
stone, amongst the ruins of fallen cliffs, in the beds of water-
courses, &c. The bones which we have had the good fortune
to dig out of the rock are perfectly sharp, and in all their
original perfection. I may here advert to a circumstance to
which the preservation of the water-worn specimens is chiefly
due. The sandstone is generally soft, but in the proximity of
the fossils it becomes ferruginous, concretionary, and so hard
as to turn the edge of the chisel ; and thus it protects the fossil
from destruction in its progress, as a boulder along the tor-
rent’s bed. These concretions are occasionally globular, and
become singularly conspicuous, by the weathering of some of
the ridges, when the mass of rock takes the appearance of
huge spherical concretions piled confusedly on each other.

Fic. 4.

DISLOCATION OF STRATA BELOW THE VILLAGE OF TULOAGPOOR.
a. Clay. 6. Sandstone. c¢. Clay. d. Sandstone.

The organic remains of this sandstone yet brought to light
belong to the following classes and genera: most of the species
are new; and the appearance of totally undescribed forms
will add very considerably to our fossil genera. The new
genera themselves will be the subject of separate accounts in
their proper place.

MAMMALIA.
Pachydermata . . Mastodon, elephant, rhinoceros, hippopotamus, hog.
Carnivora . . . Canine and feline.
Ruminantia . . . Elk, ox, deer in great varieties.

Solipeda . . . . Horse.

ee a

STRUCTURE OF SEWALIK HILLS. 39

REPTILIA.

Crocodilia. . . . Gharidl, crocodile; both very closely corresponding
with the existing species now in the rivers.
Lacertine remains indeterminable.

Chelona . . . . Emysand Trionyx; some of the fragments are of
the most gigantic proportions. Of the smaller
varieties, nearly entire specimens have been
found; upper buckler and carapax complete.
‘My cabinet also contains three heads, wanting
only the occipital portion of the cranium.

Pisces.

Genera not established.

Many other fragments have been found, but so imperfect
as to render a classification impossible. I may remark, that
there appears to be no end to the variety as well as quantity
of these remains; and we may expect to do much, even in
this remote region, in advancing the inquiries respecting
fossil zoology.

Of each genus above mentioned, with the exception of the
horse and the carnivora, I have already almost perfect skulls.
The bones of the body, however, appear to have been much
broken and mutilated; but it is a singular fact, that from
many places where the fossils have been found as mere débris
of fallen cliffs, fragments of bone have been obtained, which
have admitted of being joined, although the fractured ends
were coated with carbonate of lime, as if they had been fos-
silized separately. A beautiful example of this is exhibited
in an almost perfect rib of an elephant or mastodon, which is -
forwarded to the Society’s Museum, and which consisted of
no less than eight pieces. A perfect humerus of a ruminant
has been secured in this state; and the bones of two hind
legs, namely, the upper part of the metatarsal connected to
the lower portion of the tibia by the intermediate tarsal
bones, with, also, the os calcis entire, and all the smaller
bones of the tarsus equally so. These remains have be-
longed to an enormous animal, and, I believe, to the same
genus as that of a skull in my possession, and now under
description by my friend Dr. Hugh Falconer, of the Bengal
medical service, and myself. Although I refrain from zoolo-
gical details, J must mention that we have an animal evi-
dently forming a connecting link between the Pachydermata
and Ruminantia, or between the Tapir or Paleotherwwm, and
the latter order of mammalia. The hippopotamus of this
sandstone appears to be a new species, having six incisive
teeth, besides other peculiarities, particularly in the propor-
tion of the bones of the head: the tusks also differ from

40 FAUNA ANTIQUA SIVALENSIS.

those described by Cuvier in his ‘ Ossemens Fossiles.’ The
great depth at which the marl lies beneath the upper strata,
with the discovery in it of remains of the horse, is an inter-
esting fact. In the sandstone strata the remains of the horse
are by no means scarce. There appears, however, to be a
local disposition in the deposits of these remains. In some
places the hippopotamus, elephant, mastodon, crocodile, tor-
toise, &c., are found in abundance, with the remains of ru-
minants ; in others the hippopotamus and the water reptiles
are almost totally absent, and only the remains of ruminants
and carnivora occur; all tending to prove that these animals
were destroyed on the site of their habitats; and that this
former world was not more mysterious than the present; that
there were vast tracts of marsh and river, with their attend-
ant hippopotami and crocodiles in the waters, and elephants
and mastodons in the neighbourhood ; and that there were
other tracts free from water and marsh, and frequented by
their natural inhabitants, ruminants, carnivora, &e.

Dr. Falconer, in a note read at a meeting of the Asiatic
Society of Calcutta,! suggested the identity of this deposit
with that near Prome, some of the fossils from which have ~
been so beautifully lithographed in the Society’s ‘'Transac-
tions.’? The mastodons in the Sewalik strata are in great
abundance; and as we have perfect skulls, we are enabled to
form some opinion of the dentition and the change of teeth.
Three of these skulls, now in my cabinet, have the front
tooth worn, and the rear one coming into use; the whole
line of teeth stretching on a surface of nineteen inches, and
- forming an are of 90°. The front tooth I cannot distinguish
from that of the M. latidens figured in the ‘Geological Trans-
actions ;’ and the rear one, in the same animal, bears such
perfect resemblance to the M. Hlephantoides, that I cannot
help risking the conjecture that the M. latidens and the M.
Elephantoides are one and the same animal; the specimens,
from which the specific characters were taken, having been
detached teeth.

Bringing this forward in the way of suggestion may be of
use, in ultimately leading to truth. The question, however,
must be determined by a strict examination of specimens;
and having advanced the above supposition, I will take care
that the means shall be provided.’

1 Journal Asiatic Society, vol. iv. p.| of the elephant, so far ag relates to the
58 ; date of communication, January 3, | integrity of the apices or summits of

1835. ; their transverse ridges; and they might
2 Geological Transactions, 2nd Series, | therefore be mistaken for those of the
vol. iii. p 377 et seg.; pl. 36 to 43. M. Elephantotdes ; but the observations

3 The rear or newly-formed teeth of | which led to the conclusion that the Ava
every species of mastodon resemble those | specimens of Mastodon belonged to two

STRUCTURE OF SEWALIK HILLS. 41
The minerals in this tract, west of the Jumna, correspond
with those to the eastward, with the exception of the presence
of the gold, which occurs in the beds of the rivers in these
mountains, both eastward of the Ganges and westward of the
~¥amna; but I am not aware of its existence in the interme-
diate tract. Under Nahun, zold-washers are constantly em-
ployed during the dry months ; their daily return varying from
two annas to two rupees, or from 3d. to 4s. The process is
extremely rude. A piece of board, a long wooden trough, a
ladle made out of a gourd, a sieve of the large grass which
erows so abundantly in this part of the country, and a piece
of hollow bamboo with a little quicksilver, constitute the
portable apparatus of these most .primitive washers for the
precious metal.! A great deal of the gold-dust must neces-
sarily be lost by this method of proceeding; and all the
mercury, as it is evaporated in open air. The grains of gold
are not larger than the small scales of mica, so common in
fiver sand; and I have not heard that the metal has been
found in large masses. This gold, too, has its localities,
some streams being much richer than others. Where is the
fountain head? In some stratum of this alluvium perhaps,
yet to be discovered, as the streams have no connection with
the higher mountains. A description of the washers, and the
method practised by them eastward of the Ganges, have been
given in the ‘ Journal of the Asiatic Society ;’ and as there is
little difference in the apparatus used at both places, it is not
necessary to refer to it further.

Some of the fossils obtained by Mr. Colebrooke in the hills
near Cooch Behar, and described by Mr. Pentland,” are iden-
tical with some of those obtained by the present discovery ;
and as the former were found in the most eastern extremity
of this line, it is possible that a careful examination would
prove the existence of animal remains throughout the whole
of the intermediate mountains; all the tract being probably
tertiary. An inquiry into the difference between the hills
bounded by the Ganges and Jumna, and those westward of
the latter; and into the confused and interminable disloca-
tions, without any flats in the former, and the more scattered
ridges with the intermediate plateaux of the latter, would
distinct species were not made on de-

tached teeth. A repeated examination
of the jaws and teeth described by Mr.

evidence to the contrary is adduced.—
REFEREE Gxou. Soc.

Clift, and the knowledge that his opinion
was formed from considering the size and
number of the transverse ridges in rela-
tion to the length and breadth of the
teeth, have tended to confirm my belief in
the establishment of two species, which I
must retain until further and more definite

1 See further particulars of the gold-
washings in the Gumté river, by Captain
Cautley, Journal Asiatic Society of Ben-
gal, vol. iv. p. 279; plate 7; April,
1836.

2 Geological Transactions, 2nd Series,
vol. ili. p. 393; plate 45.

42 FAUNA ANTIQUA SIVALENSIS.

only lead to a disquisition on the general formation of the
dhoons or valleys lying between the Sewalik hills and the
great chain, which the limits of this communication will not
admit, and must be left for future inquiry. It will be suffi-
cient to state, that the general form of the mountains ap-
proaches more or less to a right angle, the long slope being
covered with vegetation, and the crest terminating in a per-
pendicular and generally mural cliff, which descends into the
beds of the torrents. The scenery from these cliffs is most pic-
turesque, and many of the passes or ghats up the torrents are
bounded by gigantic walls of sandstone, varying both in height
and character, and subject to all the tortuosities dependent on
a river’s course, forcing its channel through a complication of
mountains. This effect is considerably heightened by the
pointed and jagged style of the outline, depending on that
abundance of clay and carbonate of lime throughout the whole
formation, which permits even the shingle-beds to hold u
their pointed summits in the wildest manner imaginable. At
places, a perfectly inaccessible needle of shingle raises its
head far above the others, and is crowned by one solitary fir
tree, Pinus longifolia. This conical form, produced in the
clays by weathering, is sometimes magnificently displayed, by
the whole face of a cliff, consisting of light pink, yellow, and»
blue clays, being externally decorated with small conical spires
of the clay from top to bottom, a height of 1,000 feet from the
bed of the river. The origin of this structure is due simply
to a little coping stone, of some harder material than the
matrix of the strata, and upon which the weather does not
act so rapidly as upon the clay. Under a bright sun, the
beauty of these cliffs, with their illumined pinnacles, is exqui-
site; particularly westward of the Jumna. I have before
adverted to the presence of great abundance of lime, forming
stalactites, and the cement of the sandstone and conglomerate.
There is another striking method of deposition on the surface
of the large stones which lie in the beds of the rivers, and
which, during the greater part of the year, are in contact
with the water. The substance is very similar in appearance
to the coarse brown paper made in this country: and is pro-
duced by the water, while washing the stone, depositing its
lime, and entangling the finer particles of vegetable matter,
sand and mica, until the stone acquires a superficial coating
of a brownish yellow cclour, which on removal resembles the
substance before mentioned,

Manuxmow, near Suharunpoor :
June 30, 1835.

ELEPHANT AND MASTODON. 43

Il. ON THE FOSSIL SPECIES OF ELEPHANT AND
MASTODON FOUND IN THE SEWALIK HILLS."

BY H. FALCONER, M.D., F.R.S.

Generat Remarks,

81.

Tur fossil remains of the Proboscidean Pachydermata have
in all ages attracted more attention, both from the learned
and from the unlearned, than perhaps those of any other
family of extinct animals. Until a comparatively late period
in Europe, and at the present time in all countries where the
light of anatomy cannot be brought to bear in solving the
mystery of their indications, the enormous bones of this
tribe, when disinterred from the earth, have been regarded
as demonstrative evidence of the former existence of Titans,
Giants, and other fabulous beings handed down to us in the
records of superstition and mythology. Like the Greeks and
Romans of old, the people of India even now usually refer
such remains to the Rakshas or Titans, who hold so prominent
a place in the ancient writings of that country. The severe
investigations of modern science have expelled these fictions

1 This memoir comprises the only | rately ascertained. The order to be

portion of the letterpress of the Fauna
Antiqua Sivalensis which ever appeared
(in 1846). A continuation of the letter-
press found in manuscript among Dr.
Falconer’s papers is now added to the
part already published. The work was
prefaced by the following remarks :—
‘ The plan which we have laid down for
our guidance in the conduet of this work
is, in the first place, to determine the
remains of the extinct genera and spe-
cies, and, on the conclusion of the sys-
tematic and descriptive details, to inves-
tigate the general results to which they
lead. The advantages of this method
are so obvious that it is unnecessary to
insist upon them; for general conclu-
sions in science are of little value, if the
facts upon which they are founded be
not in the first instance rigidly and aceu-

observed in describing the different fami-
lies will depend more on the state of
preparation and convenience of the ma-
terials than upon any strict principle of
zoological arrangement. This, which
might be deemed objectionable in a
general systematic work, is of little con-
sequence in the case of a particular
Fossil Fauna, provided that the forms in
each family and genus are taken in
sequence. The great paleeontological
work of Cuvier opens with the Pachy-
dermata, the proboscidea being the first
in the order of description. Following
our illustrious guide in extinct zoology,
we shall commence with the Elephant
group, in which is most signally dis-
played the numerical richness of forms
which characterizes the Fossil Fauna of
India.’ —[ Ep.]

44 FAUNA ANTIQUA SIVALENSIS.

from the belief of civilized mankind, and reconstructed the
true forms of the animals which appear, in many instances,
to have given rise to them. Palzontology made, as it were,
its first great advance in the exact determination by Cuvier
of the mammoth of Siberia, and the mastodon of North
America. Since that time several new forms have been dis-
covered, and most of the great points connected with the
structure of the Proboscidea, fossil and recent, have been ascer-
tained. But, notwithstanding the vast amount of observation
on the subject during late years, a great difference of opinion
has prevailed among comparative anatomists and paleeontolo-
gists, down even to the period when we now write, in regard
to the degree of affinity and generic relation of the different
species of mastodon and elephant. The majority of late
authorities, including Cuvier and Owen, have regarded them
as constituting two distinct and well marked, although
closely allied, genera; others have gone the length of break-
ing up mastodon into two genera; while M. de Blainville has
reverted to the opinion of some of the earlier observers, that
the so-called mastodons and elephants are but modifications
of one common type, differing so little from each other that
all the species may, with propriety, be included within the
limits of a single genus. A still greater and vastly more
important difference of opinion has prevailed regarding the
number and characters of the species; for, while the con-
flicting views respecting the generic distinctions concern
little more than the principles of systematic classification,
the accurate determination of the fossil species affects the
value of facts, which implicate the accuracy of some of the
most weighty arguments in the geology of the later tertiary
strata, more especially such as relate to the changes of
climate which are supposed to have accompanied their depo-
sition, and the extension of the species through a wide range
of time and space. Cuvier considered all the elephant remains
which have been found in Europe, the north of Asia, and
America, whether occurring in the superficial drift of Siberia,
or in the tertiary beds of the Val d’Arno, to belong strictly to
a single species, Hlephas primigenius. Professor Owen, with
all the lights, and wielding every arm of an advanced science,
holds the same opinion. M.de Blainville does not think
that there are sufficient characters even for separating the
mammoth from the existing Indian elephant, both of which
he appears to regard as varieties of the same species.! On the
other hand, Nesti, after a careful study of the elephant re-
mains of Italy, during a period of nearly twenty years, upon

1 De Blainyille, Ostéographie; Elephants, p. 222.

eee Ee

;
:
¢
-
:
.

ELEPHANT AND MASTODON. 45

an ample collection of the best materials in the form of
crania, jaws, and teeth, insists upon the specific distinctness
of the Tuscan fossil elephant, H. meridionalis, from the true
mammoth of Siberia. Other paleontologists have gone so
far as to construct ten species out of the single species of
Cuvier, founding the distinctive characters upon the differ-
ences presented by the molar teeth. A like range of con-
flicting opinions has prevailed in regard to the mastodons.
Cuvier, Owen, and De Blainville concur in restricting the
narrow-toothed mastodons of Europe to a single species, the
geographical range of which Cuvier extended even to South
America; while Croizet and Jobert, Kaup, Von Meyer, and
others, divide them into two, WM. longirostris, and M. angus-
tidens. No less than two genera and at least ten nominal
species have been founded upon teeth, which Owen, De
Blainville, and most other authorities, attribute merely to
different ages and sexes of a single species, the M. Ohioticus
of North America. In short, the ascertained fossil species,
exclusive of those of India, according to some, are limited
to one elephant and four or five mastodons; while others
would raise the number of the former to ten, and of the latter
to upwards of twenty.

This great diversity of opinion, almost unequalled in regard
to any other section of mammalian paleontology, has in a
ereat measure arisen from the isolated and often defective
nature of the materials relating to this tribe, as they ordi-
narily come before the paleontologist. From the peculiar
mode of succession of the molar teeth, which yield the prin-
cipal distinctive characters in mastodon and elephant, by
repeated renewals from back to front, at different stages of
the animal’s growth, as the worn and exhausted grinders
drop out, a limited number only of the whole series can be
met with in any one fossil specimen, even under the most
favourable conditions. It is this peculiarity which has so
long retarded the attainment of an accurate knowledge of
the dentition of the living species. The difficulty applies
with: double weight to the fossil species,! for the teeth are
rarely met with in connection with perfect crania and jaws;
they most frequently occur detached, or connected with
mutilated fragments. It is only, therefore, from the com-

1 Tn illustration, it may be mentioned | of Dinotheriwm as a portion of the upper
that Hichwald (Nova Act. Acad. Nat. | jaw of a new species of mastodon, which
Curios. 1834, vol. xvii. p. 735, tab. lili. | he names Mastodon Podolicus. Other
fic. 2), in a memoir descriptive of fossil | equally remarkable cases of the same
remains of Elephas, Mastodon, and Di- | kind might be adduced ; so unsafe is it to
- notherium, &c., found in Poland, figures | draw conclusions regarding the fossil
and deseribes what appears to be a frag- | Proboscidea from imperfect materials.
ment of the symphysis of the dower jaw ;

46 FAUNA ANTIQUA SIVALENSIS.

parison of an extensive series of specimens, embracing every
period of life and the range of individual and sexual varieties
through which the species runs, that any safe conclusions
can be drawn regarding the distinctive characters of any one
form. In consequence, although paleontologists are agreed
on the great points relating to the construction of the head,
trunk, and extremities, hardly any two concur respecting the
number, form, and succession of the teeth in the different
species of mastodon and elephant.

The surprising number of forms belonging to this family,
embraced in the fossil fauna of India, and the immense
abundance in which their remains have been met with, have
placed us perhaps, with respect to the quantity and perfect
condition of the materials, in more favourable circumstances
for the determination of the Sewalik species than has ordi-
narily happened to the paleontologist in the case of most of
the other fossil Proboscidea. Of five of the species to be
described in the sequel, we possess nearly perfect crania of
each, and, in most of the instances, crania with teeth of all
ages, from the very young up to the adult animal, in addition
to a vast collection of the detached teeth and lower jaws, so
as to furnish us with the whole of the essential evidence
requisite for the specific determination of each of these forms.
The distinctive characters are so broadly marked, that there
is hardly room for a doubt being entertained in regard to
them. In the course of the investigation, we have been led
to examine the conclusions which have been arrived at b
writers who have preceded us upon this family. The Indian
species, and those previously described, fossil and recent,
have mutually reflected light on each other, and ranged
themselves into natural and allied groups. Instead, therefore,
of restricting ourselves merely to a description of the Sewalik
fossil forms, we shall endeavour, in what follows, to trace the
affinities, and institute an arrangement of all the well-deter-
mined species in the family.

The results to which we have been conducted lead us to
differ on certain points from the opinions most commonly
entertained at the present day; for while, on the one hand,
it would appear that the fossil species of both elephant and
mastodon have been unnecessarily multiplied by authors both
in Hurope and America, on the other, we are compelled to
think that Cuvier and others have run into the opposite
extreme of caution, and in more than one instance included
distinct forms under the same nominal species. Further, in
regard to the views which have been at different times
advanced respecting the differential characters of elephant
and mastodon, in the succession and development of the

eo? | ee

ELEPHANT AND MASTODON. 47

series of molar teeth, our conclusions are, in some measure,
at variance with those of most other paleontologists who
have preceded us upon this family.

Before entering upon the description of the species, we
shall examine, at some length, the general characters pre-
sented by the teeth ; but in order to comprehend the present
state of knowledge on this branch of the subject, it will in
the first instance be necessary to pass briefly under review
the leading opinions which have been entertained by palon-
tologists regarding the relations of mastodon and elephant to
each other, and to notice the successive steps in the discovery
of new forms which have led to the modifications of these
opinions.

It is beside our object to give anything like an historical
account of the labours of the earlier writers. Those who are
desirous of the information will find it detailed in the great
work of Cuvier, down to the period at which he wrote; and
for subsequent opinions, they may consult the ‘ Ostéographie’
of De Blainville, now in course of publication, and the writings
of Professor Owen, Bronn, Von Meyer, Kaup, and other
palzontological authors.

Notwithstanding the earlier contributions of Daubenton,
Pallas, Merck, and the elder Camper, hardly anything was
known regarding the succession of the teeth in the elephant,
except that they are repeated oftener than once during life,
by protrusion in the jaws from behind forward, till the
appearance of the memoir by Corse,' in the ‘ Philosophical
Transactions’ of 1799. This excellent and careful observer
had resided many years in India, in charge of a Government
stud of elephants in Bengal. By captures of herds of the
wild animal he had an opportunity of watching the successive
fall and renewal of the teeth from the youngest age up to
the adult, the periods of which he carefully recorded ; and
casualties supplied him with a series of upwards of thirty
crania of all ages, upon which he studied the form, size, and
the number of plates which enter into the composition of the
grinders at different stages of the animal’s life. The obser-
vations embodied in this memoir are the most valuable which
have been made on the teeth of either of the living species.
Corse first showed that the Indian elephant has ‘ milk’ tusks
(incisors), which cut the gum when the calf is about six
months old, but are extremely caducous, as they drop out
between the first and second year. He detected the position
of the capsule of the permanent tusks, which protrude about
two months after the milk incisors are shed, and go on in-

1 Corse, Phil. Trans. 1799, vol. Ixxxix. p. 205.

48 _ FAUNA ANTIQUA SIVALENSIS.

creasing in size during the rest of the elephant’s life. He
has described the variations in size, form, and direction which
the tusks present in the different sexes and castes of the
Indian species, the general character of which castes he has
accurately recorded; but the most valuable part of his
observations is comprised in what relates to the molar teeth.
He showed that they are reproduced several times during
life, and that the number of plates entermg into the compo-
sition of each molar goes on increasing as the teeth are
successively renewed. This succession he has carefully traced
up to the fourth grinder: the first cuts the gum eight or ten
days after birth, is well out at six weeks, and is composed of
four plates; the second is completely in use at two years,
and consists of eight or nine plates; the third serves the
period between the second and sixth year, and has twelve or
thirteen plates; the fourth is in use between the sixth and
tenth year, and consists, according to Corse, of about fifteen
plates. Puzzled, probably by the irregularity in the number
of plates, and the size of the rest of the molars in different
individuals, this faithful observer stops short at the point
where his observations ceased to be conclusive, and does not
attempt to define the number of plates in those which follow
after the fourth. He states, generally, that the plates go on
increasing successively up to the ‘seventh or eighth set,’
when each grinder consists of twenty-two or twenty-three
plates, being the greatest number which he had observed.!
These observations were of especial value in furnishing a
standard of comparison for the teeth of the fossil species ;
and by establishing the existence of milk incisors, they proved
that part at least of the dental system of the elephant agreed
with that of the ordinary Pachydermata. But Corse was at
fault in the conjecture that eight molar teeth are successively
developed in an antero-posterior series in this animal; for if
this were the case, the elephant would form an exception to
a general law in the Pachydermata and _ allied orders, among
which the normal number of milk molars does not. exceed
four, that of the true molars being invariably three.
Notwithstanding this objection, Corse’s numerical state-
ment was adopted by Cuvier and by all other authors prior to
1844, when it was, for the first time, challenged by M. de
Blainville in the part of his ‘ Ostéographie’ devoted to the
elephants. Bronn,? in his ‘ Lethea,’ gives eight molars on
each side of both jaws to the genus Elephas ; and Dr. Grant,3
in his memoir upon the Proboscidea, puts forward different

1 Corse, loc. cit. p. 224. 3 Grant, Geolog. Proceedings, 1842,
2 Bronn, Lethza Geognostica, 1538, } vol. 11. p. 771.
p. 1240.

ELEPHANT AND MASTODON. 49

dental formule in the molars as points of generic distinction
between elephant and mastodon, attributing eight molars in
each side to the former, and only six to the latter.

Cuvier, except in what regards his hypothetical explanation
of the formation of the dental tissues, has described with
admirable clearness and in great detail the structure and
mode of growth of the teeth in the elephant. But he had
assuredly arrived at no accurate idea of the true division of
the molar series into milk and permanent grinders. He
makes no attempt to show where the deciduous series termi-
nates and the true molars begin, although so particular upon
this point in his descriptions of the dental system in the
ordinary Pachydermata and Ruminantia. In fact, the term
‘dent de lait’ is but rarely applied in his elephant descrip-
tions ; and in these instances it has evidently reference more
to the immature age of the animal than in the ordinary sense
to a deciduous tooth which is expelled by a vertical successor.
Having adopted the excessive numerical formula of Corse,
this great anatomist was led to believe that the successional
premolars, instead of being suppressed, are developed behind
the milk teeth. The language in which this remarkable
opinion is expressed is clear and explicit :—‘ L’on pourroit
dire que les dents de remplacement de l’éléphant viennent
derriére ses dents de lait, au lieu de venir dessus ou dessous,
comme dans les autres animaux.’ !

The discovery of the entire set of molars in Mastodon
Oluoticus, by Dr. Hays,? in 1834, and in M. longirostris, by Dr.
Kaup,? in 1835, had clearly established that the antero-
posterior series in these species does not exceed six, and
paved the way for an accurate knowledge of the dentition of
the elephant, when M. de Blainville published, in 1844, the
osteographical memoir upon this genus contained in his great
work. This distinguished anatomist, in his remarks upon
the dentition of the Dugongs, first states that the elephants
(in which he includes the mastodons) have neither milk teeth
nor vertical successional teeth. His words are—‘ Ainsi l’on
peut dire que chez les dugongs et chez les autres lamantins,
ainsi que chez les éléphants, il n’y a ni dents de lait ni dents
de remplacement, comme chez la plupart des autres mammi-
férés; celles-ci poussant, pour ainsi dire, celles-la de bas en
haut, et formant ainsi deux rangées superposées; mais que
toutes les dents que doit avoir l’animal dans le cours de sa vie
entiére forment une seule ligne, un seul rang, une sorte de

1 Cuvier, Oss. Fossil, tom. i. p. 39, 3 Kaup, Oss. Fossil. pt. iv. pp. 65-89,
4to, edit. 1821. 1835.

? Hays, Amer, Phil. Trans. vol. iv. 4 De Blainyille, Ost¢ographie, ‘ Mana-
p. 820, 1834, tus,’ p. 75.

VOL. I. E

50 FAUNA ANTIQUA SIVALENSIS.

boyau cylindrique, qui se développe d’arriére en avant dans le
canal dentaire, dont l’énorme diamétre est en partie du a cette
particularité. C’est quelque chose de semblable a ce qui
existe dans les éléphants.’ This statement is subsequently
modified in the portion of the work devoted to the elephants.
M. de Blainville there prefers a claim to having first indicated
the normal number, the mode of succession, and the true
signification of the molar teeth in this genus.’ He admits,
on the evidence of Corse, milk and permanent incisors, but
sees no occasion to apply a similar division of ‘ milk’ and
‘permanent’ to the molar series. He follows their succession
from first to last, and vindicates the first part of his claim by
showing that in the existing species, and probably in the
mammoth, the number of developed grinders is neither more
nor less than six. But in regard to their signification, viewed
as a series, we cannot, with every respect for this eminent
anatomist, admit that he has been equally successful. He
divides the six molars into three sets, each consisting of two
teeth. The first includes the two anterior molars, which are
characterized by being worn out in the part of the jaw where
they first protrude; the second set is characterized by being
formed in the posterior part of the jaw, and at last pushed
out in front; and the third set, formed behind like the two
preceding teeth, is characterized chiefly by serving the adult
stage of the animal’s life, and by the greater space which the
teeth have to traverse in progressing forwards. This, how-
ever, is at best but an etal division, and cannot be received
as a philosophical interpretation of the theoretical significa-
tion of the molar series: for the second of these sets groups
together the last milk and the first permanent molar. Fur-
ther, M. de Blainville expresses, by a numerical formula,? his
view of the dental system of the elephant, which, if inter-
preted according to the exposition of his peculiar method of
symbols, laid down in the first part of the ‘ Ostéographie,’
would imply that the elephant has three premolars and three
back molars, the milk series being suppressed : whereas, in
the three species of which the dentition is known, it is in
reality the three milk molars that are invariably developed,
while the three premolars are constantly suppressed.

1 Loe. cit. p. 58. ‘Aussi dans aucun

2 Loe. cit. p. 75. The formula given
ouvrage le nombre normal, la mode de

isp + $434 4 4 2. The sig

succession des dents molaires de /’élé-
phant, et leur signification réelle, n’ont
pas été déterminés, ce que nous croyons
pouvoir faire aujourd’hui d’une maniére
positive, en nous aidant, il est vrai, des
dents fossiles que possédent nos collec-
tions en plus grand nombre peut-étre
que des dents récentes,’

nification of these figures, indicated in
the opening memoir ‘Sur les Mammi-
féres en géneral, p. 34, is + in. + 2
can. + avant mol. + + princip. +

2 .
9 arriére mol,

ELEPHANT AND MASTODON. 51

Professor Owen, down to the close of 1844, appears to have
held with Cuvier the opinion that the developed molars in the
Asiatic elephant amounted to seven or eight. This number,
adopted on the authority of Corse, is stated in his ‘ British
Fossil Mammalia.’! But it was not likely that the true
numerical formula would escape the sagacity of this eminent
comparative anatomist when directed to the teeth of the
elephant in connection with those of the ordinary Pachyder-
mata. Accordingly, in the last part of his ‘ Odontography,’
while he agrees with M. de Blainville in attributing six
molars on each side of the jaws to this genus,” he has made
a considerable step in advance of the latter anatomist in
regard to their signification. The occurrence of a vertical
successional premolar in the upper jaw of one or more species
of mastodon had previously established that the two anterior
grinding teeth in that genus are displaced, like the milk
teeth in the ordinary Pachydermata, by a vertical successor.
Professor Owen follows up this indication to its legitimate
conclusion, and infers that the third molar in the series of
antero-posterior succession in the mastodon is the last milk
molar, the vertical successional tooth by which it ought, nor-
mally, to be pushed out, usually remaining undeveloped. He
then extends this view to the dentition of the elephant, and
states that ‘it is probable that the three preceding teeth’
(namely, the three first developed molars) ‘are analogous to
the true deciduous molars of the ordinary Pachyderms.’*
The correctness of this opinion is susceptible of demonstration
by the dentition of an Indian fossil species which we have
named Hlephas planifrons, to be described in this work. The
determination of the point was of great zoological interest,
by explaining the apparent anomaly which had hitherto
divided the teeth of the elephant from those of the allied
families in the order.

Next, in regard to the establishment of the species. The
fossil remains of the mammoth had, during ages, attracted
more or less attention in every country in Europe, having
been found in England and in all parts of the Continent,
from’ Italy to Siberia. But it was only towards the close of
the last century that definite notions as to the species were
arrived at. Pallas, who had better opportunities for deter-
mining the point than any of his contemporaries, upon the
perfect remains so commonly met with in Russia, erroneously
considered the fossil teeth to be identical with those of the
Indian species. A great advance was made in the inquiry

Owen, Brit. Foss. Mammal. part v. 2 Idem, Odontography, 1845, p. 626,
p. 226. and note p. 635.
3 Idem, loc. cit. p. 634.

k 2

52 FAUNA ANTIQUA SIVALENSIS.

through the discovery by Peter Camper of the specific dif-

ference between the teeth of the Asiatic and African
elephants,! when Blumenbach and Cuvier almost simul-
taneously entered upon the investigation and arrived at the
same result, viz. that the mammoth was an extinct form, dif-
fering from both of the existing species. Struck with the
length of the cranium and of the incisive sheaths in the
mammoth, as represented in the figures of Messerschmidt’s
specimen attached to Breyne’s excellent remarks in the
‘ Philosophical Transactions,’? and connecting these pecu-
liarities with the great width of the crown, and the narrow-
ness and number of the plates in the fossil grinders, Cuvier
was conducted to his first happy conclusion. The probability
of a similar difference characterizing the species in other
fossil genera flashed across his mind, and opened to him new
views respecting the theory of the earth. Great and important
were the results ; and after they had been achieved, the illus-
trious anatomist reverted, in terms of the liveliest acknow-
ledgment, to the long neglected figures of Messerschmidt
which had helped him to-the first idea.*

After determining the specific independence of the mam-
moth, the next point to ascertain was, whether the remains
occurring in very different deposits, and in localities widely
contrasted in climate and in geographical position, belonged
to the same or to different species. Notwithstanding that the
fossil teeth from the southern parts of Europe commonly
presented wider and fewer plates, with thicker enamel, than
those of the typical form of mammoth found in Siberia,
Cuvier attached minor importance to these differences, as the
teeth agreed in certain other respects; and he ranged the
whole under the single species of Hlephas primigenius. This
opinion has been very generally adopted by subsequent
authors—among others, by M. de Blainville and by Professor
Owen, who has entered at considerable length upon the ques-
tion in his ‘ British Fossil Mammalia,’ and decided in favour
of the specific unity of the European forms. But notwith-
standing this array of authority, we cannot help thinking
that Cuvier was premature in his conclusion, and that’ the
identity of the forms has rather been assumed against the
evidence than proved by it. Had the differences in the teeth
been less considerable than they are known to be, it would
have been requisite to show that the crania at least agreed,
before this identity could be considered to have been satisfac-
torily established. But there were not sufficient materials

1 P. Camper, Descript. Anatom. d’un| 7? Phil. Trans. vol. xl. 1738, p.124.
éléphant mile, p. 16. 3 Cuyier, Oss. Fossil. tom. i. p. 178.

— = ee

ELEPHANT AND MASTODON. 53

for making such a comparison when Cuvier wrote. The
cranium of the Siberian mammoth was known to him only
through figures of five specimens, not one of which was drawn
in any exact projection; and his acquaintance with the
Italian fossil elephant, exclusive of teeth, was limited, in
regard to the head, to a single mutilated fragment, not
extending above the orbits and maxillary bones. There is
little doubt, also, that, like other writers, he was partly
swayed by the extraneous consideration of the geographic
range of the two existing elephants, the continents of Asia
and Africa having each but a single species. We have the
less hesitation in advancing these doubts, as conclusive proof
will be adduced in the sequel that, in the similar instance of
the Mastodon angustidens, Cuvier and others have included
under that name two forms which are so distinct that, in our
view, they do not even belong to the same section of the
genus; while the Sewalik fossil remains show that there were
formerly several species of elephant at one time in the same
Fauna in India.

Soon after the publication of Cuvier’s Memoir, Nesti, in
1808, proposed an addition to the European fossil species,
founded on remains from the Val d’Arno.'! He put forward
two new species: the first, resting on a lower jaw without
teeth, was characterized by a peculiar spout-shaped pro-
longation of the symphysial apophysis; the second, which
he named H. minimus, also resting upon a lower jaw, was
distinguished by its supposed small size, by the absence of
the beak apophysis, and by the rhomboid section presented
by the plates of the grinders. Cuvier refused to admit either
of Nesti’s species. The lower jaw of the first he pronounced
to belong rather to Mastodon angustidens than to an elephant;
and the second to be nothing more than the lower jaw of a
young individual.” Nesti, after a long silence, reverted to
the subject in 1825, in a memoir which embodied the results
of extensive observations carefully made during seventeen
years. He tacitly admits the justness of Cuvier’s criticism
in regard to the second species, by abandoning H. minimus ;
but in confirmation of his first species he adduces a great
mass of evidence derived from numerous crania of all ages,
from the foetus up to the adult, and from lower jaws in the
Florence and other Tuscan museums, all of which, he affirms,
show the peculiar beak-like elongation of the symphysis in
connection with elephants’ teeth. The result of the whole was
to remove every doubt or hesitation from his mind regarding

1 Nesti, Annale del Museo di Firenze, 3 Nesti, Nuovo Giornale de Letterat.

tom. i. Pisa, 1825, tom. xi. p. 119.
2 Cuvier, Oss. Fossiles, tom. i. p. 186.

54 FAUNA ANTIQUA SIVALENSIS.

the distinctness of the Tuscan species, for which he proposes
the specific name of H. meridionalis.

The value of the evidence regarding this species will be
more fully considered in the sequel. Nesti’s opinion, how-
ever, has met with little favour among paleontologists.
Croizet and Jobert have adopted it for some of the elephant
remains found in Auvergne,' and the name finds a place in the
enumeration of species given in von Meyer’s ‘ Palzologica.’?
But it is not admitted by Bronn,? De Blainville,* or Pictet ;°
and it would appear from certain passages in the ‘ British
Fossil Mammalia’ that the great weight of Professor Owen’s
authority is against it.®

In 1821 Dr. Goldfuss’ described a fossil grinder found in
a collection at Cologne, which resembles very much that of
the African elephant, in the characteristic peculiarities of the
rhomboid form and reduced number of the grinding plates.
He states that the specimen, although the precise locality
was unknown, has the cement and ivory as much decomposed
as in the fossil grinders from Siberia. In a second memoir®
he figures and describes other teeth presenting the same
characters, from the banks of the Riihr, in Westphalia, and
concludes from them that the valley of the thine had formerly
an elephant, which was more closely allied to the African
than the mammoth was to the existing Indian species. He
proposed naming it provisionally Hlephas priscus.

Cuvier threw very strong doubts upon the authenticity of
these specimens as veritable fossils, in consequence of the
ambiguous circumstances under which they were found. He
considered them to be nothing more than disguised remains
of African elephants of modern origin.’ But according to
Bronn, fossil teeth of the same description have since been
found, under circumstances fully to be depended upon,
throughout nearly the whole of Central Europe, from the
Rhine to the heart of Russia.!? Some of them have been
described by Wagner;'! and undoubted fossil teeth, presenting
similar characters, have been met with in the ‘brick earth’
beds of the valley of the Thames, at a considerable depth
below the surface. These will be noticed afterwards in con-
nection with the dental series of one of the Indian fossil
species.

1 Croizet et Jobert, Oss. Foss. du Puy-
de-Déme, 1828, p. 123.

? Von Meyer, Palzeologica, 1832, p. 69.

3 Bronn, Lethea Geognostica, 1838,
p. 1245. :

1 De Blainyille, Ostéographie; Elé-
vhants, p. 220.

> Pictet, Paleeontologie, 1844, tom. i.
p. 243.

§ Owen, Brit. Foss. Mamm. p. 239.

7 Goldfuss, Nov. Act. Acad. Leop.

Carol. Natur. Curios. vol. x. p. 485.
8 Idem, loc. cit. vol. xi. p. 485.

® Cuvier, Oss. Fossiles, tom. ii. (8vo.

edit.) p. 184.
10 Bronn, Letheea Geog., p. 1244.
1 Wagener,
p. 21,

Karsten’s Archiy. xyi.

a a ee |

ELEPHANT AND MASTODON. 55

Fischer de Waldheim, in 1829, proposed the separation
from the mammoth of no less than five fossil species of
elephant, founded upon remains occurring in Russia. These
he has severally named—LH. proboletes, EH. campylotes, H. Ka-
menskui, H. Panicus, and H. pygmeus.' Dr. Hichwald went
still further, and added a sixth species, from Poland, under
the name of H. odontotyrannus.2 But the almost universal
consent of palzontologists is against these so-called species,
which are considered to be nothing more than varieties in
the teeth dependent upon age and sex, in individuals of the
mammoth.

M. Morren, under the name of Hlephas macrorhynchus, has
lately proposed a new species for some of the fossil remains
found in Belgium; but the grounds upon which it rests do
not appear to be more valid than in the case of the Russian
and Polish species.’

Mastodon.—In regard to Mastodon: the first determined
species of this genus was the M. Ohioticus* of North America.
The abundant remains found nearly all over the temperate
parts of the United States had, as in the instance of the
mammoth of Europe, attracted the notice of observant tra-
vellers to this great extinct animal, upwards of a century ago.
But, till the time of Daubenton, hardly any progress had
been made towards a definite idea of its nature. This cele-
brated naturalist, in 1762, ascertained the close resemblance
of the femur and tusks to those of the elephant: but the
molars appeared to him to present a nearer approach to the
teeth of the hippopotamus, and he was puzzled whether to
ascribe the fossils to one or to two distinct animals.> Buffon
participated in these doubts, but inferred that a part of the
remains indicated the former existence of a terrestrial animal
which had become extinct, larger than the elephant.® Peter

1 Fischer de Waldheim, Bullet. de la
Soe. de Moscou, 1829, tom. i. p. 275;
Mémoires de la Soe. de Moscou, tom. i.
p. 285.

2 Eichwald, Nova Acta Acad. Ces.
Leop. Car. Natur, Curios. 1834, vol. xvii.
p. 723, tab. 63, figs. 1 and 2.

3 Morren, Bulletin de la Soe. Géol. de
France, tom. ii. p. 231.

4 As in the case of the Mammoth (£.
primigenius, Blum.), the specific name
given by Blumenbach to the North Ame-
rican mastodon, M. Ohioticus, is here
adopted instead either of M. giganteus
or M. maximus, the names applied by
Cuvier at different times. Blumenbach,
in his ‘Handbuch der Naturgeschichte,’
1797, had characterized the extinct ani-

mal by the form of the teeth, and called
it Mammut Ohioticum, as a species of a
peculiar genus, before the appearance of
Cuvier’s memoir (Annal. du Mus. tom.
vi. p. 260, 1805), in which the designa-
tion of M. giganteus is first applied.
This latter was abandoned (Oss. Foss.
4to edit. of 1824) for M. maximus. If
the law of priority left a choice, M.
Ohioticus would still be preferable to
either of the names given by Cuvier, as
the species is by no means the giant of
the family.

5 Daubenton, Actes de l’Académie des
Sciences, 1762; and Histoire Natur. de
Buffon, tom xi.

® Buffon, Histoire Natur. tom, xi.
p. 86.

56 FAUNA ANTIQUA SIVALENSIS.

Collinson, in the ‘ Philosophical Transactions’ for 1768, gave
some good figures of the back grinders, and shrewdly observed
that the form of their crown was adapted for crushing boughs,
twigs, and leaves, betokening the animal to have been of
herbivorous habits.'. William Hunter, struck with the great
contrast between the salient enamel ridges of these teeth and
the narrow plated flat surface in the grinders of the elephant,
inferred the American fossil to have been a gigantic car-
nivorous animal, which he proposed naming Pseudelephant.?
The great reputation so justly attaching to the name of
Hunter gave undue weight and a wide currency to this very
erroneous opinion, and further led to the mammoth of Siberia
being commonly confounded under the same carnivorous
notion with the mastodon of North America.

Peter Camper, in the first instance, made a considerable
step towards an accurate knowledge of the extinct animal,
by the inference that its molar teeth had a greater analogy
with those of the elephant than of the hippopotamus; and
that, like the former, it was probably invested with a trunk
and with tusks;* but he afterwards expressed doubts which
compromised the value of his original observations, having
been led to adopt the opinion of Michaélis, that the animal
belonged to the order Bruta of Linneeus ; that it had no tusks,
and differed greatly from the elephant.‘ This error, as has
been explained by Cuvier, arose from the inspection of a
detached palate with grinders, the posterior part of which
was mistaken, both by Michaélis and by Camper, for the
anterior.°

Pennant first ventured, in 1793, to designate the American
fossil animal in a systematic work as a species of elephant,
by the name of H. Americanus; and Blumenbach, in 1797,
erected it into a kind of genus, under the name of Mammut
Ohioticwm, which he briefly characterized by the form of
the teeth. Cuvier, in his earliest memoir on the elephant,
described it also as a species of this genus under the specific
designation applied by Pennant of EH. Americanus,® for which
Adrian Camper, entertaining the same opinion of its generic
relations, proposed the substitution of H. macrocephalus.’
But in his second extended and. elaborate memoir, published
in 1805, which formed the groundwork of what he has

1 Collinson, Phil. Trans. 1768, vol.

vii. p. 469,

2 W. Hunter, Phil. Trans. vol. lviii.

p. 38.

3 Camper, Acta Petropolit. tom. i.

part 11, p. 219.
4 Idem, loe. cit. tom. ii. p. 259.
5 Oss. Foss. tom. i. p. 212.

6 «EF. Americanus, molaribus multi-
cuspidibus lamellis post detritionem
quadrilobatis.’ Cuvier, Mém. de l’In-
stitut. Ann. vii. (1798). ‘Sur les es-
peces d’éléph. viv. et foss.’ p. 21.

7 A. Camper, ‘ Deseript. Anatom. d’un
éléphant male,’ avant-prop. Note, p.
10.

ELEPHANT AND MASTODON. 57

written on the subject in the ‘Ossemens Fossiles,’? Cuvier
separated the elephants with mammillated molars from the
ordinary forms with lamelliform molars, and united the
former into a genus which he designated Mastodon, taking
the North American species, under the name of M. giganteus,
as the type. In the interval between these two memoirs,
Peale made the important discovery of two skeletons of the
Ohio Fossil, near the banks of the Hudson River, in the
State of New York, one of which was brought to Europe in
1802, and furnished nearly complete materials for instituting
a detailed comparison between the osseous frame of the
mastodon and of the existing elephants.? Cuvier pointed
out the entire correspondence between them in the tusks,
trunk, and the whole of the skeleton, except the molar teeth.
He admitted, even in regard to the latter, that the difference
between the transverse mammillated ridges of the mastodon
and the thin plates of the elephants is merely one of pro-
portion; but insisted that there is an essential distinction in
the circumstance that the spaces between the enamel ridges
are filled with ‘cement’ in the teeth of the elephant, which
is wanting in those of the mastodon. In addition to this
supposed difference, he found corresponding modifications in
the form of the cranium, which confirmed him in his view
of a well-marked generic distinction between Mastodon and
Lilephas.2 In the same memoir, Cuvier characterized four
other species of his new genus, viz. M. angustidens, in which
he included all the narrow mastodon molars found in Europe,
together with some from America, two species from South
America, M. Andiwm and M. Humboldtii, and a small Euro-
pean species, which he named M. minutus. To these was
subsequently added, in the‘Ossemens Fossiles,’ a sixth species,
under the name of M. Tapiroides.

In regard to the number and succession of the teeth, while
he admitted eight molars on each side of both jaws to the
elephant, this great anatomist, not hazarding a conjecture
beyond the materials which had come under his eye, was not
aware of more than four on each side in the mastodon, or
sixteen in all. He was also without the knowledge of the
occasional presence of tusks in the lower jaw; but he first
made the important observation, that in the Mastodon angus-
tidens a part of the anterior series of molars in the upper jaw
is replaced by a vertical successional tooth, or true premolar,
thus bringing them under the normal law of the order of

1 Annales du Muséum d'Histoire Na- | of the Mammoth,’ &c.; and ‘Historie.
turelle, tom. viii. ‘Sur le grand Masto- | Disquisit.’ 1802 and 1803.
donte.’ 3 Oss. Foss. tom. i. p. 221.

2 T.R. Peale, ‘Account of the Skeleton

58 FAUNA ANTIQUA SIVALENSIS.

Pachydermata, by showing a division into a milk and perma-
nent set. The Dax specimen, figured in the ‘Ossemens
Fossiles,’! clearly establishes this point; and the significa-
tion of the structure is distinctly, although guardedly, ex-
plained by Cuvier in the descriptive part of the work.? But
he was less happy in the definition of his species. Under the
name of Mastodon angustidens he has included two very dis-
tinct forms, characterized by a different numerical formula
in the crown ridges, viz. M. angustidens and M. longirostris ;
and the South American teeth which he distributed among
three nominal species, M. angustidens, M. Andiwm, and M.
Humboldtu, appear to be all referable to a single form,
M. Andiuvm of M. de Blainville. We agree also with the
latter authority, in considering the tooth upon which M.
minutus is founded to be nothing more than an anterior
molar of a young M. angustidens.

Cuvier’s opinion was first called in question by Tilesius, in
1815, in his memoir upon the skeleton of the frozen mam-
moth, discovered by Mr. Adams in Siberia. He repudiates
the validity of the grounds for separating mastodon from
elephas, in terms of such strong dissent as to have excited
the indignation of the French philosopher: ‘Cuvierus in
tractatu suo de hac specie (M. giganteus) quam injuste ab
elephantorum genere separavit quamquam non solum dentes
molares, in quibus male genericam diversitam’ queesivit
lamellosee sunt structure, ut omnes reliqui elephantorum
molares sunt, sed etiam totum animal characteribus genericis
elephantorum respondeat ejusdem opinionis est.’ Cuvier’s
division, however, has been adopted by every subsequent
writer except M. de Blainville, who coincides with the view
taken by 'Tilesius.

A still more important oversight was made by the founder
of the genus, in regard to the statement which he advanced
of the entire absence of ‘ crusta petrosa,’ or ‘ cortical’ from
the molars of mastodon. It is true that this substance is
not present in an appreciable quantity in M. Ohioticus, and
that it is also but very sparingly developed in M. angustidens
and M. longirostris; but, in M. Andiwm, atypical form of the
genus, this substance exists in a layer of considerable thick-
ness, which we have observed in almost all the teeth of the
species, contained in the museum of the Jardin des Plantes,
including the specimens brought by Dombey, Humboldt, and
Gay, from Chili and Peru, and also in the rich series of

1 Divers Mast. pl. iii. fig. 2. p. 462.
2 Tom, i. p. 256. 4 Oss. Foss. tom. i. pp. 11 and 225.
3 Tilesius, Mém. de l’Acad. Impériale 5 Ste in orig.

des Sciences de St.-Pétersbourg, tom. v.

ELEPHANT AND MASTODON. 59

specimens from Buenos Ayres, lately acquired by the British
Museum. The great weight of Cuvier’s authority has given
an undue influence to his statement upon this point, which
has biassed the observations of some later writers directed
to the subject.

No other additions were made to the species of Mastodon
from the second edition of the ‘Ossemens Fossiles’ until
1826, when an important discovery was made of fossil bones,
along the banks of the Irrawaddi River, in the Burmese
Empire, by Mr. Crawfurd. These remains have been figured
and described in the ‘ Geological Transactions,’ by Mr. Clift,
who has made a valuable contribution to paleontology by
proving the existence of two Indian fossil species, which,
through the form and number of the coronal divisions of the
molars, establish connecting links between Hlephas and Mas-
todon. These he has named M. Hlephantoides and M. latidens,
and from their examination, he was led, with prescient
sagacity, to anticipate the discovery of other forms which
’ should constitute a complete transitional series between the
two genera.! But Mr. Clift, like Cuvier, overlooked the pre -
sence of a coat of ‘cement,’ which is developed in such
thickness in one of the species (M. Hlephantoides) as to be of
functional importance; and his two nominal species include
teeth which appear to belong to several distinct forms.

The next accession to the species of Mastodon was made
in 1828, by Croizet and Jobert, who described certain fossil
teeth and jaws from Auvergne, under the name of M. Arver-
nensis.2. The specimens were chiefly jaw fragments, derived
from very young animals, and the species was characterized
by these paleontologists, as distinguished from the true
M. angustidens of Cuvier, by the presence of a well-developed
front and back ‘talon’ in each of the molars, and by the
greater complexity in the composition of the crown ridges,
which are irregularly subdivided into aggregations of small
warty cones.

The observations of Croizet and Jobert were correct, so
far as they went, and to them assuredly belongs the merit of
having first recognised the distinctness of this much-disputed
species, which is most frequently met with in authors under
the name applied to it by Kaup of M. longirostris.? But the
most essential distinctive character escaped their notice. In
the year following the publication of their work, Hermann von

1 Clift, Geol. Trans. 2nd Ser. vol. ii. | was subsequently corrected in the pub-
p. 369. lished plates of the Fauna Antiq. Siv.,

2 Croizet et Jobert, ‘ Recherch. sur les | where M. longirostris and M. Arvernensis
Oss. Foss. du Départ. du Puy-de-Déme,’ | are shown to be distinct.’ See also memoir
1828, p. 133. : on Mastodon in vol. 1.

3 MS. Note by Dr. F. ‘This mistake

60 FAUNA ANTIQUA SIVALENSIS.

Meyer described under the same name a fragment of the
upper jaw of a young mastodon, from the celebrated deposit
of Eppelsheim, containing the three first molars in situ.'
After considering the characters indicated by Croizet and
Jobert, he states that M. Arvernensis is distinguished from
M. maximus (M. Ohioticus), M. angustidens, and all the other
species then known, by the circumstance that the third
molar in the order of antero-posterior succession has the
crown divided into fowr ridges, while the same tooth in the
other species presents only three ridges. In a subsequent
memoir, on the fossil remains of Georgensmiind,”? von Meyer
figured and described several mastodontine grinders, which
he referred to the true M. angustidens of Cuvier, and which
confirmed the constancy of the differential character between
that species and M. Arvernensis, in the numerical division
and form of the crown ridges, as pointed out by him in his
previous memoir. This was the first step towards a satis-
factory determination of the species, as distinct from WM.

angustidens ; and a great mass of additional materials, con-—

firming the same inference, was soon afterwards brought to
light by Dr. Kaup; but, in the interim, new observations, of
ereat interest, were made upon the mastodon of North
America, which gave an entirely different character to the
investigation from this date.

No suspicion appears to have been entertained before this
time, that any of the mastodons, more than the existing
elephants, possessed tusks in the lower jaw. Cuvier ex-
pressly affirms their absence,’ although, as has been observed
by Professor Owen,‘ he figured in the original memoir in the
‘ Annales du Muséum,’ and in the first edition of the ‘ Osse-
mens Fossiles,’ a lower jaw of an adult Mastodon, showing
what appears to be the alveolus of a persistent inferior tusk.
Early in 1830, a memoir by Dr. Godman was read to the
American Philosophical Society,? upon a mastodontoid lower
jaw, with two small tusks, which he described as character-
izing a distinct proboscidean genus, under the name of
Tetracaulodon. This jaw belonged to a young animal, and
showed four molars on each side, the anterior two of which
Godman considered to belong to the ‘ milk,’ and the rest to
the permanent series. These teeth resembled, in every
respect, molars of the same size, in other specimens of the

1 Hermann yon Meyer, ‘ Ueber Mas- | gensmiind, 1834, p. 38, tab. 1 and 3.
todon Arvernensis bei Eppelsheim.’ Noy. 3 Oss. Foss. tom. i. p. 238.
Act. Acad. Leop. Carol. Natur. Curios. 4 Owen, ‘ Odontography,’ p. 619.
1829. Vol. xv. p. 118. 5 Godman, ‘ Americ. Phil. Trans.’ New
2 Museum Senckenbergianum. Die | Ser. vol. iii. p. 478. ‘ Tetracaulodon
fossilen Zihne und Knochen von Geor- ! Mastodontoideum,’

——

ELEPHANT AND MASTODON. 61

lower jaw, which exhibited no vestige of a tusk, and which
were usually considered to belong to M. Oluoticus. Mr. W.
Cooper, in consequence, immediately questioned the accuracy
of Godman’s inference, and insisted that the inferior tusks
indicated merely differences dependent on age and probably
sex; that they were possessed by the young animals, but
were shed during the increase of age, the period of their fall
varying with the individual.! Mr. Titian Peale suggested
that these inferior tusks might be a distinctive mark of young
males.2 Dr. Harlan adopted the same view,? and referred
to the corneous appendages in several genera of the Rumi-
nantia as analagous distinctive characters between males and
females. Notwithstanding the force of these objections, Dr.
Isaac Hays, in 1831, not merely maintained the correctness
of Dr. Godman’s opinion regarding the distinctness of Tetra-
caulodon, but attempted to distinguish two additional species
of this nominal genus under the titles of T. Collinsui and T.
Godmani, besides two new North American species of Masto-
don.t| The memoir in which these opinions were advanced
is illustrated by an excellent and copious series of figures,
exhibiting the dentition of M. Ohioticus, from a very early to
the adult stage; and although Dr. Hays has entirely failed
in sustaining the genus Tetracaulodon, or the species which
he proposed, his memoir has served as an important con-
tribution to paleontology, by showing that the number of
molars developed during life, in M. Ohioticus, successively.
from behind, amounts to six. These he has traced from the
first to the last with great care, in the lower jaw, and estab-
lished the position and characters of each by the comparison
of a large number of specimens.

Whilst this discussion was taking place in America, the
discovery was made of a similar structure in a Huropean
species of Mastodon, by Dr. Kaup. This distinguished pale-
ontologist first proved the existence of two deflected and
recurved tusks of large size, in the lower jaw of his colossal
genus Dinotheriwm, the teeth of which had been referred by
Cuvier to a gigantic kind of Tapir. Soon afterwards, at
Hppelsheim, in the same arenaceous deposit which had
yielded the Dinotherian remains, he discovered an adult
lower jaw of a species of Mastodon, which presented a re-
markable semi-cylindrical and beak-shaped elongation of the

1 W. Cooper. ‘Lyceum of Nat. His- | Hays Memoir, Amer. Phil. Trans. New
tory of New York, April 1830; ‘Silli- | Ser. vol. iv. p. 318.
man’s Journ. vol. xix. p. 159; and 8 Harlan, ‘Med. and Phys. Research,’
Featherstonhaugh’s ‘Monthly American | p. 254.
Journ. of Geol.’ vol. i. p. 148. 4 Hays, loc. cit. pp. 817-337.

2 Titian R. Peale, quoted in Dr.

62 FAUNA ANTIQUA SIVALENSIS.

symphysis, forming the sheaths of two inferior tusks, while
the molar teeth exhibited the characters attributed by Cuvier
to a portion of the specimens included by him under the
names of M. angustidens. Kaup, after recognising the struc-
ture, at first adopted Godman’s genus for the reception of
his species, which he named 7’. longirostris;! but, subse-
quently, in his great work upon the Eppelsheim fossils,” he
admitted the force of the objections raised by the American
naturalists against the generic importance of the inferior
tusks in Tetracaulodon, and referred the Eppelsheim fossil to
the genus Mastodon, retaining the same specific name. He
extended the observations made by von Meyer on M. Arver-
nensis, which he considered to be the young of M. longirostris.
He traced the dental succession from the earliest to the adult
stage confirming the observations made by Hays on M.
Ohioticus, by showing that six molars are developed in the
European species during life, in antero-posterior succession.
Kaup also detected the presence of an upper premolar,
situated as a germ above the second deciduous grinder, in a
young specimen of M. longirostris, corroborating the inference
drawn by Cuvier from the Dax specimen of M. angustidens ;
but he considered this tooth as the normal successor of the
first milk molar, the second of the series beg the tooth
which it specially replaces. Dr. Kaup, in the first instance,
took a peculiar view of the affinities and systematic relations
of his most remarkable genus Dinotheriwm; but he has since
come round to the opinion advanced by other observers,
that it was a true pachydermatous form,’ closely allied to
Mastodon.

The discussion respecting Tetracaulodon, which had been
suspended in America, was renewed in England on the
occasion of Koch’s public exhibition of the entire skeleton,
and other remains of the North American Mastodon, in
London, during 1841. The ingenious exhibitor contrived a
fanciful reconstruction of the skeleton inconsistent with the
principles of animal mechanics. The huge tusks, instead of
being placed with their points directed upwards, as in the
elephant, or downwards, as had been formerly suggested by
Mr. Rembrandt Peale,* were spread out horizontally with
diverging curves, so as to resemble two great sickles. Other
corresponding extravagancies were exhibited in the apposi-
tion of the limbs, and for the grotesque form so constructed,
Mr. Koch proposed a distinct generic place, under the desig-

1 Tsis, 1832, p. 628. : 3 Akten der Urwelt, 1841.
2 Ossemens Fossiles de Darmstadt, * Cuy, Oss. Foss. tom. i. p. 239.
1835, part iv. pp. 65-89.

ELEPHANT AND MASTODON. 63
nation of Missowriwm.'! Professor Owen, on this occasion,
reviewed the whole of the evidence respecting Tetracaulodon,
and, in a masterly communication to the Geological Society,
extended the objections urged by the American naturalists, by
numerous and forcible analogies drawn from the dentition of
the Dugong and Narwahl, besides some of the ordinary
Pachydermata.2 He arrived at the conclusion that the
Mastodontoid animals of North America are all strictly refer-
able to a single species, which ‘has two lower tusks origi-
nally in both sexes, and retains the right lower tusk only in
the adult male.’ Dr. Grant entered upon an elaborate in-
vestigation of the same subject soon afterwards, and was led
to very different results? He divides the proboscidean
pachydermata into four genera, Hlephas, Mastodon, Tetracau-
lodon, and Dinotherium, to each of which he attributes a
different dental formula. He admits thirteen species of
Mastodon, and discriminates six species of Tetracaulodon
among the mastodontoid animals of North America. Mr. A.
Nasmyth adopted similar views, from a minute microscopical
examination of the structure of the tusks in these extinct
animals; but the importance of the differential marks indi-
cated by Dr. Grant and Mr. Nasmyth has not been admitted
by subsequent observers, as characterizing more than indi-
vidual and sexual varieties in different animals of the same
species. The generally received opinion at present is that
M. Ohioticus is the only mastodontoid form hitherto met with
in North America.

In the additions to the last edition of the ‘Ossemens Fossiles,’
Cuvier has recorded the discovery of fossil grinders of
mastodon, in the Lignite beds of Keepfnach and Elle in
Switzerland, which he has referred to his ‘ Mastodonte a
dents étroites.”” Some of these remains have been described
by Schinz as indicating a distinct species for which he has
proposed the name of M. Turicensis.6 M.de Blainville refers
them to M. Tapiroides, as defined in the ‘ Ostéographie.’

1 Koch has lately published a sepa-
rate memoir, in which the Missouriwm
is figured and perpetuated with all its
original exaggeration. He has given it
the appropriate specific name of M. the-
ristocaulodon the tusks being invested
with the functions of a sort of seythe.—
A. C. Koch, ‘Die Riesenthiere der Ur-
welt oder das neu entdeckte Missourium
theristocaulodon und die Mastodonten
im Allgemeinen und Besondern.’ Berlin,
1845.

2 Owen, ‘Proceedings of the Geol.
Soe.’ Feb. 1842, vol. i. p. 659; ‘ Report
on the Missourium.

8 Grant, loc, cit. June 1842, p. 770,

‘On the Structure and History of the
Mastodontoid Animals of North Ame-
rica.’

4 Nasmyth, loc. cit. June 1842, p. 775,
‘On the Minute Structure of the Tusks
of Extinct Mastodontoid Animals.’

5 Oss. Foss. 8vo. edit. 1834, tom. ii.
p. 366.

§ Schinz, quoted in von Meyer's ‘ Pa-
leologica,’ 1832, p.72, and in Bronn’s
‘ Jahrbuch’ of 1839, p. 2; mentioned in
Jameson’s Edin. New Phil. Jour. vol. v.
1828, p. 278, as a communication to the
Helvetic Soc. of Nat. Hist. in August,
18387.

64 FAUNA ANTIQUA SIVALENSIS.

Two species of Mastodon have been proposed by Hichwald,
from remains found in Poland, under the names of M. Podo-
licus! and M. intermedius ;? the former of which appears to be
founded on a disguised fragment of the lower jaw of Dino-
theriwm giganteum, and the evidence adduced in support of
the latter is insufficient as yet to establish its specific inde-
pendence. M. de Blainville refers it to M. Tapiroides.’

One of the authors has described some of the remains of a
typical species of Mastodon, from the tertiary deposits of
India, under the name of M. Sivalensis,* a detailed account
of which will be given in this work.

Professor Owen has proposed the provisional name of WM.
Australis,® for a form which rests upon a solitary specimen,
brought from Australia. We shall have occasion to refer to
this specimen in the sequel.

The grounds upon which Cuvier technically rested his
generic distinction between Mastodon and Elephas having
been invalidated by the discovery of the species named M.
Elephantoides by Clift, it became necessary for systematic
authors either to unite them under a single generic name, or
to devise other diagnostic characters for their separation.
Bronn, in his ‘ Lethza,’ gives an elaborate definition of the
two genera, founded upon the observations of his country-
men, Kaup and von Meyer, on the European species, and of
the American naturalists upon M. Olioticus. He charac-
terizes Mastodon® by inferior tusks; by the presence simul-
taneously of a greater number of grinders in each jaw; and
by the expulsion of the anterior tooth in the young animal
by a vertically succeeding premolar. The distinctive charac-
ters of Hlephas he defines to be, the absence of inferior tusks ;
a less number of more complex grinders at one time in the
jaws; and the uniform antero-posterior succession of the
whole series of molars without a vertical premolar. M. de
Blainville, in his great work,’ has given the most full and
detailed account of the species of both genera, that has
appeared since the publication of the ‘Ossemens Fossiles,’
and has endeavoured, by original observation and by the col-
lation of information drawn from every accessible source, to
make his memoir a monograph of the subject, brought up to
the state of our knowledge at the present day. Having
satisfactorily proved that the number of molar teeth, deve-

1 Vide supra, note, p. 46. p. 294.

2 Bichwald, ‘Zoologia Special” 1831, ° Ann. of Nat. Hist. xiv. 1844, p. 296.
vol, iii. p. 361. . ® Lethea Geognostica, 1838, Band ii.
3 Ostéographie, ‘Des Eléphants, p. | pp. 1233-1240. a
259. ” Ostéographie, ‘Des Eléphants, Fas-

4 Journ. Asiat. Soc. of Beng. vol. y. | cic. xvi.

ELEPHANT AND MASTODON. 65

loped in antero-posterior succession in the elephant does
not differ from that of the mastodon, as had been previously
supposed, he insists that the characters presented by the
number as well as by the form of all the parts of the skeleton
are alike in both; that the separate bones are so precisely
similar that, when met with detached, it is exceedingly diffi-
cult to decide whether they belong to mastodon or to ele-
phant; that, as regards the structure of the teeth, there is a
series of intermediate gradations forming a passage from the
one genus into the other; and that the observed differences
in these organs are systematically of no greater signification
than as indicative of the kind of vegetable food upon which
the several species subsisted.' Guided by these views, M.
de Blainville has abandoned Cuvier’s genus of Mastodon,
and, like the earlier observers, he has united it with the
elephant, under the common generic name of Hlephas, of
which he forms two sections, Lamellidontes and Mastodontes.
The former includes the elephants proper, viz. the two existing
species, with H. primigenius doubtfully admitted as a distinct
form, and H. latidens, under which name he unites the two
species M. Hlephantoides and M. latidens, described by Clift.
Of the second section, comprising the typical mastodons, M.
de Blainyille admits only four species, viz. H. (Mast.) Ohioticus,
Ei. angustidens, H. Andiwm, and HE. Tapiroides, together with
M. Swalensis as a doubtfully established species. M. de
Blainville’s work is illustrated with an admirable series of
representations of the osteology and dentition of the different
species, and he has made a valuable contribution to the
palzeontology of the Proboscidea, by defining the character
of M. Andiwm, which was distributed among several nominal
species by Cuvier. He has also thrown considerable light on
M. Tapiroides, by means of the materials collected by M.
Lartet, in the South of France, which he has combined
under this specific name with remains derived from diffe-
rent parts of Hurope. But there are weighty objections to
the rest of the details of this portion of the ‘ Ostéographie.’
Although the consideration of the teeth is of paramount
importance in every question connected with zoological
arrangement, it is to be remarked that M. de Blainville has
nowhere adverted to the occurrence of premolars in the
upper jaw of certain species of mastodon, the presence of
which—first observed by Cuvier in M. angustidens—has
been clearly established by Kaup in the young of M. longiros-
tris. The author of the ‘ Ostéographie’ describes them as

1 Loe. cit. p. 2.
VOL. I. F

66 FAUNA ANTIQUA SIVALENSIS.

equally wanting in the mastodons and in the true elephants.
With respect to the species, while M. de Blainville has judi-
ciously rejected a great many of the nominal forms which
have been proposed on slender grounds, he appears in other
cases to have pushed this numerical reduction too far, and to
have mixed up under the same name species which are essen-
tially distinct. This remark applies especially to the Indian
fossil forms, three or four of which are combined under LH.
latidens; and to the European M. angustidens, which, as
defined and illustrated in the ‘ Ostéographie,’ includes two
separate species.

Professor Owen has been engaged upon the same subject,
contemporaneously with M. de Blainville. In addition to
the memoir upon the North American mastodon previously
referred to, our eminent countryman has discussed the syste-
matic relations of Hlephas and Mastodon, in his ‘ British
Fossil Mammalia,’ and in his very valuable work upon the
teeth, lately published. In the latter work he has shown,
for the first time, that the molar teeth of the elephants and
mastodons, while they agree with each other, form no excep-
tion from the normal division into sets presented by the
ordinary Pachydermata,! and that the apparent anomaly in
the order of their succession arises from the partial or total
suppression of the successional series of premolars. In the
former work, after describing the differences in the form of
the teeth of the two genera, he adds :—

‘A more important difference presents itself when the
teeth of the typical species of mastodon are compared with
those of the elephants, in reference to their structure. The
dentine, or principal substance of the crown of the tooth,
is covered by a very thick coat of dense and brittle enamel ;
a thin coat of cement is continued from the fangs upon the
crown of the tooth, but this third substance does not fill up
the interspaces of the division of the crown, as in the ele-
phant. Such, at least, is the character of the molar teeth of
the first discovered species of mastodon, which Cuvier has
termed Mastodon giganteus and M. angustidens. Fossilremains
of proboscideans have subsequently been discovered, princi-
pally in the tertiary deposits of Asia, in which the number
and depth of the clefts of the crown of the molar teeth, and
the thickness of the intervening cement, are so much in-
creased as to establish transitional characters between the
lamello-tuberculate teeth of elephants and the mammillated
molars of the typical mastodons ; showing that the charac-
ters deducible from the molar teeth are rather the distin-

1 Supra, p. 1.

ELEPHANT AND MASTODON. 67
euishing marks of species than of genera, in the gigantic
proboscidean family of mammalian quadrupeds.

‘Two dental characters, however, exist, though hitherto, I
believe, unnoticed as such,! which distinguish, in a well-
marked and unequivocal manner, the genus Mastodon from
the genus Hlephas. The first is the presence of two tusks
in the lower jaw of both sexes of the Mastodon, one or both
of which are retained in the male, and acquire a sufficiently
conspicuous size, although small in proportion to the upper
tusks, while both are early shed in the female. The second
character is equally decisive; it is the displacement of the
first and second molars in the vertical direction, by a tooth
of a simpler form than the second, a true dent de remplace-
ment, developed above the deciduous teeth in the upper,

and below them in the under jaw.

1 Manuscript Note by Dr. Falconer.—
Professor Owen has preferred a claim
to having first pointed out the infe-
rior tusks and the vertical premolar as
distinctive generic characters between
Mastodon and Elephas. His words are:
‘I first pointed out the inferior tusks,
whether transitory or persistent, as a
well-marked generic character of Masto-
don as contra-distinguished from Hlephas
in my “History of British Fossil Mam-
malia” (p. 275); and also defined a
second character, in the displacement of
the first and second molars in the ver-
tical direction by a tooth of simpler form
than the second—a true “dent de rem-
placement,” developed above the deci-
duous teeth in the upper and below them
in the under jaw. Both these dental
characters, which are of greater import-
ance than many accepted by modern
zoologists as sufficient demarcations of
existing generic groups of Mammalia,
have been recognized in the Mastodon
giganteus of North America and in the
Mastodon angustidens, which is the pre-
vailing species of Europe’ (Odonto-
graphy, part li. p. 615. June 1845).
The passage above referred to in the
‘British Fossil Mammalia’ is thus :—
‘Two dental characters, however, exist,
though hitherto, I believe, unnoticed as
such, which distinguish in a well-marked
and unequivocal manner the genus Mas-
todon from the genus Hlephas, &c., as
in the Odontography (British Fos.
Mam. p. 274, part vi. 2 Dec. 1844). But
both of these characters are expressly
mentioned as generic points of distinc-
tion between Mastodon and Elephas by
Dr. Bronn, of Heidelberg, in his ‘ Le-
theea,’ as far back as 1838, six years

before the publication of part vi. of the
‘British Fossil Mammalia. Bronn’s
elaborate definition of the genus Masto-
don begins with these words :—‘ The
genus Mastodon (of which entire skele-
tons possessing even, as it seems, the
hyoid bone and stomach, have been found
in North America) exhibits in the skull,
in the tusks, in the number of toes, and
in all the characters of the skeleton
which admit of an inference as to the
nature of the soft parts, no other generic
difference from Hlephas, excepting the
sufficiently important ones in the grinders
and lower incisors. The dental formula
isthus: 4-4: 3 - 4: The grinders
appear in a larger number: namely,
there are present in both jaws from 1 to
4 of them contemporaneously, according
to species and age, which, as in the ele-
phant, are replaced two or three times;
not, as usual, from below upwards (ver-
tically), but as in the latter, by teeth
which are pushed from back to front—
excepting, however, the most anterior of

these teeth’ (Bronn, Lethea Geog.
Band. ii. p. 1233). Then he defines
Elephas thus :—‘ Essential character,
1-0

o.¢:% teeth; viz a huge tusk in
front, and compound grinders behind ; a
long muscular trunk; five toes before
and behind’ (p. 1239). In the continua-
tion, giving the general description, he
adds :—‘ The elephant ordinarily has
only two grinders in each jaw at one
time (in advanced age only one, pro-~
bably, as in mastodon), which are re-
placed by others, not, as usual, in a ver-~
tical, but all in a horizontal direction ;
and that not only once, but probably
three or four times,’ &c. (p. 1240). In

F2

68 FAUNA ANTIQUA SIVALENSIS.

‘These two dental characters, which are of greater im-
portance than many accepted by modern zoologists as suffi-
cient demarcations of existing groups of mammalia, have
been recognised in the species called Mastodon giganteus,
most common in North America, and in the Mastodon angus-
tidens, which is the prevailing species of Hurope.’

But the value of these alleged characters, as furnishing
certain distinguishing marks between mastodon and elephant,
is far from being absolute. It will be seen in the sequel that
premolars, instead of being invariably wanting in the ele-
phants, are developed in greater number in one typical fossil
species than they are known to be in any ascertained species
of mastodon (Hlephas planifrons) ; while, on the other hand,
they do not appear to be constant in every species of the
latter group; and, although the interior tusks have been
observed in three species of mastodon, there are other forms
in which they have not yet been detected, even in specimens
of the young animal (Mastodon Sivalensis and M. Elephan-
toides of Clift).

With respect to the European species, Professor Owen
considers, like M. de Blainville, that M. angustidens and M.
longirostris belong to a single form; and he refers the whole
of the elephantine remains which occur so plentifully in
England, whether in the fluvio-marine crag, or in the super-
ficial drift and gravel, also to a single species, H. primigenius.
He describes the dentition of the Indian species, discovered
by Clift, under the designation of ‘ Transitional Mastodons.’

We shall now proceed to the special consideration of the
teeth, as the organs which have the greatest share of influ-
ence in determining the modifications in the construction of
the cranium, and in the development of the general form
presented by the different species in the Proboscidea.

a note (p. 1218), Dr. Bronn states that:
‘Tetracaulodon, according to Kaup, has
“premolars” inthe upper jaw, which are
very similar to the back molars of hip-
popotamus, and are very caducous.’ Fur-
ther, in his remarks upon Tetracaulodon
of Godman, after stating that the species
enumerated by Hays were, according to
W. Cooper, Peale, and Harlan, only
‘individual varieties in young males of
Mastodon giganteus of different ages,’ he
adds, in regard to these inferior tusks :
‘ Mastodon giganteus, M. angustidens,
and WM. longirostris, do unquestionably
possess such inferior tusks; the other
species of mastodon occur more rarely,

and we can, therefore, only by analogy
infer their having possessed them also’
(p. 1283). Bronn appears to haye
founded his diagnosis mainly upon the
observations of his countrymen, Kaup
and Von Meyer, regarding M. longiros-
tris, Kaup (M. Arvernensis, Von M.).
Nations, as well as individuals, are
jealous on points of discovery. We are
confident that Bronn’s remarks must
have escaped the notice of our distin-
guished countryman when he wrote the
passage in the ‘British Fossil Mam-
malia ;’ and that he will concede priority
of observation, on both points, to the
German Palzontologists.—[Ep. ]

ELEPHANT AND MASTODON. 69

§ 2. On THE SrructurE anD Form or THE Moar Teeru.

Plates I., IT., and IIT. (¥.A.S.)! are intended to represent, by
careful copies of nature, the modifications in structure and form
exhibited by the molar teeth of the Proboscidea; they show,
in vertical sections, a series of gradations, commencing with
Dinotherium and Mastodon Olioticus at one extremity, and
running through the other species to Hlephas primigenius, in
which the greatest deviation from the ordinary form of a
erinding tooth is met with.

Each molar in the Proboscidea,? as is the case with all other
animals, is developed within a closed membranous sac, called
the capsule, which is lodged in a cavity of the maxillary bone,
and which forms the mould, so to speak, of the tooth. The
exterior of this sac is simple, while its internal surface is
expanded into numerous folds which determine the arrange-
ment of the ‘ivory,’ ‘enamel,’ and ‘ cement,’ entering into
the composition of the tooth. From the bottom of the sac a
gelatinous mass, the ‘pulp nucleus,’ is projected upwards,
subdivided into transverse digitated plates or segments, vary-
ing in number, length, and thickness in different species, and
more or less numerous in different molars of the same indivi-
dual, according to the age of the tooth. These ‘ pulp’
seoments are attached only by their base, and attenuate
eradually upwards to their summits, being entirely free from
adhesion, either to the opposite side of the sac or to the
contiguous ‘pulp’ divisions. The ossification of this ‘ pulp
nucleus,’ by the deposition of calcareous matter within the
cells of its tissue, constitutes the ivory core, or central part of
the tooth, being the substance called ‘dentine’ by Professor
Owen.

From the opposite or coronal side of the capsule, other
folds or induplications are given off, which proceed into the
spaces between the divisions of the ‘ pulp nucleus.’ Their
attachment is continued along the parietes of the sac, so that
on every side, except the base, they envelope the processes of
the ‘pulp nucleus,’ over which they are closely applied, inter-
locking with the latter; the two sets of processes thus form-
ing productions from the interior of the sac which are opposed
to each other in the manner of salient and re-entering angles.
The ossification of these peripheral folds in a continuous sur-

' Here and in the remaining portion ? The substance of the four following
of this memoir, the letters F.A.S. imply | paragraphs is drawn from the admirable
that the plates referred to are to be | descriptions given by Cuvier in the
found in the ‘ Fauna Antiqua Sivalensis.’ | Ossemens Fossiles, tom. i. p- 32, and
When these letters are not appended, | by Owen in his valuable systematic work
the plates are those of the present edition | on the teeth, Odontography, p. 649.
of Dr. Falconer’s works.—[Eb.]

70 FAUNA ANTIQUA SIVALENSIS.

face constitutes the ‘enamel,’ or vitreous shell, which encloses
the ivory core of the crown of the teeth, terminating abruptly
where the fangs commence.

Exterior to, and following the folds of this ‘enamel pulp,’
there is another modification of the internal membrane of the
sac, the ossification of which, during the last stage of the
development of the tooth, constitutes the external ‘cortical
layer,’ or ‘ cement,’ which, in a crust of greater or less thick-
ness, is continued over the enamel of the crown and down
upon the fangs. In the true elephants this cement substance
completely fills up the hollows between the plates of enamel.

The production of the hard tooth takes place by a process
of calcification, which commences in the summits of the
‘ivory pulp’ segments, the solidification extending gradually
downwards along the digital processes, which unite into a
transverse plate ; and these plates, at their base, are combined
so as to form the common body of ivory which occupies the
central mass of the tooth. Simultaneously with this produc-
tion of the ivory, a similar process of solidification goes on in
the corresponding and continuous portions of the ‘ enamel
pulp,’ forming a shell of enamel which is closely applied to,
and moulded on the form of, the ivory segments and their
digital subdivisions. When the calcification has reached the
common base of ivory, the enamel plates covering the conti-
guous segments of ivory unite along their lines of junction in
the bottom of the clefts between the ridges.

The basal mass of the ‘ pulp nucleus’ is not connected in
a continuous surface with the bottom of the sac, but, as it
were, by pedicles which, after the solidification of the body of
the tooth, elongate and become contracted, with more or less
of subdivision. These pedicles undergo the same process of
calcification, and form the fangs by which the tooth is
implanted in the jaws. The fangs bear a relation to the
divisions and vertical height of the crown, being few, thick,
and more or less distinct in the mastodons, while they are
numerous, slender, and confluent in the elephants.

The three constituent dental substances are structurally
distinguished by very different characters, and their combined
modifications in the molar teeth furnish the best differential
marks for the arrangement of the groups of Mastodon and
Elephas, and for the discrimination of the different species.

The molars of the North American mastodon and of the
existing Indian elephant may be selected as convenient
illustrations of the opposite extremes of form, presented by
these teeth in the Hlephantide. Taking the last tooth of the
upper jaw as the example; in the former, the crown is nearly
rectangular in outline, somewhat higher in front than behind,

ELEPHANT AND MASTODON. 71

the dimensions being about seven inches in length, by four of
width at the anterior end. The height of the crown, from the
base of the enamel to the summit of the first ridge (vide
Section, Pl. VII. fig. 2, or Pl. III. fig. 9, F.A.S.), is about
two and a half inches, decreasing a little successively to
the fourth or last principal ridge. The grinding surface is
divided in two directions, viz. longitudinally along the axis
by a narrow cleft, which, as in the hippopotamus and most
other pachydermata, bisects it into nearly equal segments ;
and transversely by four deep open hollows, alternating
with as many trenchant ridges. Each of the lateral divisions
of these ridges is composed of a pair of compressed con-
fluent points; in the upper molars, the outer division is
more elevated and slopes gradually towards the inner side,
which, although lower, is the most complex in composition,
giving off the ‘talons’ and accessory tubercles, which are
more or less developed in different species, the inner
point being commonly the largest. This relation continues
during the wear of the teeth, the inner side being more
worn by the process of trituration than the outer. The
teeth of the lower jaw agree with the upper in the form and
subdivision of the crown; but the disposition of the lateral
segments is reversed, the inner being higher, and continuing
so during the wear, while the outer are lower, but more com-
plex in composition. The higher and lower portions of the
crowns of the opposed teeth are thus brought to act against
each other, and serve more effectually the triturating function
of the teeth. The plane of the grinding surface is nearly
level from back to front, both in the upper and inferior
erinders. In consequence of the peculiar mode of protrusion
of the teeth from behind forwards, the crown ridges come
successively into use, commencing with the first, and each of
the lateral divisions is ground down to a rhomboidal disc,
surrounded by a band of thick enamel: as the wear of the
tooth advances, the separate discs unite, forming a wide
transverse depression corresponding to the configuration of
the ivory nucleus, and the shell of enamel which invests it.
The ‘cement,’ or third dental substance, is very sparingly
developed upon the crowns of the molars of the North Ame-
rican mastodon, being only distinguishable in a thin layer
under the microscope. It is more abundant upon the fangs.
The anterior ridge is supported by two stout united fangs,
and the three posterior ridges by fangs agreeing with them in
number, but more or less confluent intoanirregular hollow cone.

In the existing Indian elephant, the last grinder of the
upper jaw is of a subtriangular rhomboidal form in the vertical
section, widely different from that of the North American

72 FAUNA ANTIQUA SIVALENSIS.

mastodon. The crown is veryhigh in front,and declines rapidly
behind. In a large specimen of this tooth from Assam, the
dimensions are upwards of fourteen inches of length by eight
inches of height in front, which is reduced to about one-third
at the Jast ridge, while the width does not exceed three inches
at the anterior end, from which it narrows gradually behind.
The coronal surface is convex across, and also in the antero-
posterior direction. There is no indication of the longitu-
dinal cleft, which, in the North American and other mas-
todons, bisects the crowns of the molars into lateral seg-
ments. The ridges, which in the first-mentioned species do
not exceed four, are multiplied in the last upper grinder of the
Indian elephant to twenty-three or twenty-four thin plates,
which terminate upwards in slender, cylindrical digitations,
hence called Cheirolites by the early paleontologists, when
found separate. The cement substance enters largely into
the composition of the tooth, being interstratified with the
enamel plates in a layer which also envelopes the entire body
of the tooth. The fangs are slender and numerous, bearing
a relation to the lamellz, but they are confluent into large
hollow groups, which are of inconsiderable length, as the
tooth is held firm in the jaw by a large portion of the crown
being imbedded in the alveolus. Instead of being protruded |
in a nearly horizontal direction, as in the North American
mastodon, the teeth move forwards in the are of a circle; the
anterior plates in the upper grinders are inclined forwards,
and by the process of wear they are ground down, so that the
front part of the tooth is truncated obliquely, as shown in
(PL V. fig. 2 (Pl. I. fig.2, and Pl. VII. fig. 4, F.A.S.), long before
the posterior lamellz come into use. The plane of detrition
makes a large angle with the unworn plane of the crown,
and in the upper grinders it slopes from the inside outwards,
being the reverse of what takes place in the mastodons. In the
lower jaw the crown of the last molar is concave from behind
forwards, and convex across; the grinding plates, especially
towards the posterior end, recline backwards, and the plane
of wear, which is concave, slopes from the outside inwards,
bearing a reversed relation to that of the upper jaw. The side
of the jaw to which the teeth belong is readily distinguished
by these characters, and by the circumstance that the upper
erinders are convex on the outer, and concave on the inner
side, the reverse taking place in the grinders of the lower
jaw. The last inferior molar attains a length of fifteen
inches, and presents occasionally as many as twenty-six or
twenty-seven constituent plates in the largest sized indi-
viduals of the Indian elephant.

When the teeth come into use, the digitated summits of

o

ELEPHANT AND MASTODON. 73

the anterior ridges are first ground down into circular rings
of enamel enclosing a pit of ivory; these rings then unite
into oval groups (Pl. VIL. fig. 4a, F.A.S.) ; and as the wear
descends below their point of separation, the smaller discs
disappear in a common transverse band which is bounded by
a projecting edge, or macheris of enamel. These edges, which
represent a transverse section of the enamel plates, either run
across in straight and parallel lines, or they are minutely
erimped and undulated, or dilated into round loops, or angu-
lar expansions in the middle of the ridge: such modifications
holding with great constancy in the different species, and
yielding the characters by which they are most readily dis-
tinguished. The three constituent dental substances being
of unequal hardness, are worn unequally by the process of
trituration ; the hard enamel projects above the ivory, and the
softer cement wears more quickly than either. The grinding
plane of the tooth thus presents, throughout its period of
duration, a surface highly organized by natural inequalities,
to serve a constant triturating purpose. As the anterior
lamellz are worn down, the corresponding fangs are gradu-
ally absorbed.

Between these extremes, furnished by the North American
mastodon and the existing Indian elephant, there is a series
of intermediate forms, which establishes an almost unbroken
passage from the one into the other.

Fig. 1, Pl. IV. (fig. 6 a, Pl. IL., F.A.S.!) represents a vertical
and longitudinal section of the last upper molar of an Indian
fossil species, which we have named Hlephas insignis in this
work. It is selected as furnishing the best illustration of the
intermediate type of a proboscidean molar tooth, from which
those of the other species diverge in opposite directions. It
is in the most favourable state of age and use for showing
the characters, the four anterior ridges being affected by
wear, and the six posterior ridges entire, while the fangs are
fully developed, their mode of implantation in the jaw being
distinctly shown. The tooth is convex from back to front,
in the outline of the crown. The white mass in the centre
represents the body of ivory, which is projected upwards in
ten angular lobes terminating in a sharp edge. The height
of these lobes does not much exceed the width of their base,
and closely applied over them is seen a thick layer of enamel,
reflected up and down in a continuous zig-zag plate. The
interspaces of the five posterior ridges of enamel are com-
pletely filled up by a mass of cement, or ‘ cortical,’ much
exceeding the enamel in thickness, and in quantity in nearly
as great an amount of development as the ivory core of the

1 See note, p. 79.—[Ep. ]

74 FAUNA ANTIQUA SIVALENSIS.

ridge. This tooth belongs to one of the forms which have
been included under the name of Mast. Hlephantoides, by Mr.
Clift, and which Professor Owen names ‘ Transitional Mas-
todons.’ It is important to observe the characters presented
by the cement in this case, as its supposed absence or presence
in the molar teeth was the principal character upon which
Cuvier rested his generic distinction between Mastodon and
Elephant. Professor Owen, in his ‘ Odontography,’ states, in
regard to the teeth of this species, that ‘the interspaces are not
filled with cement, as in the true elephant: only a thin layer
of that substance is continued upon the unworn enamel, as
in the true mastodons.’! But this statement must be received
with some modification. [Fig. 7, Pl. VI., F.A.8., represents a
portion of the same section drawn to the natural size, and
comprising the sixth and seventh ridges. The cement
exhibits an appearance of lamination, of which from nine to
eleven layers may be counted, and is developed in as great a
quantity as the intervals between the ridges could admit of. |
The four anterior ridges have been well used, so that the two
first are worn down to a common disc, from which the enamel
has disappeared ; and the cement of the four anterior hollows,
being the softest of the tooth substances, has been completely,
or partially, worn away by the same process of detrition.

The dark granulated shade below the portion of the ivory

nucleus which sustains the five posterior ridges indicates, in
the figure (Pl. IV. fig. 1), the hollow of their common poste-
rior fang, which is occupied in the fossil by a core of sand-
stone ; the same matrix also fills the cells of the maxillary
bone. The anterior simple fang is shown in the section, of
much smaller size. This tooth measures 10°3 inches in length.

Fig. 6b, F.A.8., represents a similar section of the anterior
portion of an adult tooth of the lower jaw of the same species.
In this instance the two front ridges only have been touched
by wear. Instead of being convex, the common curve of the
crown is slightly concave. The ivory, enamel, and cement
present the same characters as in the upper molar, except
that the cement in the interspaces is less considerable in
quantity, although sufficiently abundant to indicate that it
was functionally serviceable in the grinding operation of the
tooth. The posterior part of the basal portion of the pulp
nucleus had not yet completed the stage of calcification, its
place being occupied in the fossil by a nest of calcareous
crystals, bounded in the figure by the undulated line. The
figure shows also a portion comprising the two last ridges of
the preceding molar, with their common fang implanted in
the lower jaw.

1 Odontography, p. 624.

Ee a ee

DESCRIPTION OF PLATE IV.

ELEPHAS INSIGNIS, HLEPHAS PLANIFRONS, AND HLEPHAS
AFRICANUS.

Vertical longitudinal sections of teeth of Hlephas insignis,
E. planifrons, and E. Africanus. Copied from drawings by
Mr. Ford in Plate IL. of the Fauna Antiqua Sivalensis; one-
third of the natural size.

Fig. 1. Is a section of the last upper molar of Elephas insignis, copied
from Plate IL, fig. 6 a, of the Fauna Antiq. Siv. The specimen
is in the British Museum. (See page 73.)

Fig. 2. Is a section of the penultimate upper molar of EHlephas plani-
frons, copied from Plate II, fig. 5 a, of the F. A. S. The
specimen is in the British Museum. (See page 75.)

Fig. 3. Isa section of the penultimate upper molar of Hlephas Afri-
canus, copied from Plate IL. fig. 4 a, of the F. A. 8. (See

page 75.)

VOL. I.
*

Vol.I. Plate 4.

“W.West imp.

el

tok

3. HE. Africanus.

2. E. plantfr ons

1. Elephas insignis.

ELEPHANT AND MASTODON. 75

The next serial modification in the disposition of the
three dental substances, and in the consequent form of the
teeth, is exhibited (in fig. 5a of the same plate, F.A.S8.) in
Plate IV. fig. 2, which represents a section of the penultimate
upper molar of another Indian fossil species, which we have
named H. planifrons. This tooth shows nine ridges, the
three anterior of which alone have been in use, the two first
being worn down to a single disc of ivory. The common
nucleus of this substance is of less thickness than in the
corresponding tooth of H. msignis, and the divisions which
are continued upwards from it into the centre of the ridges ~
are more elongated, with a narrower base, forming irregular-
shaped wedges. The layer of enamel is diminished in thick-
ness and is less uniform in outline, and the surface in contact
with the cement shows a feathered or ragged edge, indicating
superficial inequalities for the firmer cohesion of this latter
substance. ‘The enamel is reflected over the ridges of ivory,
and down into the hollows zig-zag wise, exactly as in fig. 1,
the principal difference being that the ridges are narrower,
with a greater vertical height. The cement substance attains
its maximum of development in this species, completely
filling up the wide interspaces of the ridges, over which it is
continued in a thick mass. This tooth measures 8°7 inches
in length.

Fig. 5b, F.A.S., represents a corresponding section of a
portion of the last molar of the lower jaw of the same species,
comprising nine ridges. This tooth had been longer in use
than that of the upper jaw, and all the ridges are more or
less worn, except the two last. It presents the same general
characters exhibited by fig. 5 a (PI. IV. fig. 2), in the elongated
cuneiform ivory ridges, unequal enamel, and abundant cement,
the differences being merely such as constantly hold between
molars of the upper and lower jaws and of different ages, in
the same species.

The existing African elephant furnishes another link in the
chain of modifications presented by the molars in this family.
Plate IV. fig. 3 (Pl. II. fig. 4a, F.A.S.") shows a section of the
penultimate grinder of the upper jaw of this species, which
is composed of nine principal divisions and a subordinate
‘talon’ ridge, the four anterior of which are partly worn, the
rest being entire. The elongation of the ivory segments,
which commences in H. planifrons, is carried here to a much
greater extent. The segments are produced into long narrow
wedge-shaped plates, the height of which is many times

1 We are indebted to the kindness of | 46 (Pl. II. F.A.S.) of the African ele-

Mr. Charles Stokes for the specimens | phant, the teeth of this species being com-
which have yielded the sections 4a and | paratively rare in English collections,

76 FAUNA ANTIQUA SIVALENSIS.

ereater than the width of their base. The interspaces of the
plates are proportionally deep and filled up with a copious
mass of cement, which completely envelopes the tooth. The
quantity of this substance is measured by the proportion
which it bears to the other dental materials, and it is seen to
be thicker than the ivory plates. The layer of enamel is
reflected over the ridges and down upon the hollows, as in
EH. msignis, but it is much thinner, and the attenuation is
proportioned to the elongation of the plates. The common
basal mass of ivory is greatly reduced in quantity, if com-
pared either with the sections of EH. insignis, fig. 1, or of
H. planifrons, fig. 2, there being little more of this substance
than is sufficient to establish a common connection between the
bases of the segments, and a foundation for the offset of the
fangs, which are numerous. ‘The vertical height of the tooth
is considerably greater than that of either of the two other
described species. The tooth measures 8-7 inches in length.

Fig. 4b, Pl. II., F.A.S., represents a penultimate molar of
the lower jaw of the same species in vertical section. Like
that of the upper jaw, it is composed of nine cuneiform plates.
This tooth had been a long time in use, all the plates except
the last being affected by wear. The anterior part of the
crown has been ground down to nearly one-third of its original
height, so that the enamel divisions between the two anterior
ivory plates have disappeared, and the latter are confluent
into a common mass. It is not, therefore, in the condition
best adapted to show the characters presented by a good
section; but it indicates sufficiently the correspondence of
the lower with the upper molars, in the disposition, form,
and relative proportion of the ivory, enamel, and cement
substances. It bears a very close resemblance to the section
of the lower molar of EH. planifrons, fig. 5b, keeping in mind
that the latter is taken from an older and larger tooth.
They exhibit the same kind of wedge-shaped ivory plates, a
similar amount of cement in the interspaces, and an analogous
thickness of enamel. The resemblance between the lower, in
these instances, is greater than between the corresponding
upper molars of the two species. This specimen measures
7°2 inches in length.

If the eye is carried along these sections in succession, it
will be readily perceived that they constitute a series of
eradations in form conducted from H. insignis to H. Africanus,
through EH. planfrons. The modifications are effected by the
elongation and thinning of the ridge-plates, with a corres-
ponding increase in the depth of the hollows, and in the
vertical height of the teeth; by a diminution of the basal
mass of ivory ; by a greater number of divisions in the same

ELEPHANT AND MASTODON. 717

extent of tooth surface, and a gradual diminution in the
thickness of the layer of enamel. Plate V. represents another
succession of links which, in like manner, conduct us from
the African elephant on to the extremity of the series in
H. primigenius.

Fig. 1 of Pl. V. (or fig. 3a of Pl. I., F.A.8.) shows a section
of the penultimate upper molar of an undescribed Indian
fossil species, named H. Hysudricus in this work. The tooth
is in the middle state of wear, eleven of the thirteen plates of
which it is composed having been in use, and the two anterior
ridges being worn out. The same vertical disposition of
ivory, enamel, and cement is presented as in the African
elephant, but the plates are thinner and a greater number of
them is included in the same length, nine or ten plates in
the latter being developed in the space occupied by thirteen
or fourteen plates in the equivalent teeth of H. Hysudricus.
The plates are also more vertical, the interspaces occupied by
the cement are wider in general than the ivory plates which
represent very attenuated wedges. The layer of enamel is
proportionally thicker than in the African elephant, approach-
ing, in this respect, the teeth of EH. planifrons, Pl. IV. fig. 2.
The vertical height of the tooth is comparatively less in this
specimen than in the African species, the difference being
compensated by a greater development of the basal mass of
ivory. This specimen measures 7°7 inches in length.

A portion of the last molar of the lower jaw of this
species is shown in vertical section in fig. 3b (Pl. L, F.A.8.),
comprising about fifteen plates. The entire tooth, which is
seen in figures 12 and 12a of Pl. VIL, F.A.S., in situ in
the jaw, is more elongated, and includes a greater number
of divisions than is usual in the last inferior grinder of
Hi. Hysudricus. 'The same general character, in the disposi-
tion and relative proportion of the ivory, enamel, and
cement, is exhibited as in the upper molar, Pl. V. fig. 1,
bearing in mind that the latter is a younger and conse-
quently smaller tooth. The layer of enamel, however, is
thinner than in the upper molar, owing to the unusually
large number of developed plates. The ivory segments
are curved backwards near their base, and the apices
of the posterior plates lean towards the front of the tooth,
a disposition which is still more strongly exhibited
in the lower teeth of the existing Indian elephant. The
granulated dark shade, below the undulated outline of the
ivory, indicates a core of sandstone, which occupies the place
of the unossified part of the pulp nucleus and of the unde-
veloped fangs. Both specimens, 3a and 3b, PL I. F.A.S.,
are implanted in portions of the jaws.

78 FAUNA ANTIQUA SIVALENSIS.

The existing Asiatic elephant, H. Indicus, furnishes the
next modification represented in this plate. Fig. 2, Pl. V. (or
fic. 2a, Pl. L, F.A.8.), shows a section of the penultimate
upper molar of this species. The gradual attenuation of the
plates, successively exhibited from H. imsignis to H. Hysudricus,
is here carried to excess, eighteen of these divisions being com-
prised within the space occupied by about nine in the
equivalent tooth of the African species. They are produced
vertically in the same proportion, the height of the middie
plate being about three-fourths of the entire length of the
tooth ; they, in fact, represent parallel perpendicular lamelle,
of nearly uniform thickness from the base to the apex,
interstratified with layers of cement of nearly the same
thickness. The layer of enamel is attenuated into a thin
transversely undulated brittle plate, the surface of which is
deeply wrinkled with striz, for the firm cohesion of the
cement. The general character of the section is a pectinated
arrangement of the lobes like the teeth of a comb,. which
contrasts strongly with the chevron-formed ridges of H.
imsignis, and the cuneiform plates of HL. planifrons. The mass
of ivory at the base of the tooth is much thinner than in the
corresponding molar of EH. Hysudricus, bearing but a very
slender proportion to the height of the tooth ; and numerous
small and distinct fangs are given off from its inferior edge.
This tooth had been some time in use, the anterior part of
the crown being worn off as far as the ninth plate. The plane
of the truncated portion is very oblique, being inclined
nearly at a right angle to the coronal surface of the unworn
portion. This specimen is 8:2 inches in length.

Fig. 26, Pl. L, F.A.S., represents the section of a very fine
specimen of the last inferior molar of the existing Indian
elephant of Assam, from the collection at the India House.
It is an unusually large specimen, showing as many as
twenty-seven plates, the anterior twelve of which have been
in use. Precisely the same disposition of the dental sub-
stances is observed in this case as in the upper grinder, and
they are developed in the same relative proportions. The
vertical height of the plates is still greater than in the cor-
responding lower molar of EH. Hysudricus. The upper surface
is concave, and the under very convex. ‘The anterior plates
are nearly vertical, while the posterior gradually slope back-
wards till they become almost horizontal in the hindmost
portion, with a corresponding gradual diminution in their
height. This is a mechanical arrangement arising from the
contracted diameter of the posterior part of the dental canal,
in which the back part of the tooth is developed, close under
the condyle, the plates being disposed so as to occupy the

DESCRIPTION OF PLATE V.

Exepuas Hysupricus, ELEpHAs INDICUS, «ND ELEPHAS
PRIMIGENIUS.

Vertical longitudinal sections of teeth of Hlep.1s Hysu-
dricus, E. Indicus, and E. primigenius. Copied from drawings
by Mr. Ford in Plate I. of the Fauna Antiqua Sivalensis ;
one-third of the natural size.

Fig. 1. Isa section of the penultimate upper molar of Elephas Hysu-
dricus, copied from Plate L., fig. 3 a, of the Fauna Antiq. Siv.
The specimen is in the British Museum. (See page 77.)

Fig. 2. Is a section of the penultimate upper molar of the existing Indian
Elephant, Elephas Indicus, copied from Plate L., fig. 2 a, of the
F. A. S. (See page 78.)

Fig. 3. Is a section of the last upper molar of the Mammoth, Hlephas
primigenius, copied from Plate L., fig. 1, of the F. A. S. The
specimen is in the British Museum. (See page 79.)

VOL. I.

\esuores

a
——

pce

W. West imp.

ith

1. Elephas Hysudricus. 2.H.Indicus.

mice

G.H Ford del. J.

3.5 -primigenrus.

. \
.
)/
| ¢
/ bz
« . re,
» a . a r
i M =
1 . - + .
* ..
*
~ 3
/
‘ ” ‘
’ * v - me
- faa
: ' /

ELEPHANT AND MASTODON. 79

least vertical space. The basal mass of ivory between the
plates and the fangs is reduced to a small quantity. This
tooth measures 154 inches long in a straight line.!

Fig. 3, Pl. V. (or fig. 1, Pl. 1, F.A.S.), represents a section
of the last upper molar of HE. primigenius, from an English
specimen in the Museum of the Geological Society, found
near Kingsland. We arrive here at the last link in the
chain of modifications, exhibited by the molar of the typical
elephants. The section bears a close resemblance to that of
the corresponding tooth of the Indian elephant, but the
ivory segments are more vertical, thinner, and more approxi-
mated, there being about twenty-two plates in the space
occupied by eighteen or nineteen in the latter; and the layer
of enamel is still more reduced in thickness. The disposition
of the plates presents the extreme degree of ‘ pectination’
seen in the molars of any known species of elephant. The
differences observable in the vertical section are, however, so
inconsiderable, that, if regarded in this light merely, the
mammoth and the Indian elephant might pass for the same
species. But when the grinding surface of the crowns of
their molars is examined, the transverse plates of enamel in
the Indian elephant are seen to be thicker and very closely
undulated, with the flexures deeply wrinkled for the attach-
ment of the stratum of cement; while in the mammoth the
crowns of the teeth are broader, the enamel plates are thinner
and less undulated, so much so that they are frequently
described as being transverse and straight. Such at least
is the character of the typical form of grinder in EH. primi-
genius. This tooth measures ten inches in length, being
considerably under the size attained by the largest specimens
of the mammoth. It has not been deemed necessary to give
a figure of the section of an inferior molar, which differs in
no respect from the upper, more than in the ‘case of the
existing Indian species.

These are the principal modifications in the construction
of the teeth of the elephants. Although, at first sight, the
molars of H. imsignis and H. primigenius appear to “be very
different, the other intermediate forms constitute a series
which establishes a passage between them. ‘The species
have been traced in a retrograde order, from the simpler to
the more complex forms, with the object of making the
descriptions more intelligible. It is interesting here to
observe how the existing species are intercalated: the serial

1 The artist has drawn this figure re- | fig. 6aof Pl. I1., F.A.S. In Pl. L, F.A.S.,
yersed, as compared with the other sec- | the Indian elephant has been named in
tions, the worn end of the tooth being | the reference Z. Asiaticus (Blum.) instead
to the right. The same remark applies to | of #. Indicus.

80 FAUNA ANTIQUA SIVALENSIS.

order of structural development in the teeth does not corre-
spond with the order of succession of the species in time.
The extinct mammoth exhibits the greatest amount of com-
plexity, constituting the terminal link of the chain; and next
follows the existing Indian elephant, interposed between two
fossil species. In like manner, the existing African elephant
is placed between two extinct species, H. Hysudricus and
E. planifrons. The figured sections include only the prin-
cipal forms requisite to establish the passage. Another
extinct Indian species, H. Namadicus (to be described in the
sequel), which is closely allied to the existing Indian form,
comes between it and H. Hysudricus, together with a European
fossil species, which we believe to be distinct from the mam-
moth; and the gap between the existing African elephant
and H. planifrons is filled up by another well-marked
European fossil species, H. priscus (?), Pl. XIV. fig. 7, F.A.S.,
which is closely allied to the former. This species will also
be noticed in a subsequent page.

We shall now revert to EH. insignis, and endeavour to trace
the forms which diverge from it in an opposite direction
through the mastodons, the tendency in this series being to-
wards a greater simplicity in the construction of the grinders,
and a reduction in the number of coronal segments.

Fig. 1, Pl. VI. (or fig. 7 a, Pl. IIL, F.A.S.), represents a sec-
tion of the last upper molar of an undescribed Indian fossil
species, named H. Ganesa, in this work. 'The crown consists
of ten principal ridges, with a subordinate ‘talon’ ridge in
front and behind. The anterior seven ridges have their
summits worn, the two in front being ground down to the
common base of ivory, the tooth having been a considerable
time in use. A small portion is broken off at the anterior
end. The disposition and relative proportions of the ivory,
enamel, and cement, bear the closest resemblance to those of
the corresponding tooth of H. imsignis (Plate IV. fig. 1), and
the number of ridges agrees. The section presents the same
chevron-formed character in the ridges, but the interspaces
are narrower, the cement is in less quantity, and the layer of
enamel is thicker. The common grinding surface of the
crown is also less convex. But these differences are so
inconsiderable, when taken into account with the range of
variation through which the molars run, that they are practi-
cally insufficient for the discrimination of the two species.
To guard against error, the sections have been taken in both
instances from specimens consisting of the palate with a
double line of teeth; and, notwithstanding that the molars
agree so closely, the crania are remarkably different in the
two forms ; that of H. insignis, Pl. XV., F.A.S., being singularly

ELEPHANT AND MASTODON. 81

modified so as to bear an analogy to the cranium of Dinotherium
giganteum; while the head of EH. Ganesa does not differ much
from the ordinary type of the elephant. In fact, we have
entirely failed in the detection of any good characters by
which the teeth of these two species can be distinguished
satisfactorily when met with in fragments, as is most gene-
rally the case. A similar agreement in the form of the teeth
is observable in certain closely-allied species of mastodon.
The tooth represented in this section measures nine inches
and a quarter in length.

Plate III. fig. 7b, F.A.S., shows a section of one of the
posterior molars of a lower jaw, which we infer to belong
to the same species. A small portion of the anterior end of
the crown has been broken off; but the presence of the
anterior fane proves that the section includes the whole
leneth of the tooth except the first ridge, the posterior end
being entire. It appears to have consisted of eight principal
ridges, with a ‘talon’ ridge behind, and a subordinate ridge
in front. Five of the ridges have been in use, the anterior
two in the section being worn down close to the common base
of ivory, while the three last ridges are entire. Like the
upper molar of H. Ganesa, it bears a close resemblance to the
corresponding inferior tooth of H. insignis (Pl. IV. fig. 1), in
the form of the ridges, thickness of enamel, and proportion
of cement. This specimen measures about seven inches and
a half in length.

The next serial modification in the form of the molars
occurs in another extinct Sewalik species, named H. bombi-
frons in this work. It is not included among the sections
in Pl. IIL., F.A.S8. This species, of the distinctness of which
we are assured by possessing several crania containing per-
fect teeth, belongs to the-same group as the two species last
described. The crown is divided into similar transverse
ridges, composed of numerous mammille, which yield a
corresponding chevron-shaped section, and the interspaces
are occupied by a thick coat of cement; but they differ, in
being broader and less elevated, with more open hollows.
The principal ridges of the last molar do not exceed eight
in the upper jaw and nine in the lower; while in L. insignis
they amount to ten in the former, and reach as many as
thirteen in the latter. The last tooth of the upper jaw
measures eleven inches in length, by four and a half in width.
This species will be described in detail in a subsequent
chapter.

The same group comprises a fourth extinct Indian species,
named in this work H. Cliftii, which furnishes the next link in
the chain of forms presented by the molars of the Hlephantide.

VOL. I. G

82 FAUNA ANTIQUA SIVALENSIS.

It is not figured among the sections in Pl. ITT., F.A.S. In our
view, the tooth represented in Pl. XX XIX. fig. 6, of Mr. Clift’s
memoir in the ‘ Geological Transactions,’ under the name of

Mastodon Elephantoides, and the palate specimen represented in —

Pl. XX XVI.of the same memoir, under the name of M. latidens,
belong to this species.'_ The reasons for this opinion will be
given along with the detailed description of the species.
The penultimate and antepenultimate molars in the upper
jaw have only six transverse ridges, continuous and chevron
shaped, with numerous mammille, as in H. insignis and
FE. Ganesa; but the cement does not fill up the interspaces of
the ridges, being reduced to a comparatively inconsiderable

quantity in the bottom of the hollows. H. Cliftvi, in the ©

reduced number of the coronal ridges, and in the other cha-
racters of the teeth, appears to constitute the dental link
which forms the immediate passage from Hlephas into Mas-
todon. Mr. Clift, in reference to his M. EHlephantoides and
M. latidens, has justly remarked that, ‘On an examination of
the structure of the teeth, this discovery’ (viz. of these two
species) ‘will be found to have still higher claims to attention;
for it illustrates the gradual shades of difference by which
nature passes almost imperceptibly from one form to another,
and helps to fill up the interval which has hitherto separated
the mastodon from the elephant.’?

The three species last described, along with H. imsignis,

constitute a peculiar section of Hlephas, of nearly equivalent —

value to the section which includes H. primigenius, H. Indicus,
and H. Hysudricus. That they belong to Elephas proper,

rather than to Mastodon, is clearly indicated by all the prin- |

cipal characters of the "teeth: viz. the crowns are divided
into many transverse ridges, consisting of numerous mam-

mille resembling the digital terminations of the plates in the —
Indian elephant; the hollows are occupied by a more or less —

abundaut layer of cement; and, as in the typical elephants,
there is no appearance of the longitudinal cleft along the

axis, which, in almost all the species of Mastodon, bisects the —

crown into lateral divisions. The same direction of affinity
is indicated by the characters presented by the crania.
We here take leave, for a time, of the proper elephantine

forms; and from this point the complexity in the molars —
radually diminishes till they assimilate to the character ex-_

hibited by the ordinary Pachydermata.
Fig. 2 of Pl. VI. (Pl. IIT. fig. 8, F.A.8.) shows a section of
another of the specimens described in Mr. Clift’s memoir

1 Geol. Trans. 2nd Ser. vol. ii. . 869. 2 Idem, loc. cit. . 370.

er Oe eee ee

DESCRIPTION OF PLATE VI.

ELEPHAS GANESA, MASTODON LATIDENS, DINOTHERIUM

InpIcUM, AND DINOTHERIUM GIGANTEUM.

Vertical longitudinal sections of the teeth of Hlephas Ga-
nesa, Mastodon latidens, Dinotheriwm Indicum, and D. gigan-

tewm.

The three first copied from drawings by Mr. Ford in

Plate III. of the Fauna Antiqua Sivalensis, the fourth copied
from a drawing by Mr. Scharf in the Quarterly Journal Geol.
Soc., vol. i., Pl. XIV., fig. 2a; the first two, one-third, and
the last two, two-thirds of the natural size.

Fig. 1.

Fig. 2.

Fig. 3.

Fig 4.

Is a section of the last upper molar of Elephas Ganesa, copied
from Plate IIL., fig. 7 a, of the Fauna Antiq. Siv. The spe-
cimen is in the British Museum. (See page 80.)

Is a section of the last two upper molars of Mastodon latidens,
copied from Plate III., fig. 8, of the F. A. 8. The specimen was
discovered by Mr. Crawfurd in Ava, and is now in the British
Museum. (See page 82.)

Is a section of the penultimate lower molar of Dinotherium In-
dicum, copied from Plate III., fig. 11, of the F.A.S. (See pages
85 & 396.)

Is a section of the penultimate lower molar of Dinotheriwm
giganteum. (See pages 85 & 396.)

VOL. I.

Se eT

Vol. 1. Plate 6.

GHFord dal. J.Dinkkel lith. W.West imp.

1Blephas Ganesa. 2. Mastodon isiidens. 3.Dinotherium Indicum. 4.D. siganteum

——

o. .
Bee

—— we . a

{
a oes Wl i

ELEPHANT AND MASTODON. 83

under the name of Mastodon latidens,! and represented by
him in Pl. XX XVII. fig. 1. It consists of the two last molars
of the upper jaw. ‘The figure (in F.A.S.) is drawn on a scale
of two-thirds of the natural size. The last tooth shows five
principal ridges with a posterior talon ridge and a subordinate
ridge in front. The ridges are transverse, and divided by a
longitudinal cleft into two pairs of principal points without
intermediate mammillz in the hollows. The enamel is very
thick, and the cement is reduced to a thin layer which is
only observable in the bottom of the hollows. The ivory lobes
resemble those of H. Ganesa, Pl. VI. fig. 1, but they are less
elevated with a broader base. The artist has been eminently
successful in his representation of the texture of the two dental
substances in this specimen. The anterior tooth had been a
long time in use, and the ridges are nearly all worn out. They
were four in number, in this as well as in the two teeth which
preceded it in the jaw. We believe this to be a small or
dwarf variety of M. latidens, a species the adult teeth of
which generally attain a large size. The last tooth figured
in the section measures 54 inches in length. WM. latidens, of
the known forms of mastodon, is that which is most nearly
allied to H. Cliftii, and through that species to the true
elephants. One or more intermediate links perhaps still re-
main to be discovered. It closely resembles the Huropean
M. Arvernensis (M. longirostris of Kaup) in the form of the
molars. The correspondence is so great, in the last milk
molar and in the antepenultimate and penultimate true
molars, that they have been regarded as identical species.
PL VIlL.fig.1 (Pl. IIL. fig.10a, F.A.8.) represents a section of
the last molar of the upper jaw of an Indian fossil species
named Mastodon Sivalensis in this work. The ridges in this
species are more complex in their composition than in MW.
latidens. The crown is traversed by a longitudinal furrow
which bisects them, each division being composed of a pair
of contiguous or connate conical mammille, placed more or
less alternately. The hollows are in consequence interrupted.
This tooth, like its equivalent in M. latidens, consists of five
principal ridges, with a subordinate ridge in front, and a
‘talon’ ridge behind. Hight divisions of the ivory may be
counted in the figure, the smaller segments arising from the
direction in which. the section has been made through the
alternate mammille. The ridges are approximated, and the
layer of enamel bears a large proportion to the conical lobe
of ivory which it invests. The cement is entirely wanting,
1 This valuable specimen, discovered | President of which has liberally allowed

by Mr. Crawfurd in Aya, belongs to the | a section of it to be made for the illus-
collection of the Geological Society, the | tration given in fig. 8 of Pl. IIL, F.A.S8.

G2

84 FAUNA ANTIQUA SIVALENSIS.

except in the bottom of the clefts. This tooth measures
about seven inches in length.

Fig. 105, Pl. IIT., F.A.S., shows a section of a fragment
comprising the greater part of the last lower molar of the
same species. There is a similar alternate arrangement of the
mamunille, and the tooth differs from the corresponding upper
molar only in being complicated with an additional ridge.

The teeth of this species bear an exceedingly close resem-
blance to certain of the European fossil grinders, which have
been described under the indefinite name of M. angustidens. The
three species, M. latidens, M. Arvernensis, and M. Sivalensis,
with perhaps a fourth, of doubtful determination, constitute
a particular section of Mastodon, characterized by the same
numerical division of the crown ridges in the last deciduous
molar, and in the first and second true molars in both jaws.

PL VII. fig. 2 (Pl. ITT. fig. 9, F.A.S.), as previously described,
represents a section of the last upper molar of Mastodon
Ohioticus. It consists of four principal ridges and a small
talon lobe. The successively increasing simplicity of form
which has been traced from EH. imsignis attains its extreme
limit in the molars of this species. The ridges are trans-
verse, terminating in a trenchant edge; the ivory segments
are in regular angular lobes, the layer of enamel is of uni-
form thickness, and the hollows between the ridges are very
wide and open, being almost rounded at the bottom. The
cement is present only in an exceedingly thin crust, con-
tinued over the fangs in greater thickness. The common
plane of the grinding ridges of the crown is nearly horizontal,
while it is more or less convex in all the previously noticed
species. It has not been deemed necessary to give a delinea-
tion of the section of an inferior molar, which differs in no
respect from the upper, except in being complicated with an
additional ridge.

To the same group belong two other species, M. angustidens
and M. Andiwm, and probably a third, M. Tapiroides, the
dentition of which is but imperfectly known. The molars of
the two first differ from those of M. Ohioticus, in the same
manner that M. Arvernensis and M. Sivalensis differ from
M. latidens; viz. the crown ridges, instead of being trans-
verse, are composed of mammille, which are placed more or
less alternately, projecting into the interspaces and inter-
rupting their continuity. The teeth of M. Andiwm are re-
markable in being invested with a coat of cement, which fills
up the bottom of the hollows and is extended over the
mammille in a considerably greater quantity than occurs in
any other species of true Mastodon. These three species,
M. Ohioticus, M. angustidens, and M. Andium, constitute a

DESCRIPTION OF PLATE VII.

Mastopon Srvautensis, Mastopon Oxioticus, AND DINo- ©
THERIUM GIGANTEUM.

Vertical longitudinal sections of the teeth of Mastodon
Sivalensis, Mastodon Ohioticus, and Dinotherium gigantewm.
Copied from drawings by Mr. Ford in Plate III. of the Fauna
Antiqua Sivalensis; the two first, one-third, and the last,
two-thirds of the natural size.

Fig. 1. Is a section of the last upper molar of Mastodon Sivalensis,
copied from Plate III., fig. 10 a, of the Fauna Antiq. Siv. The
specimen is in the British Museum. (See page 83.)

Fig. 2. Is a section of the last upper molar of Mastodon Ohioticus, |
copied from Plate III, fig. 9, of the F. A. S. The specimen is
in the British Museum. (See page 84.)

Fig. 3. Is a section of the penultimate lower molar of Dinotherium
giganteum, copied from Plate III., fig. 12, of the F. A. 8. The
specimen is in the British Museum. (See page 85.)

VOL. I.

Fig. 1. Vol. I. Plate 7.

GH Ford del. J.Dinkel lith.
W. West imp.

1.Mastodon Si is. 2.M
; a = ry} 2 7}
Sivalensis. 2.M.Ohioticus. 3.Dmotherium gsiganteum

ELEPHANT AND MASTODON. 85

distinet section, agreeing in having the three molars which
precede the last, viz. the third deciduous molar and the
penultimate and antepenultimate true molars, uniformly cha-
racterized by having their crown divided into three ridges in
both jaws; while the same teeth in the preceding group have
each a crown with four ridges. The teeth of M. Andiwm and
M. angustidens are not included in the plates of sections.
Following the same serial arrangement which has been ob-
served throughout, their place would be between M. Sivalensis
and M. Ohioticus, the latter of which forms the terminal link
in the chain, establishing the nearest passage into Dino-
theriwm, and through that genus into the ordinary Pachyder-
mata.

Fig. 3 of Pl. VI. and fig. 3 of Pl. VII. (PL IT. figs. 11 and
12, F.A.8.), represent sections of the penultimate lower
molar of two species of Dinotheriwm, the former a fragment,
showing the posterior half of the tooth in D. Indicwm, and the
latter the whole tooth in D. gigantewm. The sections exhibit the
same arrangement of the dental substances as in M. Ohioticus.
The tooth, Pl. VII. fig. 3, consists of two transverse crenulated
ridges, and a talon ridge, while in the equivalent molar of
M. Ohioticus there are three principal ridges. Corresponding
to the smaller number of divisions, the ridges in D. gigantewm
are more widely separated, less elevated, and broader at their
base, while the interspaces are also wider and more open
than in the North American mastodon; the layer of enamel
is of similar thickness, and there is no appreciable crust of
cement. The correspondence is followed out in the form of
the subordinate heel ridge. D. Indicwm is the species which
is most nearly allied to M. Ohioticus; and all the ascertained
evidence regarding it tends to prove that it belonged to a
true Proboscidean Pachydermatous genus like the latter.

§ 3. ON THE SuccEssion or THE MoLars, AND THEIR CHARACTERS AS
INDICATING SECTIONAL GROUPS OF SPECIES.

The molar teeth, developed during the course of life in the
ordinary Pachydermata, are divisible into three well-marked
sets; the milk or deciduous molars, the false molars or succes-
sional premolars, and lastly, the true molars. The milk teeth
are sodistinct, in theirtransitory character, from the permanent
series, that the consideration of the former is usually omitted
in the construction of generic definitions; but in Mastodon
and Hlephas, the succession of the teeth is so modified, and
the premolars are so completely or partially suppressed, that
the triple division is rendered very obscure, and it has
commonly been found necessary by systematic authors, to

86 FAUNA ANTIQUA SIVALENSIS.

include the whole series, in framing the expression of the
dental formula. In fact, till the appearance of Professor
Owen’s ‘Odontography,’ the normal division and theoretical
signification of the different molar teeth in these genera were
not understood.

In the ordinary Pachydermata, which the rhinoceros may
be supposed to represent, there are normally four milk or
deciduous molars in both jaws, the hindmost of which has the
complicated form which characterizes the last true molar of
the adult animal. They are frequently reduced to three in
other genera, by the suppression of the anterior tooth, which
is the most variable and rudimentary in form. These teeth
are succeeded vertically by an equal number of premolars,
the last of which is always of a simpler form than the tooth
of which it takes the place; and they, in like manner, are
subject to a numerical reduction by the non-development of
one or more of the anterior teeth. Behind the premolars are
the true molars, the normal and developed number of which
is invariably three, this set being exempt from the partial
suppression to which the others are subject. They are dis-
tinguished from the premolars by greater complexity of
form; they come into place like the milk molars, in antero-
posterior succession, and the first of the series is protruded
and in use before the appearance of the last premolar, which
immediately precedes it in position in the jaw. In the
adult animal, in most genera, the whole of the premolars and
true molars are simultaneously present and in use in the
jaws. A remarkable exception from this rule takes place in
the subgenus of the hog tribe, called Phacocherus, in which,
in consequence of the complicated form and large size of
last true molars, there is not room in the jaws to accommo-
date the whole number at one time, and the first true molar
is worn down and pressed out before the last molar is pro-
truded. This last molar is gradually pushed forward, caus-

ing the anterior teeth to be shed, so that the number of —

molar teeth in the upper jaw, which at one time in the
adolescent animal amounts to five on each side, is finally
reduced to one or two in the advanced age.? Precisely ana-
logous conditions take place in the true elephants, in which
this kind of exception from the ordinary mode of dental suc-
cession is carried to the greatest known excess.
Dinotheriwm.—The first and most simple deviation from
the usual Pachydermatous type, in the dentition of the Pro-
boscidea, is presented by Dinotherium. In this genus only
two milk molars, viz. the penultimate and last, have been

1 Ante, p. 61. 2 Owen, Odontography, p. 550.

a: ©

q

ELEPHANT AND MASTODON. 87

met with in both jaws, one or two of the anterior teeth of
this set being suppressed. The last milk molar, above and
below, is three-ridged, while the penultimate has only two
ridges. These teeth are replaced vertically by an equal num-
ber of premolars, which represent the penultimate and last,
the two anterior teeth of this series being also suppressed.
The last premolar, as well as the penultimate, is only two-
ridged, conforming to the ordinary rule of being simpler in
form than the milk molar which it succeeds. Of the three true
molars, the first or antepenultimate, in both jaws, is three-
ridged, repeating the complex form of the last milk molar,
while the penultimate and last are only two-ridged. This isa
very remarkable anomaly, of which no other example is
known among the Pachydermata, as it is commonly the last
true molar which reiterates the form of the last milk molar.
The first true molar is protruded and in use before the last
molar is shed, so that in the adolescent animal there are two
contiguous teeth, which have each three ridges. We have, in
this circumstance, the first essential proboscidean character,
which at once distinguishes Dinotheriwm from the Tapirs and
allied genera, and indicates its near relations to Mastodon.
In regard to the number of teeth which are in place and in
use at the same time, Dinotherium is less aberrant than
even Phacocherus, as the two premolars and three true
molars in the adult animal are simultaneously present in
both jaws. The molar formula in Dinotherivm is therefore
2 premol. + 3 mol. — 5 in each side of both jaws; and the
number of ridges in the different teeth, according to their
successive position in the jaw, is 2 + 3 in the milk molars;
2 + 2 in the premolars, and 3 + 2 + 2 in the true molars.
Mastodon. Sect. Trilophoden.—M. Ohioticus.—The next
degree of deviation from the ordinary dental rule is presented
by Mastodon Ohioticus. In this species, which appears to be
the most nearly allied of the well-known forms to Dinothe-
rium, there are three deciduous molars in both jaws, the
most anterior of the series being suppressed. Of these, the
antepenultimate, or anterior tooth (being theoretically the
second), in the upper jaw, measures 1:4 inches in leneth, by
about 1:4 in width; and the penultimate, or second (theoreti-
cally the third), measures 1:7 by 1°75 inches. These teeth are
of the same form, each consisting of four points, which are dis-
posed in two transverse ridges; and they further correspond
with the same teeth in the ordinary Pachydermata by differ-
ing but slightly in relative size. The third milk molar, as in
Dinotherium, consists of three transverse ridges, each com-
posed of two pairs of confluent points. It measures three
inches in length, by 2:4 in width. The milk molars of the

88 FAUNA ANTIQUA SIVALENSIS.

lower jaw differ in no important respect from those of the
upper, except in being narrower in proportion to their
length, and in the greater development of the anterior and
posterior subordinate talon ridges.

With respect to the premolars, the statements which have
been advanced regarding them are conflicting. They have
never been observed in either of the jaws by Godman, Hays,
Cooper, Harlan, or any other of the American naturalists
who have described the dentition of M. Ohioticus; nor has
their presence been noticed by Dr. Grant. But Professor
Owen, in his ‘ British Fossil Mammalia,’ affirms that they have
been recognized in this species; and in his ‘ Odontography,”!
he figures and describes a tooth as the penultimate premolar
of the upper jaw. It is there stated to be composed of two
bifid transverse ridges, girt by a basal cingulum, and to be
of a simpler form than the second deciduous molar; the
crown being broader in proportion to its length, and mea-
suring one inch five lines, by one inch four lines.

Professor Owen also gives a figure of the hypothetical
position of the same tooth in the lower jaw,? the presence of
which he admits has not yet been established in the species.
The accurate determination of this point is of considerable
systematic importance, as the occurrence of this premolar
constitutes one of the two characters upon which (failing
those advanced by Cuvier) Professor Owen founds his generic
distinction between Mastodon and Hlephas. Had the tooth
been observed in situ in the jaw, as in the Dax Specimen of
M. angustidens figured by Cuvier, and in the specimens of
M. longirostris figured by Kaup, its occasional presence in
the upper jaw of M. Ohioticus would have been placed beyond
doubt; but the tooth described by Professor Owen appears
to have been a detached specimen, and no characters are
attributed to it inconsistent with its being the first milk
molar of the upper jaw. In order to arrive at a certain
determination of the point, we have been permitted to make
a section of a specimen consisting of the entire palate of a
young Mastodon Ohioticus in the British Museum, containing
the second and third milk molars, with the first true molar
protruded and the second true molar in germ. A section
was made both along the palate, and along the outside of
the jaw; but not a trace of a premolar was visible, although
the cranium was exactly of the age when a premolar, if
developed, ought to have been shown. A similar negative
result attended a corresponding section of a specimen of the
same age of the lower jaw. The only other evidence which

1 Loc. cit. p. 260. Pl. 144, fig. 3, p. 1. ? Ibid. Pl. 144, fig. 7, p. 1.

—— se

|
;
’
P

ELEPHANT AND MASTODON. 89

could establish the case would be the finding of an unknown
tooth in front of the third milk molar. But, so far as we are
aware, no instance of this sort has been recorded, notwith-
standing the great number of young specimens which have
- been described by different observers; and the result of the
whole evidence at present is, that, ordinarily, the premolars
are entirely suppressed in M. Olioticus, in both jaws.

There is nothing, therefore, in the mode of succession of
the teeth in this species, to show where the deciduous series
terminates and the true molars begin. The last milk molar
is followed in antero-posterior succession, as in Dinotherium,
by a tooth which has its crown divided also into three ridges,
and is thus indicated to be the antepenultimate, or first true
molar. It measures four inches long, by about three in width,
differing only in size from the tooth which precedes it. The
penultimate, or second true molar (being the fifth in the
order of succession), consists also of three ridges, and mea-
sures about five inches by three and a half. The third, or
last true molar, consists of four principal ridges, and a small
heel ridge, which varies considerably in amount of develop-
ment. This tooth measures 7°25 inches or upwards, by about
4°5 in width.

The inferior true molars in M. Ohioticus, agree with the
upper in the form and division of their crowns, except the
last, which has usually five principal ridges. They are
narrower in proportion to their length, and the subordinate
talon ridges are more developed. The molar formula in this
species is, therefore, 3 milk molars in the young animal; and
0 premol. + 3 true molars — 3 in each side of both jaws of the

adult; the number of ridges in the different teeth, accord-

2+2+3 ..

ing to their succession, being 5; in the milk molars, and
34+3+4 re

34375 1 the true molars. With regard to the number of

teeth which are simultaneously present in the jaw; the lower
jaw of Tetracaulodon figured by Godman,' shows the three
milk molars in use, and the first true molar in its alveolus,
there being four out of the whole number of six teeth at one
time in the jaw. These are ultimately in advanced age
reduced to the last tooth, the others being shed.

M. angustidens.—The dentition of M. angustidens is in-
volved in great confusion, in consequence of most authors
who have written on this species having mixed up, under
this name, two distinct forms, the one characterized like M.
Ohioticus, by a ternary, the other, by a quaternary division, in
the ridges of the middle teeth. It will be necessary to

1 Amer, Phil. Trans. New Series, vol. iii. pl. xviii.

90 FAUNA ANTIQUA SIVALENSIS.

enter at some length upon the evidence on this point, more
especially, as the two latest authorities of weight, Prof. Owen
and M. de Blainville, do not admit a specific difference
between WM. angustidens and M. longirostris. On the other
hand, Dr. Kaup has, in some instances (as in the case of the
Stellenhoff lower jaw found near Vienna!) excluded from his
M. longirostris, specimens which assuredly belong to it; while
in others (viz. the Georgensmiind mastodon teeth described
by von Meyer) he has transferred to this species molars
which appear to pertain to MW. angustidens.

The first point to determine, under these circumstances, is
the form to which the specific name of M. anqustidens is pro-
perly applicable. Cuvier’s description of the species com-
mences with the Simorre tooth? which has the crown divided
into three ridges, with a back talon of two tubercles, measur-
ing 4°5 inches in length by 2°35 in width. The next specimen
which he describes as belonging to it is the Dax fragment
(Oss. Foss. Pl. ITT. fig. 2) containing two teeth implanted in the
palate on one side, the anterior of which is the unworn germ
of a premolar, and the posterior, nearly of the same size as
the Simorre tooth, like it consists of three ridges and a small
talon of two tubercles. A third tooth which he immediately
afterwards attributes to this species, is another Simorre speci-
men (Oss. Foss. Pl. IIT. fig.3) measuring 3°6 inches by 2°6, and
having its crown also divided into three ridges. It is therefore
to a species having the intermediate molars distinguished by
a ternary division of the crown, as in M. Ohioticus, that the
specific name of M. angustidens is strictly applicable, so far as
priority of description and reference to original types can be
taken as the guides to a decision on the point.

Of the other specimens referred by Cuvier to his M. angus-
tidens, and represented in the four plates devoted to ‘ Divers
Mastodontes,’ the South American teeth (figs. 6 and 7 of PI. I.
and fig. 4 of Pl. III.) appear to belong to M. Andium, as has
been advanced by M. de Blainville, and nearly all the rest,
which are susceptible of determination, belong to M. Arver-
nensis (M. longirostris of Kaup), with the exception of figs. 1
and 2 of Pl. L, fig. 11 of PLIL., and fig. 14 of Pl. III., which
are probably to be referred to M. angustidens.

We have already stated the grounds’ upon which Von Meyer,
following up the observations of Croizet and Jobert, distin-
guished M. Arvernensis from M. angustidens, and that Kaup
was led by his researches to the same conclusion. It would
appear, from a communication in Bronn’s ‘ Letheea,’ that the
ridge formula which Kaup attributed with doubt to M. angus-

1 Cuv. Oss. Foss. edit. 1834, tom. ii. | Mastodontes, pl. i. fig. 4.
p. 363, pl. ii. figs. 4 and 5. 3 Ante, p. 60.
2 Oss. Fossil. tom. i. p. 255; Divers

ELEPHANT AND MASTODON.

tidens is Se
true molars.’

M. de Blainville has entered at great length, in his ‘ Ostéo-
eraphie,’ on what had been previously written regarding
M. angustidens, and he has given a beautiful series of illus-
trations of all the teeth in succession, in both jaws, as he
conceives them to be developed in this species. The rich
collection of specimens, discovered by M. Lartet and others
in Gascony and along the flanks of the Pyrenees (a large
portion of which is displayed in the palzontological gallery
of the Paris Museum), furnishes ample materials for estab-
lishing the specific independence of M. angustidens and M.
longirostris. But M. de Blainville has not attached sufficient
importance to the constancy of the ridge formula; he has
throughout his illustrations intercalated Hppelsheim teeth of
the latter species, having four ridges, with Gascon specimens
of the former, having three ridges. In consequence, the teeth
of the two species are not merely intermixed, but a wrong
position in the jaw is in many instances assigned to M.
Lartet’s specimens of the true M. angustidens. This remark
applies, without exception, to the determinations of the two
last teeth of the upper jaw. Giving a numerical expression
to M. de Blainville’s descriptions of the different teeth, the
auris 2 Ore a8,
24243? 3+44+6
respectively, in the three deciduous and three true molars
on each side of both jaws. It is apparent that the lower
numbers do not coincide with the upper, and that, followed
in sequence, they deviate widely from the uniform succession
of three ridges presented by the last deciduous, and the first
and second true molars, in M. Ohioticus.

Professor Owen has on two occasions described in detail
the dentition of M. angustidens, and the result stated in his
‘Odontography’ is, that he has seen as yet no evidence that
the teeth described by Cuvier and by Kaup characterize diffe-
rent species. In his ‘ British Fossil Mammalia’ he identifies
English specimens with some of the typical forms figured by
Cuvier, and the number of ridges which he assigns to the
different teeth, according to their succession, is 2+38+3
to the deciduous molars, 2 to the small premolar, and 38+4-+4 5
to the true molars.?, This formula is liable to the same ob-
jections as that put forward by M. de Blainville. In his
‘Odontography,’ however, published subsequently, Professor
Owen describes the teeth of M. angustidens in a different
manner, and the number of ridges assigned by him to the

6 ° ° 3?+4+5 .
in the deciduous series, and 5 in the

P?+4+4

ridge formula in M. angustidens would be

1 Letheea Geognost. p. 1239, 2 British Fos. Mam. p. 286.

92 FAUNA ANTIQUA SIVALENSIS.

successive molars in the upper and lower jaws may be

2+ 34 4°. °
expressed thus: 5—~3-,4 im the deciduous molars, and
44+4+65.

7.7773 im the true molars." This formula, with the excep-

tion of the number attributed to the last milk molar of the
lower jaw, is precisely the same as that assigned by Dr. Kaup
to his M. longirostris, Professor Owen having referred, in
almost every instance on this occasion, to Kaup’s figures,
which he quotes as the types of his descriptions. But he
still alludes to Cuvier’s Dax specimen of M. angustidens as
identical with Kaup’s species, although it is represented in
the original figure, and described by Cuvier, as three-ridged ;
and he states, in the ‘ British Fossil Mammalia,’ that the
rich series of analogical facts in the dentition of M. giganteus
(M. Ohioticus), would ‘now appear to complete the demon-
stration of the specific identity of the Mastodon longirostris,
and Mastodon angustidens.’ *

From these details it will be seen how various and opposed
the opinions of the best authorities are, up to the present
time, regarding Mastodon anqustidens. In eonsequence of its
rarer occurrence in the fossil state, the available materials
for tracing the dentition of this species are less numerous
and complete than in the case of M. Ohioticus. The following
descriptions are chiefly derived from specimens in the Paris
Museum, the most of which have been figured by M. de
Blainville.? Of the milk or deciduous molars in the upper
jaw, the third only has yet been met with i sitw in the
palate. It is well shown, on the right side, in the posterior
tooth of the Dax specimen figured by Cuvier! and referred to
above, the crown consisting of three transverse ridges, and
an accessory talon of two tubercles, each of the ridges being
composed of two pairs of confluent mammille. A single
tubercle juts out into each of the hollows between the ridges
alternately with the principal points, causing the trefoil-
shaped dises, which the worn teeth present in this species, so
different from the lozenge-shaped discs of M. Ohioticus. The
dimensions of this tooth are not mentioned by Cuvier, but it
may be gathered from the context of his description that it
measured a little above three inches long by about two in
width. The same tooth, of the left side of the upper jaw, is
seen in a most instructive specimen found by M. Lartet, near
Sansans, in the department of Gers, containing two molars in
situ, both of which are three-ridged. Of these the anterior,

1 Odontography, pp. 619-28. liberal permission of MM. de Blainville

2 British Fos. Mam. p. 290. and Laurillard.—[H. F.]

3 During a visit to Paris, I had the 4 Oss. Foss. pl. i. fig. 2.
freest access to these specimens, by the

ELEPHANT AND MASTODON. 93

which is the third milk molar corresponding with the Dax
tooth, is in an advanced stage of wear, the ridges of the crown
being ground down into three disc surfaces. No back talon
is distinguishable; if originally present, it has merged into
the wear of the last ridge. This tooth measures 3:15 inches
long, by 2 of width in front, and 1:75 behind, narrowing a
little towards the posterior end. It is figured by M. de
Blainville.'!| The same collection possesses another detached
specimen from M. Lartet, of exactly the same size, but less
worn, which shows three distinct ridges and a small subordi-
nate talon. The grinder described and figured by Von Meyer,
in his memoir on the fossil remains of Georgensmiind,? appears
to furnish another example of the third milk molar of the
upper jaw, left side, of this species. The crown is divided
into three ridges, with a small posterior talon. It corres-
ponds closely to the Gers specimens in dimensions, being
three inches long by two in width. Von Meyer describes
this tooth as the second milk molar of M. anqustidens, but the
size would seem to be conclusive against the correctness of
this determination. Kaup compares it to the third upper
molar of his M. longirostris.*

With regard to the first and second upper milk molars,
neither of these teeth having yet been observed in situ in the
jaw, we are unable to refer with confidence to any specimens
for their characters. But we are inclined to regard the tooth
described by Von Meyer‘ (PI. I. fig. 4) as representing the penul-
timate, or second, and fig. 2 of the same plate as the first. The
former measures 2°2 inches by 1:4, and is composed of three
ridges, which are so far advanced in wear as to furnish no good
diagnostic characters. Von Meyer refers it with doubt to the
last milk molar of the lower jaw, while Kaup considers it to
be the second upper of the left side of his M. longirostris.
The specimen ® here regarded as the antepenultimate, or first
milk molar, has a square crown composed of four points. It
measures 1°6 inches in length by 1:4 in width, resembling
closely in form and dimensions the small Simorre specimen
figured by Cuvier,’ which is also about 1:6 inches long by
1-4 wide, and is regarded by M. de Blainville as the first
upper molar of M. angustidens. This eminent paleonto-
logist assigns the same place to several other specimens
from M. Lartet and others; but such of these figures as are
susceptible of exact determination, from their being found
im situ in the jaw, are derived from Auvergne and Eppel-

1 Ostéographie, pl. xv. fig. 3 ¢ sup. ° Oss. Foss. de Darmst. pt. iv. p. 73.
2 P. 38, tab. ii. fig. 7. ° Fig. 2 of Von Meyer's plate.
2 Oss. Foss. de Darmst. p. 81. 7 Diver. Mast. pl. i. fig. 2.

‘ Georgens. p. 38, tab. i. fig. 4.

94 FAUNA ANTIQUA SIVALENSIS.

sheim specimens of M. longirostris. The same remark applies
to M. de Blainville’s figures and descriptions of the second
milk molar in both jaws.

Of the inferior milk molars, the two anterior, like the
upper, have not yet been found im situ, and the specimens
which have been assigned to them are, in consequence, in a
great measure conjectural determinations. The first was
probably a simple tooth consisting of a pair of cusps; and
the second, reasoning from the analogy of the same tooth in
the nearly-allied M. Andiwm, was probably three-ridged. The
third is represented by the ‘dent de Saxe,’' upon which
Cuvier founded his nominal species of M. minutus, but which
M. de Blainville with reason attributes to M. angustidens.
It is of the left side of the lower jaw; the crown is divided
into three ridges, each composed of two pairs of confluent
points, with a well-developed back talon of two tubercles, and
one or two subordinate tubercles in the spaces between the
ridges. The dimensions of this specimen are 3°25 inches
long, by 1:25 of width in front, and 1°65 behind. An unworn
germ, of unknown origin, in the British Museum, of the
same size as the Saxon tooth, and exactly resembling it
in the ternary division and form of the crown ridges, fur-
nishes another example of the third inferior molar. M.
de Blainville? attributes the same place to a worn three-
ridged tooth, from the collection of M. Lartet, found near
Sansans.

We have seen that the premolars, of which two are deve-
loped in Dinotheriwm, appear to be entirely suppressed in M.
Ohioticus. But there is no doubt about the presence of one
in the upper jaw of M. angustidens. A beautiful illustration
of this tooth is furnished by the Dax specimen, previously
referred to. As figured in the ‘Ossemens Fossiles,’? it is
shown as a germ of a square form and composed of four
points. It is proved to be a premolar, and to be protruded
vertically in the ordinary manner, by being unworn, while
the third milk molar behind it has the three ridges well
affected by wear. This circumstance is clearly indicated by
Cuvier in his description of the specimen. Von Meyer refers
to the same tooth the Georgensmiind specimen represented
in tab. 1, fig. 1, of his Memoir, which resembles the Dax
specimen in form, and in the crown being composed of four
points ; it measures about 1:6 inches square. This upper
premolar, as has been pointed out by Professor Owen, takes
the place of the second milk molar; it therefore represents
the penultimate of this series. There is no evidence that the

1 Oss. Foss. pl. i. fig. 11. 8 Pl. iii. fig. 2 a, 0.
2 Loe. cit. pl. xv. fig. 3 0. 4 Ossemens Fossiles, tom. i. p. 256.

ELEPHANT AND MASTODON. 95

third milk molar of the upper jaw in this species is followed
by a corresponding vertical successor. It is of importance to
observe this apparent irregularity in the order of suppression.
In Dinotherium the two last premolars are developed, the
two anterior being suppressed; in M. Olioticus the whole
four remain undeveloped ; while in M. engustidens the penul-
timate alone is developed, the two anterior and the last
being suppressed. A similar order of suppression has been
observed in the premolars of M. longirostris.

In regard to the lower jaw, there is no evidence yet that a
premolar is included in the dental succession of the inferior
grinders. Von Meyer, with doubt, assigns this place to a
detached tooth which he figures,! but the determination is
merely conjectural, Kaup referring it to his M. longirostris ;
and it is by no means certain that this specimen does not
belong to the upper, rather than to the lower jaw. That a
rudimentary lower premolar may have been developed in this
species is highly probable; but we are not warranted, in the
absence of direct proof, to hazard any inference respecting
organs which are liable to be entirely suppressed, and which,
when developed, are so rudimentary in form as not to be of
functional importance in this tribe of animals.

The materials to illustrate the dentition of the adult animal
have been found in sufficient abundance to leave no room for
doubt respecting the characters and succession of the true
molars. The antepenultimate, or first,? is seen in the San-
sans specimen from M. Lartet, in situ in the left side of the
upper jaw, along with the third milk molar which we have
described. It is an oblong tooth, in the condition of an
almost unworn germ, having the crown divided into three
distinct ridges, with a well-marked basal cingulum on the
inside, and a small back talon. It measures 4°13 inches in
length by 2°75 of width in front, and 2°75 behind.

Another example of this tooth appears to be furnished by
fie. 5, of tab. 1, of Von Meyer’s memoir. The crown has
the same three-ridged form as the Sansans specimen, with
which it agrees very closely in dimensions, being 4°2 inches
long by 2°7 in width. Von Meyer refers it with doubt to
the third molar of the lower jaw, right side of this species,
while Kaup assigns to it the same position in the lower jaw
of his M. longirostris.*

The penultimate, or second true molar, is shown in situ
alone with the last, in another instructive Gascon specimen
from M. Lartet, displayed in the Paris Museum. ‘This frag-
ment, likewise, is of the left side of the upper jaw. Of the

1 Loe. cit. tab. i. fig. 2. 2 De Blainville, Ostéographie, pl. xv. fig. 4.
$ Loc. cit. p. 81.

96 FAUNA ANTIQUA SIVALENSIS.

two teeth which it contains, the anterior (or penultimate)
had been a long time in use, and is very much worn. It is
nearly rectangular in form, and the crown is distinctly
divided into three discs, which indicate the same number of
ridges. No back ‘talon’ is distinguishable, the abrasion of
the last ridge being far advanced. The dimensions of this
tooth are 4°5 inches long by 2°75 of width in front, and 2°6
behind. It is described by M. de Blainville as the fourth,
or antepenultimate.! The posterior tooth in this specimen,
being the third, or last true molar, like its equivalent in
M. Ohioticus, is more complicated in form than the two which
immediately precede it. The crown consists of four ridges,
each composed of two pairs of confluent points, arranged
somewhat alternately, and there is no distinct heel ridge
appended to the posterior extremity. This tooth is wide in
front, and contracts very considerably backwards, a character
common in most species of mastodon, to the last molar of the
upper jaw. The dimensions are—length, 6°25 inches; width
in front, 3°25; width behind, 2:25. The paleontological
gallery of the Paris Museum contains numerous other speci-
mens of the last upper molar of M. angustidens, four of which,
from different localites, have been admirably figured? in the
‘Ostéographie.? They all agree in having the crown inva-
riably divided into four. ridges; the only variety which they
present being in the greater or less development of the ‘ talon’
appendage of the last ridge. Of these, the superb Tournans
specimen,’ which comprises the palate with one tooth on
each side, and the greater part of the lower jaw, shows a
third upper true molar, which resembles very closely the one
described above. It is entire; of the four ridges the two
anterior are worn, and the two posterior intact. This tooth
measures 6°25 in length by 8°25 of width in front, and 2°5
behind, dimensions which are almost identical with those
yielded by the other specimen. M. de Blainville describes all
these teeth as ‘penultimates;’ and he adds, ‘11 est trés
singulier, que dans la grande quantité de dents d’H. angustidens
que M. Lartet a envoyées au Muséum il n’y a pas une seule
sixiéme d’en haut.’® But, in our view, the teeth, which are
figured and described in the ‘ Ostéographie,’ as representing
the ‘fifth,’ do, in reality belong to the ‘sixth’ of antero-
posterior succession in this species, M. de Blainville’s idea of
the sixth or last upper grinder, in M. angustidens, being derived
from a tooth of more complicated form in another species.

1 De Blainville, oc. cit. p. 296, pl. xy. 4 Here terminates the portion of let-
fig. 4 } sup. ter-press already published.—[ Ep. ]

2 Ibid. fig, 5 a, b, ec, d, sup. SP 5298)

3 bid. pl. xiv.

ELEPHANT AND MASTODON. 97

In the lower jaw, the first, or antepenultimate true molar
(being the fourth in the order of antero-posterior succession),
like the-milk molar which precedes it in position, has a crown
composed of three ridges, with a hind talon of two tubercles.
In a detached Gers specimen from M. Lartet, which is con-
siderably advanced in wear, it measures 4 inches in leneth,
by 2 of width in front, and 2-4 behind. It differs from
the corresponding tooth of the upper jaw chiefly in the talon,
and in being broader behind than in front; the reverse of
which takes place in the upper. This dilatation of the pos-
terior part is very constant in all the lower molars of M.
angustidens except the last. M. de Blainville refers to this
character in describing the third lower molar in this species
as being somewhat ‘en forme de gourde.’!

The second, or penultimate, occurs in situ along with the
third in the fine lower jaw figured by M. de Blainville,?
belonging to the Tournans cranial specimen previously
referred to. The penultimate on both sides, in this case,
is very much advanced in wear, the ridges being abraded
down to the common base of ivory, so that the discs are
partly confluent; but the division of the crown into three
segments is distinguishable. This tooth on the left side
measures 4 inches long by 2:5 in width. Another example
of the same tooth is presented by a fine Gers fragment from
M. Lartet, comprising the anterior part of the left half of the
lower jaw of an adult individual. It is well advanced in wear,
but the discs of the crown afford distinct evidence of a
division into three ridges. There is an ill-defined basal
collar along the outside of the two posterior ridges, which
Sweeps around the last so as to form a small terminal talon.
This tooth is considerably broader behind than in front, the
dimensions beine—length, 4°5 in.; width in front, 2°25, and
behind, 2°63 inches. This specimen corresponds very closely
in size, age, mineral condition, and external character gene-
rally, with the palate specimen, also from M. Lartet, which
furnished the first illustration of the last molar, along with
the penultimate of the upper jaw; and it is not improbable
that they may have been derived from the same individual.
Several other examples might be quoted in proof of the con-
stancy of the ternary division of the crown, and of the rela-
tive proportions in this tooth. Among these is the Simorre
Specimen figured by Cuvier,’ which is composed of three
ridges and a talon of two tubercles; the dimensions being—
length, 4:5; width in front, 2; and behind, 2°55 inches.

1 P. 299, % Oss. Foss, tab. i. fig. 4.
* Ostéographie, pl. xiv, and xv.
VOL. I. H

98 FAUNA ANTIQUA SIVALENSIS.

Another example is furnished by a beautiful specimen ot
unknown origin, in the British Museum, which consists of
three ridges and a talon appendage of two tubercles. The
two anterior ridges are slightly affected by wear, the last
being intact. One intermediate tubercle is developed in each
of the hollows alternately with the points of the outer divi-
sion. This specimen belongs to the left side of the lower jaw ;
it presents the characteristic ‘gourd-shaped’ expansion at
the posterior extremity, and the dimensions correspond
exactly with those yielded by Lartet’s Gers specimen—viz.,
length, 4°5 ; width in front, 2-1; and behind, 2-6 inches.

The third, or last, inferior true molar (being the sixth in
antero-posterior succession) occurs along with the penultimate
in the Tournans lower jaw specimen already referred 'to.! The
crown is divided into four ridges, with a large talon forming
a fifth and terminal ridge. The tooth is entire, and the
animal to which it belonged must have been aged, as the
three anterior ridges are well worn. The plane of the grind-
ing surface shows a considerable amount of concavity from
back to front, and the crown narrows very much towards its
posterior end, being the reverse of what is seen in the two
molars which precede it in position. The dimensions of this
tooth are—length, 7°25; and width in front, 3 inches. It
is figured by M. de Blainville, and enumerated in the refe-
rences as the fifth, or penultimate ; but it is described by him
as the sixth, or last, in the text devoted to the dentition of
M. angustidens, although the corresponding last tooth of the
upper jaw of the same specimen, and which, in fact, was
opposed to it in use, is described as the ‘fifth.’? The occurrence
of this last inferior molar mm sitw behind the penultimate,
taken in conjunction with the form and dimensions, furnish
conclusive proofs that this tooth is the third true molar of the
lower jaw; and the evidence yielded by the division of the
crown ridges is equally demonstrative that the species from
which it was derived is distinct from the M. longirostris of
Eppelsheim, a conclusion which is further supported by the
enormous elongation of the beak of the symphysis in the
lower jaw of M. angustidens.

The dental formula of the molars in M. angustidens appears,

eos URE eal 0+140

therefore, to have been ;——; milk molars; >~—;-~» Pre-

1+1+1 .
; true molars; and the number of ridges

1+1+
24+ 37?+3
ZOPEE Biren S!

molars; and

in the different teeth, according to their succession

1 De Blainville, Ostéographie, pl. xiv. ? Idem, loc. cit. p. 302.
and pl. xv.

ELEPHANT AND MASTODON. 99

in the milk molars; 2 in the premolars; and — : : in
the true molars. Omitting the consideration of the two
anterior mill molars, which are only conjecturally fixed, and
of the premolar, the ridge formula furnished by the four last
teeth is exactly similar to that of M. Ohioticus—viz., 3 in the
last milk molar, and 3 + 3 + 4 in the true molars.

We have deemed it necessary to go so much into detail on
this point, as the definition of the ridge formula constitutes
the basis upon which the species of Mastodon are arranged
in this work ; and the position could not have been considered
as established till the exception presented by the teeth of M.
angustidens, as ordinarily described, was explained.

M. Andiuvm.—This species, as defined by De Blainville,
includes the teeth, upon which Cuvier founded his ‘ Masto-
donte des Cordilieres’ and ‘ Mastodonte Humboldien,’ besides
some South American specimens which the great anatomist
erroneously identified with the European M. angustidens. It
is closely allied to the latter species, and there are fortunately
sufficient materials available to establish the succession and
character of the principal teeth. We shall first describe the
grinders of the lower jaw, of which the specimens are most
complete.

Among the fine collection of remains of this species from
Buenos Ayres, lately acquired by purchase for the British
Museum, there is a beautiful specimen of the left half of the
lower jaw, broken only at the symphysis and coronoid process,
of a young M. Andium, corresponding in relative size, and in
the development of the teeth, with a sucking Indian elephant
of about two years old. (See Plate VIII. fig. 1, copied from
Plate XL. fig. 13, F.A.8.) It contains the second and third
milk molars i situ (the first being broken off), together with
the empty alveolar cavity, in which the pulp nucleus of the un-
developed first true molar was lodged. The second milk grinder
is fully protruded, but had barely come into use, the two front
ridges being but slightly abraded ; the third is in the state of
an intact germ, and, although fully formed, had not penetrated
the gum when the animal died. These teeth are both three-
ridged, with a subordinate crest in front and a small bituber-
cular talon behind. They are exactly alike in form, narrow
in front but broader backwards. The ridges, as in M. angus-
tidens, consist of two pairs of principal points, which, instead
of being nearly simple, as in the latter species, are subdivided
into a vast number of superficial warty tubercles, which jut
into the valleys, forming a bridge or connection between the
contiguous ridges, and interrupting the transverse continuity
of the valleys. In this respect they bear a greater analogy

H 2

100 FAUNA ANTIQUA SIVALENSIS.

with the young teeth of M. longirostris than with those of M.
angustidens. 'The dimensions of the second milk molar are—
length, 2°6 inches, width in front 1:3, behind 1-5; and of the
third—length, 3°5 inches, width in front 1:7. M. Andium,
therefore, differs from M. Ohioticus, and diverges more from
Dinotherium and the ordinary Pachydermata than does that
species, by having a more complex crown in the second milk
molar. It supports the presumption that the same tooth in
M. angustidens was also three-ridged. Neither in this speci-
men, nor in the more advanced one next to be described, is
there any indication of an inferior premolar.

The first and second true molars (being the fourth and fifth
in the order of succession) are equally well seen in a fine and
hitherto undescribed specimen, also of the left half of the
lower jaw of an adolescent M. Andiwm from Chili, belonging
to the Canterbury Museum (Plate VIII. fig. 2, copied from
Plate XL. fig. 15, F.A.S.).1. This fragment contains both of
these teeth in situ, with the pit of the fang of the third milk
molar, which had dropped out, and the formed alveolus of the
last true molar. The anterior tooth is somewhat worn, and
consists of three ridges of complex composition which have
rubbed down under the process of mastication into deeply
notched trefoil-shaped, or occasionally quadrilobed, discs, to-
gether with a talon of two points. The posterior (being the
penultimate) is intact, and also has its crown composed of
three principal ridges, with a hind talon. Both of these teeth
are more rectangular in form and relatively broader in front
than the same grinders of M. angustidens. In this respect,
their proportions resemble those of M. Olwoticus. 'The ante-
rior tooth or antepenultimate true molar measures in length
A inches, width in front 2°45, and behind 2°55. The penulti-
mate is partly concealed in the alveolus; the estimated length
is about 5 inches, and the width in front 2°75 inches.

The perfect lower jaw belonging to the nearly entire adult
cranium of this species, from Buenos Ayres, now displayed in
the British Museum (Pl. VIII. fig. 3, or Pl. XXXV. fig. 3,
F.A.S.), completes the evidence regarding the inferior teeth.
by presenting the two last teeth in situ. The anterior of these
confirms what is shown by the Canterbury specimen respecting
the penultimate. It is in an advanced stage of wear, but
exhibits distinctly the discs of three ridges. The crown is
nearly rectangular in form, the dimensions being 5:1 inches

1 My best acknowledgments are due | months. The excellent public museum
to Alderman Masters, of Canterbury, for | in Canterbury, highly creditable to a
enabling me to examine this very valu- | provincial city, owes its origin and pre-
able specimen with leisure in London, by | sent condition chiefly to the well-directed
putting it at my disposal during several | exertions of Mr. Masters.—[H. F.]

~. a.

Fig. 1.

DESCRIPTION OF PLATE VIII.
Mastopon ANDIUM.

Portion of left side of lower jaw of a young Mastodon Andium,
containing the second and third milk molars in situ, one-third of
the natural size, and copied from a drawing by Mr. Ford in
Plate XL., fig. 13 a, of the Fauna Antiqua Sivalensis. The

specimen is in the British Museum, and was obtained from

Fig. 2.

Buenos Ayres. (See page 99.)

Left half of lower jaw of an adolescent Mastodon Andium, con-
taining the first and second true molars, one-third of the natural
size, and copied from a drawing by Mr. Ford in Plate XL.,

' fig. 15, of the F. A.S. The specimen belongs to the Canterbury

Fig. 3.

Museum, and was obtained from Chili. (See page 100.)

Perfect lower jaw of a nearly adult Mastodon Andium, con-
taining the second and third true molars in situ, one-third of
the natural size, and copied from a drawing by Mr. Ford in
Plate XXXV. of the F. A. S. The specimen is in the British
Museum, and was obtained from Buenos Ayres. (See page 100.)

VOL. I.

pe a ht

I. Plate 8,

Yo

W.Westimp.

Mastodon Andnm.

GH¥ord del. J Dinkel lith.

ELEPHANT AND MASTODON. 101

in length, 2°85 inches of width in front, and 3 inches behind.
The posterior tooth, which is the last or third true molar, has
the crown composed of four principal ridges, and a convex
subtriangular heel of several points. The three anterior
ridges are partly worn, and finely exhibit the characteristic
complex trefoil discs of wear; the two posterior are intact,
and the sinuous hollows between them show the very con-
siderable layer of cement, which, as previously noticed, is
present in a greater quantity in this than in any other species
of true mastodon. The dimensions of this tooth are about
8 inches in length by 3°5 inches of width in front, whence it
narrows gradually towards the posterior end. Another de-
tached specimen in this collection exhibits the same form,
and is very nearly of the same size.

These three specimens, each presenting two molars in situ,
and respectively derived from the very young, the adolescent,
and the aged animal, furnish the clearest demonstration
regarding the dentition of the lower jaw in M. Andiwm.
Additional confirmation is dgrived from-the fine specimen
of the lower jaw, containing two entire three-ridged teeth
im situ, described by M. Laurillard, in M. d’Orbigny’s work
on South America;' and by Gay’s specimen from Chili, figured
by De Blainville, which consists of the greater part of an
adult lower jaw, with the second and third true molars in
situ, these teeth corresponding in form and dimensions with
the penultimate and last molars of the adult lower jaw in the
British Museum. ‘The specimen figured by Cuvier? appears
to be the last molar of the lower jaw.

The materials illustrative of the molar series of the upper
jaw in M. Andium are less complete; but the uniform cor-
respondence between the four last teeth in the upper and
lower jaws of M. Ohioticus and M. angustidens would, a priors,
lead to the inference that a similar agreement had held in
WM. Andiwm. We are not aware of the existence of specimens
or figures of the first and second upper milk molars of this
species ; but we are warranted in inferring that the second
was three-ridged, as in the lower jaw. The third deciduous
molar is well represented by Cuvier (tab. ii. fig. 5). The
original, brought by Humboldt from Chili, is the specimen
upon which the illustrious anatomist founded his M. Hwm-
boldtvi. It is an oblong and nearly rectangular tooth, pre-
senting the same square proportions as in M. Ohioticus, and
having the crown divided into three ridges, which are far
advanced in wear, and without either a front or a back talon.

1 Alcide d’Orbigny, Voyage en Sud | and 2.
merique. Palzontologie, pl. x. figs. 1 | ” Oss. Foss. tab. iii. fig. 4.

102 FAUNA ANTIQUA SIVALENSIS.

Tt measures 3°15 inches in length by 2°35 inches in width.
As yet there is no evidence of a premolar in the upper jaw.
Of the true molars, the first or antepenultimate (being the
fourth in the order of succession) appears to be represented
by fig. 7 of Cuvier’s Plate I. of ‘ Divers Mastodontes.’ The
anterior part of the first ridge is broken off; the crown consists
of three ridges which are far advanced in wear, together with
a talon of three tubercles. The dimensions, inferred from
the description given by Cuvier, would be about 4°5 inches
of length, by 2°6 in width. This specimen was brought by
Dombey from Peru; and it is referred by Cuvier to WM.
angustidens. The penultimate, or second true molar, is repre-
sented by the Imbaburra specimen discovered by Humboldt,
in the volcanic region of Quito, and seen in fig. 1, tab. 11. of
the ‘Divers Mastodontes.’ Like the two preceding molars,
it is of a broad rectangular form, having three ridges to the
crown. The dimensions stated by Cuvier are, 4-7 by 3°35
inches. The third or last true molar occurs in situ, in the
cranium belonging to the British Museum, as a solitary
tooth on either side, the penultimates having been worn out.
The crown is composed of four ridges and a large complex
heel. The anterior ridges are well worn, and present deeply
notched trefoil discs. This tooth, on the left side, presents
the following dimensions, viz. length 8-75 inches, width in
front 3°75, and behind, 3-25. Another detached specimen in
the same collection, and in nearly the same stage of wear,
measures only 6°75 inches in length, by 3°5 in front, and 2°75
behind. It consists also of four ridges and a heel; the
sinuous hollows between the ridges display a decomposed
layer of cement of considerable thickness. Other examples
of this tooth are shown in De Blainville’s Ostéographie ;
and the same position may be assigned to Dombey’s
specimen from Peru,! which exhibits four ridges and a
compound double heel. Cuvier refers it to his M. angustidens.
We have ascertained that the old palate specimen preserved
in the British Museum, which is figured and described by
Peter Camper,? referred by him to M. Ohioticus, and by
Cuvier to M. angustidens, belongs in reality to this species.
Camper, in this instance, as in the case of Michaeli’s palate
fragment of M. Ohioticus,’ took the posterior extremity of
the palate for the muzzle end, and he has in consequence
described the back tooth as the front one. This specimen, of
which the precise origin is unknown, appears to have been
presented to the British Museum by the Earl of Shelburne,
with specimens of M. Ohioticus, sent by Croghan, from North

1 Oss. Foss. tab. i. fig. 6 of ‘Divers | 7? Nova Acta Petrop. tom. il. tab. vill.
Mastodontes.’ | % Loe, cit. tab. ix.

ELEPHANT AND MASTODON. 103

America, some of which have been figured and described by
Peter Collinson and William Hunter.'!
The ridge formula, in the successive molars of M. Andiwm,

may therefore be safely expressed as eae : : in the mill
84+3+4

molars, and ;—-3—z; in the true molars, the only difference

from M. Ohioticus being an additional ridge in the second
milk molar. The constancy of the ternary division in the last
milk molar and the first and second true molars in both jaws
of these species, and of M. angustidens, defines a well-marked
. group of mastodons, for which we propose the sectional
name of Trilophodon. Of these M. Ohioticus was the most
colossal form, next M. Andiwm, and last M. angustidens,
which appears to be the smallest known species among the
elephantine Proboscidea. While the first of all the species
indicates the nearest affinities to Dinotheriwm, the molar teeth
of the two last present close analogies to those of Hippopota-
mus, to gigantic forms of which genus the earlier palzon-
tologists were led to refer them. To the same section M.
Tapiroides appears also to belong, so far as the limited infor-
mation regarding the dentition of the species will warrant
an inference respecting it.

M. Tapiroides.—Cuvier founded this nominal species upon
a single tooth from Montabusard, near Orleans,’ the crown
of which, divided into three lobes, is not bisected along the
axis by a longitudinal furrow, as in M. angustidens and the
other species, these eminences being continued across, and
their edges simply cremulated with small regular denticula-
tions, as in the teeth of Dinotheriwm. It has also a small
talon, which exhibits the same cremulated character. This
tooth would correspond in size with the second milk molar
of M. Andiwm. Regarded per se, it does not furnish suffi-
cient evidence to establish a distinct species of Mastodon.

Buffon, in 1778, had described two large grinders, the one
stated to have been found in Little Tartary, and the other in
Siberia, both of which Cuvier referred to M. Ohioticus, at
the same time that he questioned the accuracy of the locali-
ties attributed to them; having restricted the geographical
range of this species to the North American continent.
Doubts were also expressed by him regarding a three-ridged
grinder, figured by Pallas,? and described by the latter as
belonging to the species of the Ohio, although found in the
Ural mountains. In the additions to the last edition of the
‘Ossemens Fossiles,’ Cuvier relinquished his doubts regarding

1 Phil. Trans. vol. lvii. | § Oss. Foss. tom. iil. addit. p. 376.
2 Oss. Foss. p. 267, tab, ill. fig. 6. |

104 FAUNA ANTIQUA SIVALENSIS.

the occurrence of M. Ohioticus as a Huropean species, upon
the evidence of an Italian fossil grinder, described by Borson,
as having been discovered in the hills of Villanova, near
Asti; but the great anatomist throws out a query, whether
it might not belong to a distinct species? Since that time
similar remains have been discovered in Switzerland, and in
different parts of the South of France, which M. de Blainville
has brought together and described as a distinct species,
under the name proposed by Cuvier, of M. Tapiroides. Lauril-
lard, in‘a note appended to the posthumous edition of the ‘Osse-
mens Fossiles,’ had previously expressed his opinion that these
remains indicated a species different from the North American
mastodon, but he has not characterized it by any name.
The whole of the known materials attributed to this species
are at present inadequate to show what its dental system
really was. The back grinders are those in regard to which
there is least doubt. Of these the specimen described by
Professor Borson is one of the most characteristic. Like the
last upper molar of M. Ohioticus, the rectangular crown is
composed of four transverse trenchant ridges, which are ob-
scurely divided into two pairs of principal points; but their
direction is more oblique than in that species, and they are
not distinctly bisected by a longitudinal furrow as in the
latter. The two first ridges are worn, the second exhibiting
a rhomb-shaped disc like that of the American species,
while the two last are intact. The talon is broken off. The
hollows between the ridges are transverse, and free from any
tubercular processes of enamel. The dimensions are 6-1 inches
of length by 3-2 inches of width in front. The enamel is
stated by Borson to be two lines in thickness. He considers
it to have been a back molar of the upper jaw, while M. de
Blainville assigns for it the place of the fifth or penultimate
of the lower jaw. Whether upper or lower, the form indi-
cates it to have been the last true molar of the jaw. The
large grinder figured by Buffon’ presents a rectangular crown
with four trenchant transverse continuously-edged ridges and
a small crenulated heel. It appears to have been the last
upper true molar. Another specimen of the same tooth, from
Antroy, in the department of the Upper Saone, of nearly
similar form, is figured by De Blainville,? along with speci-
mens from Alan and Sansans. No well-determined examples
of the anterior grinders of this species have yet been recorded.
Schintz has described the remains of two kinds of mastodon,
from the lignite mines of Ellg and Kcepnach in Zurich. Of
these, four grinders are referred to M. angustidens; the other

1 Loe. cit. pl. i. and il. p. 411. | 2? Loc, cit. p. 318, pl. xvii. fig. 6 a.

ELEPHANT AND MASTODON. 105

teeth are stated to belong to a species different from any of
those described by Cuvier, and having only a distant resem-
blance in form to the teeth of the North American mastodon.
The large teeth have always three rows of tubercles, the
small two rows. The ridges are transverse and trenchant,
and the terminal lobes of the contiguous ridges are reflected
in a decurrent crenulated crest. These teeth probably belong
to M. Tapiroides. Similar remains, from Koeffnach, have been
described by Meissner, anc were known to Cuvier. Von
Meyer considers them to indicate a distinct species, for which
he has proposed the name of M. Twricencis. Cuvier mentions
that a portion of a tusk, the ivory of which was invested with
a layer of channeled enamel, was found along with the
erinders noticed by Schintz from Keeffnach. M. de Blainville
has described as the first upper molar of this species the
small tooth figured by Kaup;! but it was overlooked by this
eminent anatomist when he made this determination, that
the tooth in question, although drawn detached, occurs. in
situ as a premolar germ, above the second milk molar in the
young palate specimen of M. longirostris, figured by Kaup in the
same plate ;? it has been described in detail as such by Kaup.*

This comprises all that can, at present, be safely adduced
respecting the dentition of M. Tapiroides. The large adult
erinders, which the size and form indicate to be the last true
molars, being uniformly four-ridged, appear to justify the
inference that the last milk molar and the first and second
true molars would have been three-ridged, as has been shown
to be the constant rule in the three previously described
species. This conjecture is further supported by the three-
ridged grinders described by Schintz; while the Sansans
specimen from M. Lartet, attributed by De Blainville to the
third molar (being the third or last of the deciduous series)
of this species, is also three-ridged. We therefore refer it
provisionally to the Trilophodon group. The species cannot
well be confounded with any other, except M. Oluoticus, from
which it is sufficiently distinguished by the form of the crown
ridges. It is not improbable that, when better known, M.
Tapiroides will prove to be the species of the genus which is
most nearly allied to Dinotheriwm.'*

M. Australis.—Professor Owen has described, under this
provisional name, a fossil grinder brought by Count Strzlecki
from Australia. The specimen is an entire tooth, the crown
of which ‘supports six principal mastoid eminences in three

1 Oss. Fossil de Darmst. pl. xvi. fig. 8. | ‘ The above inferences arrived at in 1846

2 Loe. cit. pl. xvi. figs. 1 and 1a. | were confirmed by M. Lartet’s observa-

% Loe. cit. p. 70. tions published in 1851..—[Eb.]
4 Subsequent Note by Dr. Falconer.— |

106 FAUNA ANTIQUA SIVALENSIS.

transverse pairs,’! with a narrow subordinate basal ridge in
front, and a quadritubercular talon behind. A pair of small
tubercles is placed in each of the valleys, in the long axis,
forming a connection between the transverse ridges. The
dimensions are stated to be 4 inches 10 lines in length,
width behind 2 inches 11 lines, and height of the middle
ridge from the base of the crown 2 inches 6 lines. According
to Professor Owen, the specimen presents a generic and
nearly specific identity with Cuvier’s representation of M.
angustidens, as exhibited in Pl. II. fig. 11, Pl. ITI. fig. 2, and
Pl. I. fig. 4 of the ‘Ossemens Fossiles.’ He considers that it
differs as much from M. angustidens as the latter does from
M. Andiwm, and from the M. longirostris of Kaup in the
principal transverse ridges being more compressed antero-
posteriorly in proportion to the height, while they taper to
sharper summits. The dimensions correspond very nearly
with those assigned above to the penultimate of the lower
molar of M. Andium.

The occurrence in the Australian continent of fossil
marsupial types of the large Pachydermata, the dentition of
which so closely assimilates that of the latter that Dipro-
todon was in the first instance referred to Dinotherium,
might raise a question whether the specimen under con-
sideration might not be a marsupial representative of Mas-
todon ; but Professor Owen, with good reason, believes it to
belong to a true Mastodon, founding his opinion upon the
complex form of the molar. A question of more weight
arises respecting the authenticity of the specimen as really
of Australian origin. It was not found im situ by the emi-
nent traveller, but procured by him from a native of New
Holland, at some distance from the asserted locality. The
tooth itself does not furnish characters sufficient to distinguish
it from M. Andiwm; and until further evidence is adduced,
the specific independence of M. Australis and its Australian
origin must, under these doubts, remain in abeyance. The
almost cosmopolitan range of Mastodon makes the presump-
tion more probable of a species of this genus extending to
New Holland than perhaps of any other genus among the
Pachydermata. The specimen belongs to the Tvrilophodon
section of Mastodons.

No species of this group has yet been discovered in the
fossil state in India.

Mastodon, Section Tetralophodon.—The group of species
for which we propose this sectional name is defined by
characters, in the numerical disposition of the crown ridges

1 Owen, Ann. of Nat. Hist. vol. xiv. 1844, p. 269.

Oe

ELEPHANT AND MASTODON, 107

of the grinders, equally constant with those which have been
shown above to hold in the section Trilophodon: a quaternary
division of the crown being presented by every tooth in
which a ternary arrangement occurs in the latter. This
section, so far as it is known to us at present, comprises four
fossil species, two of which are European and two Indian.
The two former have hitherto been confounded together by
most authors, as well as with M. angustidens. Their accurate
discrimination is of great importance, as they appear to
belong to faunas of two distinct periods.

M. longirostris.—The materials requisite to illustrate the
dentition of this species, thanks to the indefatigable re-
searches of Dr. Kaup in the fossiliferous sand of Eppelsheim,
are as complete as in the case of M. Ohioticus. The series
of teeth is beautifully represented in the ‘Ossemens Fossiles
de Darmstadt.’ There were three deciduous molars in the
upper jaw; of these, the antepenultimate or first (theoreti-
cally the second) is of an oblong form, broader behind than
in front, consisting of two ridges disposed in four points ;
the dimensions being about 1:2 inches long by 0-9 inches
broad. The second or penultimate milk molar is composed
of three principal ridges, with an obscure talon crest behind,
measuring 2 inches long, by about 1:4 of width in front, anc
1-6 inches behind. The third or last milk molar has a crown
composed of four transverse ridges, with an accessory crest
in front, and a talon ridge behind. A sinuous longitudinal
channel bisects the crown, along the axis, into an outer and
inner division, as in the North American Mastodon, each
division of the ridges presenting a finely lobed or denticulated
edge, composed of from four to five minute points. These
edges are gradually depressed from either side towards the
centre, so that they unite at a very open angle. The anterior
crest is given off from the principal point of the inner side of
the first ridge ; it descends obliquely outwards to the base of
the outermost point, and is thence reflected backwards in a
basal collar, bounding the inner side of the crown. There is
no corresponding collar along the outer side. The valleys
are transverse, but their continuity across is somewhat
interrupted by small mammille near the axis, which form a
sort of bridge between the contiguous ridges. These charac-
ters are minutely and carefully described by Von Meyer.
This tooth measures 2°7 inches long by about two inches
broad. For the determination of these teeth we have had
access to the original specimen, now in the possession of the
Earl of Enniskillen, figured by Kaup (Pl. XX. fig. 2),'and toa

1 Also figured in Fauna Antiqua Sivalensis, PJ. xl. figs. 6 and 6 a,—[Ep.]

108 FAUNA ANTIQUA SIVALENSIS.

cast of the Darmstadt specimen, also figured by Kaup (Pl. XVI.

fig. 1), but first described by Von Meyer, under the name of

M. Arvernensis. These specimens—the former belonging to

the right and the latter to the left side of the jaw—are very

nearly of the same age.!
% * * % * * *

We now get out of the true Mastodons, and the numerical
definition of the crown ridges of the intermediate molars
becomes gradually more and more irregular. Instead of
characterizing groups of species, as has been shown to hold
good with respect to the ternary and quaternary divisions,
the next ascertained numerical increments become distinctive
marks of individual species only; and as the cypher in-
creases everything like correspondence between consecutive
teeth ceases, the number of ridges augmenting with the age
of the tooth, till at last they become indefinite.

The next numerical formula after the ternary and quater-
nary might be expected to be quinary, in the ridges of the
last milk molar and the first and second true molars; but the
number five is not met with in these teeth in any known species.
There is an abrupt transition from four to six, which occurs
in Hlephas Cliftvi, the species which, in our view, constitutes
the first of the elephants, as distinguished from Mastodon,
and establishes the immediate passage between the two.?

The accessible materials for illustrating the dentition of
this species are limited, at present, to the last milk molar
and the three true molars of the upper jaw, and the last
molar of the lower. The specimens which furnish these are
all derived from the kingdom of Ava. The last milk molar
is shown in a palate specimen along with the anterior portion
of the first true molar, brought by Colonel Burney from the
banks of the Irrawaddi, and presented by him to the British
Museum.? But the crown is worn down to the common base
of enamel, and the number of ridges does not admit of being
distinctly made out. The first true molar is beautifully
shown in fig. 6 of Plate XX XIX. of Mr. Clift’s memoir in the
‘Geological Transactions.’ It is a very broad rectangular
tooth, having the crown divided into six ridges, with a small
heel ridge. The ridges are continuously transverse, as in
the plates of the true elephants, with no indication of the
longitudinal furrow which, in all the true Mastodons, divides
the crown along the axis into two halves.‘

1 There is a deficiency here in the | Mastodon Sivalensis indicated a Penta-
Manuscript, but further remarks on | lophodon type. See vol. ii—[Ep.]

M. longirostris and M. Arvernensis will ° See Fauna Antiq. Sival. pl. xxx. figs.
be found in yol. ii—[Ep.] | 1, 2, and 3; and also Appendix.—[Ep.]
* This was written in 1846. Dr. Fal-| +4 The Manuscript ends here.—[Ep.]

coner subsequently determined that |

ELEPHANT AND MASTODON. 109

APPENDIX TO MEMOIR ON ELEPHANT AND MASTODON.

J.—Descriptions By Dr. FALCONER oF Fosstt Remarns or ELEPHANTS IN
Museum or Asiatic Society, BENGAL, EXTRACTED FROM THE
Museum CaTALOGUE.

A. From the Sewalik Hills.

[ Of the numerous specimens of Fossil Elephant from the Sewalik hills,
described in the catalogue, the following are most noteworthy.—Ep. |

No. 7. Elephas insignis Beautiful specimen ‘of the upper maxilla,
right side, of a very young animal, showing the second milk molar
nearly entire im situ, with the fang cavities of an anterior milk molar in
front: also the lower half of the alveolar canal with some of the outer
laminz of the tusk adhering to it. The tusk for the age is of very large
size, probably indicating the animal to have been male. The end of
the alveolus of the third milk molar is seen behind, but no part of the
tooth. The specimen includes also the palatine portion of the maxillary
bone, and part of the jugal apophysis of the same bone. The specimen
is a good deal covered with argillaceous matrix which conceals the sur-
faces of the bone.

Inches
Leneth of fragment 4:3
Length of molar . : : : : 5 : : : oy eit
Width at second ridge . 5 ‘ é : 2 : : a isto
Greatest width behind . : ¢ : . : é 16
Height of crown at fifth ridge. : . A 4 : il

The enamel of the ridges is deeply grooved vertically, forming
sinuous folds upon the plane of wear. The first five ridges are more or
less worn, showing that the tooth had been in full wear; the last ridge
intact, exhibiting the tips of five mammille. Diameter of alveolus of
tusk, 1:2. This is a very valuable and instructive specimen, as re-
gards the early dentition. It corresponds very closely with the spe-
cimen figured in Pl. XIX. fig. 1, of Faun. Ant. Siv., but is still more
perfect.

No. 8. Elephas insignis.—Fine specimen comprising the anterior
half of the horizontal ramus, lower jaw, left side, of a young animal
corresponding very nearly in age with No. 7, and showing the same
teeth. It is broken off vertically in front near the symphysis, and
behind posteriorly to the molar. It shows the second milk molar in
situ nearly entire, with the first five ridges more or less worn, the last
ridge barely touched. The two first ridges partly broken off. The
tooth is the second milk molar, and it agrees so exactly in every respect
in form and characters with the corresponding tooth of No. 7, that it
is unnecessary to give details: it might even have belonged to the
same animal. Two empty fang-pits of the fallen out first milk molar
are seen in front.

Length of specimen. : : 4 : a : :
Height of jaw to alveolar margin, at first milk molar . : 3 One:
Thickness of jaw behind ; ; : ; : : EL
Length of molar p 2 r Q : : < : 6 Pl
Width of crown at second ridge 10
Ditto ditto, behind 16

110 FAUNA ANTIQUA SIVALENSIS.

It will be seen from the last measurement that the tooth contracts
from behind forwards very considerably.
A valuable and instructive specimen.

No. 10. Elephas (Stegodon) bombifrons (doubtfully referred to this
species).— Fragment comprising the upper maxilla, left side, containing
a large tooth in situ; the greater part of the outer surface of the max-
illa is shown vertically, together with the extremity of the pterygoid
apophysis behind the alveolus. The posterior half of the crown of the
tooth is broken off, leaving only four white discs of wear in front. The
first ridge is entirely worn out, and the tooth must have belonged to a
very large-sized adult animal, for it appears to have been the last true
molar.

Inches
Entire length of tooth 5 : . s ; : - eon
Width of crown at fourth ridge 2 F : mee
Breadth of ivory disc of second ridge . : 7 12

No. 19. Elephas (Loxodon) planifrons.—Fine fragment, comprising
the superior maxilla, left side, detached, with two molars 7m situ and the
greater portion of the palate ; broken off behind immediately to the rear of
the posterior border of the palatum, and in front near the commence-
ment of the diastema. The anterior tooth is well advanced in wear,
and shows about seven ridges, the two anterior of which have been
ground down to a common disc. The plates are expanded somewhat
in a lozenge form in the middle. The enamel for an elephant is thick
and a good deal waved (or plaited) in this specimen; the crown is very
broad, the talon consists of about four mammille. The posterior molar
is broken off in a line with the posterior border of the palate, and only
shows four very thick ridges, which are in germ and completely covered
over with cement. The greater part of the section of the tooth shows
the low elevation of the ridges characteristic of L. planifrons and the
other Loxodons. This specimen is much weathered, and there is
hardly any matrix upon it.

Dimensions.

Length of fragment 12
Ditto anterior molar ‘ . < ; 7
Greatest breadth of crown of ditto ; 4 : os
Height of crown behind at the last ridge inside of ditto - a
Length of fragment of last molar : P i)
Height of second ridge of ditto : : “ DMs
Width of last ridge of ditto measured at the section A sot

No. 21. Elephas (Lox.) planifrons.—Fine specimen, lower jaw, left
side, comprising the greater portion in length of the horizontal ramus,
broken off in front about the middle of the diastema, and in rear be-
hind the offset of the coronoid apophysis, with two molars zn situ. The
Jjirst is entire, being surrounded by the alveolar border and shows seven
ridges, the two anterior of which have their surface broken off;
the rest, with the exception of the last, being more or less worn; the
ridges show the discs expanded in the middle with very thick enamel
plates, which are nearly free from plaiting, and the mammille appear
to have been thick and few in number, there being five to the last
ridge. This tooth appears to be the antepenultimate or first true

—-r

ELEPHANT AND MASTODON. 111

molar—inferred from the comparatively small size of the jaw indicating
a young adult, from the small size of the tooth and from the one behind
it. The second is broken off about the middle and is entirely in germ,
no part of it having emerged above the border of the alveolus; it shows
the greater part of six ridges; the mammille are few in number and of
large size and with very thick enamel, nearly as thick as in the stego-
dons: one mental foramen is visible. This is a highly instructive
specimen, both as regards the middle-aged dentition and as showing the
transition into the Stegodons, the enamel being very thick and the ridges
comparatively low.
Dimensions.

Inches
Length of specimen “ : 5 ; : : : é 13°5
Ditto of anterior molar. 6°
Width of crown in front ; oF
Greatest width at fifth ridge : 3
Length of fragment of posterior molar 4:7
Width of base of crown of last ridge 3°)
Height of enamel plate of fourth ridge 2°9

The specimen is entirely cleared of matrix, but is weathered into
the reddish ochreous colour common in many of the Ava specimens.
Unluckily there is no information as to whence it came.

No. 22. Elephas ( Lox.) planifrons.—F¥ine fragment of the lower jaw,
comprising nearly the whole of symphysial portion of both sides, the
beak of which is slightly broken off: the whole of the horizontal ramus
and greater part of ascending ramus right side; the condyle and
coronoid apophysis alone are wanting. ‘Two molars are contained in
the jaw. ‘The anterior one is well worn, showing about six ridges and
a heel; the three anterior ridges being ground down into one common
disc, and the three posterior also worn down close to the base of the
crown. ‘The discs of wear are wide, and a little expanded in the
middle, the enamel thick and with very little tendency to plaiting. The
posterior molar shows six ridges emerged from the alveolus, the two
anterior of which are slightly abraded, the rest being quite intact and
covered over with a large quantity of cement. It is not apparent how
many ridges more are contained in the back part of the alveolus. Both
molars exhibit the characteristic form of Hlephas planifrons in a well-
marked manner.

Dimensions.

Inches
Extreme length of fragment . , 21-5
Height of jaw to the surface of anterior ridge of first molar. 77
Ditto behind to third ridge of second molar. v5
Length of front molar ; 5 - 5 . : ° : 53
Greatest width of crown of ditto : 6 : : a : 36
Length of emerged part of second molar . ‘ ’ 5d
Length of diastema from anterior border of penultimate molar

to upper edge of extremity of symphysis 7

The characters of the anterior part of the jaw differ very considerably
from those of the existing Indian Hlephant. In the latter, the diastema
descends nearly vertically from the anterior extremity of the molar,
the beak is very short and the outline of the two rami is rounded,
whereas, in the fossil the diastema descends obliquely forward at an
acute angle with the inferior border of the ramus, entailing a long

112 FAUNA ANTIQUA SIVALENSIS.

spout to the symphysis, and this part of the ramus is thickened and
much swollen out, so that the outline forms a narrow oval, instead of
being round. In these respects it resembles the African Elephant, but
the symphysial portion is still more elongated than in that species.
The inferior border of the jaw is considerably arched anteriorly; five
scattered mental foramina are visible, the largest being at the base of
the beak.

No. 28. HElephas Hysudricus.—Mutilated cranium, broken across
about the middle of the sheath of the incisive bones, with the loss of the
whole of the frontal, temporal and occipital regions; perfect in the left
maxillary, the whole of the palate, posterior part of diastema on both
sides, and the inferior part of the incisive bones; the right maxilla
broken off obliquely outwards from the inner margin of the alveolus,
the fracture having removed the greater part of the molar of that side;
the posterior opening of nasal fossa entire. Two molars on the left side,
the anterior perfectly entire, showing ten ridges, the eight anterior of
which are more or less worn, but none of the discs confluent. The
enamel plate of the front accessory ridge also remains. The discs
of the three anterior ridges have the enamel a good deal plaited with a
narrow loop in the middle. The next three ridges have their digital
extremities worn into three distinct divisions, and the mammille of the
posterior ridges are distinct. The enamel is thick as compared with
the Indian Elephant, but less so than in EZ. planifrons. The crown of
the tooth forms a narrow oblong, with little obliquity in the place of
wear. Of the posterior molar all the plates are in germ and concealed
by the pterygoid bone, the three anterior alone being exposed by a
fracture. They are seen to be very much deeper than in /. planifrons
and to be curved forwards at the apex. The section of the incisive
bones shows that the tusks were of about medium size, namely four and
a half inches in diameter.

Dimensions.

Inches
Length of palate from broken extremities of incisives : : 12:
Width of palate in front : 2 ‘ : : : ; 3°
Ditto behind . : 4:
Length of anterior molar (penultimate) 2 : : 8
Width of crown in front : A : : c 4 : 31
Ditto in middle é c : , 37
Ditto behind é : : c : c - 3.5
Estimated length of last molar i : : : = c 115

Height of second plate . : : . - : : ay

This is the only portion of the cranium of his species of considerable
size contained in the collection. The molars are inferred to be the
penultimate and last. They show well the characters of the species as
distinguished from H. planifrons.

No. 41. Elephas Hysudricus.—Very fine specimen of the lower jaw
left side, comprising the greater portion of the horizontal ramus broken
off in front of the mental foramen, and of the whole of the ascending
ramus as high as the neck of the condyle; containing the last molar in
situ, nearly worn out, and the whole length emerged from the alveolus;
the anterior part of the tooth would appear to have been worn away;
the remains of twelve ridges are visible, the first three of which are

ELEPHANT AND MASTODON. 113

confluent into a continuous disc; all the rest except the two last are
more or less affected by wear; the discs are narrow and the plates
closely approximated, showing little plaiting of the enamel and not much
expansion in the middle; both these characters being inferred to be
owing to the very advanced stage of detrition.

Dimensions.
Inches

Ue

Length of specimen from anterior fracture to posterior border of
ascending ramus ’ : : : : : : :
Height from plane of lower border of horizontal ramus to
summit of fractured condyloid apophysis : : : Wied

Antero-posterior diameter of ascending ramus : . : 8
Vertical height of horizontal ramus, measured to outer margin

of alveolar border. : : : : : : - ue
Length of crown of molar : : . - : : . 8:
Width of ditto in middle : : : : - - : 31

This is the most perfect specimen in the collection, so far as the lower
jaw is concerned.

No. 642. Elephas Cliftii.—Fine fragment, comprising horizontal ra-
mus of lower jaw right side, from the middle of the symphysis on to
near the middle of the ascending ramus, with one finely preserved
molar. This tooth shows seven ridges in full wear, with about two
ridges broken off, dises a good deal depressed, and the enamel much
plaited. An interrupted basal cingulum on the outside at the extremi-
ties of the valleys. The diastema slopes downwards and forwards at an
acute angle with the inferior border of the ramus, and is a little concave
in its outline. Three mental foramina in a nearly horizontal line,
differing from what is seen in /. planifrons, the most anterior being
the largest.

Inches
Length of molar . : : : : ; : “ : ei ee
Greatest width \ é ‘ 4 : : ‘ 5 > 3:8

Inferred to be the last true molar of an old animal. Presented by
Col. Colvin: No. 5 in his Catalogue in ‘Journ. As. Soc.’ vol. v. p. 181.

B. From Ava.

No. 2. Elephas (Stegodon) Cliftii—Fine specimen consisting of
the superior maxilla, right side detached, with the last true molar in
situ, the whole length of the alveolus shown with a small part of the
diasteme in front, and a portion of the remains of the penultimate true
molar which had been worn out. The specimen is impregnated with
black ferruginous infiltration, and is sparingly covered with a gritty
sandstone matrix: the last molar is nearly entire and in fine preserva-
tion, showing seven ridges and a heel: the first three ridges are well
worn, the mosi anterior being nearly ground out, and they show a con-
tinuous transverse excavated fossa, surrounded by a continuous more
or less waved belt of enamel. The 4th ridge is only slightly abraded
at the apices of the mammille, and the 5th ridge is barely touched.
The 6th and 7th, with the heel, are quite intact.

The plane of wearing slopes obliquely from the outside inwards, so
that the interior side of the first three ridges is ground down much
lower than the outer side. There is no mark of a longitudinal line

VOL. I. I

114 FAUNA ANTIQUA SIVALENSIS.

bisecting the tooth, as in the true Mastodons, into an outer and inner
division. The ridges are a little convex in front and concave behind,
determining a similar form to the valleys between them, which run
across without interruption, there being no accessory mammille deve-
loped so as to encroach upon the valleys. The mammille are obtuse
and closely packed; they are obscurely separated from each other by
shallow grooves or fissures, and nine or ten of them may be counted on
the intact ridges. Hardly any crusta petrosa is visible except between
the 5th, 6th and 7th ridges. The ivory is very thick, with a rugose
surface very much like that of Mastodon latidens. Overlapping the
front ridge there is a small portion of the ivory of the penultimate
molar, the remains of which had not yet dropped out.

Dimensions.

Inches
Extreme length of crown of molar. . . ‘ : » 9°35
Breadth of ditto at second ne - - : é - = » 43
Ditto at fifth ridge : - : . : ° . . 406
Ditto at seventh ridge ‘ : : : : : . 36
Ditto of heel : : - - eis,
Height of crown of third ridge, taken at outside : : . 13
Ditto at inner. : ao ie

Two large fangs are Hetbls on sithet site in dale and another out-
side below the third ridge.
The crown of the tooth capers gradually and slightly from in front

backwards.
The ridges of enamel, even where intact, are very low for the size of

the tooth.

Inches
Extreme length of specimen is . : . 13
Height of ditto from surface of fifth ridge to fracture . : iy

The characters of the crown of the tooth agree very closely in every
respect with the tooth of the lower jaw figured in Faun. Antiq. Siv.
Pl. XXX. fig. 5, and it is inferred to be the last true molar of the upper
jaw, represented in fig. 3 of that plate. It is possible that the fragment
in front which is attributed above to the penultimate molar may be an
anterior ridge of the last true molar which had got detached from
wearing out, in which the normal character of the tooth would be to
have eight ridges and a talon instead of seven as described above.
Supposed to be one of the seven jaws mentioned in the Journ. As. Soc.
vol. i. p. 365.

No. 16. Elephas Cliftii.—¥ine fragment, comprising the horizontal
ramus lower jaw, left side, with nearly the whole length of the last molar
embedded in it; broken off in front at the symphysis showing a part
of the diastemal ridge, and behind, a little in front of the offset of the
coronoid, the fracture vertically including a portion of the fang; a
single mental foramen is shown in front of considerable size. The
whole of the anterior two-thirds of the crown of the molar is entirely
worn out into a hollow surface of ivory without any enamel. The ivory
is seen to be of great thickness. Two ridges and a part of a third are
seen behind, the front one well worn, the last barely touched. An
intact mammilla is seen bounding each of the rear valleys at the out-
side, and a considerable quantity ” of crusta petrosa fills up the valleys.
heron is no indication of a longitudinal bisecting line along the crown.

ELEPHANT AND MASTODON. 115

The enamel is very thick and resembles that of the upper jaw, specimen
No. 2, also from Ava.

Inches
Length of fragment : : c : ; ° 12:
Height in front, at commencement of diasteme 5 . é a ise
Height of jaw to inner margin of alveolus behind . a 0 s E
Diameter of jaw behind at bulge of coronoid : 3 eG
Length of molar . : : : : : . 0 : a EP
Greatest width behind 4

C. From the Nerbudda.

No. 40. Hlephas Namadicus.—A superb cranium of Elephas Nama-
dicus of enormous size, exhibiting the whole width of the occiput and
occipital fossa, the foramen magnum, the temporal fossa on both sides,
also both maxillaries with the last molar of each in situ; also part of
both incisives, with the empty sheath of a huge tusk in each, and the
inter-incisive fossa; also the whole length of the palate and a con-
siderable portion of the diastema. The frontal region mutilated from
the bosses of the vertex on either side on to the incisive border of the
nasal opening, involving the loss of the greater part of the parietal and
frontal bones and the whole of the nasals which are broken off, together
with the zygomatic arches and orbits. A portion of the deeper part of
the orbital fossa is shown on the right side, but considerably within the
broken off rim of the orbit. The orbital ala of the sphenoid is also
shown together with the spheno-frontal fissure, leading to the optic
foramen and the common spheno-orbital and round foramen. ‘The
whole of the spheno-palatine region is preserved, but the occipital con-
dyles are more or less mutilated, and the rim of the foramen magnum
injured. A portion of the jugal apophysis of the temporal on both
sides remains, but the glenoid surface is removed, with the exception of
the inner portion on the right side near the base of the apophyses; the
auditory foramen, broken off externally, is also distinctly visible on the
right side. The surface of the specimen, where entire, is black and
shining, as if coloured with iron, but it yields a white fracture. It is
almost entirely denuded of enveloping matrix, but the incisive sheaths
are filled with a hard gritty calcareous sandstone, which also occupies
the cancellated cavities of the diploe where shown denuded. The skull
had fortunately escaped crushing, and the mutilated part only involves
that portion of the frontal region which, from its slight power of resist-
ance, would yield first to the effects of exposure to the weather.

Dimensions.

’ tes Feet Inches
Extreme length from broken margin of incisives to the pro-

jecting part of the occipital boss behind ; : 1 2 Ost
Length from anterior border of occipital foramen to com-

mencement of diastema : P 5 ‘ ‘ « A 5 20:25
Extreme width of the fragmented portion of the skull

behind : ; : é é ; ¢ : » 2, 11:5
Width from middle of occipital crest to fragmented border

of cranium, left side , ¢ 6 : : : se ee Op:
Length from base of occiput at posterior margin of foramen

magnum to commencement of the diastema : Ay oe his)

Length from the protuberance of occipital boss to the inte-
rior margin of the nasal opening : : : :
Height from posterior end of grinding surface of right molar
near the pterygoid to summit of right occipital boss . 2, 6°24

12

116 FAUNA ANTIQUA SIVALENSIS.

Feet Inches
Height from anterior end left molar to base of left incisive
bone, at border of nasal opening , nies Ber (o
Height from inferior margin of occipital foramen to apex of
vertex, right side 2 1 ,, 6:25
Length of alveolar cavity measured from the inferior (distal)
end of pterygoid process to commencement of molar in
front, right side . 3 oa! 1
From ditto to floor of intermaxillary fossa 2 Dene Gite
From the intermaxillary suture to the outer border of the
maxillary portion of the incisive alyeolus . : a 0
Length of incisive fossa remaining - 5 out!)
Width of incisive fossa near apex 5 < : 5 cm Olas.
0
0
0

a

Ditto anteriorly at fracture of incisive
Transverse diameter of the incisive alveolus
Antero-posterior ditto
Depth of occipital fossa from the plane of the bosses at the
vertex . . 5 0
Greatest width BE ditto 2 F : ~ Oc
Narrowest diameter ditto between the bosses 0
Greatest length of fossa from Pee foramen to anterior
fragmented margin. 0 ,, 12°5
Depth of jaw from lower margin of maxilla to orbital 1 pro-
cess of sphenoid near the middle . 0
From posterior border of palate to broken edge of incisives 1
Width of palate in front between alveoli 5 5 id)
Ditto behind . ; : “4 : : : ; 0%
Length of crown of molar . 1
Width of ditto in front 0
Ditto about middle 0

Both of the molars are in situ and appear to have been well
worn, the crown on to the back ridges being ptotruded; the ante-
rior plates had been worn out. The grinding surface is a good
deal damaged by blows and abrasion, but the plates are seen to
be closely “ approximated with narrow discs and plaited enamel.
The discs expand into a loop in the middle; about 22 plates can be
counted, besides some of the most anterior which have probably been
worn away. The teeth are the last true molars of a large adult, and
probably male, judging from the size of the tusks. An entire cranium
im very good preservation of #. Namadicus is figured in Faun. Antiq.
Siv. Pl. XII. A.; the original i is lodged in the United Service Museum,
Charing Cross, London ; it was probably a female. ‘The specimen
above described agrees so entirely with that figure, that there can be
no doubt of its belonging to the same species. The enormous width
and depth of the occipital fossa which form so prominent a character
in the Tae are still more exaggerated in the fossil, together with
all the other characteristic marks. Unfortunately the anterior bulge of
the forehead is broken off, or it might have been expected to have been
still more largely developed. The crowns of the molars are better
preserved than in the figured specimen, which shows only 10 or 11
plates, Faun. Ant. Siv. Pl. XII. B. fig. 8. This cranium is fully equal
in size to correspond with the huge specimens of fore and hind legs
from Sejouni, in the Museum, presented by Dr. Spilsbury. This speci-
men was found at Brimhan Ghat.— Vide Journ. As. Soe. vol. xii. p. 165.

No. 54. Elephas Namadicus.—Rare specimen, comprising a large
portion of the sternum of enormous size. The mass is wedge-shaped

ELEPHANT AND MASTODON. 117

in section, with a deep keel below and broad above, where it shows two
pairs of costal articular surfaces, and the commencement of a third
pair, the bone being there broken off. The uppermost pair of discs
appear to represent the junction with the first pair of ribs? The
specimen was enveloped in a mass of calcareous sandstone, which left a
small portion exposed. The episternal portion is wanting.

Dimensions.
Inches
Length of the fragment . : : ; é 0 : eliza
Depth at anterior end . : : ; F ; 5 : . 63
Ditto in the middle , : ‘ 3 F : F ‘ wore
Width of upper surface, anteriorend . i : . . 49

The upper (vertebral surface) forms two hollows bounded by ridges,
i.e. two sternal elements.

No.1. Elephas insignis.—Lower jaw, left side, comprising the anterior
portion of the horizontal ramus, truncated in front of the symyhysis,
containing six ridges of a molar tooth, the three anterior of which are
worn, the rest intact; broken off behind near the commencement of
the ascending ramus; covered over with soft sandy matrix. Vide
Journ. As. Soc. vol. xix. p. 489. This specimen is in the usual mineral
condition of the Nerbudda fossils as contrasted with those from Ava and
the Sewalik hills, viz. being white, soft and friable, and adhering to the
tongue, there being no ferruginous or calcareous infiltration. It is of
great importance, as being the only specimen of this Sewalik Stegodon
from the Nerbudda in the Asiatic Society’s collection.

IJ.—Descrirptions py Dr. Fatconer or Fossiz Remarns oF Mastopon
In Museum or Asiatic Society oF BENGAL.

A. From the Sewalik Hills.

No.1. Mastodon Sivalensis.—Fine specimen of the upper maxilla
left side, comprising the greater portion of the palate and two molars
embedded in the jaw, with four empty pits, marking the situation of
the fangs of the second milk molar which had fallen out. The third
milk molar is shown nearly entire, with the enamel crown broken off
at the outside of the first two ridges, and the first three ridges are seen
to be touched with wear forming depressed cups. The crown of the
tooth is bisected longitudinally into an outer and inner division, and
the groups of mammille are seen to alternate, instead of being trans-
verse with accessory mammille in the valleys (See Pl. IX. fig. 2).
This is the normal character of the species distinguishing it alike
from M. latidens and M, Perimensis (Pl. IX. figs. 3, 4, 5, and 6).
The surface of the enamel is deeply grooved vertically, so that
the ridges, when worn down, present a very complex pattern. Behind
the fourth ridge is a talon consisting of a complicated group of
small mammille. To the rear of this tooth, the anterior portion of the
first true molar is visible in germ, and the posterior part of it is still
concealed in the alveolus; it shows three ridges, presenting the same
complex form as the anterior tooth, but is very considerably larger in
every proportion. The four empty fang cavities in front are nearly
square, showing that the crown of the tooth had a similar form, viz. that
it was short and broad; these cavities are well apart from each other.

\

118 FAUNA ANTIQUA SIVALENSIS.

lal
16

“
a

Extreme length of fragment .

Height from molar surface in front to upper edge of fracture
Length of crown of third milk molar

Width of ditto in front

Ditto ditto behind ;

Ditto of back molar at third ridge

Length of alveolar space of second milk molar (fallen out)
Width of ditto in front ;

Ditto ditto behind

RRP DPN wr
Awp ae © ty &

This is a very important specimen from showing so much of the early
dentition. It is evidently a young MW. Sivalensis, but there is nothing
to indicate the locality whence it came.

No. 2. Mastodon Sivalensis.—Fine specimen showing the upper
maxilla, both sides, with the last milk molar and first true molar in
situ, closely resembling specimen No.1. The specimen is broken
lengthways along the palate into two fragments, and it is a good deal
covered with sandstone matrix. The anterior molar on the left side
has the greater part of the crown hammered off by attrition. On the
right side, what remains of it shows discs of considerable wear. The
back molar, on either side, is ina state of germ. It resembles No. 1
so closely, that further details are unnecessary.

B. From Ava.

No. 1. Mastodon latidens.—Fine specimen of the upper jaw, both
sides containing one entire molar in each, and the anterior ridge behind
of another molar germ. The specimen is broken off horizontally about
four inches above the base of the crown of the molars, showing the floor
of the nasal fossa. The palate.is perfect from the posterior border to the
commencement of the diasteme, where it is abruptly broken off. The
teeth are in the most perfect state of preservation; the enamel is thick,
rugose at the sides, and presents on the crown a clouded pearly
appearance; the anterior of the molars consists of four ridges and a
talon or heel ridge; the three anterior ridges are more or less worn,
the fourth ridge only slightly touched in the middle; each tooth
exhibits longitudinally a line bisecting it into an outer and an inner
half; the ridges are transverse, the furrow intervening being unbroken,
with a small mammillary process bounding the interior termination of
the furrow and interposed between the ridges. These mammillary pro-
cesses are chiefly seen in the first and third furrows; they are absent from
the second. Each ridge near the middle throws out one or two accessory
adpressed mammille behind, which encroach on the furrows without
disturbing their continuity, causing loops of wear when the ridges are
ground down; there are about six mammille to each ridge, the inner-
most of which j is the largest. The talon on both sides forms a rather
complicated ridge, more largely developed than in most other species
of mastodon. The first and second ridge of the left molar are worn
down, so as to show continuous depressed discs of ivory surrounded
* with a border of enamel. The third ridge shows three small discs, the
outermost mammilla being hardly touched; the fourth ridge and the
talon are hardly touched; on the right side, the anterior ridge forms
also a continuous depressed disc, the second ridge is distinctly bipartite
into dises of an irregular pattern, with a loop in each behind, caused.

DESCRIPTION OF PLATE IX.
Mastopon SIvaALENsIs AND Mastopon PERIMENSIS.

Figs. 1 and 2. Show in plan and profile the last upper molar of Mastodon
Sivalensis, one-third of the natural size, and copied from draw-
ings by Mr. Ford in Plate XXXVL, figs. 6 and 6a., of the
Fauna Antiqua Sivalensis. The tooth has six ridges and a hind
talon and shows well the alternate disposition of the crown-
mammille. There is a cast of the specimen in the British
Museum. (See pages 117 & 467, and vol. i. p. 29.)

Figs. 3 and 4. Show in plan and profile the antepenultimate molar of the
upper jaw, left side, of Mastodon Perimensis, one-half of the
natural size. The specimen was obtained from Perim Island,
and is in the Museum of the Asiatic Society of Bengal. The
figure is copied from a pencil drawing executed by Mr. Claude
Augier for Dr. Falconer, in 1856, and the identification of the
tooth is in Dr. Falconer’s handwriting upon the drawing. (See
page 122.)

Figs. 5 and 6. Show in plan and profile the penultimate upper molar,
right side, of Mastodon Perimensis, one-half of the natural size.
The specimen, like the last, was obtained from Perim Island,
was identified by Dr. Falconer, and is in the Museum of the
Asiatic Society of Bengal. The figure is copied from a drawing
by Mr. Claude Augier. (See page 122.)

VOL. I.

WAIL ‘Tep zeisy'9 9 Prog HO

mth a

age Os, oan yr ‘ey 4
rs : aly a

ELEPHANT AND MASTODON. 119

by the accessory mammille. The third ridge shows three discs, the
outer mammilla being nearly intact; the fourth ridge is barely touched
in the middle, and the talon is entire, The talon on the right side
shows seven mammille, large and small, that on the left only four.
The posterior molars are in a state of germ, and the first ridge only is
preserved on the left side; on the right side it is broken in the middle.
The posterior tooth is seen to be of much larger size than the anterior
one; the first ridge consists of about five mammille, with an accessory
mammuilla behind the first and second. A portion of the enamel border
of the anterior ridge of the left molar is broken off on the outside.

The enamel is very thick; there is no appearance of crusta petrosa
on the furrows. The rugosity of the surface runs in transverse wavy
grooves or meshes somewhat as in the enamel surface of Rhinoceros,
but more marked.

In front of the right molar, inner side, there is a nearly obliterated
pit of one of the fangs of the molar preceding it which had been shed.
There is no trace of this on the left side, where the molar is more
advanced in wear, having been protruded earlier.

The bony surface of the specimen when HS is generally tinged
of a brownish red colour.

This specimen is singularly free from miatrie of which not even the
smallest portion is visible upon the discs of wear, or in the crevices
between the teeth. In this respect it resembles a fresh macerated bone.
A black shining surface covers the upper side irregularly, upon which
thin gold leaf has been freely dispersed. This, when tested by a hot iron,
proved to be caused by a coat of Thee-tsee varnish so commonly used by
the Burmese for lackering. This was evidently an artifice to enhance
the supposed value of the specimen, which appeared to the Burmese
King worthy of being made _a royal present, and was so offered to the

Governor-General of India, by the Burmese Embassy at Calcutta, in
December, 1854.

Dimensions of Molars.
Right Left
Length :

A ° * A , . 6°35 595
Width of crown in front : . : - : . 3d 3°45
Ditto ditto behind . 3 . 5 . 365 3°6
Height of crown in front exteriorly . : 5 - 1:26 115
Ditto ditto interiorly . : 2 rds 0°85
Ditto ditto behind exteriorly . : 5 - 2°05 21
Ditto ditto ditto interiorly . - - 2 2S 176
Width of first ridge posterior molar. : A Ave 4:15
Exterior height of ditto ditto : n ie 2°3

Other dimensions.
Breadth across the two molars anteriorly - - : . 83
Ditto ditto posteriorly . ; ; 5 . 10°83
Width of palate between molars anteriorly . - . : - 15
Ditto ditto posteriorly . é : ; . 315

- Length of palate c : . : . 8&1
Extreme width of jaw behind - . 10°7
Greatest height from crown of molar behind to upper surface of

fragment . 6°5
Thickness about middle between surface of # palate and floor of
nares 0 A

2°35

120 FAUNA ANTIQUA SIVALENSIS.

No. 8. Mastodon latidens ?—Specimen consisting of the horizontal
ramus of lower jaw left side, truncated obliquely downwards and back-
wards in front of the molar and behind about the middle of the ascend-
ing ramus, of which it shows a part. The specimen is dark-coloured
from ferruginous infiltration and is sparingly covered with a thin crust
of gritty calcareo-arenaceous matrix, resembling that found on the sur-
face of many of the Ava fossils. The whole length of the molar tooth
is visible 7m situ, but unfortunately all the enamel surface of the crown
is broken off, so as to deprive us of any direct and conclusive evidence
as to the species, but it is evident that the tooth is the third or last true
molar, as the section behind shows no marks of the nucleus of another
successional tooth to come after it; but from the narrowness of the
crown, it is confidently inferred that the specimen belonged to a true
Tetralophodon mastodon (either M. latidens or M. Perimensis) and not
to E. (Stegodon) Cliftit. The jaw is characterized by its great transverse
diameter behind and great height in front. The removal of the sym-
physis has carried away the outer orifices of the mental foramina, but
two canals are seen in the section which also exhibits a nearly cylindrical
core, which is probably the residuum of the fang of a shed molar or a
tusk of the lower jaw?

Dimensions.
Inches
Extreme length of fragment from broken ae vid symphysis to
broken edge behind. 16°7
Height of jaw measured outside immediately i in front of molar 85
Ditto ditto behind . : 5 67
Greatest diameter behind, taken at bulge of ascending ramus 8°
Diameter in front in a vertical line with the anterior edge of the
molar . : : : . 54
Extreme length of alveolus and crown of molar : 7 4 : 10:
Diameter of crown of molar in front : ; : - ; 5 2-4
Ditto where widest behind ; : : : 7 : ; ; 2°8

The tooth narrows very much from behind backwards, as seen in the
section; the residuary portion of the crown of the molar viewed from
the inside in front is elevated 1 to 6 inches above the alveolar border.
There is no portion of the enamel visible except an edging on the
outer side.

The form of the lower jaw is much thicker and shorter in proportion
than in the existing Indian elephant, further supporting the conclu-
sion that it belongs to a true mastodon.

This is a valuable and characteristic specimen ; and it is much to be
regretted that there is no tracing its origin in the records of the Asiatic
Society. Is it one of Col. Burney’ s specimens from Ava?

No. 4. Mastodon latidens——Fragment of lower jaw left side, com-
prising part of horizontal ramus from base of coronoid apophysis to
about 64 inches forward, containing the posterior half of a well-
preserved molar, which is inferred to be the last or third true molar
from the absence of the indication of any other forming behind. It
shows four ridges with a talon, the three anterior of which are well
worn, showing that the animal was fully adult. The tooth agrees in
the closest manner in size, form, amount of wear and every other
respect with the corresponding specimen figured in Faun. Ant. Siv.
Pl. XXX. fig. 6, which, however, has a ridge more in front preserved.

ELEPHANT AND MASTODON. 121

The anterior ridge of the specimen is worn into a deep transverse disc,
of which the anterior boundary edge of enamel is wanting. The second
and third are also worn into continuous excavated discs; the fourth
shows the caps of three large mammille slightly touched, and the talon
consisting of three mammille in a transverse row is intact.

The accessory mammille described as bounding the valleys on the
inside of the Burmese Embassy specimen (No. 1) are shown of very large
size in this specimen.

Dimensions.

Inches
Extreme length of fragment c 0 c : : ' o GF
Height of section in front . : - : 0 ; . of Ue
Ditto ditto behind 6:2
Diameter behind : : : 2 : : ; : . 6:55
Length of fragment of molar , ; . : : : . 61
Greatest width in front 4 c 4:1
Height of enamel crown in front . : “4

The colour of the specimen differs from most of the Ava fossils in
absence of a ferruginous tint, and the enamel is greyish and pearly,
looking as in No. 1: very little matrix upon it.

No.5. Mastodon ?—Fragment of lower jaw, left side, divided
longitudinally nearly in the middle, jet black in colour, with a small
quantity of sandy matrix; no indication of a tooth or alveolus but seg-
ment of a canal (dentary?) running longitudinally near the inferior
border. The fragment is in two joined pieces.

Extreme length, 114 inches.

Identification very obscure.

No. 6. Mastodon latidens.—Fragment of molar divided vertically
and lengthways, consisting of the back part with three well worn ridges
and a talon, and one great posterior fang. The end of the talon bears
upon its ivory surface the disc of pressure of a molar behind. The
enamel is very thick, and in every respect as in specimen No. 4. The
worn surface of the discs agrees exactly with those of the Ava speci-
mens figured in the Fauna Antiq. Siv. Pl. XXXI. figs. 7, 8.

Inches
Extreme length of specimen . b : . : . . 3d
Height from fang to surface of crown . p . : : . 3d

No.7. Mastodon?—Fragment consisting of the last ridge and talon
of a molar tooth, with the mammille in germ and entire. The
mammillz numerous and closely aggregated together. Species in-
determinable.

No. 28. Mastodon.—Proximate phalanx of hind leg, very thick and
massive, highly coloured with iron infiltration.

No. 29. Mastodon.—Mutilated fragment of a large massive acetabu-
lum in two pieces, black and heavy like a mass of iron.

1 The engine turned-like marking, re- | ivory of the tusk does not differ in
sembling that of a watch-case, which | structure from that of the molar. The
Owen states to be ‘ peculiar to the tusks | term ‘dentine’ is therefore untenable:
of Proboscidian Pachyderms,’ is dis- | Vide Owen, ‘Odontography,’ p. 11. Intro-
tinetly visible at the posterior end of the | duct. and p. 627.
ivory in this molar, showing that the

122 FAUNA ANTIQUA SIVALENSIS.

No. 30. Mastodon.—Dorsal vertebra; apophysis broken off; body
compressed and showing very approximated costal articulations. Trans-
verse diameter of body 3-8 inches; antero-posterior diam. 3°5 inches;
thickness 2 inches. Very small for a proboscidean vertebra; highly infil-
trated with iron.

C. From Perim Island.

No.1. Mastodon Perimensis——Permanent molar, upper jaw, left
side, with four ridges complete, first ridge worn (antepenultimate).
Characteristic specimen. (See Pl. IX. figs. 3 and 4.)

No.2. Mastodon Perimensis.—Upper molar, right side, probably
penultimate, with four ridges, the first two worn. Characteristic

specimen. (See Pl. IX. figs. 5 and 6.)

No. 13. M. Perimensis——Molar, unworn, with four ridges and
talon ; anterior portion wanting; highly characteristic.

No. 16. Mastodon ?—Humerus, lower end, left side, huge size, in
two pieces. The largest known.

No. 17. Mastodon ?—First vertebra of neck (atlas) of large size,
very complete.

No. 18. Mastodon or Elephant.—Cervical vertebra, in two pieces ;
left half of neural arch broken off and lost, belonging to an animal of
enormous size. :

No. 19. Mastodon or Elephant.—Dorsal vertebra (anterior) nearly
complete. Spinous process of enormous size broken off; costal cup
present on left side.

No. 20. Mastodon or Elephant ?—Lower extremity of a femur, right
side, in two pieces, of very remarkable form, with a considerable portion
of the shaft attached. The condyles are covered with Perim matrix,
which has become partly engrained with the bone, but the general
contour is visible, agreeing closely with the elephants and mastodons.
The shaft, however, gradually contracts from above the condyles into an
attenuated and comparatively slender shaft, such as is unknown among
these types, fossil or recent. The section of the shaft is a compressed
oval, with a very sharp edge at the outer side. Can this be a femur of
Dirotherium? There are no ascertained remains of that genus in the
collection.

Dimensions.
Inches
Length of fragment : ; . 5 . . :
Greatest width above condyles : 4 : 5 5 : . Td
Antero-posterior diameter of condyle . : 5 : . > 6°
Transverse diameter of shaft at fracture ; c : : eal
Antero-posterior ditto . ; . . 4 4 : . - 25

No. 102. Mastodon Perimensis.—Fragment comprising a portion of
the horizontal ramus of the right lower jaw broken off vertically in
front of the penultimate molar, and behind at the commencement of
the offset of the coronoid process, showing the entire molar alveolus,
together with the hollow alveolus of another molar behind. The
molar unluckily has the crown very much mutilated, but it distinctly
shows the masses of four ridges, bounded by a large fang in front and

ELEPHANT AND MASTODON. 123

also behind. A depressed disc, bounded by a belt of enamel, shows
that the first ridge was partially affected by wear; and a similar cha-
racter shows that the second ridge was slightly touched. A great part
of the enamel on the right side of the ridge remains entire. The third
and fourth ridges are completely denuded of their shell of enamel, but
they were probably quite intact. The first and second ridges, which
remain, show the longitudinal bisecting line. The valleys are trans-
verse, but it is seen that they are interrupted in the middle by an
accessory lobule in front and behind each ridge. Their outer termi-
nation is bounded by a-large depressed mammilla exactly as in M.
latidens. The enamel is very thick, and shows deep rough lines of
longitudinal grooving. The mammille of the ridges appear very
high, and it is seen that the accessory lobules in the furrows form a
projecting loop to the disc of depression exactly as in M. Perimensis.
Dimensions.

I
Length of fragment

ne
8
Height from lower surface of ramus to alveolar margin, outside 7
Transverse diameter of ramus at posterior end of tooth ; SOs
Length of molar . i
Width about middle 3°
Height of crown at second ridge 2°
Nothing more than the cup of the last rable is owe behind: The
lower edge of the ramus is broken off, and shows a canal of about 1} inch
in diameter, running longitudinally. The nature of this is doubtful.
This jaw is seen to be of less diameter proportionally than the jaws
attributed to M. latidens, more especially the Ava specimen No. 3. There
is hardly any matrix to indicate whence the specimen came, and there is
nothing of the yellow marly conglomerate generally found on Perim Island
fossils. But the characters of the tooth are so close to those of M/. Peri-
mensis that it is attributed to that species. It is much to be regretted
that no record has been kept of the history of this specimen, or where
it came from. [After having gone over all the collections, the opinion
arrived at is that this and the next specimen, No. 103, are more likely
to have been from Ava than Perim Island.—25rd Feb. 1855. |

No. 108. Mastodon latidens.—Fragment of the horizontal ramus,
lower jaw, right side, broken off near the symphysis in front, and im-
mediately behind the molar in rear; broken also at the lower border.
It contains an entire molar zn situ, being the first of the true molars,
and the remains of a worn-out milk molar in front. The crown of the
tooth, which is well worn, shows four distinct ridges, the three anterior
of which are ground down each into a continuous broad disc, showing,
about the middle, the characteristic loop of enamel in front and behind,
which indicate the accessory mammille situated about the middle of
the valleys. The last ridge shows two discs, which are distinct,
although nearly confluent. The talon is barely touched by wear.

5
6

Inches
Length of fragment : F : a : 5 GF
Height ditto to crown of molar in front . 5 5 : . APT
Ditto ditto behind . : ; : : . 53
Length of molar . . : ; é 5 Ae ot
Width of ditto at second ridge : . : : . : 5 dk8
Ditto ditto fourth ridge : 2°2
Length of fragment in front of molar 1:7

124 FAUNA ANTIQUA SIVALENSIS.

No mark. There is some sandy matrix sparingly covering the
fossil, not like the ordinary Perim matrix. But the characters of the
tooth are entirely those of Mastodon latidens ; a younger tooth closely
resembling it contained in the Scinde collection, No. 2.

D. From Scinde.

Wo. 1. Mastodon latidens.—Fragment of large upper (?) molar,
comprising two ridges, little touched by wear. The only complete
furrow present is transverse without interruption, as in the Ava speci-
mens. The longitudinal bisecting cleft is very distinct. 'The weathered
appearance is peculiar, differing from anything else in the collection.
The teeth appear to have been embedded in a yellowish clay.

No.2. Mastodon latidens.—Third milk molar, lower jaw, left side (?),
showing the posterior two and a half ridges and the heel, all well
worn. The two last ridges are supported on a large fang. The an-
terior, with the whole of the first and half of the second ridge, are
broken off. The valleys are transverse, without interruption. Vide
Perim Island specimen, No. 103.

No. 3. Mastodon latidens.—Portion of a second milk molar, show-
ing two ridges and a low talon ridge supported on one fang. The
valley very open. Found in a low range of sandstone breccia, com-
posed of angular pieces of nummulitic limestone cemented with clay, at

Sehwan on the north side of the Jukkeo Hills.

III.—Mastopon (Tritopn.) Panpronis.!_ Description By Dr. FaLconer
or Foss, Moiars From THE DECCAN, PRESENTED BY COLONEL
Sykes To THE InprA House CoLuecrion.
(Extracted from Note-book: 24th December, 1856.)

No. 1.—The principal piece is a penultimate molar, upper jaw,
left side: so determined from comparison with a germ specimen from
M. Lartet of an antepenultimate. The crown of the tooth is perfectly
entire, the front ridge alone being a little touched by wear on the
inner side; but the fangs and base of the tooth are broken off right
across on a line with the termination of the enamel shell. It exhibits
three well-defined ridges, with a thick strong front talon, and a hind
talon confluent with the last ridge. Itis a true and unmistakeable
Trilophodon—the only one yet yielded by India, and very different in
its crown characters from all the Sewalik, Ava, or other fossil Mas-
todons of the East.

The general form of the crown resembles very strongly that of
Triloph. angustidens, the principal difference being that the ‘col’ of
vallecular flanking mammille is still more developed than in that
species. The crown is traversed, as usual, by an indistinct longitudinal
cleft along the axis, marking off an outer and inner division. Each of
the three ridges has the outer division simple and composed of a thick

' Dr. Falconer described the first { of Godavery.)
specimen as belonging to a new species, The second, he regarded as identical
Mastodon Andaranus. (From Pliny, | with a specimen of M. Pandionis in the
Nat. Hist. vi. cap. 19, ‘ Validior deinde | collection of M. Lartet, and its deserip-
gens Andare plurimis vicis, &e. The | tion is headed Mast. Andaranus, nune
‘Andhra’ race of kings (Wilson), south | Mf, Pandionis—[Ep.]

ELEPHANT AND MASTODON. 125

conical transverse mass, the summit of which is somewhat compressed,
and indistinctly bi- or tri-lobed, by corresponding longitudinal furrows.
The inner division is more massive and complex, each ridge throwing
out from its anterior and posterior surface—the former diagonally
forwards, the latter diagonally backwards—a ‘col’ of robust tubercles,
which meet in a chevron form in the middle of the valley, so that,
when the inner division of the crown is regarded in plan apart from
the outer, it presents, in conjunction with the outlying tubercles, a
series of zig-zags closely resembling the letter W. The complexity of
pattern is further increased by the salient apex of the connecting ‘ col’
being continued outwards towards the margin in a single line of cylin-
drical mammillz, which completely obliterate the bottom of the outer
half of each valley; while the inner half, corresponding with the re-
entering angle between the large inner cones, forms a gorge which is
entirely free from tubercles. The anterior talon forms a subordinate
ridgelet, which is thrown off in the usual manner from the anterior
portion of the inner cone, and is continued outwards towards the
margin, with less inclination downward than is ordinarily the case in
the other allied species. It is composed of about four robust com-
pressed tubercles, which are separated from the anterior ridge by a well-
marked chasm. The posterior talon consists of a cluster of indistinct
confluent tubercles thrown off from the posterior part of the inner
tubercle of the last ridge, and so adpressed to the ridge that it does not
yield the defined and separate appearance seen in the talon of T’riloph.
angustidens. After a little wear, the posterior talon would be involved
in the disc of detrition of the last ridge.

No. 2.—Among the teeth presented by Col. Sykes there is also a
small two-ridged Mastodon tooth, with very smooth enamel, which
in form, through every detail, agrees so exactly with a specimen of
Lartet’s, that I unhesitatingly consider them to be homologous teeth of
the same side, and nearly of the same age.

Dimensions of Premolar Lartet’s Indian

Extreme length : : : : : : 17
Width of front ridge é : : A . 11
Ditto of back ridge : 14

Lartet’s specimen is a detached tooth, labelled in his list, ‘ Last lower
premolar, left side.’ It consists of two ridges, both of them worn ; but
the outer and inner discs not continuous, and the middle of the valley
occupied by a tubercle, which is worn low down, leaving a circular
disc. There is a well-marked anterior talon, of two worn tubercles,
but deeply impressed by an anterior disc of pressure against a pre-
ceding tooth in position. Behind, there is also a talon, but very
strongly impressed by a disc of pressure, so that the posterior talon
only exists on the inner side. The crown presents the cucumber
shape, so remarked on by De Blainville—i. e. the first ridge is narrow,
the second broad.

The Indian fossil shows precisely the same characters—i.e. two
ridges worn, and two talons, with a connecting tubercle between
the ridges. The back talon is marked with a disc of pressure,

126 FAUNA ANTIQUA SIVALENSIS.

but the anterior talon consists of two confluent prominent tubercles,
free from any mark, showing either that there was no penultimate
premolar, or that it was very caducous, and dropt out without pressure
from behind. The anterior ridge is narrow; the posterior broad, as
in Lartet’s; but the ridges are more worn, and the discs confluent.
The crown slopes from the inside which is higher, to the outside
which is lower, but less so than in Lartet’s. The intermediate
tubercle is worn down as in his, and the posterior talon is only
exhibited free on the inner side.

The specimens are so exactly alike that they might have been
taken for the same species, but that the Indian is a little larger.

IV.—Nove on THE TEETH OF THE MASTODON A DENTS ETROITES OF THE
SEwaLik Hitus. By Caprain P. T. Cautuey.}

(Read at the Meeting of the Asiatic Society of Bengal, June 1, 18386.)

Without further preface I refer the reader to the 1st volume of the
Ossemens Fossiles, page 268. Figs. 1 and 2, Plate IV., under the head
of ‘ Divers Mastodons.’

These drawings were presented to Cuvier by M. Faujas, and the
fossil was found near Asti in Upper Italy.

Cuvier merely alludes to this fossil as one of the varieties into which
the true Mastodon a dents étroites passes by a greater subdivision and
an irregularity of position of the mammille; the proportions of length
to breadth of the tooth retaining their full and perfect character.

By comparing the accompanying drawings with the figures above
alluded to, there can be no demur, | imagine, in identifying the Sewalik
variety of Mastodon now under review with the Asti fossil. It remains
therefore simply to note the peculiarities in form of the tooth: al-
though it may be a point for consideration hereafter, whether, as the
character of the tooth is so marked, and its peculiarities so rigidly
adhered to throughout the whole of the remains found in the Sewaliks,
it may not be placed under a sub-genus, that of ‘ angustidens,’ with
the specific denomination of M. Sivalensis.

There is no cortical substance or crusta petrosa; the tooth consisting
of enamel and ivory only, the former being very thick and massive, as
is normal in the mastodons.

The coronal surface consists of a double line of conical and obtusely
pointed mammille: those on the external side being in most cases per-
fect, whilst those on the inner side are divided by a fissure or fissures
into two or three irregularly formed obtuse points. These mammille
are not, as in the true Mastodon angustidens, placed transversely or at
right angles with the line of surface, but meet each other from right to
left alternately, so that the furrow on one side is interrupted by the
mamumilla on the other; and the mammille onthe wholeline of tooth lock
into each other in the same way that two serrated edges opposed to each
other might be supposed to do, were they placed in contact. (Pl. IX. fig. 2.)

The outer surface of the enamel is smooth, and the space or furrow
between each mammilla both on the external and internal surface is
marked by a small tubercle, the presence of which however does not
appear to be constant.

1 Reprinted from the Journal of the Asiatic Society, vol. v. p. 294.—-[Ep.]

ee ee

eee

ELEPHANT AND MASTODON. 127

The surface of the tooth of the lower jaw wears obliquely and out-
wardly on the grinding surface, as in the ruminants, in which respect
it differs entirely from the elephants.

The wear of the coronals is marked at the commencement by irre-
gularly lobed figures, which, as the detrition advances, become con-
fused and gradually unite, until the mammille are worn away entirely,
when the tooth is left with merely a surface of ivory surrounded by
enamel.

The drawings are intended to represent the tooth at these different
stages;! from the state of germ, to the old and worn down, tooth,
showing the intermediate state of detrition at different ages.

I wish to draw attention particularly to the alternating position of
the mammille, which I consider to be the chief specific character, and
which is distinctly marked throughout the whole series; and, referring
again to the Asti fossil as figured in Cuvier, I think that a clear iden-
tification is established. (See Pl. IX. fig. 2.)

As my object in writing this note is simply to point out the dis-
tinctive characters of the teeth of the mastodon & dents étroites, which
has been found in the Sewalik hills, it is unnecessary to make any
further remarks until we can enter upon a general description of the
fossil mastodons and elephants of these hills; noting, however, that
from the half of a lower jaw of this species, with its ramus attached,
which is now in my possession, we may look forward to some pecu-
liarities of form, differing very materially not only from the fossil and
existing elephant, but also from the other species of mastodon.

Up to this period I am only aware of the discovery of two species of
mastodon in the Sewalik hills; namely, the variety of M. angustidens
which is the subject of this note, and the MW. Elephantoides of Clift.
The former is very rare, and the latter in very great abundance.

V.—Nore on Mastopons oF THE SewaLixs. By Caprt. P. T. Caut ey.?

In the present state of the researches into the fossil remains of the
Sewaliks, it will be interesting to note any discovery of peculiar in-
terest, without entering upon a description in detail. Such a descrip-
tion may, with propriety, be reserved, until the possession of a more
perfect and a more numerous collection of remains enables us to enter
upon the description with greater confidence: whilst, in the meantime,
to those who are interested in the study the periodical announcement
of progress made in our operations cannot be devoid of interest; under
this idea I did myself the pleasure of forwarding to your Society the
note on the dentition of.the Mastodon angustidens (variety of), and
now send you one on a skull of another variety of Mastodon which has
been lately received. The sketches are drawn on transfer paper, and
will, I hope, be intelligible.

Figs. 1 and 2 are representations of the fossil skull—Fig. 1 being the
front, and Fig. 2 the profile or side view. Figs. 3 and 4 are similar

1 The reader is referred to the de- ? The former is the Mastodon Siva-
scriptions by Dr. Falconer, in the Fauna | /ensis; the latter includes the Hlephas
Antiqua Sivalensis. Plates xxxii. to | insignis and EL. Cliftii of the Fauna An-
xxxyll.—[Ep.] tiqua Sivalensis.—[ Ep. |

* Reprinted from the Journ. As. Soc. Dec, 1836, vol. v. p. 768.—[Eb.]

128 FAUNA ANTIQUA SIVALENSIS.

outlines of the existing elephant, on a scale of one-eighth on linear
measurement.!

The fossil is exceedingly perfect in some respects. The left orbit
and maxillaries are as sharp and well defined as in the recent skull;
the frontal and nasals are tolerably perfect, the specimen is fractured
obliquely, removing the temporal swellings and diploe of the cranium,
together with the occipital condyles and foramen magnum; the curve
of the occipital on its external surface is however retained, and
although sutures are altogether wanting, and the alveoli of the tusks
are mutilated, the specimen may be considered as sufficient to give a
perfect idea of the form of the skull; and, as a form perfectly unique
amongst the proboscidean pachydermata, will be looked upon with
satisfaction by all those who take interest in the additions that have
of late years been so rapidly made to paleontology, and to the catalogue
of animals now no longer existing on the globe. The present skull
derives additional interest from its being so different from the only
type of the same genus or co-genus (for it may be permitted so to
designate the elephant) which has been left to us—so different, indeed,
as to completely modify the construction of the head, and the arrange-
ment of the muscular and fleshy matter that must have belonged to it.

Without entering into any minutie of detail on the peculiarities of
the head, of which the drawings will give a representation, and which
detail will be reserved until our collections enable us to bring under
one view all the varieties of this genus that the Sewaliks may contain,
it will be sufficient, in announcing this very interesting addition to our
cabinet, to draw attention to a few leading points.

In the skull of the existing elephant, the excess of longitudinal
measurement, over that in the contrary direction, owing to the great
development of the superior portion of the cranium, is one of the
most marked peculiarities of its form; the height from the external
nasal opening to the top or apex of the cranium is immense, although
undergoing modification from age; this excessive development not
being derived from any increase of size in the cerebral cavity, but from
a wide space composed of cellular bone or diploe, giving an external
and deep covering to all that space occupied by the brain; the size
of the orbit is small in comparison to the temporal region; the
large external nasal aperture is situated between the orbits; and the
front in the Indian species is slightly depressed. Now in turning to
the fossil, we find that the whole of these peculiarities are either
reversed, or modified in an extraordinary degree.

The elevated and massive cranium does not exist, the slope towards
the occipital and foramen magnum commencing from the top of the
external nasal opening and falling off to the rear in an abrupt angle;
the size of the orbit is large, and its encircling bones massive and
prominent; the space between the orbits to the front continued up
to the nasal opening is depressed to an enormous extent, and the two
lines of alveoli of the tusks strongly marked; the temporal fosse are
small in comparison to those of the existing elephant, and the
temporal bones, which although broken off in the specimen from
which the drawing is taken, exists in another skull in our possession,

The original figures have not been | Sivalensis is represented in Plate x.—
reproduced, but a fine specimen of M. | [Ep.]

DESCRIPTION OF PLATE X.
Mastopon SIVALENSIS.

Cranium of Mastodon Sivalensis, one-fifth of the natural
size. Copied from a drawing by Mr. Scharf in Plate XXXII.
of the Fauna Antiqua Sivalensis. The specimen is in the
British Museum. (See pages 129 & 464.)

vou. I. b

a ae

Vol.1. Plate 10.

G.Scharf del. J. Dinkel ith.

W. West imp.

Mastodon Sivalensis.

ELEPHANT AND MASTODON. 129%

appearing to be large and composed of cellular bone. The angle
formed by the tusks with the grinding surface is more obtuse than in
the existing elephant, and the form of head, instead of possessing the
proportion assimilating the skull of the elephant to that of man, may
be considered as nearly square, or perhaps possessing a breadth in
greater proportion than the length. The height of the maxillary
bones, which is great in the elephant, is here much exaggerated, and
the form and profile especially is so peculiar, that a glance at the
sketch will, by comparison with that of the existing elephant also
given, be sufficiently striking.

The suborbitary foramen is by no means large; the proportion of
diploe in the upper part of the cranium bears no comparison with
that in the existing elephant, these differences, combined with the
peculiarity of form and position of the external nasal aperture, may,
in ail probability, modify the extent to which this variety of Mastodon
was provided with a trunk; but to forbear from surmises or speculations
in the present imperfect state of the inquiry, it will be sufficient to
place this as a second to the angustidens formerly noted.

P.S.—A letter this moment received from Captain Cautley an-
nounces the discovery of a superb specimen of the Mastodon angusti-
dens, a skull with both lines of molars, palate, and one orbit entire.!
He adds: ‘We have much still to learn of these Mastodons. With
regard to the Mastodon elephantoides of Clift, there are evidently two
species, of the same character as to dentition, but with a remarkable
difference in the form of the cranium, one of which has the flat and
the other the elevated crown.’—Zd. Journ. As. Soc.

1 A fine specimen of Mastodon Siva- | copied from Pl. xxxii. of the Fauna An-
lensis is represented in Pl. x., which is | tiqua Sivalensis—[Ep.]

VOL. I. K

130 FAUNA ANTIQUA SIVALENSIS.

Iv. ON THE FOSSIL HIPPOPOTAMUS OF THE
SEWALIK HILLS.!

BY HUGH FALCONER, M.D., AND CAPTAIN P. T. CAUTLEY.
(Read 3rd February, 18389.)

From the abundant remains of this genus that have been
procured from the Sewaliks, and particularly the perfect
specimens now in our possession, we are at no loss in recog-
nizing the characters which distinguished the Sewalik species
so remarkably, not only from the existing Hippopotamus of
Africa, but also from the fossil species hitherto found and
described.

The great pointof peculiarity is that the Sewalik fossil has six
incisors of a character peculiar to itself, independent of the form
of the cranium, which differs very materially from other varie-
ties. The numerous fragments in our collection enable the pro-
portions of the bones of the head and face to be very tolerably
ascertained; and these, added to three nearly entire skulls,
one of which is that of an animal just approaching adult, and
the other two of a more advanced age, are so perfect as to
leave no doubt of the characteristic distinctions of one or
more new species.

To the fossil variety now to be described, we propose the
name Sivalensis, a name so far applicable as attaching it to
its locality, and commemorating the region in which its re-
mains have been scattered in such profusion.

In the African Hippopotamus, figured by Cuvier and so
fully described in the first volume of the ‘Ossemens Fossiles,’
we find the incisors consisting of four slightly curved teeth
in the upper, and in the lower jaw four straight teeth project-
ing forwards at an obtuse angle with the plane of the grinding
surface, the two centre ones being of considerably larger pro-
portions than the others, and being formidable weapons either
for tearing the roots and weeds, from which the animal de-
rives its nourishment, or for defence. In the fossil Hippo-
potamus before us, these large and powerful teeth are replaced
by others of a smaller size but in a greater number, there
being no less than six, those in the upper jaw being slightly

1 Reprinted from the ‘Asiatie Re- | potamus dissimilis will be found in the
searches, vol. xix. p. 89. Numerous | Fauna Antiqua Sivalensis (Plate lix.
drawings illustrating almost every bone | e¢ seg.) The illustrations in this re-
in the skeleton of the Hippopotamus | print are copied from the Fauna.—[Eb.]
(Hexaprotodon) Sivalensis and Meryco-

DESCRIPTION OF PLATE XI.
Hippopotamus (HEXAPROTODON) SIVALENSIS.

Figs. 1, 2, and 3. Upper, palatal, and lateral views of cra-
nium of Hippopotamus Sivalensis, one-eighth of the natural
size. Copied from drawings by Mr. Ford in Plate LIX., figs.
1, la, and 16, of the Fauna Antiqua Sivalensis. The
specimen is in the British Museum. (See pages 131 & 499.)

VOL. I.

Vol. I. Plate 11.

G.H Ford del. J Dinixel lith. é W.Westimp .

Hippopotamus (Hexaprotodon) Sivalensis.

HIPPOPOTAMUS. 131

eurved downwards, and those in the lower projecting forwards
(Pl. XII. fig. 3); the diameter of these teeth, which are cylinders
with truncated.ends, is less in the upper than in the lower jaw,
and the central teeth may be considered as being in some degree
larger than those on the right and left. When we advert to
the uses to which the incisive teeth of this unwieldy animal
are applied, the means of tearing up the food, and the sieve
to cleanse that food afterwards,! we see in this form of tooth
and this arrangement of the muzzle, an adaptation to the
wants as perfect as, although for defence less powerful than,
in the existing species. With the six incisors our fossil
animal has the canine teeth of the upper jaw with a reniform
outline in transverse section, whilst that of the lower jaw is
pyriform, or pear-shaped. The molars resemble those of the
existing species, and are numerically the same; the first milk
or deciduous tooth which, as in the horse, falls and is not
again replaced, is here also conspicuous.

In proceeding to a comparison between the fossil head and
that of the Cape Hippopotamus, we are at once struck with
the position of the orbit of the Sewalik fossil. Viewing it in
profile, the orbit is considerably more advanced, and the
general contour of the head thereby modified (Pl. XI. fig 3);—
taking a measurement from the posterior extremity of the oc-
cipital condyle to the anterior ridge of the orbit, and from that
point to the front of the muzzle, we have in the existing animal
a proportion of 3 to 5, and in the fossil 9 to 134, giving to
the orbit of the latter a more central position on the face; this
peculiarity leads to the muzzle and the zygomatic arch bein¢
separated by a hollow much more abrupt and much shorte.
on its antero-posterior line than in the Cape Hippopotamus
(figs. 1 & 2). The anterior termination of the zygomatic arch
on the malar angle is more acute, and the general form of this
arch more prominent. The temporal fosse are longer, and
the temporal apophysis in its descent to join the malar bone
is slightly inclined forwards, placing the posterior angle of
the zygomatic arch in a more advanced position, and more
in front of the occipital surface, than in the existing animal.
The occipital crest is also more elevated, and the general
appearance differs, owing to this position of the orbit;
which, as will be naturally concluded, leads to a different
proportion in the bones of the head, those of the cranium
being lengthened, whilst those of the face are shortened in
proportion respectively. In the suture separating the
temporal apophysis from the jugal, we see the same direc-
tion and inclination as in the existing animal.

1 Vide Lancet: Professor Grant’s Lectures.
K 2

132 FAUNA ANTIQUA SIVALENSIS.

We will here introduce the table of measurements, in juxta-
position with Cuvier’s of the Hippopotamus of the Cape and

of the European fossil.

Existing Hip- ee
Bopolaiiins Fossil Hippopotamus
Dimensions of Skull Africa Europe Sewalik Ist | Sewalik 2nd
Inches |Métres Inches |Métres Inches Métres| Inches | Métres
Length from the posterior
surface. of occipital condyle
to the alveolus of the mid-
dle incisors a ne <0¢ eee |22°6 | °566
Length from the upper mar-
gin of one orbit to the other,
to the rear. 9°85 | °249 |11°85| -300} ... eee [12-4 | °315
Ditto greatest width of zygo-
matic arches 15°75 | °400|17°7 | 450] ... --. [14:4 | *366
Width of head over the sub-
orbitary foramen. 4°75| °120| 5-7 | °145| 3°75| -95 | 4°34) -110
Height of ditto ditto from
the border of alveoli . 51 | :180) 6:3 | 160) 4°45} -114) 5°3 | +134
Distance of posterior extre-
mity zygomatic apophysis
of malar from sub-orbitary
foramen . 10°65 | *270|13-4 | °340 11:25] -285
From ditto to the middle of
occipital crest ; 10°25| °260|12-°6 | -320 10°26 | °260
Antero-posterior diameter of
orbits 2°8 | -070| 3°55] -090} 2°30} -058| 2°63) -067
Greatest interval between i in-
ner side of zygomatie arch
and surface of cranium .| 51 | :130| 5:1 | -130| ... «. | 3°8 | -096
Height of head from poste-
rior border occipital fora-
men totop ofoccipital crest | 5°5 | *140| 7°5 | -190| 5°7 | -144| 6-8 | -173
Width of head between in-
ferior angles of occipital
crests : - {11-0 | *280)12°8 | -325] 88 | -224| 9-7 | -246
Length of occipital foramen . 2°0 | 050) 2°6 | :066}| 2°0 | -050| 2°48) -062
Width of ditto -| 16 | :040] 1°6 | -040| 1°3 | 0382) 1:5 | :038
Length of line of molars 10-25 | -260 |11°85| °300|10°5 | -266| 9-48} -240
Distance between alveolus of
first molar and canine 4°39) PULON 627 | LE ZO Taos -. | L4 | 035
From summit of occipital
crest to alveolus of middle
incisives . |25°2 | -640/29°9 | 760] ... ee. (23°15 | ‘58S
From ditto to anterior mar-
gin of arbit ase) | | see --- | 9°0 | °228/10°2 | -250
From anterior margin of
orbit to alveoli of middle
incisive : =) 550 cae eee ach eve [12-4 | °314
Vertical diameter of orbits il emtors Sy meee eee | 24 | 06 | 2°3 | -058
Interval between alveolus of
first or deciduous molar and
middle incisors . 6:85) 2h70)| 58:3" | P20 «- | 38 | :096

Width of cranium in rear of
the frontal angle

3-7 | 7093] 4:4 | :112

HIPPOPOTAMUS. 133

The Sewalik fossil, noted as No. 1,' is a perfect skull, with
the exception of the incisive bones, and fortunately exhibits
the sutures on the upper surface ; a second specimen, consist-
ing of the occipital and parietal regions with the frontal as
far forward as the front of the orbits; and a third fragment,
consisting of the incisive bones and teeth with the anterior
extremity of the nasals and maxillaries, are those from which
we draw a comparison of the bones on the upper and lower
surface, and of the form and position of the molars.

On the upper surface of the fossil, the chaffron, instead of
running in a flat line, slightly concave as in the existing
animal, is considerably depressed in the region between the
orbits, the superior ridges of which are elevated in proportion,
and stand considerably forward on the cranium (PI. XI. fig.3).
From the remarks on the elongated form of the temporal fossa
it may be hardly necessary to advert to the similar extension
of the sagittal crest, which is proportionally longer and more
marked, with a greater elevation at its junction with the
occipital. The broken and fractured boundaries of the nasal
aperture in all our specimens of skulls will not admit of our
measurements extending to that point, but we are able, from
a fragment above referred to, containing the incisive bones
and nasal aperture, to note, that the nasal bones are advanced
as far forward as those in the living animal, so that a straight
line touches their anterior extremities, drawn from the front
of the canine alveolus on one side to that on the other. The
nasal bones do not expand so much towards the rear as in
the existing Hippopotamus, and that part connected with
the frontal is more blunt and rounded; the distance between
the nasal bone and the orbit and the lachrymal juncture is
comparatively larger. The lachrymals descend upon the
jugal much the same as in the existing animal, but they
appear to advance considerably more forward on the face,
the anterior extremity in conjunction with the nasal and
maxillary being exactly over the last vicarious molar, whereas
that figured by Cuvier represents this point as over the second
true molar. The sub-orbitary foramen is also more advanced,
and the hollow in which it is situated, formed by the bulge
of the jugal and canine alveolus is, as we before remarked,
more abrupt. The figure of the muzzle is very similar to
the African variety, with a modification in the form of the
incisives adapted to the particular form of the teeth (fig. 2).
The width of the muzzle is comparatively greater, but the sepa-
ration of the whole into four bluff swellings with the spaces

No illustrations appeared with the | short communication published in the

original memoir, but this want was sup- | same volume of the ‘Asiatic Researches.’
plied by Messrs Baker and Durand in a | [Ep.]

134 FAUNA ANTIQUA fIVALENSIS.

intervening for the incisive sutures is a point which has a
close resemblance in the existing animal.

The frontal angle is more acute in the fossil; the coronal
crest runs more obliquely backwards, and the antero-posterior
length of the frontal is twice as much as in the African.
From the rounded form of the nasal suture in its contact
with this bone, the anterior part of the frontal forms a tongue
bounded by the lachrymal in front and by the nasal and orbit
on the two sides. From the depth of the temporal fossa, as
in the existing animal, the width of the cranium is somewhat
less than that of the muzzle over the sub-orbitary foramina,
and the interval between the inner side of the zygomatic arch
and the surface of the cranium is somewhat less than the
width of the cranium.

On the lower surface we are, unfortunately, not so well
provided with sutures to guide us in our comparative dimen-
sions; for, with the exception of those between the lines of
molars which are in themselves not very distinct, there are
none whatever. The position of the bones in rear of the
palatal sinus appears to correspond with that of the existing
animal, although the relative dimensions and proportions
will, it is supposed, be modified by the peculiarities described
in the upper surface and dependent on the lengthened form
of this region. The basillary mastoid apophyses and the
slightly concave surface of the gienoid cavity appear to
resemble those of the Cape Hippopotamus; this latter cavity
is more in rear of the most salient projection of the zygo-
matic arches than in the living animal. In the form and
position of the molars the only remark that may be made is
on the non-parallelism of the lines. Cuvier describes those of
the Cape Hippopotamus as parallel but slightly curving out-
wards towards the front (wn peu ecartées en avant). We see some
difference in our different specimens, but in all there is a curyv-
ing outwards both in front and rear, the middle of the palate
being the most contracted (Pl. XI. fig. 2). This curving out-
wards is most shown towards the front, where the lines of
molars appear to attempt a parallelism with the outer line of
the maxillary bone, instead of running parallel to each other.
The space between the most advanced molar and the canine
is very much smaller in the fossil than in the existing animal,
a point that may depend, perhaps, on the substitution of the
six small incisors requiring but small alveoli; for the large
ones (especially the two central) require a much larger sur-
face and a much greater depth, to admit of their being securely
fixed. The palate is, as in the living animal, marked by a
deep fissure in front, between the incisive bones; and the
suture appears similar, though this is not very distinct in the

Fig. 1.

Fig. 3.

Fig. 4.

VoL.

DESCRIPTION OF PLATE XII.

Hippopotamus (HExXAPROTODON) SIVALENSIS.

Right side of upper jaw of Hippopotamus Sivalensis, showing
the three true molars, and the second, third, and fourth pre-
molars, one-half of the natural size. Copied from a drawing by
Mr. Ford in Plate LXIL,, fig. 1, of the Fauna Antiqua Sivalensis.
(See page 136.)

. Anterior portion of lower jaw of Hippopotamus Sivalensis, show-

ing six horizontal incisors of nearly equal size, two canines, and
the anterior premolars, one-fourth of the natural size. Copied
from a drawing by Mr. Ford in Plate LXL, fig. 7, of the F. A. S.
The specimen is in the British Museum. (See page 136.)

Another view of same specimen taken from one side, showing

the peculiar curve upwards of the canines.

Descending process of ramus of lower jaw of Hippopotamus
Sivalensis, one-fourth of the natural size. Copied from a drawing
by Mr. Ford in Plate LXL, fig. 9, of the F. A. 8S. (See page
135.)

1.

(
i
i
4
y

; VoL 1. Plate at

—

eee

GH Ford ddl. J-Dinkel Lith. WWest imp.
Thppopotamus (Hexaprotodom) Sivalensis.

———————

HIPPOPOTAMUS. 139

fragment from which we draw our comparisons; the two in-
cisive holes are very distinct, but those referred to by Cuvier
as commencing on the edge of the maxillaries in a small
channel and terminating on the incisives by another hole, are
not so distinctly marked, although it is by no means impro-
bable that in clearing the fossil which is embedded in a hard
and crystalline sandstone, the two holes have been made into
one; we have before noted the fissure separating the incisive
bones, and those (not so strongly marked but equally open
outwardly) of the junction between the incisives and maxil-
laries, or that space between the canine and the third incisive.
The extremity of the muzzle in front of the two canines forms
part of a circle; if this segment be divided into seven equal
parts, and one part given to each echancrure (of which there
are three), and two parts to each of the incisive bones con-
taining the alveoli of the incisors, a tolerable idea of the
proportions of this region will be obtained. The incisors of
the upper jaw, as before remarked, are in diameter smaller
than those of the lower; they project but slightly from the
alveoli, are directed downwards, and obliquely truncated on
their internal faces.

It now merely remains with us to compare the occipital
face with that of the African animal, which may be best done
by a reference to our table of measurements. We note, how-
ever, the great difference in the proportions in breadth to
height, which in the latter animal are as 2 to 1, whereas in the
Sewalik fossil the proportion is as 3 to 2, showing, as was
before remarked, an increased height of the occipital crest.
To proceed, therefore, to the lower jaw :—

In comparing the lower jaw with that of the existing animal,
independently of the additional incisors, we have a marked
differenceand distinction in the form of the ramus, the enormous
descending process of which is, if anything, more extravagantly
developed (Pl. XII. fic. 4). This strange appendage peculiar
to the genus, and formed for the attachment of the masseter
and temporal muscles, is here of a form less tapering and
more deep and massive in its proportions than in the existing
animal; the posterior margin is more round and the anterior
or that descending from the base of the maxillary bone,
which in the existing animal is curved and pointed forwards,
is here blunt and unmarked by any peculiarity of form. This
angle is inclined outwards, and the outer surface is as de-
pressed for the reception of the muscles as that of the living
Hippopotamus. We observe no increase of height in the
coronoid process, but it differs from the living animal in not
being projected so much forward. There appears to be no
difference in the condyles nor in their position with reference

136 FAUNA ANTIQUA SIVALENSIS.

to the form of the jaw; the line of the grinding surface (the
specimen from which we draw this description is a lower jaw
joined at the symphysis, and only broken at the posterior ex-
tremities) is inclined to the outwardly curved direction, described
as a peculiarity in the upper surface (Pl. XII. fig. 1) : the teeth
do not appear to differ from those of the animal now living,
but the space between the front molar and the canine is, as in
the upper jaw, more contracted (fig. 2). The canines protrude
from the alveoli considerably, in a curve slightly inclined back-
wards at the point (fig. 3), which is obliquely truncated on the
internal surface, from the root or point where it leaves the
alveolus to the tip. The space for the incisors and the in-
cisive teeth themselves differ, as was before remarked, from
those of the existing animal, the large central incisors of which
are here replaced by much smaller ones. The number of in-
cisors in the fossil is six, of nearly equal dimensions, cylindrical,
inclined outwards at an obtuse angle to the plane of the
grinding surface, and sharply truncated at the internal side
at the point (fig. 2). In taking the dimensions of the incisive
teeth of the upper and lower jaw from two specimens of
adult animals we find their proportions as follows :—

: Inches Metres
Diameter of incisor—lower jaw . : : ; 0-9 0-022
Ditto ditto upper jaw . : . : 07 0-018

Tf there is any fixed difference in the size of the teeth of
each jaw, it exists in the second incisor being a little less
than the others. It may be necessary to note here with re-
gard to the number of molars in the lower jaw, that amongst
the great number of specimens before us of animals of all
ages, we see no mark or vestige of the first milk tooth, or
that which, as was mentioned before, falls and is not replaced ;
and the space between the adjacent molar and the canine is
so contracted as hardly to admit of room for another tooth ;
but as this tooth exists in the upper jaw in every specimen in
our possession, we may infer that its non-presence in the
lower jaw is accidental.! In viewing the lower jaw in profile,
we see that the anterior angle below the canines is somewhat
more abrupt, and more inclined to the form represented as
belonging to the European fossil species; the depth of the
inferior maxillary is more regular, and the form of the pos-
terior branches, as before described, very different. The lower
surface exhibits a width of symphysis equal to that of the
living animal, and the angle formed by the branching off of
the two sides is also similar. The width across the muzzle
from the exterior side of the canine alveolus to the other is

1 See Description of Plates in Fauna Antiqua Sivalensis (PI. Ixii. figs. 2and 3).—[Eb. ]

HIPPOPOTAMUS. 137

comparatively greater in the Sewalik fossil, and the extreme
width of jaw over the penultimate false molar less. It will
be seen that these differences of form correspond with those
of the skull, the advanced position of the orbit and the con-
traction of the sinus, in which the infra-orbitary holes are
situated, leading to a modification in the whole form of the
erinding surface.

Having made the comparison with the Cape and existing
Hippopotamus, we will cursorily note the differences that strike
us when comparing it with the fossil described by Cuvier as
belonging to the cabinet of the Grand Duke of Tuscany, and
figured in the first volume of the ‘Ossemens Fossiles.? The
distinctive differences will be, perhaps, best observed by a
reference to the table of measurements; we see, however, that
our fossil in the gradual slope of the malar process towards
the cheek corresponds, but differs completely in the hollow
formed, at this point, between the jugal bone and canine
alveolus, which in our fossil is more abrupt and marked. The
length of the parietal region of the European fossil is even
less than that of the existing animal, and its proportions
relatively with the bones of the face less. In the Sewalik
fossil the advanced position of the orbit completely modifies
the whole form, and, by equalizing the proportions of the
anterior and posterior divisions, gives a new style of appear-
ance to the cranium. In the fall of the occipital crest towards
the region between the orbits and a consequent height of
occipital surtace, the Sewalik and Florence fossils agree.
In the proportion of the frontal surface to the area of the
rest of the skull, the resemblance also holds good; but we
have the same difference in the relative position of the
canines to the molars; the Florence and African species
corresponding in this respect. The grand distinction of the
incisives and canines, both in form and number, is peculiar to
the Hippopotamus Sivalensis. In the lower jaw, the space
between the two branches and the angle which is internally
formed by them do not resemble those of the Florence
fossil, but, as we before remarked, are more assimilated to
those of the existing animal, in being round at the angle, and
the whole interval space being more open: the descending
process of the ramus differs, as explained before; and the
form of the anterior angle of the jaw below the canines is
somewhat similar and not so gradually rounded off as in the
living animal. The difference in size and number of the in-
cisors leads to a difference which, as before noted in the
comparison with the living animal, needs not be made the
subject of further remark here.

With the Hippopotamus Sivalensis and that figured in the

138 FAUNA ANTIQUA SIVALENSIS.

‘ Reliquie Diluviane,’ described as found in a peat bog in
Lancashire, and of which a drawing is given in Professor
Buckland’s work alluded to, little resemblance is recognizable.
The Lancashire fossil has the four incisors, with a lower jaw
of proportions apparently quite unique, and with a prominency
of arch in the nasal bone equally so. We may, however, re-
mark the elevated occipital crest, and the fall towards the
space between the orbits which exists in the Lancashire
fossil, as this appears to be general in the fossil species, re-
lieving the head from that straightness of chaffron which
is noted as one of the peculiarities of the African Hippo-
potamus.

Having concluded our remarks regarding the Hippopotamus
Sivalensis, we now come to another and a smaller species of this
genus, which appears to have been less numerous, but with the
remains of which we are sufficiently provided, although in the
possession of only two fragments; one the imperfect skull of
an old animal with the teeth much worn, and the other the
right side of the lower jaw, showing an unusual contraction or
narrowness in the symphysis; this latter fragment contains five
molars, the rear one perfect, and the last false molar sufficiently
marked to establish the age of the animal; this was past adult,
the first and second advanced cylinders of the rear molar being
worn, and the third or rear one in the state of germ, but fully
out of the alveolus. The form of this tooth differs from the
ereat Hippopotamus in the absence of the trefoil, the wear of
the coronals of each pair of collines taking a crescentic form
outwards, not unlike that of ruminants,’ the grinding surface
sloping outwards, very similar to the description given by
Cuvier of the Hippopotamus minutus (Pl. XIII. figs. 1 & 4).
The form of the jaw, however, is peculiar, the marked features
consisting of a general slenderness of proportions, and an
inequality in the depth, which being contracted at the
point of the descending process, gets gradually deeper, and
diminishes again still more gradually up to the symphysis
(Pl. XIII. figs. 1 & 2): in the great Hippopotamus we have a
straight, thick, massive jaw. The foramen for the artery
distinctly exhibited in the fossil enters just behind the last
tooth on the internal face of the ramus, and shows itself

1 This ruminant character of the teeth
afterwards induced Dr. Falconer to con-
stitute this species intoa new subgenus,
under the designation of Merycopotamus.
In a letter to Captain Cautley, dated
January 27th, 1844, he writes thus: ‘I
intend bringing out the Hippo. dissimilis
as a new genus, under the name of Mery-
copotamus, merico having reference to

the ruminant character of the teeth’
As Merycopotamus the species is figured
in the Fauna Antiqua Sivalensis, in Plate
Ixviii., of which the resemblance between
the teeth of Merycopotamus and those of
Anthracotherium is pointed out. Subse-
quently, Dr. Falconer distinguished two
species of Merycopotamus. See yol. ii. —
[ Ep. ]

7

DESCRIPTION OF PLATE XIII.
MERYCOPOTAMUS DISSIMILIS.

Figs. 1 and 2. Palatal and upper surfaces of a cranium, in the British
Museum, of Merycopotamus dissimilis, one-third of the natural
size. Copied from drawings by Mr. Ford in Plate LXVIL,
figs. 3 and 3 b, of the Fauna Antiqua Sivalensis. (See pages
138 & 506.)

Figs. 3 and 4. Lower jaw, right side, seen in plan and profile, of Mery-
copotamus dissimilis, one-third of the natural size. Copied from
drawings by Mr. Ford in the F. A. S., Plate LXVIL, figs. 4
and 4 a. The specimen is in the British Museum. (See pages
138 & 506.)

[In page 138 it will be noted that the figures are not cor-
rectly numbered. |

VOL. I.

Vol. I. Plate 13.

GH Ford del, J:Dinizel ith WWest imp.
Merycopotamus dissimilis.

F r ” 2
ad : 5
:
\
‘
’
4
- =
. f
foi
re

HIPPOPOTAMUS. 139

again on the opposite side just between and under the fourth
and fifth molar, in a markedly large hole from which, to the
space between the tusk and the most advanced molar, there
is a deep channel or indentation running upwards in a curved
line parallel to the lower face of the jaw. The anterior and
posterior portions of this beautiful fragment are unfortunately
wanting, but a small part of the symphysis, at which point
the fossil terminates, is distinctly marked, as well as the
transverse section of the canine or tusk which, as in the large
animal, is pear-shaped. A considerable portion of the anterior
extremity i is wanting, and with the tusk the fracture shows
only one alveolus or “hollow for an incisive tooth ; the exist-
ence of two, however, can hardly be doubted, but the
narrowness of the front may make a greater number than
four between the two canines problematical.! The ramus of
this specimen is strongly marked on its anterior part by an
elevated ridge pointing angularly forwards, and pushing
forward a nearly flat surface to the centre of the rear tooth ;
the descending process is unfortunately too much broken to
allow of our speaking decidedly, but the angle of departure
from the straight line of the jaw is abrupt (Pl. XIII. fig. 3).
The other remains of this smaller species to which we have al-
. luded consist of a skull, the front and rear of which is broken
off, and one line of molars with the palate only perfect. The
superimposed cranium would appear to be contorted by pres-
sure, as is by no means uncommon, but this circumstance
would lead us to refrain from an attempt at characterizing
its peculiarities. The molars consist of the three rear per-
manent ones, and the last false molar, this latter one exhibit-
ing the crescentic form of wear on its coronal surface, described
as peculiar to the first fragment. The other molars are much
worn, and, therefore, with the exception of the encircling
ridge of enamel, we have not those flexures which would
have brought us to a correct conclusion. These molars are
remarkably broad in proportion to their antero-posterior
dimensions, and have an oblique grinding surface, as before
described in the other fragment. We may remark, that,
should these two remains belong to a small Hippopotamus of
the same species, the great difference in the breadth of the
erinding surface in the upper and lower jaws, as marked as
it # in the Rhinoceros, would establish a species with (in this
respect) rather unusual peculiarities. To this smaller species
we propose the name of dissvmilis, from the differences of
form from the rest of the genus.

From the above additions to the species of the Hippo-

1 Subsequent specimens showed that there were six incisors. See Plate xiii.
fig. 4.—[Ep.]

140 FAUNA ANTIQUA SIVALENSIS.

potamus, and from the marked distinctions in the incisive
apparatus of the Hippopotamus Sivalensis, we shall, perhaps,
be justified in at once establishing a new subgenus in this
genus of mammalia, fixing the subgeneric characters on the
incisive teeth. So marked a distinction in the form, number,
and character of the incisors, will, we imagine, admit of such
an arrangement, with every advantage to science, and in
taking this step we place the new subgenus in the following
position and order ! :—
Genus—HIPPOPOTAMUS.

1st Subgenus—HEXAPROTODON.
1. Species—H. Sivalensis (Nobis).
2. Species—/. dissimilis (Nob.). An hic, vel infra, potius referendus ?

2nd Subgenus—TETRAPROTODON.
. Species—H, amphibius. Recent.
. Species—H. antiquus (Cuv.). Fossil.

1
2
3. Species—H. minor (Cuv.). Fossil.
4. Species—H. medius (Cuv.). Fossil.
5

. Species—H. minimus (Cuv.). Fossil.

The specific characters of the first species of our new sub-
genus are as follows :——
Genus—HIPPOPOTAMUS.
Subgenus—HExaProTopon.
Species—Sivalensis.
Cuar.—H. dentibus primoribus utrinque sex, subequalibus ; la-
niariis difformibus : superioribus nempe quoad sectionem transversalem

reniformibus: inferioribus pyriformibus; cranio elongato; oculo ad
medium caput feré attingente ; facie ad latera valdé sinuata.

Before closing this paper, we may make a few general re-
marks on the remains of this genus, which, with the exception
of the Mastodons and Elephants, are by far the most numerous.

As may be imagined in such an extensive collection we
find the remains of animals of all ages, with teeth in every
variety and state of detrition; from the young animal with
the complicated and triple cylindered milk tooth, to the old
and worn down molar without any mark of the trefoil, and
with a simple encircling ridge of enamel. In the fossil skull
described as approaching adult (from which the measurements
noted as No. 1 have been taken) we have a beautiful exhibi-
tion of the teeth in that state when the animal has just lost
its last milk tooth, and the new molar or ‘dent de remplacement’
is just showing itself in germ, whilst the last permanent molar,

' See antea, page 21; also Synopsis of Hippopotami in vol. ii—[Ep. ]

a a a ea en

HIPPOPOTAMUS. ° 141

or that most posterior, is in the same state of advancement,
having just pierced the bone: the oldest tooth in the head or
the first permanent molar is just worn to that state when the
development of the trefoil crown is most perfect; the second
permanent molar is just showing this appearance on its two
front pillars; the front false or pointed molars are unworn,
and exhibit in all their perfection the richly embossed surface,
which is peculiar to these teeth in the Hippopotami. The
first false molar or milk tooth seems to have retained its posi-
tion in many of our fossils, long after the fall of the other
milk teeth, and long after the arrival of the animal at the
adult state. In some of our skulls, which are the remains of
very old animals, we observe the alveolus of this tooth very
- distinct, and having the appearance more of having been
broken off in the fossil than of having been lost previous to
the death of the animal, in which case, moreover, a filling in
of the pit from the growth of the bone would be more or less
evident in the fossil. From the natural wear of the tusks
upon each other, the truncated extremity of the upper
one, which in the Hippopotamus Sivalensis is described as
reniform, occurs on the convex or outer side of the tusk;
and this must be the case wherever the tusk belongs to the
upper jaw. Amongst a very extensive and very large collec-
tion, containing, as we before remarked, three perfect skulls,
with a number of fragments of nearly perfect lower jaws, with
a great number of pieces of both more or less mutilated, the
reniform tusk is an invariable appendage to the upper, and
the pyriform to the lower jaw. Our collection, however, ex-
hibits one solitary instance of the anterior extremity of a
reniform tusk truncated on the inner or concave surface ;
this, unfortunately, is a separate fragment, unattached to
any portion of the jaw, and bearing in itself no further mark
of its having existed in the lower jaw than this truncation of
the extremity. It is difficult to imagine any fortuitous cir-
cumstance that would have produced such an anomaly, and
it is at the same time difficult to come to a conclusion con-
trary to the facts elicited by such an extensive collection of
remains, in which we see no sign of the reniform character
of the canine in the lower jaw; should the truncation alluded
to not be accidental, or be caused by some deformity in the po-
sition in the alveolus, we have yet to discover a variety of the
Hippopotamus with the reniform tusk in the lower jaw. The
fact of the existence of this fragment, however, may be as
well noted, as we observe peculiarities of form in other frag-
ments of the bones of the head that may ultimately prove to
belong to different species. We have contented ourselves
with drawing our comparisons from the bones of the head,

142 * FAUNA ANTIQUA SIVALENSIS.

without any reference to the osseous structure generally of
the animal, in which our collections, however, abound, espe-
cially in vertebree and the solid articulating extremities of
the bones. A more lengthened period of search and exami-
nation will add much to the value of an inquiry upon this
point, and a comparison with the actual bones of the Cape
Hippopotamus, instead of Cuvier’s drawings, will render any
attempt at a discrimination of existing differences easier, and
when completed and worked out, doubly valuable.
NortHern Doss: Nov, 15, 1835.

APPENDIX TO MEMOIR ON HIPPOPOTAMUS.

J.—Descriptions By Dr. FALCONER OF SPECIMENS OF Fosstt Hippo-
POTAMUS IN THE MusEeuM oF THE ASIATIC SOCIETY OF BENGAL.

A. Specimens of Hippopotamus (Hexaprotodon) LIravaticus, from
Burmah.

No. 803.—Lower end of left radius detached from the shaft and
ulnar extremity broken off; of small size, and corresponding very
closely with the specimen figured in the ‘Fauna Antiqua Sivalensis,’
Pl. LXV. fig. 18. This specimen is notable as being one of the very few
referable to Hippopotamus contained in the Ava collection.

Supposed to have come from Ava, from the ferruginous matrix, and
from differing from the Sewalik species.

No. 807.—Lower end of left femur, showing the articulating con-
dyles and a part of the shaft—both much weathered and the surface
abraded, so as to render the character indistinct. The bone is pro-
portionally of much smaller size than Hippopotamus Sivalensis, and
wauld thus agree with the dimensions of the radius No. 303, and with
specimens assigned to Hippopotamus Jrdvaticus in the ‘ Fauna Antiq.
Sival.’ The shaft is slender.

The specimen is believed to have come from Ava, being black from
iron infiltration, and very heavy.

B. Specimens of Hippopotamus (Hexaprotodon) Sivalensis. Mostly
from the Sewalik Hills.

No. 193.—A nearly entire cranium, in three pieces, showing the
occiput and occipital crest, the sagittal do., and the whole of the chevron
region on to the border of the incisives. Both zygomatic arches broken
off, and the rim of both orbits damaged. These parts show marks of
recent fracture, and were probably perfect when presented. The true
molars of both sides broken off. The four premolars on the left side
are present, and the three posterior premolars on the right side nearly
entire, and with the crowns much worn, showing the animal to have
been old. A section of the right canine is also seen zn situ. This is,
on the whole, a very perfect specimen—nearly as much so as fig. 1,
Pl. LIX. of ‘Faun. Antig. Siv.’ This subdivision of the genus is dis-
tinguished by the length of the sagittal crest and the slight projection of

HIPPOPOTAMUS. 143

the orbits above the plain of the frontal ; both of which characters are
well shown when the skull is compared with Hipp. Palwindicus.
‘Nerbudda Cat.,’ No. 17. The specimen is partly covered with hard
sandstone matrix. No mark, but inferred to be Sewalik, and probably
from Col. Colvin.

No. 194.—Cranium broken across near the constriction in front of
the orbits, but very perfect behind, showing the zygomatic arches,
temporal fossz, occiput, and condyles; the orbits a little broken in
their rim; molar teeth protruded, but broken off. This must have
belonged to an animal of large size, as the distance between the outer
surface of the zygomatic arches is 13-7. inches. No mark or number,
but embedded in hard sandstone, and inferred to belong to the species,
from the length of the sagittal suture and the little prominence of the
orbits above the plain of the frontal. Inferred to be Sewalik.

No. 196.—Like the last. Occipital portion more perfect, with edge
of crest preserved. The sagittal crest and both zygomatic arches
broken off. Right orbit a little mutilated behind, but showing a large
lachrymal tuberosity. Left orbit nearly entire behind, showing the
whole of the rim. Four molars on the right side in the last stage of
detrition, showing the animal to have been extremely old. Covered
with sandstone matrix.

No. 197.—Another, like the last; mutilated in front of orbits, but
showing the tympanic fosse. The last true molar on either side is not
much emerged above the alveolus, and the sutures are open, showing
the animal to have been young. This specimen is more mutilated
about the orbits than the three preceding, and the temporal fosse are
filled with hard sandstone matrix.

Wo. 198.—Another cranium, nearly like 194, with two zygomatic
arches and orbits. The last true molar on each side emerged but

unworn. All the cavities filled with hard sandstone matrix. A young
adult.

No. 199.—Cranium mutilated at the muzzle in front of the second
premolar, and with the parietals and occiput obliquely broken off
towards the left side, so as to expose the cerebral cavity. Right zygo-
matic arch and glenoid surface entire; orbits a little damaged in the
rim; nasal bones stove in by pressure. The teeth on either side are
all present, except the anterior premolar. The last true molar is in a
state of germ; the penultimate shows well the basal cingulum of
enamel. ‘The animal was a young adult. ‘The cavities and fosse are
filled up with a hard gritty sandstone. This specimen shows the cha-
racters of the upper molars better than any of the five preceding
specimens (194 to 198).

No. 200.—Mutilated cranium, broken off behind, about the middle
of the temporal fosse, and in front, through the first true molar. Zygo-
matic arch present on the right side, with the greater portion of both
orbits. The three last molars on the left side present, and the two last
on the left; the former well worn, showing the animal to have been
old. Hard sandstone filling the cavities and fosse.

144 FAUNA ANTIQUA SIVALENSIS.

No. 201.—Fine fragment of the muzzle, broken off a little behind
the fosse of the canine alveoli, showing on the right side three pre-
molars, well worn. The whole tuberosity of the canine alveolus on
the left side, together with the empty alveoli of three incisors, are
present ; both broken off on the right side. Belonged to an old animal.
An instructive specimen, as showing the characters of the canines and
incisors of the upper jaw. Hard sandstone matrix.

No. 202.—Mutilated fragment of cranium, broken off from behind
the right orbit, obliquely forwards to in front of left orbit and ante-
riorly about the middle of the muzzle constriction, showing on the right
side the crowns of the back molars, well worn and in tolerable pre-
servation. Belonged to an old animal. Hard sandstone matrix. No.
201 and No. 202 unite, and belong to one cranium.

No. 203.—Fragment of cranium, comprising the upper maxilla of
both sides, broken off on a level with the floor of the nasal fossa, show-
ing on either side the six back molars, very much worn, proving the
animal to have been very old. From Ava.(?)

No. 204.—Fragment of upper maxilla, left side, showing the two
Jast premolars and three true molars in situ; the last molar is a little
worn, showing the characteristic trefoil discs of wear. Half of palate
also shown. Some ferruginous coloured matrix on the palate like the
Ava specimens.

No. 205.—Superior maxilla, right side, containing the five pos-
terior teeth cn situ of a very old animal. Specimen much mutilated.

No. 206.—Fine specimen of the cranium, broken off in front about
the middle of the facial portion, showing the occiput nearly entire, as
also the prominent edge of the occipital and sagittal crests; the two
zygomatic arches and temporal fosse, together with both orbits, the
right nearly entire: the teeth all broken off, and the spheno-palatine
region with occipital condyles wholly abraded by attrition. Hard
sandstone matrix in the temporal fosse and orbits, the most perfect
specimen in the collection, so far as regards the strictly cranial charac-
ters of the head.

No. 207.—Fine specimen comprising the greater part of the lower
jaw, both sides, in two pieces. The left ramus containing all the
molars, except the anterior premolar, which had dropped out; the right
containing the anterior premolar and the next five teeth, the last true
molar alone being wanting. The chin portion is perfect to the border
of the incisive alveoli. The two canines are present, broken off only at
their tips, and presenting the discs of wear on their posterior surface.
The six incisors are also present, perfect nearly to their tips; they
emerge nearly horizontally, the outer being a little lower, and they are
subequal in size. The left ramus shows the last three true molars, of
which the anterior is much worn, the penultimate less so, and the last
barely touched; the premolars are all broken off, except one. On the
right side the base of the anterior premolar is in situ, with but a very
short interval between it and the canine, not much exceeding that
between it and the second premolar; the body of this tooth is broken

HIPPOPOTAMUS. 145

off, as are also the second and third premolars. The fourth is present
and barely touched at its apex, showing that the animal was adult, but
not old. This specimen is probably the most perfect that has been
found in the Sewalik hills, being deficient only in the ascending ramus
and the expanding disc at its base. The difference in the length of the
diastema between it and H. amphibius is very remarkable. Probably a
female. This specimen (from Conductor Dawe’s Collection) was very
perfect and valuable, but it has been hardly dealt with in the Asiatic
Museum, most of the teeth having been broken off, and the fragments
searched for everywhere in vain.

No. 208.—Exserted portion of the upper canine right side, showing
the disc of wear complete nearly to the tip.

Wo. 209.—Inferior maxilla with the symphysial portion perfeet on
both sides on the border of the incisive alveoli, showing part of the
canines in situ and the outer incisor on either side, the four inter-
mediate ones having dropped out; the left horizontal ramus exhibiting
the broken off bodies of seven molars, the right side being shorter and
showing only the four premolars, also broken off. The alveoli of the
four middle incisors are much larger in diameter than in specimen
No. 207, indicating probably that the animal was a male.

No. 210.—Mutilated fragment of lower jaw, comprising only the
symphysial part and a short portion of the right ramus; the canine
alveoli on either side broken off; the two middle incisors very large,
being 1:25 inch in diameter. The alveoli of the next two incisors show
them to have been not more than half this diameter: the outer-
most on each side nearly as large as the middle ones; probably an old
large sized male. The premolars very much worn. This specimen is
much weathered and greyish white in colour, although thoroughly
mineralized and heavy. It differs very considerably in the large size
of the middle incisors from any other specimen in the collection, but
there is no mark or label to indicate whether it is from Ava or the
Sewalik hills.

No. 212.—Lower jaw, right side, comprising the posterior half of
horizontal ramus, containing the last three molars, also the ascending
ramus with the coronoid process and condyle, nearly entire: the leafy
expansion below broken off; the last true molar shows a well-developed
talon behind: attached to sandstone matrix:

No. 214.—Fragment comprising the lower jaw, both sides, at the
symphysis, showing the canines and six incisors; at other parts much
mutilated; belonged to a young or half-grown animal, with milk
dentition? No mark or number. Can this be the Ava specimen from
Col. Burney.—Journ. As. Soc. vol. vi. p. 1009?

No. 230.—Detached fragment upper maxilla, right side, containing
the alveoli of a canine and three incisors, and the three anterior pre-
molars broken off: the greater part of the canine is seen in the alveolus.

No. 231.—Axis with odontoid process, broken off.
VOL. I. L

146 FAUNA ANTIQUA SIVALENSIS.

No. 232.—Right femur, upper portion, comprising head and tro-
chanters.

No. 235.—Lower jaw, right side, posterior half with portion of
ascending ramus and leafy expansion, also the last three molars.
Very black and heavy. Doubtful if from Sewalik hills or Ava.

No. 237.—Lower end of united radiusand ulna, right side.

No. 239.—Lower end of radius and ulna, right side, considerably
smaller than No. 287.

No. 240.—Upper end of femur, left side, top of trochanter broken off.
No. 241,
No.-244,—Astragalus,

No. 710.—Sixth cervical vertebra, nearly perfect, of a young adult,
the posterior articular epiphyses not being synostosed: the neural arch
and both superior transverse processes together with both vertebral
canals present: the latter are very short.

Patella very entire.

No. 711.—Another sixth cervical vertebra of an older animal, and
tolerably perfect: vertebral canal very short.

No. 712.—Dorsal vertebra nearly entire, with a considerable portion
of the spinous apophysis resembling fig. 3 of Pl. LXIV. of the Faun.
Antiq. Sivalensis.

C. Specimens of Merycopotamus dissimilis, from the Sewalik Hills.

No. 246.—Lower jaw, right side, comprising the horizontal ramus
and part of the ascending ramus; the expanded dise below lost by a
recent fracture; contains the last three molars 7m situ, and little worn.
The three premolars dropped out, but their alveolishown. The canine
also in situ. The two last teeth exhibit well the rugous surface of the
enamel, with the basal cingulum and the ruminant-like pattern of wear
on the crown, which are characteristic of the genus, which is nearly
allied to Anthracotherium in the teeth.!

No. 247.—Fragment, comprising the posterior part of the ramus,
left side, containing the three last molars im situ, but very much con-
cealed by matrix. Ofa larger size than No. 246.

D. Specimens of Hippopotamus (Tetraprotodon) Paleindicus, from
the Nerbudda.

No. 16.—Specimen comprising the entire cranium, truncated ante-

riorly about the middle of the nasals; containing three back molars —
on each side, and on the left side the alveolus of the last premolar. —

1 See Journ. As. Soe. vol. vi. p. 899, | Sival. Plate Ixvii. fig. 6. The most cha-
and yol. v. p. 293. The entire speci- | racteristic part of it has been subse-

men, as presented’ to the As. Soc. Mu- | quently lost,
seum, is figured in the Fauna Antiq.

ae tp ee

HIPPOPOTAMUS. 147

Occiput, occipital and sagittal crests, zygomatic arches and orbits
nearly entire; also the glenoid articulating surfaces perfect ; occipital
condyles broken off. Teeth: the two anterior true molars worn, and
showing the pattern characteristic of the species; the last molar in
germ, and not protruded through the gums, indicating a young adult
animal; nasals raised above the plane of the front, at the naso-frontal
suture. An exceedingly fine specimen, more perfect than the original
Specimen upon which was founded the species as figured in Faun.
Antiq. Siv., Pl. LVII., fig. 1, and Pl. LVIIL, fig. 4. It shows well the
great saliency of the occipital and sagittal crests, and the projection
of the orbits above the plane of the frontal.

No. 17.—Short fragment, comprising the chin portion of the lower
jaw, both sides, near the anterior extremity of the symphysis, contain-
ing the alveoli in half section of four incisors and two canines. Dia-
meter of central incisors, 1:5 in.; of the external incisors, 1°85 in.
The true character of the specimen was recognized by Dr. Spilsbury.
This is a very valuable specimen, as it conclusively refutes the opinion
put forward by De Blainville (liv. xxii., ‘ Hippopotamus,’ page 240), that
the Nerbudda Tetraproto:on is identical with the living African species,
the middle incisors being the largest in the latter, while the converse
holds in the former ; and the relative proportion of the incisors differs
very remarkably.}

Il.— Nore upon a Youne Suit or Merycoporamts DissIMILIs, FROM
Burman. Broucur py Proressor OLDHAM:

(Note-Book; 28th Aug. 1862.)

The skull is of an adolescent animal,with all the sutures open: penulti-
mate, antepenultimate (1st and 2nd m.) in place, present on right side,
crown broken, and p.m. 4 just emerging; also p.m. 3 emerging ; other
teeth broken off or covered by matrix on right side; on left, only part
(inner) of t.m. 1 and 2 seen—the rest in front all hammered off; the
dilated part of imuzzle (inside portion) also broken off. Matrix of ver
hard and compact sandstone. Fossil dark reddish brown. (See Pl. XV.
figs. 1 and 2.)

Cerebral boite broken off. Cavity of anterior lobe of cerebrum seen
below (fig. 1 a), quite clean ; occipital and parietal portion broken off,
with the whole of temporal and parietal regions and zygomatic. Frontals
bb’ broken above; behind, the orbits send down a process on either side
between lachrymals and nasals; one concave mass between orbits, but
upper edge of rim of orbits broken on both sides, so that extent of con-
cavity concealed. Orbits evidently of large size, but rim partly broken
on both sides. (Frontal and orbits hippopotamine.) A foramen on
either side at middle of frontals at (c c).

Lachrymal ¢ e most perfect on right side, of a narrow oblong, let in be-
tween frontal apophyses and nasal above and maxillary below. Length,
1-2 in.; width, 45 in. Nasals, broad at base (dd), contract into a
narrow muzzle forwards. Suture with maxillary bone very open on right
side (marked by white band, f). g g, maxillaries long; extremities or

* Another species of Hippopotamus | (Fale. and Cautl.), is figured in the
from the valley of the Nerbudda, with | Fauna Antiqua Sivalensis. See Plate
si@ incisors, Hexa-protodon Namadicus | liii. fig. 1, &e.—[Ep.]

L2

148 FAUNA ANTIQUA SIVALENSIS.

nasal aperture not seen, and incisives broken off. Infra-orbital foramen
large, and about middle of height of muzzle, situated right above the
interval between p.m. 4 and 38. Palate long, but character not seen.
Surface of enamel very rugous.

Dimensions.
Width of nasals at base 1:9
Extreme length of fragment : : : b : c » 75
Greatest contraction of muzzle : ‘ : - : m « 1.2
Length of two (2 and 3) true molars : : : = 49
Ditto of two last premolars c ; ° : : about 1:3

Last premolar consists of a single barrel, with one pyramidal cusp
in front, and another behind.

TETRACONODON

V. DESCRIPTION

OR CHGROTHERIUM.

149

OF A FRAGMENT OF A JAW

OF AN UNKNOWN EXTINCT PACHYDER-
MATOUS ANIMAL FROM THE VALLEY OF

THE MURKUNDA.!

BY H. FALCONER, M.D.

Tae fossil is a portion of the right side of the upper jaw,

containing the two posterior molars.

The last tooth is

nearly entire, and very perfect; the other is broken off at its
anterior third. The teeth are not mineralized. The enamel
retains its ight and pearly colour; but the osseous structure

1 The specimen to which this memoir

refers was found by Messrs. Baker and |

Durand in the tertiary hills between
the Murkunda Pass and Pinjore, and
was figured by them in the ‘ Asiatic Re-
searches’ (vol. xix. plate vy. figs. 2a, 2,
‘and 2d), with the following description :
‘The specimen is the only one of the
kind hitherto met with. It is a fragment
from the jaw of some pachydermatous ani-

mal, but differs materially from all with |

which it has been compared. Further
discoveries will, it is hoped, throw light
on this interesting fragment.’ To this
description, the following note by Mr.
James Prinsep, the editor of the ‘ As.
Researches,’ is appended :—‘ The draw-
ing of this fragment so much resembled
Cuvier’s plates of the hippopotamus,
that I wondered at the authors’ mis-
givings on the subject, and wrote to
interrogate Dr. Falconer previous to
putting the present page to press. Dr.
F., however, assures me that the frag-
ment undoubtedly does not belong to
that animal; but, as Licuts. Baker and
Durand had rightly conjectured, to a new
pachydermatous animal, to which Capt.

Cautley and himself have from other |

specimens given the name of Chwrothe-
rium; ‘the engraving is imperfect, and
so much like the hippopotamus that it
might easily be mistaken. The differ-
ence in the original tooth, however, is well
marked. There is no real trefoil on it ;

the appearance is spurious ; the plane of |

| rused the memoir.
| prising that this memoir, which was

wearing is oblique; the spur is strongly
bifid ; and the ecollines, or mammillary
processes, are wide apart.”—J.P.” In
another paper on Sewalik fossils, pub-
lished by Messrs. Baker and Durand, in
the Journ. Asiat. Soe. for May 1836 (vol.
vy. p. 298), the opinion is expressed that
the Cherotherium of Falconer ‘is a new
species of Anthracotherium, a view which,
in the following memoir, Dr. F. opposes.
In a manuseript ‘ Synopsis of Fossils
from the Sewalik Hills,’ by Dr. F., I find
the following :—‘ Tetraconodon (or Che-
rotherium) grandis, Nobis, Dadoopoor col-
lection. There can be no doubt, then,
that the fossil described as Ti traconodon
was the same as Cherotherium, and this
opinion is now corroborated by Sir Proby
Cautley, who has a perfect remembrance
of the specimen, and has carefully pe-
It is somewhat sur-

probably written about the year 1835,
was never published, and that the fossil
is not figured in the Fauna Antiqua Si-
valensis. In a letter, however, written by
Dr. Falconer to Capt. Cautley, in Janu-
ary 1844, I find the following:—‘ We
must have a Cherotherium, for Agassiz
quotes it under our authority in his
“ Nomenclator Zoologicus.”’ The figure
is taken from a drawing by a native
artist, labelled ‘The Dadoopoor Tooth,’
and found among Dr. Falconer’s papers.
What has become of the fossil, I have
been unable to ascertain.—| Ep. |

150 FAUNA_ ANTIQUA SIVALENSIS.

of the jaw and alveoli has disappeared, and given place to an
argillaceo-calcareous incrustation of a pale yellow colour,
which also fills the discs of detrition of the grinding surface.

The teeth had belonged to an adult animal; the posterior

molar having come into use, and the anterior one being con- _

siderably worn down. They are known to be the two last
molars, from the posterior one having the accessory spur or
process which characterizes this tooth. Their form and state
of detrition show that they were not the last and penultimate
milk molars.

They have the figure of a rectangular or nearly square
shaft, rounded at the corners, surmounted by a crown
divided into four large conical or mastoid processes brought
together in transverse pairs. The body of the tooth on
either side is contracted in the middle by a perpendicular
fossa down to the root, dividing it into two portions, each of
which has a rounded or convex surface; so that altogether
the tooth has the appearance of four cylinders in apposition
by pairs, terminating in a crown of as many conical processes.

THE DADOOPOOR TOOTH, FROM A SKETCH BY A NATIVE ARTIST; OF THE
NATURAL SIZE.

There is no bulge or ridge of enamel making the circuit of
the shaft, so as to form a neck or collet; but the crown is
complicated by small transverse ridges of enamel, one at each
end of the tooth, and a third in the angle of junction of
the pairs; these are chiefly developed along the outer col-
liculi, and are more or less complete forwards.

The plane of detrition is oblique, the teeth sloping at a
considerable angle from their outer margin inwards; so
that the colliculi of the inner lines are truncated, while the
outer remain more or less conical.

TETRACONODON OR CHG@ROTHERIUM. 151

The external outline of the crown viewed in profile gives a
sweep of salient and re-entering angles getting more and
more obsolete forwards, from advanced detrition, with a
small prominence in the open angles or swell in the col-
liculi, marking the commencement of the transverse ridges.

The vertical outline externally of the outer colliculi, from
the root to the apex, is but slightly convex.

Besides the complication of the spur in the last tooth
there are other differences between the two, dependent on
form and detrition.

To begin with the last, which is the more perfect and least
worn :—

This tooth, exclusive of the spur, is a little longer than
broad; and the distinction of the crown into two pairs of
processes, and of each pair into separate colliculi, is well
marked.

The two outer processes have the same form, differing only
from wearing, the posterior being more pointed, and their
outer surface rises to the apex with little of a curve; they are
flattened laterally, sloping right and left. Their inner surface
inclines from the point, downwards, at a considerable angle ;
the anterior one is worn down to a disc, which is oblong
transversely, and surrounded by a complete belt of enamel;
in the posterior one, the enamel is scarcely touched, the
surface is marked with slight inequalities, the disc is very
small, and the posterior edge of the process is notched into a
few tubercles. Around the base, at their connected aspect,
both processes are bounded by a fissure, which separates them
from the rest of the crown.

In the angle between these colliculi lies the middle trans-
' verse crest or ridge of enamel. It is notched, narrow, and
little prominent. It arises by a slender pillar in the external
fossa, where the shaft is contracted, runs forward a short way,
and is then lost in detrition. The posterior crest arises in a -
considerable lobule or swelling of the hind process at its base.

It is well notched, and runs forward in a sweep between the
process and the spur, terminating at the outer third of the
latter. The anterior ridge arises also in a swelling of the
corresponding process. It is partly worn and has a flattish
surface. It stretches across (with an accidental breach of
continuity in the fossil) along the margin of the crown, and
descends in a curve, so as to turn round a little way on the
shaft.

Now, in description of the inner divisions:—The protube-
rance of the colliculi has disappeared; they are truncated
from wearing down to a flat disc. The corresponding por-
tions of the shait, from which the processes rise, are more

152 FAUNA ANTIQUA SIVALENSIS.

convex than those of the outer side, having also a swell
vertically. The anterior one has the form of a cylinder
compressed on the sides; it is more worn than the hind one.
Its disc is transversely oblong, surrounded by a belt of
enamel, which at its outer anterior angle runs into the
corresponding transverse crest (there is an accidental fissure
in the enamel here), so that in the continuation of the latter
it appears to double back on itself, descending in a sweep
upon the middle of the cylinder, where it terminates obtusely
at the upper third of it, forming a marked feature in the
tooth.

The posterior division is more irregular in the outline of
its crown; the disc of detrition is less; the cincture of
enamel is wider and at its anterior outer angle projects in a
broad surface into the hollow between the outer processes,
throwing out also a heel in the opposite direction into the
fossa between the inner divisions. Posteriorly it is notched
by an angular fissure into a sort of cordiform outline, with
the appearance of an additional pillar joined on to it opposite
the clefts of the spur.

In the fossa of contraction between the divisions there
arises a considerable pillar of enamel, tuberculated laterally.

The spur is a large convex bifid process, like two convex
cones in apposition, increased laterally by tubercles. The
apices formed by the cleft are circular and a little worn.

The anterior or penultimate molar has the same general
form as the last one, but it is more simple, and its dimensions
are somewhat less. It is less deeply contracted in the mid-
dle, and the outline of the divisions of the shaft is less convex.
Its crown is much more worn. The anterior half is muti-
lated by a fracture across, at a third of its length, and the
pair of processes have been worn down to one continuous
transverse disc. The posterior half has the divisions still
distinct. The outer hillock resembles the adjoining one of
the last tovth; its dise is oblique, transversely oblong, and
encircled with an entire belt of enamel. The inner division
has none of the complexity of the corresponding portion of
the last tooth. It has a simple broadly-oblong outline and
disc. Its belt of enamel in the middle of the crown is
confused with the adjoining one of the anterior half. There
is no accessory pillar of enamel in the fossa between the
inner divisions, as in the last tooth.

The middle transverse crest is seen in the furrow between
the outer hillocks, advancing but a short way forwards.
The posterior one is also present, its surface flattened by
wearing. The mutilation of the tooth anteriorly leaves it
doubtful whether the anterior crest was present, or not,

et

TETRACONODON OR CH@ROTHERIUM. 158

and whether it swept round on the inner side of the shaft, as
in the last molar.
_ The length of the tooth is greater on the coronal surface
than at the base of the shaft.
The dimensions of the fossil are as follows :—

Inches

Length of the fragment : : : : : 5 - . 3°38
Ditto of the last molar 0 5 : . - : : . 2°05
Ditto ditto excluding the spur . ; : ; . 168
Width of ditto (greatest) . : : ° : é : . 148
Height of the outer hillocks of ditto (greatest posterior) . omy ee)
Ditto of the inner side ditto : meni

Length of the anterior molar : : ; : : seeelc2.
Width of ditto . : , : i : ‘ : P Series
Anterior width of hind molar, top of coronal P 5 A

Now to determine the animal to which the teeth belonged.

The fragment is a unique specimen in a large collection of
fossil bones, got within the last few months in the valley of
the Murkunda and other small adjoining valleys, from the
upper deposit of the tertiary line of hills, west of the Jumna,
forming the flank of the Himalayahs. Of heads and jaws
alone we have upwards of 300 fragments, more or less perfect,
of three species of Hippopotamus, two of which are as-
certained to be distinct, and a third doubtful; as many of
the Mastodon Elephantoides and fossil Elephant respectively ;
several of a Rhinoceros; of Hogs, of a great variety of Rumi-
nants, and of some other animals. The fragments of other
bones amount to several thousands. This will give an idea
of the richness of the collection, yet among the whole there
is no other specimen with teeth resembling that now described.
There are, therefore, at present, but the two molars to go
upon; but these are sufficiently well marked to discriminate
the animal.

That the fossil belonged to an herbivorous animal of the —
family Pachydermata is at once evident from the form and
detrition of the teeth.

The subdivisions of the family, dependent on dentition,
give further aid in making it out. From all Pachydermata,
which have transverse or flexuous lines, or crescentic plates
of enamel in the teeth, such as the Horse, Rhinoceros,
Anoplotherium, Palotherium, &c., it is distinguished by the
division of the coronal into four hillocks or tuberosities.

The comparison is, therefore, reduced to those genera
which have rounded prominences on the back molars.

Of these it differs from the Tapir and Lophiodon in the
prominences not being in the transverse crests, which form
so distinctive a character of these genera.

154 FAUNA ANTIQUA SIVALENSIS.

It differs from the existing Hog tribe and the Cheeropo-
tamus, in the coronal not being an aggregation of tubercles or
rounded cylinders, but divided into four distinct portions, two
of which’ are certainly conical.

At first sight the fragment has a considerable resemblance
to certain species of Mastodon, in the form of the protuber-
ances and discs of detrition, such as the M. Andium; but
the nearly square form of the shaft, the presence of only
two pairs of hillocks in a back grinder on a coronal along
with a spur, and another tooth of nearly the same form placed
before it, are proofs of molars in a series of several, and alone
sufficient to distinguish it from every species of that genus.

The Anthracotherium has a coronal divided into four dis-
tinct pyramidal hillocks, with deep furrows between. In
these respects there is some resemblance in our fossil. But
the two differ materially. In the Anthracotherium the shaft
has a ridge or collar of enamel in relief around it; and it is
strongly convex in its vertical section. The hillocks are a
little elevated, and their sides converge rapidly in a blunt
pyramid ; the surfaces by which these pyramids regard each
other throw off connecting sharp and sometimes bifurcating
edges, which make the coronal angular. Further, the upper
molars have four accessory tubercles, projecting from the
collar, alternating with the bases of the large processes. All
these characters are absent in our fossil, which has, in addi-
tion, positive points of distinction not present in the Anthra-
cotherium, such as the transverse crenulated ridges of enamel,
&c. These characters combined strongly mark the Anthra-
cotherium and the fossil as distinct genera.

The only genus remaining for comparison is the Hippo-
potamus. Ata casual examination the fossil seems sufficiently
distinct. But the essential points of structure in both are
so analogous as to require a strict comparison.

The resemblance is this :—

The last teeth of both have a square shaft surmounted by
a coronal, divided into four conical hillocks, brought together
by pairs, a transverse ridge or crest in front of the anterior
pair, and in the last molar an accessory solitary hillock or spur.

The points of difference are these :—

In the Hippopotamus the pairs of conical processes are
brought together, and separated by a longitudinal irregular
cleft, of which the sides are nearly vertical. The processes
are marked on each side by a vertical narrow furrow, so that
in detrition the discs are trefoiled when the processes of each

1 This reservation is made, because, | detrition of the last tooth leayes the

although the four hillocks may well be | character apparent but in two.
inferred to be distinct and conical, the

2 ee

TETRACONODON OR CH@ROTHERIUM. 155

pair are distinct, and of a quadrilobal figure when they have
run together. This character is very marked in the large
species, and the structure on which it depends is more or less
present in all, although in the small species described by
Cuvier, it is so marked by the obliquity of the plane of de-
trition as to be little apparent. In all the species the
undivided portion of the shaft is low, and the processes are so
much in relief from it, that the vertical height of the latter
is several multiples of the height of the shaft. The curvature
of the processes is directed inwards, so that the apices of each
pair are brought considerably within the margin of the shaft.
The spur is a simple hillock; there is but one transverse. crest
or ridge, the anterior one.

In the fossil the height of the tooth, compared with its
other dimensions, is much less than in the Hippopotamus.
The shaft is not so low, and the prominence of the processes
in relief from it is less than the height of the shaft. Their
form is, consequently, shortly conical. The longitudinal cleft
is not a vertical fissure: it is an open hollow, and the sides of
the hillocks slope outwards at a considerable angle with it.
The curvature of the hillocks is but slight, and their points
are not brought much inwards, so that they are, as it were,
seated marginally. They are not furrowed vertically so as to
give anything approaching a trefoil in detrition, and their
grinding plane is very oblique, compared with the large species
of Hippopotamus, fossil and recent, although not so much in
contrast with the smaller. They have a transverse crest in
the hollow between the outer hillocks, and another between
the outer posterior hillock and the spur, neither of which is
present in the Hippopotamus. The spur is strongly bifid,
so as to resemble two hillocks in apposition, like that of the
large Mastodon, and not simple, as in the Hippopotamus.

These points of distinction are so marked as to establish
the generic difference of the fossil from the Hippopotamus.

It appears, therefore, that there existed formerly in the
North of India, associated with the Mastodon, the fossil
Elephant, the Rhinoceros, and the Hippopotamus, a pachy-
dermatous animal of large size, distinct from all known fossil
or recent species, and adding another genus to this numerous
fossil family. Our knowledge of its structure is at present
confined to the posterior molars, which connect it with the
Hippopotamus, the Hog, and the Anthracotherium. But till
the entire composition of its dentary system is known we
must remain in the dark regarding its strict analogies and
position among the Pachydermata. From the active and
extensive search, however, which is at present made near
the valley of the Murkunda by fossil collectors employed by

156 FAUNA ANTIQUA SIVALENSIS.

several independent inquirers, we may expect that, amidst
the immense quantity of bones which are continually met
with, the débris of this animal will be found in sufficient
abundance to lead to a knowledge of all the main points of
its osteology. From the form of the posterior molars we
have designated the animal 'ctraconodon, with the specific
name of magnum, from its large size,' which appears nearly
to have approached that of the Hippopotamus and the Indian
Rhinoceros, as will be seen by a comparison with the corres-
ponding teeth of these animals.?

Large
Large Fossil Hip-
Tetraco- \ Existing |Fossil Hip-| popotamus
nodon Hippo- |popotamus| of the | Rhinoceros
magnum | potamus of Murkunda.| Sivalensis
Europe H. Siva-
lensis

Length of the last molar
of upperjaw 2°05
Width of ditto . 1:48

2°3
2°5

bot
a

bo ty
Oo co
me bo
lo oi Ww)
mm or

The occurrence of such an animal is, perhaps, not more
than might have been looked for. The discovery by Crawfurd
and Wallich, on the Irrawaddi, of two new species of Masto-
don, peculiar to India, gave reason to expect corresponding
new results among other Pachydermata. Our researches in
the Murkunda have borne out this idea. The Tetraconodon
magnum is not the only new animal which we have found in
this vast Golgotha; but we have not yet had the opportunity
or leisure of examining the bones in detail fit to make the
results public. We can only allude to them at present.
Among others, we have several portions of the jaws and heads
of what we believe to be a new species of Hippopotamus, in
which the canines, instead of being furrowed longitudinally
by a deep fossa on the inner side, have the sides meeting in-
wards in a sharp edge.

The fragment of the jaw of the Tetraconodon was got by a
native collector employed by Lieuts. Baker and Durand, and it
is now in the joint collection of these gentlemen, who have
kindly permitted us to make use of it for publication in the
journal of the Asiatic Society.

1 From térpa, k@vos, and é80vs. This ? The measurements, which were
name must not be confounded with the | omitted in the manuscript, have been
Tetracaulodon of some later American | made by me in the Museum of the Royal
authors, applied to individuals (?) of the | College of Surgeons and in the British
large Mastodon, with incisives in the | Museum, with the assistance of Mr.
lower as well as the upper jaw. Flower and Mr. Dayis.—[Ep.]

RHINOCEROS. 15

bat |

VI. ON THE SPECIES OF FOSSIL RHINOCEROS
FOUND IN THE SEWALIK HILLS.

[Three species of fossil Rhinoceros from the Sewalik hills are
figured in the Fauna Antiqua Sivalensis (Plates LXXII. to
LXXIX.), viz.: R. Sivalensis, R. Palwindicus, and R. platy-
rhinus, the two former characterized by the curved line of the
upper plane of the head, the last by the upper plane of the
head being straight and broad. (See Pl. XIV.) No complete
description of the species was ever published or written by
Dr. Falconer. The following account, which appears to
refer mainly to R. Sivalensis (although the possibility of there
being two species is hinted at), is extracted from a memoir
on Sub-Himalayan Fossils, by Messrs. Baker and Durand,
published in the Journal of the Asiatic Society for August,
1836, vol. v. p. 490. The annexed illustrations are copied
from the Fauna Antiqua Sivalensis, and the reader is referred
to the descriptions of the Plates in the Fauna for further
particulars. Professor Owen, in his ‘Odontography’! makes
the following statement. ‘In one of the extinct species of
Rhinoceros from the Himalayan tertiary beds, Dr. Falconer
informs me that there are six incisors in both jaws; the typi-
cal number was, therefore, retained in this ancient species, as
in the contemporary Hippopotamus of the same formations.’
The species referred to was evidently R. Sivalensis, for Plates
LXXII. fig. 4b, LXXIV. fig. 4, and LXXV. fig. 10, F.A.S.,
of the Fauna Antiqua Sivalensis, show that the remark does
not apply to R. platyrhinus or R. Paleindicus. (See Pl. XIV.
fig. 4.) The following note, also, on Colonel Baker’s large
specimen of the skull of the Rhinoceros platyrhinus in the
British Museum, was written by Dr. Falconer in his Note-
book for 1860: ‘The molars are in fine condition, six on
either side. The last true molar is only just touched by wear.
The last true molar is exactly like R. hemitechus, in having
a posterior basal funnel-shaped pit; while the penultimate
and antepenultimate true molars, and the penultimate and
antepenultimate milk molars, have each three distinct fos-
settes, as in Rhinoceros tichorinus! The vertical ridges of the
outer side are very well pronounced in three valleys. The
animal had two large incisors above and four below: of the
latter, the two outer are big, the two inner small, as in
the existing Indian Rhinoceros.’ In the ‘Fauna Antiqua
Sivalensis,’ there are likewise illustrations of fossil Rhinoceros
from Perim Island (R. Perimensis), and from the valley of the
Nerbudda.—Ep. ]
1 Vol. i. p. 589.

158 FAUNA ANTIQUA SIVALENSIS.

J.—DescripTion By Messrs. Baker AND DuRAND OF THE FossIL
RHINOCEROS OF THE SEWALIK HILLs.

(Reprinted from the Journal of the Asiatic Society for August, 1836.")

Cranium.—We shall commence with the fossil which, being the most
perfect, affords the best means of instituting a comparison with the
skulls of described species.

The fossil cranium is imperfect in the following parts. The extre-
mity of the nasal and intermaxillary bones is broken off; the zygo-
matic arches are both fractured; the left occipital condyle is wanting ;
the following molars have either dropped out prior to the envelop-
ment of the head by the matrix, or have been broken off subsequently
to its fossilization, viz. the fifth of the right, the first and seventh of
the left, maxilla. In addition to these losses, the cranium has under-
gone, when in the stratum, the common fate of Sub-Himalayan relics,
and is cracked in several directions; the crush, however, which pro-
duced these cracks has not materially altered the form of the head;
the chief effect produced has been the forcing the left half of palate at its
anterior extremity a little above its proper level; this the longitudinal
crack passing through the left orbit enabled it to accomplish; the
displacement resulting may be best observed in the profile view of the
skull, fig. 3. The transverse cracks are accompanied by a small
hollow and a consequent neighbouring bulge, both so partial and of
such small relief, that in the profile their places can only be observed
by paying attention to the jagged outline at the depression of the
frontals. With the above exceptions the specimen is perfect.

A glance at Pl. XV. will be sufficient at once to determine the species
with which this fossil rhinoceros must be compared. The depression
of the frontals causing the deeply curved outline of the upper planes of
the head, the slope of the occiput, the septum, and the nasal arch all
separate this cranium from the existing and fossil bicorn species. The
existing unicorn species is that, therefore, to which recourse must be
had in order to establish a comparison.

In the unicorn rhinoceros of Java the height to which the crest of
the occiput rises above the palatal plane, and also the thickness and
prominence of the nasal arch supporting the horn, are less than in the
Indian rhinoceros. A line drawn at a tangent to the crest of the occiput
and the highest point of the nasal bones will, in the unicorn species of
India, be more raised above the plane of the frontals than is the case in
the Javanese rhinoceros. In the foregoing respects the fossil associates
itself with the Indian, and differs from the Java, species. The com-
parison may, therefore, in general be confined to the former.

With the view of bringing at once under the eye, the discordance
which occurs between the relative values of analogous dimensions, the
subjoined table is here inserted. The modulus chosen is the space
occupied by the seven molars, because on this measurement the
development of the bones of the head must, to a certain extent, be
dependent. The measurements given in Cuvier’s ‘Oss. Foss.’ have
afforded the proportions of the existing species; and the table of
dimensions which closes this paper has given the proportions of the fossil.

1 The illustrations referred to are those in the ‘ Journ. Asiatic Society..—[Ep. |

DESCRIPTION OF PLATE XIV.

RHINOCEROS SIVALENSIS AND RHINOCEROS PLATYRHINUS.

Fig. 1.

Fig. 3.

Fig. 4,

Profile view of skull of Rhinoceros Sivalensis, with three true
molars and three posterior premolars, one-eighth of the natural
size. Copied from a drawing by Mr. Dinkel, in the Fauna An-
tiqua Sivalensis, Plate LXXIII., fig. 2 a. The specimen is in
the British Museum. (See pages 157 & 514.)

. Fragment of upper jaw, right side, of Rhinoceros Sivalensis, with

three true molars and two last premolars, one-fourth of the
natural size. Copied from a drawing by Mr. Ford in the F. A. 5.,
Plate LXXV., fig. 5. The specimen is in the British Museum.
(See pages 157 & 516.)

Profile view of cranium of Rhinoceros platyrhinus, one-eighth
of the natural size. Copied from a drawing by Mr. Dinkel, in
the F. A. S. Plate LXXIL., fig. 1. The specimen is in the
British Museum. (See pages 157 & 513.)

Lower jaw, with incisive border of Rhinoceros platyrhinus,
showing two outer large and two central small incisors, one-
fourth of the natural size. Copied from a drawing by Mr. Ford,
in the F. A. S. Plate LXXV., fig. 10. The specimen is in the
British Museum. (See pages 157 & 517.)

VOL. I,

Vol. 1. Plate 14.

G-HFord del. JDinkel ith. W.West imp.
1,2. Rhinoceros Stvalensis. 3,4. R platyrhinus.

RHINOCEROS. 159

salen ae Ind. Hnin,

Space occupied by the seven molars assumed equal to. ap aleOo 1:00
Height of occiput from lowest edge of occipital foramen to

summit of crest of occiput. 5 c : . : 1:02 0°80
Greatest breadth of occiput é f “ : f 1-11 1:05
Least thickness of cranium across temporals . e | 0:45 0°38
Breadth across at post-orbital apophysis of frontals . . | 0°83 0-78
Distance from anterior of orbit to auditory foramen . . | 1:02 1:00
Breadth across the occipital condyles . i : : . | 0147 0-60

Referring to the table of dimensions it will be observed, that the
height of the occiput is in the fossil less by mét. 0-021 than the
corresponding measure of Cuvier’s Indian rhinoceros ; but the greatest
breadth of the occiput is mét. 0-036 in favour of the fossil; relatively
to the space occupied by the seven molars, these two measurements
attain a less development in the fossil than in the existing animal.
The difference in the occipital condyles amounting to mét. 0:065 in
excess of the Indian rhinoceros causes a marked discordance in the
ratios of these dimensions; but, as the left condyle and the adjacent
parts are wanting in the fossil, the measure was obtained by doubling
what appeared to be the exact half dimension; this of course is not so
satisfactory as if the condyles had been perfect; any inaccuracy con-
sequent on this circumstance could not, however, amount to a quantity
which would materially alter the deduced proportion. The occiput,
figs. 8, 9, Pl. XVIL., is fortunately very perfect ; from its dimensions,
which prove it to have belonged to a smaller animal than the cranium
of Pl. XV., it may also be concluded, that though inferior in size to
Cuvier’s specimen of the Indian rhinoceros, which in greatest breadth
of occiput exceeds it by mét. 0-039, yet the space occupied by the
condyles is 0°010 in favour of the small fossil occiput. In both of the
fossils the depressions near the summits of the occiputs on each side of
the mesial projections are deeper than those of the existing species.

The zygomatic arches not being entire, and the matrix being
uncleared from the portions which remain, no particular remarks can
be passed on them.

The sutures cannot anywhere be traced; a circumstance which
precludes the notice of particulars frequently of importance in the
comparison of species.

The least thickness of the cranium is but met. 0001 greater than
that of the Indian rhinoceros; and therefore in proportion to the
modulus, yields a less ratio than that species.

The breadth at the orbits is met. 0-024 greater than in the existing
species; consequently the skull does not in this part present any
material dis:ordance of proportion.

The length between the auditory foramen and the anterior of the
orbit is 0:043 mét. greater in the fossil; this measurement affords a
proportion only differing met. 0-002 from that obtained from the
existing species.

The infra-orbital foramen is situated similarly to that of the Indian
rhinoceros,

160 FAUNA ANTIQUA SIVALENSIS.

The nasal arch is massive and much developed; the spring of this
arch is perpendicularly over the anterior of the second molar; that is a
little more retired than in the Java or Indian rhinoceros skulls, given
in Cuvier’s Pl. IV.

The breadth of the palate has not been given in the table of dimen-
sions, because the first and seventh molars not being perfect on both
sides, measurements corresponding to those of Cuvier’s could not be
obtained. It is comparatively less than in the existing species, but the
great breadth of the teeth compensates for this difference.

Having detailed the essential differences and the points of resem-
blance observable in the fossil Indian rhinoceros when compared with
Cuvier’s dimensions of the existing Indian rhinoceros, we must be
permitted to add, that additional measurements from skulls of the latter
species are requisite before anything certain can be pronounced as to
the amount of difference or correspondence between the two species. We
are induced to make this remark in consequence of having been favoured
with the examination of two craniums which presented considerable
variation of proportions when compared with Cuvier’s and with each other.

It appears to us desirable, therefore, to ascertain the limits within
which individual variations range before anything positive can be
asserted. The foregoing remarks will have shown a great general
resemblance, accompanied by a departure of proportions in some corre-
sponding parts; the latter may be sufficient for the establishment of a
new species—at least for the present, until more data are obtainable
whence to determine the bounds by which the individuals of one species
are limited in their variations. For the sake of distinction, therefore,
and present convenience, at the same time keeping in view the type
to which it is a near approach, we have termed the species under con-
sideration the R. Indicus fossilis.

Teeth —The remark has been already passed, that the greater num-
ber of fossils obtained from the Moginund deposit are the remains of
young animals; with the rhinoceros this has been particularly the case.
We accordingly find ourselves better able to illustrate the early stage
of dentition than that more advanced. ;

Fig. 1 contains the four milk molars of the left maxilla; the
fourth being but just cut is unworn; but the palate being broken away
from the base of the tooth, more of it is seen than would otherwise be
the case; in the right half of the specimen, where the palate is whole,
the fourth molar is more concealed. ‘The first molar is also unworn,
but the second and third have suffered detrition. The two rows of
teeth have their internal base lines parallel to each other, and the lines.
which would circumscribe their exterior much curved, in consequence
of the difference of breadth which exists amongst the teeth. The upper
part of an unworn tooth, measured exteriorly, is much longer than the
lower; for the anterior of each molar projects beyond the posterior
extremity of the one immediately in its front by the gradual enlarge-
ment of the external line of enamel from the base to the summit. As
the molars wear down, this outer development is reduced, the internal
sides of the teeth come more into use, and breadth is gained in com-
pensation for the diminished length of surface in wear.

Fig. 5, Pl. XIX. The sixth molar from a left maxilla. The spur,
which occupies no inconsiderable part of the hollow between the

wi

OG ee

RHINOCEROS. 161

anterior and posterior transverse hillocks, is here less curved than that
of the Indian rhinoceros; and there is wanting altogether the small
salient of enamel, which in the Indian rhinoceros occurs between the
starting point of the above-mentioned spur and the point of junction
of the exterior and anterior main lines of enamel. It may also be
mentioned, that the exterior and posterior lines of enamel being less
thick than the corresponding parts of the sixth molar of the Indian
rhinoceros, there is a greater space between the two. Such modifica-
tions of form are however fortuitous, differences of equal amount being
observable in the teeth of animals of the same existing species.

This fossil measures in length . 3 in. 2°50 mét. 0°0645
3 in breadth . : » 2°62 » 0°0675

Fig. 6. The 5th molar, derived from a left maxilla. The outline
of its enamel accords with that of the similar tooth of the Indian rhino-
ceros, the only difference being in the dimensions and in the emamel-
lated edge of the short beading at the anterior side of the tooth.

It measures in length . F 5 ; in. 2:08 mét. 0°053
» 10 breadth i j é » O27 » 0°0835

Fig. 7 is the 7th molar, and from a right maxilla; the point of the
small spur is broken, as also the anterior extremity-of the external line
of enamel; but the tooth is sufficiently perfect to show a close resem-
blance to the analogous molar of the Indian rhinoceros.

It measures in length . : : : in. 2°88 mét. 0:0735
» 10 breadth - é 99 2°58 » 0°065

Fig. 8 is the 7th molar of a left maxilla. The difference observ-
able between this and the foregoing specimen consists in the great
development which the small anterior spur here attains. In the former, it
is scarcely observable ; in fig. 8 it is very prominent. Variations to an
equal amount may, however, be observed in the minor salients, &c., of
enamel in teeth appertaining to skulls of the same existing species. No
weight can therefore be attached to such unimportant modifications.

This fossil measures in length , 3 in. 2°95 mét. 0-075
be in breadth . : » 2'55 » 0:065

The cranium Pl. XV. has its molar teeth so much worn down that
the configurations of the enamel cannot be traced. The table of dimen-
sions gives the length and breadth of each tooth, and shows that although
the lengths do not materially differ from those of the corresponding
teeth of the existing species, the breadths exceed those of any hitherto
described. A

Without complete illustrations of the milk-teeth of existing species,
it would be dangerous to attempt a comparison between them and the
fossil Indian rhinoceros. We have therefore avoided the endeavour ;
but we must be allowed to notice the upper jaw fig. 4, Pl. XIX., which
offers peculiarities when compared with figs. 1, 2, and 3 (of the same
plate), deserving of remark.

The right half of the specimen is figured in the plate, the left half
having lost the first tooth. With respect to age, this jaw nearly cor-
responds with fig. 3, the fifth molar being in both on the point of
appearance. The following departures from the tracing of enamel in
figs. 1, 2, and 3, may, however, be observed. The second molar of

VOL. I. M

162 FAUNA ANTIQUA SIVALENSIS.

fig. 4 has this peculiarity, that instead of the anterior portion of the tooth
being one continuous offset from the exterior line of enamel, it only
assumes that appearance after considerable detrition, consisting at
first of a short offset and an isolated pillar, as shown in the drawing.
The two sides of the jaw have been very unequally worn, in conse-
quence of which the opposite side to that delineated has the pillar and
offset conjoined. The third molar also presents a marked difference
when placed in juxtaposition with the corresponding teeth of the
other three jaws: the two spurs which occupy the central hollow of
the tooth are of a different shape from that which occurs in the other
specimens. In other respects, fig. 4 corresponds with them: its rows
of molars are parallel to each other, and the dimensions offer but
trifling variations. The modifications of form above alluded to, unless
fortuitous, which is perhaps improbable, denote the existence of
another species—a fact corroborated by the examination of the milk
molars of the lower jaws in our possession. Upon the consideration
of these we now enter, but are able to offer but few and unsatisfactory
remarks.

Lower Jaws.—With the exception of the fine fragment, fig. 6, Pl. XVL.,
submitted to our inspection by Conductor Dawe, and the fragment,
fig. 9, the specimens of lower jaws are all from the Maginnud deposit,
and all the remains of young animals.

Fig. 1, Pl. XVI., represents a fossil which has lost the interior of its
symphysis, the second molar on the right, and the first molar on the
left side of the jaw, asalso both the rami, which are broken off. Four
molars have appeared, the second and third of which are worn; but
the first and fourth have their enamel intact. The sections of fracture
expose germ teeth. The two lines of molars have a gentle conver-
gence, which is effected, not by a curve in the rows of teeth, for these
are set in a perfectly straight line, but by the gradual approach of the
two rows, which make a small angle with the median line of the jaw.
The section shown by the break of the symphysis and the interval
between the front molars argues the existence of a prolonged sym-
physis. The fourth molar is characteristic, having an isolated point
or low pillar in the centre of the chord of its posterior crescent.

Fig. 4 is the right half of the lower jaw of a young rhinoceros,
but of one somewhat older than the animal to which fig. 1 belonged,
for the fourth molar has in fig. 4 suffered detrition. Notwithstanding
the difference of age being in the favour of this specimen, the space
occupied by the four molars is less than that of the four in fig. 1.
The fourth molar is here devoid of the low isolated pillar in the pos-
terior crescent, and has the central enamel, or junction of the two
crescents, larger than in fig. 1. There are no means of ascertaining
whether or not the opposite rows of molars were parallel, but in the
position of symphysis and set of the teeth in a perfectly straight line,
this specimen corresponds with the foregoing.

Fig. 2 has its fourth molar just disclosed, and rising into the line
of molars. It is devoid of the isolated pillar; but in size corresponds
with fig. 1, instead of fig. 3, to which latter it assimilates itself by the
fourth and second molars.

It is difficult to ascertain the degree of importance to be attached
to such points of difference. In no specimen from the jaw of an adult

RHINOCEROS. 168

animal has any trace of the isolated pillar been hitherto found. Oc-
curring as this peculiarity does in a deciduous tooth, should nothing
similar take place in the permanent tooth which replaces it, the only
chance of determining the question will be the discovery of an entire
head. We have noticed an upper jaw, fig. 4, Pl. XIX., which indicates
the probability of the existence of two species. The examination of
the above lower jaws rather confirms this supposition; but in the
event of such slight modifications denoting specific distinctions, we
are unable, in consequence of the paucity and incompleteness of speci-
mens, to decide which are the milk-teeth of the fossil Indian rhino-
ceros. Nor are we fortunate with respect to the lower maxilla of the
adult animal ; figs. 6, 7, and figs. 8, 9, being all that we can bring for-
ward. ‘The sections of these two fragments differ, in consequence of
their being derived, one from the posterior, the other from the anterior
part of the jaw, which thickens as it approaches to the symphysis.
These two specimens resemble the corresponding portions of the
lower jaw of the Indian rhinoceros, but are too imperfect to afford
any satisfactory measurements for grounds of comparison.

Anterior Extremity.

A scapula in our possession is not sufficiently perfect to give accu-
rate measurements, but it bears as great a general resemblance to
that of the Indian rhinoceros as do the other parts of the skeleton.

The humerus, figs. 1, 2, Pl. XVII, having its radius and ulna at-
tached, was discovered by ourselves very close to the place whence we
excavated the femur and tibia forming the subject of Pl. XVIII. With
the exception of the deltoid crest, this humerus is perfect, and has
afforded the dimensions which enter into the first column of the table.
For the purpose of comparison, the five following columns are
added. The proportions of the Indian and Sumatra small species of
rhinoceros are deduced from Cuvier’s table; those of the fossil speci-
mens are of course from the Table of Dimensions. The length of the
bone is assumed as the unit, and the measures of other parts referred to
it, in order to obtain their comparative values.

4 Cuvier’ é a .
Cuvier’s nals Fig. 1, Pl.| Fig. 5, Pl. | Fig. 6, Pl.
Measurements Ind. Rhin.| gyan Sp. |: 17, fossil | 17, fossil | 17, fossil

Rhin, | Ind. Rhin.| Ind. Rhin.| Ind. Rhin.

Length of humerus from tub-

erosity to external condyle | 1-00 1-00 1:00 1:00 1:00
Ditto ditto ditto internal ditto | 1-03 0:95 0:91 0°94. nc
Greatest anter. post. diameter

at top : 3 : " 0:44 0°30 av 0°44 0°43
Breadth across condyles -| 0°36 0°31 0°35 0°37 da
Ditto of articulating pulley .| 0-25 0-19 0°22 0°22 0°26
Least diam. of the body of the

humerus . = . - | 015 0-13 0-14 aac 0:15
Length of radius . : | 0-79 0°78 0°76 BA ai
Breadth at top. - -| 0°26 0°20 0:23 re 550
Ditto at bottom . : - | 0:25 0718 0:23 aa
Length from articulating head

to bottom of internal con-

is | eee ae nit 0°82 | 0-81 | 0-87

M 2

164 FAUNA ANTIQUA SIVALENSIS.

The Sumatra rhinoceros (small species) concurs with the - fossil
Indian rhinoceros in having the length taken to the external condyle
longer than that taken to the internal. The Javanese and the larger
Sumatra species also accord with the fossil in this respect, but not so
nearly as the small Sumatra species, which has consequently been intro-
duced into the above table.

The length of the fossil humerus, figs. 1, 2, Pl. XVII., exceeds that
of any of the existing species: its thickness is, in proportion to the
length of the bone, intermediate between the Sumatra and Indian species.
The articulating pulley also possesses a development intermediate in
value to those of the two existing species. The breadth at the condyles
is in the same proportion, or nearly so, as that of the Indian rhinoceros.
The radius is in length, considered with reference to length of femur, a
little less than in the Indian, and somewhat in excess of the small
Sumatra species. The remaining two dimensions of this bone yield
values intermediate to those of the two existing rhinoceroses. These
remarks apply to the deductions for fig. 1; nor would it be necessary
much to alter them in speaking of fig. 5; but fig. 6 presents such
a close approximation to the Indian rhinoceros, that it is much to be
wished that the specimen had not been so broken as to prevent
additional measurements being derived from it. Excepting in the length
from the articulating head to the bottom of the internal condyle, it does
not much differ from fig. 5. The bone, however, being imperfect,
must be omitted in drawing a comparison between the fossil and existing
species.

Fig. 1 varies most from the Indian rhinoceros in the proportion of
the length taken to the internal condyle—an anomaly difficult of expla-

nation. We must here repeat, that there exists a necessity for a greater
~ number of tables of dimensions taken from the skeletons of the Indian
rhinoceros. The anterior extremity of a rhinoceros, with the exami-
nation of which we have been favoured, yielded proportions so nearly
corresponding with those deduced from the fossil humerus, figs. 1, 2,
as to prevent our drawing more positive conclusions than those ex-
pressed at the close of the remarks on the cranium, Pl. XV.

Posterior Extremity.

The femur and tibia, P]. XVIII., were dug up in such close proximity
to the humerus and radius, fig. 1, Pl. XVIL., that little doubt could be
entertained of their having belonged to the same animal. Being perfect,
except at the lower part of the great trochanter, the specimen affords
ample means of comparison with the femur of the existing species.

On reverting to the Table of Dimensions, it will be observed that this
fossil exceeds, as did also the humerus, any of those in Cuvier’s table
of existing species. The following columns show in what respects the
proportions of the bone vary from those deduced from Cuvier’s Indian
rhinoceros. The length of the femur is here the modulus.

From a comparison of the two first columns in the annexed table there
results that the fossil has a greater development at its upper, and a
somewhat less development at its lower extremity, than is the case in the
Indian rhinoceros. The third trochanter is set lower down, and the
inferior extremity of the small trochanter higher up than in the existing

RHINOCEROS. 165

species; the articulating head is larger in proportion in the fossil than
in the Indian rhinoceros. None of these modifications, however, are
excessive; on the contrary, they are less than those which exist
amongst the fossils themselves, which are all three undoubtedly of the
same species.

2x) . | Fossil 3ra| Fossil 5th
Measurements Cuvier’s Fossil | in table of | in table of
Ind.Rhin.| Pl. 18 |qimensions\dimensions

Length of femur from articulating

head to bottom of internal condyle . | 1:00 1:00 1:00 1:00
Breadth from head to most salient a

of great trochanter . : : 0°38 0°43 cbc one
Breadth across condyles. - | 0:29 0°28 0°26 one
Antero-post. diam. of internal condyle . | 0°34 0°34 mae ove
Ditto ditto of external ditto. : 0:27 0°26 oud ses
Distance between bottom of 3rd tro-

chanter and top of lst . 0°59 0°61 SoC Soc
Ditto ditto ditto small trochanter and

top of head of femur . ; - | 0°46 0°41 0°46 0°42
Diam. of articulating head of femur .| 0-18 0°19 0°16 0°17
From lower'side 3rd trochanter to bot-

tom of external condyle . noe 0°38 0°38 sale
Length of femur from articulating head

to bottom of 3rd trochanter. =or 0°72 0-71 0°64
Length of tibia from anter. tubero. to

anter. edge of inferior ee ae

surface . > 0:6 0°70 eco 200
Greatest transverse diam. at top. -| 0:25 0:25 ec0 otic

Antero-post. diam. from antero-post.
tubero. to post. ext. of internal con-

mies: ra 029 2 0-81 a ue
Transverse diam. at pattem: A . 0-21 0°20 AOE ane
Diam. antero-post. of internal side .| 0-14 0713 nbc sea
Length of fibula . : . : - | 062 0°65 cic “
Breadth at bottom 5 : ‘ 0710 0:10 oie

From the manner in which the lower and exterior part of the great
trochanter is broken there is every probability that a descending
point protruded from the fractured surface towards the third trochanter,
the ascending point of which is very perfect.

The third ie Gchanter, however, differs from that of the existing species
as figured in Cuvier’s ‘ Oss. Foss.,’ in not possessing the double point ;

_ for it has a single well-defined ascending process, without any sign of
the bicuspid termination. The lower edge of this trochanter, instead
of ascending with a gradual swell towards the point, as in the existing
species, has a counter curvature to that of the upper edge. The chief
dissimilarity between Cuvier’s plate and the fossil occurs in this part
of the bone, the third trochanter assuming a different shape, and
offering a variation more distinctive than any other presented in either
extremity. This circumstance, together with some of the proportions
of the cranium, has led us for the present to distinguish these re-
mains by appendifg the word fossil to the name of that species of
which they are the prototype. But we dwell on the necessity of more

166 FAUNA ANTIQUA SIVALENSIS.

extended research, and the collection of a greater series of tables of
dimensions of the Indian rhinoceros, before anything absolutely con-
clusive can be pronounced with regard to the fossil and existing
species.

We have had no hesitation in ascribing the two limbs dug up in
such close neighbourhood to the same animal. An additional confirma-
tion of the correctness of the assumption may be derived from the
proportion which exists between these two extremities, when compared
with that which occurs in the Indian rhinoceros.

Ind. Rhin. femur and tibia mét.0°960 humerus and radius mét. 0°868
Fossil Ind. Rhin. do. do. » 1:056 ditto ditto » 0947

In the first, the humerus and radius are to the femur and tibia in the
ratio of 1 : 1:10; in the fossil, the ratio is 1: 1-11.

The analogy which exists between these fossil extremities and those
of the Indian rhinoceros being no less striking than that which was
observed between the cranium, Pl. XV., and the skull of the existing
species, we have considered such correspondence sufficient to prove
that the fossil anterior and posterior limbs appertained to an animal of
the same species, and of about similar size to the one of which the
cranium in question is a relic.

Even in the event of a much closer approximation of symmetrical
proportions than that given in this paper being obtained, we are aware
that identity of species could not be presumed. It could not be
assumed that the skin and the external appearance of the animal were
precisely similar to those of the existing species. The. fossil Indian
rhinoceros must, however, have presented a figure bearing a strong
general resemblance to the uncouth symmetry of its present repre-
sentative.

Measurements of Anterior Epi ee Eve Sake eae
Extremity.
Mét.| In. | Mét.| In. |Mét.| In. |Met.| In. | Mét In.

Length of humerus

from tub. to exter-

nal condyle . . | °538] 21°20] 488] 19°22) 482) 19-0 | ... | 22 | oe
Do. do. do. internal do | *492} 19°38) -461| 18°15} ... foe aea Seoalesee
Greatest anter. post.

diam. at top . a) coc «.. |°218] 8-60} *208] 8-20}-200} 7-90) ... ae
Breadth across con-

dyles . F .|°1938) 7-60) :183) 7:22) ... aot tees --. |°176) 6°94
Breadth of the articu-

lating pulley . .|°119} 4°70) °111} 4:40)°121} 4°80) :104) 4:10) -109] 4:30
Least diam. of the body

of the humerus’ .|‘078} 3:07] ... «ee | 073] 2°90) -071) 2°82) 069] 2-75
Length of the radius . | 409) 16-10} ... Bro Nace aes, eee avs Lleeee oe
Breadth at top . . | °124) 4:90) ... Beilacas eal ieee aon Rawe Ears
Ditto at bottom . = SEN AEC es seer see one
Length of humerus

from art. head to

internal condyle. | -441) 17-40) 393) 15-51) -420) 16-55) -389| 15-35] -398)15-70

167

RHINOCEROS.

eee

0&-€ | €80-

OT-6 | L&s-

9¢-F1| 698:

‘aL “QT
‘dg

eee eee ooo eee eee eee oe ooo $G.Z 790.
eee eee eee eos aes eee oe see 66-1 GOF-
eee eee see oon aoe aoe eee aoe OF-E 980:
see eee eee eee eee eee wee oe G0-¢ QZI-
eee eee see nee eos aoe wee eee OL-L G6I-
eee oe eee eee eee eee eee cee CL-9 9CT.-
see eee eee eee eee aoe eee eee CL-LI Cet.
co “+ 10G.0T| 992 | 03-8 | 80G- |00-£ | LLT- | 69-6 | 6FG-
0¢-€ | 680. | ~~ “107-8 | 980. | °~ S97 | Site
09-8 | GIG | °* “* 108-6 | 673 | °° “** 106-01 | 696-
eee eee eee see eee eae eee eee OL-GI ege.
nas cor iS eOe 1G Sie | eee ae OF-9 | GOT- | GE-9 I9T-
no * 1¢¢-9 | 99T- | ~~ bese: OL-8 | 166: | SF-8 FIG:
Cae “1 G).G | OFT. | 69-9 | EFT oer “| 68-9 eLT-
eee eee eee eee aes eee eos ovo 09-01 693:
01-06] O19. | ~~ *** 1@G-16 | 689. | ~~ “1 GF-FG | 169-
F6-G1 | 88 | * “* /OLGT| 688 | °°" “1 OL-LT | 6FF-
‘ul “QW “ul ‘RIL “ul ‘RAL ul “QW “Ul “RAL
‘dg ‘dg ‘dg ‘dg ‘dg

. “ ° ° . . e

T0430q 4B YIpRerg
. . . . . e ° . ernqy jo yyeueT
opis [eutezur *ysod-orezueR Jo “weRIC,
d : : : . * 110}}0q 4¥ “WUeIp esdoAsURLY,
. . . ° . . . . e,Apttoo yeu
ela jo "7x0 maod 0} ‘qny ‘toque WLOIF “Merp “ysod-or10jUy
oi 0 doy 4B “UIRIP OsOASMRA} 480}ROT0,
0 a 0 ° * — gaRyIns ‘oryare
“TOFU Jo edpe TeqUEB 04 ‘ozeqny “Teque ULOTF IQIy JO qysuery
: : * epApuoo
Toure} x2 jo U10}}0q 0} JoyURYDOTY pig epIs TeMOT WOT
0 . aTvUrey Jo poy qepnorjae Jo “weI,
0 0 0 : 0 ‘ * «mules Jo pray
jo do} pue seyURyOOI} [TRUS Jo WOYJOq TEA Joq eDULISTC,
ST jo doy pus IoJULYIOTY PIE JO MOJJoq UeeAJoq does
9 c e[Apuoa yeuseyxe “Op ‘op 0791.
* afpuoo yeuseyzut Jo ‘wip ‘qsod-o10yuy
G : 0 y : nepfpuoa SsO10B YpRelg
5 0 5 ‘ ; ; - zoqueyo
0m ede 70 qed querTEs qaou oF peat morIy YpRergq
* — e,Apuoa
yearoyur jo 103304 4 peoy “que wOay IMuUaF Jo WysueryT
* gaquryo
-01} pig JO WOJOq 0} pRey “jUR WOI; IMWEy Jo YyduerTT

AgTWMEI,XY IOL10}s0g JO syuoMernseayy

168 FAUNA ANTIQUA SIVALENSIS.

Measurements of the Head

Height of occiput from lowest edge of occipital

foramen to top of crest . 0-259 | 10°20 | 0-223 | 8°78
Greatest breadth of Bsc a behind auditory
foramen. . . |0°341 | 13-44 | 0-266 | 10°50
Least thickness of cranium at temporal bones. | 0°126 495 | ... coe
Breadth between post. orbital apophysis of fron-
tals : . |0°254 110-00] ... os
Distance from anterior of orbit to auditory
foramen. ¢ - . |0°325 | 12°80 | ... one
Space occupied by the seven molars - A . |0°324 |12°75 |... sec
Breadth across occipital condyles. = - |0°195 | 7:70 | 07140 | 5°51
Ditto of occipital foramen . - : 7 oilecares ee =| 0°0575) 2°25
Height of ditto ditto . “ ose --- |0°049 | 1:90
Distance between internal extremities of glenoid
facets of temporal . . see se» |0°0735) 2°88
Ditto from lower edge of occipital foramen to me-
dian post. extremity of palate . : 0°368 | 14-50 * eee
Ditto from post. of right occipital condyle to
spring of nasal arch : - : 0°539 | 21:22] ... Bs
Ditto ditto ditto to anterior of orbit : 0-449 17°71 | 4. os
Depth from edge of maxilla at 5th molar to upper
surface of frontals . 0-289 | 9:42 | ... eee
Greatest transverse width of nasals at horn site. |0°174 | 6°86 = >
Ditto external breadth at 6th molar . : 0246 | 9°72) ... eee
Thickness of cranium over the median post. ex-
tremity of palate . 0204 | “8:06 | ~.c5 oer
Height of highest point of nasal arch above an-
terior of palate . : 0°238 | 9°38 eee
Perpendicular from a line tangential to the sum-

mit of crest and vertex of nasal arch to the de-
pression of frontals . : . : : . |0°099 | 3°91

Measurements of [st Sp. 2nd Sp. 3rd Sp. 4th Sp.
Upper Molars a |
Mét. | In. | Mét.| In. | Mat. | In. | Mét. | In.

Greatest length

Molar1]| ... ««- |°030 | 1°19 | :0295] 1:14 | -030 | 1-20
035 | 1°36 | -034 | 1:335]-038 | 1-49 | -0395) 1°53
045 | 1°75 | -0475) 1-85 | -053 | 2-07 | -056 | 2:17
049 | 1°92 | -058 | 2:26 | -061 | 2°39

“ID Cr > Go bo

Greatest breadth

Molarl} ... «-- |°024 | 0-95 | -024 | 0-95 | -0285] 1-09
2|-059 | 2°31 |-0885) 1-5 |-036 | 1-40 |-041 | 1:58
3 |-080 | 3:15 |-049 |1:9 |-045 |1-88 |-053 | 2-05

RHINOCEROS. 169

Sp. 1 Sp. 2 Sp. 3
Measurements of Lower Molars
Mét. In. Met. In. Met. In.

‘016 | 0°61 one Sec
0387 | 1:44 | 0335} 1:30 | °038 | 1:29
053 | 2°09 | -050 | 1°98 | -0425} 1°67
"047 | 1°82 | :056 | 2°18 | -046 | 1°79

Greatest length of Molar

Greatest breadth of Molar .

020 | 0-77 | -021 | 0-81
-o26 | 1-01 | -027 | 1-05 | -025 | 0-98
029 | 1:12 | -029 | 1-10

IMA SOPWNEPATMBMHMPwWNHe

IL—Descrietion sy Dr. FaLconer oF Fossiz REMAINS OF RHINOCEROS
in Museum or Asiatic Society oF BENGAL. REPRINTED FROM
CATALOGUE OF MusEumM.

A. From the Sewalik Hills.

No. 269. Rhinoceros Sivalensis ?—Fragments comprising the greater
part of the cranium broken off behind about the posterior parts of the
zygomatic arch, the fracture having removed the whole of the occiput
and the left zygomatic arch. The specimen had also suffered from a
crush acting from above downwards from right to left ; the greater part
of the parietal and the whole of the frontal, and also the united nasals
are present; the right orbit broken off; the left nearly entire. The
right maxillary shows the remains more or less of seven molars, the
last broken off, the penultimate well worn; the anterior teeth have all
their crowns broken off nearly on a level with the alveoli; on the left
side, the crowns are all broken off; the palate seems narrow, but this
may be probably owing to the crush; the tip of the nasal shows the
rugous gibbosity of the base of a very large horn. The species was
evidently unicorned. From the Sewalik hills near Nahun.

No. 270. Rhinoceros —— ?—Lower jaw, left side showing greater
part of horizontal ramus, but broken off in front and behind, with the
remains of four molars, the crowns all broken off.

No. 271. Rhinoceros ?—Lower jaw, right side, broken off in
front at commencement of symphysis and behind at the coronoid, with
remains of five molars, much mutilated. In condition like No. 270.

No, 272. Rhinoceros ?—Fine fragment comprising the lower end
of tibia and fibula, right side, attached to each other and to the bones
of the tarsus in their natural position, together with the greater part of
the length of three metatarsals also united, and attached to the carpus:
the inferior apophysis of the calcaneum is broken off, the tibia bent
nearly at right angles with tarsus and metatarsus. All the bones are
held together by argillaceous matrix in their natural relative position

170 FAUNA ANTIQUA SIVALENSIS.

in a manner which is remarkable for the fossil state among Sewalik
remains. From the clay-marl of Maginnud. Figured by Messrs.
Baker and Durand in Journ. As. Soc., vol. v. Pl. XVII. fig. 19.

No. 273. Rhinoceros ?—Upper extremity of humerus showing
the head and upper trochanters; the descending spine of the large
tuberosity broken off: of large size.

No. 274. Rhinoceros ?—Upper extremity of humerus, right
side, showing the head and both tuberosities, as also the middle apophysis
of the upper end,

No. 277. Rhinoceros ?—Right femur, articulating head with
part of shaft attached; leafy expansion of third trochanter broken off.

No. 278. Rhinoceros ?—Shaft of femur, left side, articular
epiphysis and trochanters broken off, base of leafy expansion remaining.

No. 280. Rhinoceros ?—Lower end of femur, left side, showing
the condyles and trochlear pulley with a short portion of the shaft
attached.

No. 283. Rhinoceros
pulley and part of olecranon,

No. 285. Rhinoceros
cular surface nearly entire.

No. 287. Rhinoceros
No. 288. Rhinoceros
No. 289. Rhinoceros

entire.

No. 302. Rhinoceros ?—Entire humerus, left side found em-
bedded in argillaceous matrix, which has been partly removed, laying
bare the articular surfaces of both extremities together with the tuber-
osities, and a great part of the shaft on one side; the lower jaw of a
horse, both rami, together with the lower end of the left femur of the
same animal united to it on the other side by matrix. The humerus is
of large size, and equal to Nos. 273, 274.

?—Top of ulna, left side, showing articular

?—Upper half of right tibia with arti-

?— Astragalus, very perfect, of right side.
?—Calcaneum, left side, nearly entire.

?—Middle metacarpal of right fore leg

Dimensions.
Inches
Length from tuberosity to external Be bes : ° . ° . 194
To internal ditto . : : : 5 : - 18:8
Width of condyles : - . : : o- 44

There is great obliquity in the plane of the ‘distal end of the lower
articulation, the outer condyle projecting very much beyond the inner,
as in Baker and Durand’s figure, Journ. As. Soc. vol. v. Pl. XVII. figs. 6
and 7.

A notable specimen as indicative of the manner in which the bones
of different animals were washed together into the mud-beds of the
Sewalik strata: the Maginnud bone-bed being a stratum of argillaceous
earth: the bones soft and white and adhering to the tongue. The mud
matrix penetrates into the cores of the hollow bones.

No. 324. Rhinoceros -—— ?— Molar of upper jaw, nearly entire, with
part of palate attached.

RHINOCEROS. 171

No. 326. Rhinoceros
right side.

No. 764. Rhinoceros

No. 771. Rhinoceros
surface.

No. 773. Rhinoceros ?2—Astragalus (left) of a small-sized
species nearly entire (vide No. 287).

No. 782. Rhinoceros —~?—Scaphoid and Semi-lunar of the left
carpus of the same individual fitting together and connected by matrix,
both of large size.

No. 786. Rhinoceros
No. 787. Rhinoceros ?—Cuboid of left tarsus agreeing in size
nearly with Rh. unicornis, but differing in form.

No. 788. Rhinoceros ?—Fragment of left scapula, comprising
glenoid cavity; tuberosity, and neck spine and lamina broken off; of
the size of the Indian unicorned Rhinoceros.

?—Entire outer incisor, lower jaw

?—Lower end, right tibia, of large size.

?—Top of right radius with articular

?—Cuneiform bone of left carpus.

B. From Perim Island.'

No. 29. Rhinoceros Perimensis—Lower jaw, left side, including
part of horizontal ramus, posterior angle and ascending ramus, and
portion of the last molar: coronoid and condyle broken off: of very
large size. Presented by Lieut. Fulljames——See Journ. As. Soc.,
Wie 19.

No. 30. Rhinoceros Perimensis.—Fragment consisting of the supe-
rior maxilla left side, containing two molars, well worn.

No. 31. Rhinoceros Perimensis.—Lower jaw, right side, consisting
of posterior angle with one unworn molar, and portion of ascending
ramus, in three pieces: of much smaller size than No. 29, and probably
of a young animal.

No. 83. Rhinoceros Perimensis.—Humerus, inferior end, right side,
with articular surface.

No. 109. Rhinoceros Perimensis?—Fragment of left scapula com-
prising the glenoid cavity, neck and tuberosity together with the low
commencement of the spine of very large size: the whole of the lamine
broken off. The curve of the glenoid cavity on the antero-posterior
direction is very great: and the cup is much less circular than in the
most of the known forms of rhinoceros. Doubtful if from Perim
island: the matrix resembles that of specimens from Ava.

No. 110. Rhinoceros Perimensis.—Part of an upper molar vertically
broken through transversely about the middle, agreeing in size with
No. 30.

No. 111. Rhinoceros Perimensis—Metacarpal, outer toe, left fore
leg with both articular surfaces. The bone is short and the inferior
articular surface very thick and oblique.

1 See also Fauna Antiqua Sivalensis, Pl. lxxvi.

172 FAUNA ANTIQUA SIVALENSIS.

C. From Scinde.

No. 5. Rhinoceros ?—Tnner metacarpal of left fore leg, a little
mutilated at the upper articulation. It is proportionally very short
and the upper articulation deep as compared with the Indian unicorned
rhinoceros.

No. 6. Rhinoceros ?—Fragment of the left scapula of a very
large species, showing the greater portion of the glenoid cavity, neck
and a part of the spine: the greater part of the blade broken off. The
lower margin of the glenoid cavity partly broken off. Tuberosity of
very large size.

No. 8. Rhinoceros ?—Distal extremity of middle metatarsal,
of large size, shaft broken off.

No. 10. Rhinoceros ?—Fragment of left scapula of a small-sized
rhinoceros, showing glenoid cavity, neck and trochanter: blade broken
off. Ofa much smaller size than No. 6.

D. From Ava.

No, 23. Rhinoceros ?—Fine fragment comprising the lower
half of the right humerus with the articular surface very perfect : an old
animal, very nearly of the size of the Sumatran Rhinoceros.

No. 24. Rhinoceros ?—Lower end of radius right side, show-
ing the articular surface with part of the shaft of very large size. Width
of articular head being 5°7 inches: Antero-post. diam. 3 inches.

No. 25. Rhinoceros
size.

No. 26. Rhinoceros ?—-Axis, a good deal mutilated, showing
the greater part of the body, but the apophyses broken off; posterior
articular surface cup-shaped ; odontoid process thick and massive: be-
longed to an animal of large size. In mineral condition and wearing
this resembles some of the Perim Island fossils.

No. 245. Rhinoceros
mutilated.

?—Fragment of Os innominatum of large

?—Detached molar, very much worn and

FOSSIL RHINOCEROS OF TIBET. 173

VII. ON THE FOSSIL RHINOCEROS OF CENTRAL
TIBET AND ITS RELATION TO THE RECENT
UPHEAVAL OF THE HIMALAYAHS.!

BY H. FALCONER, M.D.

Tuat fossil bones occur on the Hioondés or elevated plain of
Tibet, at the northern face of the Himalayahs behind the
sources of the Ganges has long been well known. They are
brought to Almorah by the Bhoteah merchants, and sold as
talismans or charms under the name of ‘Bijli ki har’ lightning
bones; ammonites, from the crests of the neighbouring
snowy passes, called ‘ Chakar futteer’ and venerated all over
Hindostan as the sacred Salagram, are generally found
mixed up with them. The occurrence of these organic
mammiferous remains appears to have been first established
by Captain Webb and Mr. Traill; but little or no attention
has yet been paid to the determination of the species, the
circumstances under which they are found, or the general
results to which they lead.

Some of these fossils belong to a large species of Rhinoceros,
others to a bovine ruminant, as large as the Indian wild
buffalo ; and when it is remembered that the bed of the Sutlej,
where it flows through the Hioondés or Steppe of Chang-tang
at a lower level than the situation of the stratum in which
the bones are found, is elevated 15,000 ‘feet above the sea,
and that the natural vegetation at present hardly anywhere
attains the size of a shrub—not to mention the Polar
severity of the climate—it will at once be seen that the case
involves important considerations regarding the physical
changes which must have taken place in this part of the
Himalayahs since the Rhinoceros remains were entombed in
the stratum where they are now met with. But to give any
value to the results, it is necessary that all the facts of the
case be subjected to a rigid investigation.

1 This interesting paper, which was | and Equus, from the Niti Pass, are to
probably written about the year 1839, | be found in the British Museum, and
is now for the first time published. | are figured in the Fauna Antiqua Siva-
Fragments of bones of fossil Rhinoceros | lensis, Plates Ixxvi. and Ixxxiy.—[Ep.]

174 FAUNA ANTIQUA SIVALENSIS.

And, first, in regard to the fact of fossil bones occurring in
the Hioondés. No competent European observer has as yet
seen them in situ. Moorcroft and Hearsay are the only tra-
vellers who have traversed the tract where they are said to
occur. They went over the Niti Pass, and thence north
across the plain of Chang-tang by Dhapa to Gortope; thence
eastward to the lake Manasarovara and back to Niti by
another route along the Sutlej, the course of which they
followed to Dhapa. Their journey embraced about a degree
of longitude and latitude through the tract where the fossil
bones are said to be found. But Moorcroft nowhere makes
any mention of them; ‘ Bijli ki har’ are not even noticed in
his narrative. He describes lofty gravel and clay precipices
near Dhapa, and states his disappointment at not finding
traces of marine remains in them. * He also mentions having
found abundance of ammonites at the Changlu river, under
the Niti Ghati, on his return route. Captain Webb ascended
to the crest of the Niti Pass and procured fossil bones
brought from the plain of the Hioondés, some of which, to be
noticed in the sequence, are figured in Royle’s Illustrations
of the Botany of the Himalayahs, Pl. III. Mr. Traill,! in
his Bhoteah and Kumaon reports, mentions the occurrence of
fossil bones, and says they ‘would appear to have belonged
to some large animal of the ox species, probably the Yak.’
He further states ‘that the Bili ki hdr are chiefly found at
the crest of the Niti Pass.’ Mr. Batten, in his most graphic
account of a visit to the Niti Pass,? says he advanced about
two miles beyond the ravine of the Sianki river on the
Steppe of the Hioondés and came upon the Ammonite Fossil
Ground. He subsequently mentions having ‘a good many
fossil bones from the interior of Tibet and the Mana Pass;’
but it does not appear that he saw any of them in situ. The
fact, therefore, of their occurrence still wants the important
testimony of direct observation; but the other evidence to
the point is so good as to leave no room for reasonable doubt
on the subject. This evidence is as follows:

1. The concurrent statements of good observers, such as
Webb, Traill, and Batten, supported by specimens, that fossil
bones are found in the northern faces of the Niti and Mana
Passes, and the Steppe of the Hioondés.’

2. The direct testimony of the Bhoteah merchants who

1 Asiatic Researches, vol. vii. p. 17. | brought down from a very high eleya-

2 Jour. Asiatic Soc. vol. vii. p. 310. tion to the Commissioner of Revenue in

3 Mr. McClelland does not appear to} Kumaon, during my residence in the
have had an opportunity of examining | province; but not having inspected the
these fossils, but he states that ‘a skull, | fossil, I cannot answer for the fact’”—
said to be that of an elephant, was | Kumaon Inquiries, p. 216.

FOSSIL RHINOCEROS OF TIBET. 175

bring the fossil bones to Almorah; they state that they are
found in ravines in the plain below the Snowy Passes.

3. The universality of the belief at Almorah, where the
Byli ki har are brought, that they come from the plains of
Tibet, and from nowhere else.

4. The absence of any grounds tending to discredit the
evidence in favour of the fact.

Next in regard to the geological features of the fossil
tract. Mr. Batten,! from whom the most of what follows is
derived, describes the rocks from the southern side up to the
crest of the Niti Pass: tale and clay slates predominate near
Malari; quartz rock, mica, schist, gneiss, and granite between
Malari and Gumsali. The granite contains abundance of
tourmaline and kyanite, as is the case all along the culmi-
nating axis of the mountains between the valley of the Spiti
and the Hastern sources of the Ganges. Above Gumsali the
road leads along granite and gneiss precipices. At Niti the
formations appear to alter, clay slate rising into hills with
a rounded outline, and a compact uncrystalline blue limestone
succeeding the granite series, and higher up an arenaceous
quartzose rock. From the source of Dhauliriver to the crest
of the pass the road leads up through crumbling of crags
of blue limestone, the top of the pass being strewed with
blocks of this rock and arenaceous quartz. The blue and
mottled grey limestone here noticed has an extensive range
of distribution all along the northern face of the Himalayah
chain abounding in Ammonites, Terebratule, Belemnites,
Zoophytes, &c., which have been met with in the valley of
the Spiti by Dr. Gerard, at the head of the Ganges by Mr.
Batten, and at Muctinath on the Gandaki river in Nepaul.?
Several of the species have been determined by Mr. Sowerby
not to differ from fossils of the English oolite. It is hardly
necessary to add that this limestone has no other relation
with the deposit which contains the fossil bones, besides
contiguity of place.

The top of the pass, which is round and open, commands
a view of the plain of Hioondés. - ‘ Right in front,’ says Mr.
Batten, ‘stretched a dreary plain, shrubless, treeless, and house-
less, terminated along its whole northern side, at a distance of
about 20 miles, by a low range of rounded brown hills, utterly
without shrub or tree or jutting rock, but very broken into
ravines and perpendicular faces on their Southern side. Had
there been heather instead of stone, it would have resembled
a highland moor.’ Its level was hardly anywhere lower than
the pass. He further states his opinion that ‘The Niti Pass

1 Batten, loc. cit. 2 Colebrooke, As. Research. vol. xii. Append. p. xxi.

176 FAUNA ANTIQUA SIVALENSIS.

is only the highest point of the Tartaric plain, running up to
the Himalayah peaks.’

From the details given by Moorcroft, it is very clear that
the upper stratum of this great plain consists of a deep allu-
vial deposit—whatever the age of the alluvium may be—
composed of beds of clay and gravel. He was struck, on en-
tering the country, with the broad flat channels of the rivers,
bounded by lofty steep banks, as contrasted with the narrow
angular beds on the Hindostani side of the mountains, being
precisely the shape that would be washed out by a torrent
running through soft unconsolidated strata. His description
further gives good reason to surmise that the alluvium rises
in successful steps like the parallel roads of Glenroy. He men-
tions broken ground with ravines near Dhapa, rising into pyra-
mids and buttresses, ‘ bearing no unapt resemblance to ruined
castles and fortifications in piles above each other.’ A ravine
near the Tiltil river yields a section of beds of indurated clay
and gravel above 300 feet in elevation; the heights are
broken into all manner of fanciful shapes, spires, buttresses,
&c., the sides being excavated into habitations. There is but
little variation from the above in his account of the country
between Dhapa and Gortope, or between the latter place and
the Manasarovara Lake, except at Tirthapuri, where he states
that ‘steep craggy limestone rocks in a state of decomposition
immediately overhang it (the village). Still higher, and
losing their heads in the clouds, are pointed mountains, which
from their brilliant whiteness appear to consist of chalk,
covered here and there with a layer of yellow ochre.’ Near
this spot he describes, in very characteristic terms, an enor-
mous bed of travertine, forming a table of about half-a-mile
in diameter, deposited from hot springs now in operation.
At Kienlung he met with other great travertine deposits,
perhaps not exceeded in extent by those hitherto observed in
any other part of the globe. ‘The vast walls and masses of
rock which have been formed by the action of hot springs in
this neighbourhood show an antiquity that baffles research
and would afford food for sceptics.’

So much for the general geological features of the Hioondés
plain. Of the particular beds which yield the fossils we have no
accounts, besides the meagre details which may be gleaned
from the Bhoteah merchants, who describe them as occurring
in broken ground with ravines, upon the surface of which they
are seen projecting or strewed over patches where the earth
has been washed away by rills formed by melting snow. The
specimens have rarely any of the matrix attached to them,
but where it exists it is usually of coarse sand or gravel,
agglutinated by a calcareous paste which effervesces strongly
with the mineral acids.

FOSSIL RHINOCEROS OF TIBET. 177

Judging from the quantities which find their way to
Almorah, the fossils are by no means scarce. They are rarely
seen entire, consisting generally of fragments three to six
inches long; sometimes the contents of a collection are no-
thing but bits of bone hardly an inch long. They usually
present a clean and sharp or splintery fracture, wearing the
appearance of having been fragmented after the mineraliza-
tion was complete. They vary greatly in the amount of
fossilization, and, consequently, in specific gravity. The
infiltrated mineral in most cases is carbonate of lime. The
specimens adhere more or less to the tongue. In some of
them the cancellated tissue has the cells entirely filled with
the infiltrated mineral; in others the cells are empty. It
is rare to see any tinge of iron about them, a character so
prevalent in the Sewalik fossils of the arenaceous beds.
One class of them has very much the appearance of bleached
bones, with the fracture also white; their fossil character
resting on a core of crystallized carbonate of lime and the
increased specific gravity. In another class the specimens
yield a dark blue fracture, and weather with very much of
the greyish white leprous appearances which chalk fluids
exhibit. They effervesce strongly with nitric acid, and
treated with a weak solution of it, the greater portion of
them dissolves ; they retain few or no traces of animal matter.

Our materials for the elucidation of the speciesare but scanty.
They are; first, a set of specimens in Captain Cautley’s col-
lection at Subarunpoor, received from Captain Corbet of
Almorah; second, specimens received from Mr. Batten of
Kumaon; third, specimens procured from a Bhoteah mer-
chant, said to have been collected by himself on the Hioondés;
fourth, Pl, ITI. of Royle’s ‘ lustrations,’ which contains some
figures of fossil bones procured from the northern face of the
Himalayahs by Captain Webb and Mr. Traill.

Rhinoceros Remains.\—These are, fortunately, very decisive.
Fig. 3, Pl. ITT. of Royle’s ‘Tllustrations,’ represents the greater
portion of a tooth evidently derived from a Rhinoceros, and
probably the fifth or sixth molar left side of the upper jaw;
but this is a point not to be determined by the figure, and we
have not yet had access to the letter-press relating to it.

The next specimen is a fragment in Captain Cautley’s col-
lection, consisting of the left half of the body, with nearly
the entire ala of the atlas or first cervical vertebra of a
Rhinoceros. The upper and lower articulating surfaces are
complete, and the bone is so characteristic as to leave no
doubt about its identification. There is one remarkable cir-
cumstance about it, viz. that there is no hole for the passage

1 See Plate xv., figs. 3 to 11.—[Ep.]

VOL. I. N

178 FAUNA ANTIQUA SIVALENSIS.

of the vertebral artery, the transit of that vessel to the head
having been outside, and not through the bone. But this is
merely an abnormal variation in the individual nowise affect-
ing the species. The bone differs somewhat in form from
that of the Indian Rhinoceros and is smaller, indicating a
distinct species.

A second specimen in my possession happens also to be-
long to the left side of the atlas of a Rhinoceros. It shows
more of the body but less of the ala than the other, and has
the arterial hole in the usual position. The form of the bone
and size confirm the distinctness of species indicated by
Captain Cautley’s specimen.

A third specimen is fortunately also very characteristic.
It consists of a fragment of the left temporal bone, showing
the posterior half of the zygomatic arch, the entire glenoid
articulating surface, the external auditory foramen, a portion
of the petrous bone and part of the temporal fossa. The
styloid and petrous apophyses are broken off. It appears to
have belonged to rather a young animal, as the commissure
between the base of the zygoma and the petrous bone is not
completely ossified. The fragment adheres to the tongue,
and is but imperfectly fossilized. The characters yielded by
it bear out the difference of species, indicated by the other
specimens, between the Indian Rhinoceros and the Tibet
remains. The glenoid articulating surface—a very character-
istic structure—has a different outline from that of the Indian
animal; the base of the zygomatic process has less vertical
height in proportion, and the dimensions are somewhat less.

The collections contain other fragments referable to the
Rhinoceros, but too much mutilated to afford any good
character for description or comparison. There are no traces
of any other Pachydermatous animal ; but Elephant remains
will probably be found hereafter, when the ground is well
examined, if they have not been already met with.'

It is a point of much interest as regards the genéral
bearing of the inquiry, to determine whether these Rhino-
ceros remains differ specifically or not from the fossil species
of the Sewalik range; but the available materials, in both
cases, are too imperfect to warrant any safe conclusions on
the subject. It appears sufficiently clear, however, that the
Tibet fossil species differ from the existing Indian Rhi-
noceros.

Ruminant Remains.—These are the most abundant in species
and in the numerical ratio of specimens. Fig. 1, Pl. III.
of Royle’s ‘ Illustrations,’ represents a very perfect cranium

1 Vide M‘Clelland’s Kumaon Inquiries, quoted above.

DESCRIPTION OF PLATE XV.

MERYCOPOTAMUS DISSIMILIS, Fosst, RHINOCEROS OF
Niti Pass.

Figs. 1 and 2. Represent the palatal and upper surfaces of the fragment
of a young cranium of Merycopotamus dissimilis, from Burmah,
sent by Dr. Oldham to Dr. Falconer. The figures are copied
from drawings made by Mr. Dinkel for Dr. Falconer, and are
one-third of the natural size. The palate surface shows the two
last premolars and the two first true molars: a, cavity for ante-
rior lobe of cerebrum ; 0 4, frontal bones; ¢ c, foramen in centre
of frontal bone; dd, nasal bones; f, suture between nasal and
maxulary bone; gg, maxillary bones. (See page 147.)

Figs. 3 to 11. Fragments of fossil Rhinoceros bones from the Niti Rass
in Tibet, one-fourth of the natural size. Copied from drawings
by Mr. George in Plate LXXVI. of the Fauna Antiqua Siva-
lensis. Figs. 3, 5, and 8 represent a fragment of the scapula,
including the glenoid cavity and coracoid process; fig. 4 is a
fragment of the left humerus near upper end; figs. 6 and 7 re-
present another fragment of a humerus; and figs. 9, 10, and 11
show a fragment of the lower end of a femur. The specimens
are in the British Museum. (See pages 177 & 517.)

VOL, I.

6 VoLI. Plate 15.

oe del. I Dinkel ith.
* 1,2. Merycopotamus dissmmihs, from Burmah.
. < % 3toll. Fossil Rhinoceros, from Niti Pass, in Tibet.

FOSSIL RHINOCEROS OF TIBET. 179

of a ruminant with the pedicles of a couple of horns attached
to the frontal. The saliency of the occipital crest, the sweep
of the parietals and the position of the horn pedicles show
that it belongs to the Cervine group of the family. But not
having the letter-press to refer to, and in ignorance of the
scale of dimensions on which the figure is drawn, it were
useless to hazard or guess about the affinities of the
species.

Fig. 2 of the same plate represents the left line of molars
of the upper jaw of a ruminant. Judging from the figure,
which shows no internal pillar between the barrels of the
molar, the specimen belongs to the Caprine group.

In Captain Cautley’s collection there is a specimen of the
articulating head of the lower end of a femur of a bovine
species. The dimensions fore and aft, between the articula-
ting extremities, are six and a-half inches, exactly equal to
the corresponding measurement of a full-sized wild buffalo
(B. Arna) killed in the Shahjehanpoor forests. The existing
Yak of Tibet is a much smaller animal. Another specimen
in the Suharunpoor collection is the fragment of a scapula,
corresponding in size with the femur. There are numerous
other fragments of ruminant remains in the Suharunpoor
collections, but none of them sufficiently characteristic to
merit mention, except the detached core of a twisted sheathed
horn belonging to some member of the Caprine group. The
horn which it bore must have been twisted on its axis, like
the ‘ Markhor ’ wild goat of the Baltistan Mountains (Little
Tibet), a large and undescribed species.

There are no remains in the collection which can safely be
referred to other mammiferous families except a solitary and
detached Hyzena tooth procured from the Bhoteah mer-
chant. It appears to be the third molar of the upper jaw,
and is of large size. The whole of the specimens of this set
are very much fragmented. They are white and have a very
recent appearance, but they have lost their animal matter,
have a considerable specific gravity, and the tubes of the
cylindrical bones are occupied by crystallized cores of
carbonate of lime, affording strong presumption of their
being honest fossils. The Hyena molar in question has the
pipes of the fangs and the centre of the tooth filled with a
nest of calcareous crystals.

This concludes what specially regards the determination of
the fossils. It is very evident that the list is incomplete, for
on a tract which could afford sustenance and a climate suited
to the Rhinoceros a great variety of species might be
expected. In what follows we put aside the consideration
of the others, and address ourselves to the Rhinoceros.

N 2

180 FAUNA ANTIQUA SIVALENSIS.

The Steppe of Hioondés has been shown by Captain Webb
to be upwards of 15,000 feet above the sea, close on the
limit of perpetual snow; it is bounded on one side by the
Himalayah Mountains, and on the other by the Kailasa range,
of enormous height, some portions being, on a rough ap-
proximation, 30,000 feet above the sea. The tract, in the
emphatic language of Batten, is shrubless and treeless—a vast
waste supporting a few furze bushes and a sprinkling of the
most Alpine vegetation; and the climate is one of Polar
severity.

It is very certain that no Rhinoceros of the present time
could exist for a day in such a habitat; and if we suppose
the Tibetian species to have been clothed with a dense fur,
like the Siberian species the carcase of which was brought to
Pallas from latitude 64° on the banks of the Lena, still the
tract could never have subsisted it, for although it has been
urged by Dr. Fleming that the simple analogy of anatomical
structure in the living species is not sufficient to guide us to
a conclusion, or even a conjecture, as to the habits, geo-
graphical distribution, or food, of extinct species, so clearly
shown in the lichen food of the Reindeer, still there is a
limit to the force of this objection, and it only applies to
certain cases. In the case of the Rhinoceros the incisive
teeth are deficient in number, and the greater portion of them
rudimentary in form and even deciduous. It may, therefore,
be very safely predicated of all the species, fossil or existing,
that they could never subsist by browsing on a herbaceous
vegetation ; they want the nippers which enable the horse
and ruminants to subsist on low grass; and their food must
either be derived from large reeds, shrubs, or trees, none of
which are now found in Tibet.

The Siberian Rhinoceros remains are found on the shores
of the frozen ocean, under conditions of climate more severe
than those of Tibet; and it has been shown by Lyell how
these remains might have found their way by changes in the
physical geography of Siberia, by transportation in ice blocks,
and by periodical migrations. But these conditions will not
apply to the Hioondés; the Rhinoceros could neither have
migrated to its mountain-locked plain, from the side of
Hindostan by the passes, where men and goats can hardly
find their way save by the artificial aid of scaffoldings, nor is
it apparent how the bones could have been transported to
their present resting place from a higher tract.

The only explanation of the case that suggests itself, which
appears admissible, is a depression of the plain of the Hioon-
dés to a much lower level than it has at present; and to
clothe it with a vegetation resembling that of England now,

FOSSIL RHINOCEROS OF TIBET. 18]

which, on the supposition that the Rhinoceros was not a
migratory visitor but a permanent resident of Tibet, and
clothed in a warm fur, is perhaps the utmost limit that
could safely be conceded for its habitat. The plain of the
Hioondés would require to have been not higher than 7,000
or 8,000 feet above the level of the sea. The mean level of
the Hioondés which is known at Dhapa to be 15,000 feet, and
estimated to be not much less than 17,000 near Manasarovara,
may be considered as 16,000 feet. ‘To reduce it, therefore, to
the circumstances above inferred would involve the con-
sequence that the northern face of the Himalayahs and (as
elevating movements are nowhere known to be confined to
narrow belts), probably a considerable portion of the chain
itself, have been elevated 7,000 to 8,000 feet since the tract
was tenanted by a species of Rhinoceros and several rumi-
nants allied to existing species.

There are unquestionable proofs on the southern side of
the chain that important elevations have taken place within
a very late period, geologically speaking. The Sewalik for-
mations are continuous with the Himalayahs, constituting in
physical confirmation but the outermost belt of the chain.
They bear, in fact, the same relation to the southern face
that the Steppe of Hioondés does to the northern. The fos-
siliferous strata attain a height of about 3,500 feet above the
sea, and some parts of the belt about 5,000, the plains at their
foot being about 1,000. These strata have not only yielded
numerous extinct mammalia, but, besides Quadrumana and
Camels, they have been shown to contain the remains of at
least two existing species of Crocodile, viz. the Magar and
Gharial, so common all over India; and the fluviatile shells (to
which the testaceous remains are limited) have been pro-
nounced by Mr. Benson not to differ specifically from recent
types, common in the northern part of Hindostan.! This
would show the upheavement, beyond all question, to date,
geologically speaking, since the commencement of the present
order of things ; and if so grand a movement has occurred on
the southern side of the chain within a late period, there is
no reason why a similar upheavement should not have taken
place on the Northern face.

Mr. McClelland has found proofs that a movement of ele-
vation has taken place in the opposite prolongation of the
chain in the valley of the Brahmapootra, in a marine deposit
of considerable height abounding in shells on the Kasia
hills. We are not informed what proportion of recent species
has been found in these shells, and consequently, as to the
age of the formation.

1 Stated on the authority of Mr. Everest.
#
Nn 3

182 FAUNA ANTIQUA SIVALENSIS.

If it is admitted that there are good grounds for the be-
lief that the plain of the Hioondés has been elevated several
thousand feet within a late period, it is necessary that we
should consider what further consequences are involved in the
supposition, and it will be evident that the entire line of
mountains from the Lake Manasarovara to the southern bend
of the Indus near Gilgit, in the parallel of Attock, must
have partaken in the movement. For as the course of rivers
from Manasarovara is due west, through a long intramontane
tract, had the Hioondés been 7,000 feet lower than it is now,
and the western prolongations of the river beds not been
proportionally depressed, the waters would have been held
up, and we should have traces of vast lacustrine formations
somewhere along the course of the Sutlej and Indus in
Ladakh, which, so far as our information at present goes,
does not appear to be the case. But as the great water-head
of the western and eastern drainage of the Himalayahs is in
the neighbourhood of Manasarovara, it is quite philosophical
to imagine that the centre and greatest force of the upheave-
ment was at the culminating point, and gradually decreased
westward.

That upheavement of the southern face of the Himalayahs
was inthis manner is almost susceptible of direct proof. The
Sewalik hills run west skirting the foot of the Himalayahs,
beyond the western banks of the Jhelum; and the character-
istic Sewalik fossils have been dug out of the strata between
the Jhelum and Chenab, near Bimber, where they exist in
abundance; they are also found between the Ganges and
Gogra, and it is almost certain that the formation extends
at least as far as the Gogra, giving a protraction in length of
270 miles, between the Jhelum and the Gogra. The greatest
height of the fossiliferous strata is between the Jumna and
Ganges, the elevation diminishing westward. It is, therefore,
a matter for inference that the greatest force of the upheave-
ments was at the culminating point, and was feebler as it
extended westward.

It is a matter of much interest to determine whether these
upheavements of the northern and southern faces were con-
temporaneous events. There do notappear any good grounds
for coming to a satisfactory opinion on the subject, but there
ean be very little doubt that they belonged to the same
geological era.

With these undoubted proofs (in the Sewalik hills) before us
of comparatively late uprisings of the Himalayah mountains,
it naturally occurs to the mind to inquire if the chain has
been in a state of quiescence, as far as level is concerned,
since the historical period, and if it is so in our own times.

FOSSIL RHINOCEROS OF TIBET. 183

The proof is embarrassed with immense difficulties in all
mountainous tracts at a distance from the sea, which alone
affords a certain standard for comparison; and this difficulty
affects the central portion of the Himalayahs. But we shall
endeavour to show that there are grounds sufficient for en-
tertaining the presumption at least, that the Himalayahs
are now undergoing a process of upheavements.

In Mr. Traill’s excellent report on the Bhoteah Mehals, or
region of the Tibet passes, occurs the following passage,
which is so important to the point that it is given at full
length.

‘The paths to the passes’ (the Mana, Niti, Juwar, Darma,
and Beeans passes) ‘continue along the upper part of the
rivers above mentioned, till near the crest of the ridge, which
is crossed in parts offering least difficulty in the ascent, and
it is here only that snow is not met with during the season of
intercourse. Roads of communicationthrough the Himalayahs
unite the passes from Hast to West, but they are passable
during a few days only in each year, and are considered at
all times dangerous by the Bhoteahs themselves. Roads of
this description formerly used are now impracticable, owing to
the increase of snow. The interior of the Himalayah, except at
the passes and paths in question, is inaccessible, and APPEARS
TO BE DAILY BECOMING MORE SO FROM THE GRADUAL EXTEN-
SION OF THE ZONE OF PERPETUAL SNOW. The Bhoteahs bear
universal testimony to the fact of such extension, and point out
ridges now never free from snow, which, within the memory of
man, were clothed with forest and afforded periodical pastures
for sheep; they even state that the avalanches detached from
the lofty peaks occasionally present pieces of wood frozen im their
centre.’

Now these statements are of much importance, and their
value is enhanced by the circumstance of their coming spon-
taneously from an unprejudiced inquirer. Mr. Traill attempts
no explanation; he simply records the proofs and the universal
belief that the zone of perpetual snow is descending lower.
Tt is true that, before any conclusions could be safely drawn
from them, the asserted facts will require to be verified and
the observations extended, but they are at present sufficiently
plausible to justify some speculations on the subject.

The circumstance of most weight is the assertion that pieces
of wood are found frozen in the centre of the avalanches de-
tached from the lofty peaks. Now it is very evident that
this could only happen by a descent of the perpetual snow
zone upon tracts where forests once grew, for it is difficult, if
not impossible, to imagine how pieces of timber could at such
enormous elevations be transported from below, so as to be

184 FAUNA ANTIQUA SIVALENSIS.

embedded high above in a mass of snow. But a descent, so
to speak, of the snow zone could only occur in two ways,
either by the line of perpetual snow being actually lowered
to the level of the sea, or, supposing it to maintain a constant
mean height, by an elevation of the mountain belt into the
snow zone; either of which would produce, in appearance,
the same effects.

Now, in regard to the first supposition of the lowering of
the line of perpetual snow, the conditions which regulate the
limits of that line are only very imperfectly understood, but it
may safely be asserted that there are no grounds to believe, so
far as our knowledge at present goes, that it oscillates more
than the mean temperature of a place does; and the variation
in this case does not extend beyond a few degrees of ‘Fahr.
Humboldt found that in the Andes, under the crater, the
oscillation of the line of perpetual snow does not exceed thirty
fathoms. In the Himalayah Mountains the present elevation of
the line of perpetual snow is a huge anomaly, the plane being
upwards of an English mile in excess of the amount yielded by
calculation, with a formuia for the latitude and height above
the sea.' If, therefore, we suppose that the pieces of timber
mentioned by Mr. Traill got enveloped in an avalanche by
a lowering of the zone of perpetual snow, it would necessarily
be implied that the plane of congelation was formerly more
elevated, and would involve a still greater irregularity than
the enormous extent at present ascertained, a position which
it would be unphilosophical to admit, except on the strongest
grounds.

On the second supposition, that the mean altitude of the
plane of congelation is nearly constant, and that the moun-
tains have been elevated into the snow zone, the instance of.
the enveloped timber would admit of two explanations ; either
that it belonged to the age when the Himalayah Mountains
had their elevation increased by the Sewalik and Tibet up-
heavements or that the tract on which it grew had been
subsequently raised up into the zone of congelation. That
these mountains, before their summits attained their present
elevation, were clothed with forests high up on the tract
which is now covered with perpetual snow, is but consonant
with the course of nature to suppose; and wood once en-
veloped in a snow bed would retain a freshness unim-
paired for countless ages; we might, therefore, in a piece

Feet
? Perpetual snow level Niti Pass, Lat. 31° . ‘ . 17,000
Caleulated height of ditto by Professor Leslie’s formula
for Lat. 31° : : : : ‘ P . 115253
Difference a : : : . 5,747 feet

paced 4a

FOSSIL RHINOCEROS OF TIBET. 185
of green wood, which descended from the higher peaks
in an avalanche, light upon a remain which had a con-
temporaneous existence with the Sivatherium in Hindostan,
or the Rhinoceros in Tibet; and it would be a matter of
extreme if not insurmountable difficulty to determine to
what period of the interval between these upheavements and
the present time its envelopment in the snow should be
referred.

The other circumstances mentioned by Mr. Traill, viz.
that roads of communication from EH. to W., between the
passes formerly used, are now impracticable; that the zone
of perpetual snow is gradually extending ; and that ridges
which, within the memory of man, were clothed with forest
and afforded periodical pasture for sheep, have an obvious
and important bearing on the question.!

1 Memoranda from Mr. Edgeworth, ex-
tracted from Dr. Falconer’s Note-book.—
‘1. On the Vishnoo Gunga, between
Bhadra Nath and Pundoo Kesur, there
is an artificial mound, at a place called
Kutlean Kotee, which the Puharees say
is the remains of a large hill city, that
became deserted in consequence of the
increased cold or descent of the snow
zone. Charcoal and remains of pottery
are found in it, and Edgeworth says the
mound is, beyond all doubt, artificial.
2. There is a current tradition that for-

merly there was a straight path between
Bhadra Nath and Kedar Nath, which
has become impassable, so that a detour
of several days is now necessary.—3.
There was formerly a pass up the Bhil-
lung river, which led into Tibet. It
was last crossed more than fifty years
ago, during the Goorka first invasion.
Since then an attempt was made to
eross it, but the party, of whom Edge-
worth’s informant was one, were struck
with snow-blindness and nearly lost, so
that they had to return,’—[Ep. |

156 FAUNA ANTIQUA SIVALENSIS.

VIII. ON THE FOSSIL EQUIDA OF THE SHEWALIK
HILLS.

No memoir on the fossil Equidee of India was ever pub-
lished by Dr. Falconer. Six new species are figured in the
‘Fauna Antiqua Sivalensis, two of which are from the
Sewalik hills, viz. Hguwus Sivalensis and Hippotherium Antilo-
pinum, the latter presenting the form of a horse with the
attenuated proportions of an antelope. Two other species,
Equus Namadicus and Equus Paleonus,' are from the valley
of the Nerbudda, and there are two doubtful species from
Ava and the Niti Pass. The reader is referred to the
descriptions of the Plates in the Fauna (LXXXI. to LXXXYV.)
and to a memoir on sub-Himalayan fossils by Messrs. Baker
and Durand, published in the ‘ Journal of the Asiatic Society’
for October 1835, vol. iv. p. 568, and also to the following
extracts from the catalogue of the Museum of the Asiatic
Society in Calcutta.—[ Hp. |

Descrirtions BY Dr. FaLtconer oF Fossir Equip&z In THE Musrum
OF THE AsiaTIC Society OF BENGAL.

A. From the Sewalik Hills.

No. 803. Equus Stvalensis.—Fine specimen, connected by matrix
to another fossil comprising both horizontal rami of the lower jaw from
behind the molars on to the middle of the symphysis, in two pieces;
the rami united by their inferior border to the shaft of a rhinoceros
humerus, leaving their sides and the lines of molars free and in relief.

The whole series of six molars is present on either side, adult and
well-worn but not aged, those on the right side having little matrix on
the crowns of the first four; the two backmost, more or less covered,
as also the greater part of the crowns on the left side. The left ramus
is depressed somewhat by a crush below the level of the right. The
teeth resemble in size and pattern of crown those figured in the Faun.
Antiq. Sival. Pl. LXXXII. fig. 2. The front fragment passes nearly
through the middle of the symphysis, showing in section the included
portions of two canines, one on each side. The jaw is nearly the size
of that of an English horse cranium in the Asiatic Society’s Museum,

1 At page 22, the specimens of Equus | toM. Lartct, Z. Paleonus is probably the
Paleonus are erroneously stated as being | young individual of either 2, Sivalensis
derived from the Sewalik hills. According | or £. Namadicus. See note 2, p. 22.—[ Ep. |

EQUIDA. 187

the principal difference being that the height of the diastemal portion
is greater in the fossil.

The following are the principal dimensions.

Extreme length of the fragment
Depth of jaw behind the last molar .
Ditto in the middle . - 5 :
Ditto at commencement of diastema .
Length of 6 molars .

Length of diastemal portion

a

RO ATO oo eH os
We AME a

No. 304. Hquus Sivalensis.—Lower half of left femur, attached by
the anterior surface longitudinally to the shaft of the humerus of the
rhinoceros, No. 802. (See page 170.) Showsa portion of the shaft, both
condyles, the rotular pulley and the characteristic pit on the shaft above
the outer condyle. Probably belonged to the same animal as the jaw.

No. 305. Equus Sivalensis.—Fragment comprising the sympihysial
portion of the inferior maxilla, broken off on the right side close to
the symphysis, the greater part of the diastema remaining on the left,
presenting the six incisor teeth 7m situ of a young adult (say five or six
years old) the permanent teeth being all out; and the bases of the cups
of the central incisors not yet worn off, the middle ones being large ;
unluckily the crowns of all these teeth have been more or less injured
by attrition, and the form of the outer incisors is not seen, except that
there is a rather deep vertical groove upon the inner and posterior sur-
face of the outermost differing from what is seen in the English horse.
A shallow vertical furrow is also seen upon the outer surface of the two
outer incisors. ‘The size of the specimen corresponds very closely with
thatof the samepartsin the English horse cranium compared with No. 303.

No. 306. Equus Stvalensis—Lower end of femur, right side, show-
ing the condyles and articular surfaces nearly entire, with a small part
of the shaft attached. The condyles are quite entire, and the inner
border of the rotular pulley presents the characteristic large salient pro-
jection.

No. 807. Equus Sivalensis.—Fine fragment comprising part of the
horizontal ramus right side, ‘broken off immediately behind the last
molar and in front through the posterior half of the third premolar ;
showing three entire teeth, and part ofa fourth, viz. the three true molars
and last premolar; the last molar is only partially worn, showing the
animal to have been a young adult. The size and pattern of the crown
agree very closely with the specimen No. 303. The fossil is black,
hard, and heavy, and the teeth finely preserved. The large flexures of
the enamel plates well shown.

No. 315. Hippotherium Antilopinum.—Lower end of radius, right
side.

No. 316. Hippothertum Antilopinum.—Fragment comprising the
carpus, left fore-leg, and metacarpals united in the natural manner, the
metacarpus broken about the middle of the shaft; the upper part of
the lateral metacarpals shown. Of small size.

No. 369. Equus Sivalensis.—Lower end of right humerus, showing
only the articulating head somewhat crushed, and therefore doubtfully
determined, but in the absence of the third pulley agreeing with Horse.

188 FAUNA ANTIQUA SIVALENSIS.

No. 391. Equus Sivalensis.—Articulating head of right humerus very
perfect, but detached from the shaft.

No. 676. Equus Sivalensis.—Mass of matrix containing caleaneum of
Equus Sivalensis with two vertebre adhering.

No. 696. Equus Sivalensis—First cervical vertebra (atlas). A fine
specimen, nearly entire, and nearly of the same proportions as in the exist-
ing horse, but the body somewhat shorter ; the spmous tuberosity less
prominent, and the posterior inferior margin free from a hitch or emargi-
nation. The specimen is only defective in the marginal expansion of
the transverse processes, where the edge is broken off.

No. 697. Equus Sivalensis.—Fourth cervical vertebra. Body nearly
entire, processes more or less mutilated.

No. 698. Equus Sivalensis.—Fine specimen of the sixth cervical
vertebra. Body entire; neural arch broken off; inferior transverse
process on left side entire and very broad and massive, more so than
in the recent horse; broken off on right side.

No. 700. Equus Sivalensis—Very fine specimen of the third
cervical vertebra; the body nearly entire, and formed closely, as in the
existing horse, with the same broad expansion of the transverse processes
delated posteriorly ; the lips of both superior and inferior transverse
processes together with the neural arch broken off. The form generally
resembles that of the same bone in the ‘ Nyl Ghau;’ but it is at once
distinguished from the ruminant type by long and nearly concealed
vertebral foramina, the entrance of the artery being close to the margin
of the spinal canal.

No. 701. Equus Sivalensis—Seventh cervical vertebra. Body
nearly entire, neural arch present, but spinous process broken off; dis-
tinguished by the absence of a vertebral foramen ; a costal articulation
shown on one side of the posterior surface ; closely resembles the same
bone inthe recent Hq. Caballus. The anterior articulating surface broader.

No. 702. Equus Sivalensis.—Fifth cervical vertebra. Body nearly
entire; posterior articular surface having the lateral margin broken on
the left side; neural arch present; spinous axis styles broken off. Ver-
tebral canal with a very large opening; a good deal of matrix on the
specimen. od

No. 705. Equus Sivalensis.—Mutilated astragalus.

No. 706. Equus Sivalensis—Proximal phalanx, fore-leg, of small
size, nearly entire.
B. From Perim Island.

No. 384. Equus.—Upper jaw, left side, containing the two anterior
premolars, well worn, corresponding in characters with Hy. Sivalensis,
Faun. Sival., Pl. LX XXII.

No. 35. Equus.—Astragalus, small size.

No. 36. Equus.—Lower jaw, left side, containing the four anterior
molars, well worn; truncated in front and behind; agrees with No. 34.

No. 37. Equis.—Lower jaw, right side, containing the three anterior
molars, with portion of the diastema in front, and truncated behind.

——

in ed

a WE

EQUIDA. 189

No. 66. Equus.—Atlas of a very small-sized horse—Hippotherium
Antilopinum? Faun. Ant. Siv.

No. 67. Hippotherium Antilopinum.—Axis, nearly entire, inferior
cup wanting ; corresponding in size to No. 66.

No. 68. Hippotherium Antilopinum.—One of the middle cervical
vertebrx, with only the superior oblique process, left side, and body
otherwise much mutilated.

No. 85. Equus.—Left calcaneum, nearly entire, of small-sized
species.

No. 100. Equus.—Lower end of humerus with inner half of arti-
culating surface broken off.

No.111. Equus.—Fragment of upper maxilla, right side, containing
the penultimate and the next two adjoining anterior molars; the crown
covered by matrix, but of the size of No. 34.

C.
[Specimens of fossil ‘Equus’ from Ava and the Nerbudda are

also described by Dr. Falconer in the Catalogue of the Asiatic Society.—
Ep. |

Manuscript Note by Dr, Falconer.

Horse, large Sewalik | Horse, large Nerbudda
specimen specimen

Length Width Length Width

Extreme length of 6 molars . é

75 Ae ue ate
1st molar ‘ . c 5 17 11 1:6 1:0
Di ees : - 7 : “ 1-4 1:25 1:2 11
Sie LY ee es A E 5 : 8 1:2 1:2 11 119
Athe- 5 C 5 5 “ : 11 Noll ie ils
Oba 7, ‘ : . é i 12 1:15 1:05 1:
6th. ,. é a é - ‘ 1:25 9 1:25 95

FAUNA ANTIQUA SIVALENSIS.

IX. ON SOME FOSSIL REMAINS OF ANOPLOTHE-
RIUM AND GIRAFFE, FROM THE SEWALIK
HILLS.!

BY H. FALCONER, M.D., AND CAPT. P. T. CAUTLEY.

In continuation of their former researches on the fossil
remains of the Sewalik hills, the authors, in their present
communication, establish, on the clear evidence of anatomi-
cal comparison, certain discoveries which, in previous pub-
lications, they had either merely announced, or had supported
by proofs professedly left incomplete. They now demonstrate
that there occur in the remarkable tertiary deposits of the
Sewalik range, together with the osseous remains of various
other vertebrate animals, bones belonging to the two genera,
Anoplotherium and Giraffe: the former genus determined by
Cuvier from parts of skeletons dug out from the gypsum
beds of Paris; the latter genus known only as one of man’s
contemporaries, until, in the year 1838, the authors gave
reason for believing its occurrence in the fossil state.

The specimens now figured and described form part of the
collection which was made by the authors on the spot, and is
now deposited in the British Museum. They were found,
together with remains of Sivatherium, Camel, Antelope,
Crocodile, and other animals, in the Sewalik range to the
west of the river Jumna.

The bones are found embedded either in clay or in sand-
stone. When clay is the matrix, they remain white; and,
except in being deprived more or less completely of their
animal matter, they have undergone little alteration. The
bones in this state the authors have elsewhere designated as
the ‘ soft fossil? When sandstone is the matrix the animal
matter has completely disappeared, and the bone is thoroughly

1 The memoir, of which this is an
abstract, was communicated to the Geo-
logical Society of London, on Noy. 14,
1848. The abstract and illustrations
are copied from the ‘Proceedings,’ No.
98. The Anoplotherium Sivalense was
afterwards described in a separate me-
moir, under the designation of Chalico-
therium Sivalense, to which the reader is

referred. Subsequently to the publica-
tion of this memoir, the bones of the
anterior and posterior extremities of
Camelopardalis Stvalensis were disco-
vered, and were figured in an unpublished
plate of the Fauna Antiqua Sivalensis.
See descriptions of Plate E, and also
appendix to paper.—[ Eb. ]

ANOPLOTHERIUM AND GIRAFFE. 191

mineralized and rendered nearly crystalline by the infiltration
of siliceous or ferruginous matter, and acquires a corre-
sponding hardness, or tinge of iron, with increased specific
gravity. The matrix in contact with the bone is rendered
compact and crystalline in texture. The remains in this
state have been designated by the authors as the ‘hard
fossil.’

The remains of Anoplotheriwm and of the larger species
of Giraffe, described in the present communication, belong
to the ‘ soft fossil; those of the smaller species of giraffe to
the ‘ hard fossil.’

Anoplotherium.— The occurrence, in the Sewalik deposits,
of bones belonging to this genus, was announced by the
authors in their ‘Synopsis of the Fossil Genera from the
upper deposits of the Sewalik hills,’ published in vol. iv. of
the Journal of the Asiatic Society of Bengal, in the year
1835; and the same fact was afterwards referred to in
vol. vi. p. 358, of that journal. In these communications the
species was not described, but was named provisionally, A.
posterogenium. In a communication made to the Geological
Society in the year 1836, descriptive of a quadrumanous
fossil remain, and published in vol. v. of the second series
of their ‘ Transactions,’ the same species was mentioned under
the name of A. Siwvalense, a term which the authors propose
to retain, in accordance with the principle they adopted in
the cases of the horse, camel, hippopotamus, &ec., of con-
necting the most remarkable new species of each fossil Se-
walik genus with the formation itself.

In their present communication the authors purposely ab-
stain from entering on the anatomical characters of this new
species further in detail than is barely sufficient for its deter-
mination; and they therefore confine their notice to two fine
fragments of one head, one fragment (Pl. XVII. fig. 1) be-
longing to the left upper jaw; the other fragment (Pl. XVII.
fig. 2) to the right upper jaw.

By a happy chance the teeth are beautifully preserved.
The age of the individual, which was just adult, was the
best that could be desired to show the marks characteristic
of the genus; for the teeth had attained their full develop-
ment, though the two rear molars had hardly come into use.

Plate XVII. fig. 1, is a horizontal view of the left upper
jaw comprising the six back molars. These teeth were sub-
jected to a rigid comparison with a cast from the jaw of
Anoplotherium commune, figured by Cuvier in vol. iii. of the
‘“Ossemens Fossiles ’ (Plate XLVI. fig. 2), and also with casts
from the corresponding molars of Chalicotheriwm Goldfussi,
figured by Kaup in the second ‘livraison’ of his ‘ Qssemens

192 FAUNA ANTIQUA SIVALENSIS.

Fossiles’ (Plate VI. figs. 3-5 and 8-10), between the teeth of
which two extinct quadrupeds those of the Sewalik fossil are
intermediate in size. In general form and in the principal
distinctive marks they agree closely with the teeth of the
typical European species of Anoplotherium, as described by
Cuvier; but they differ from those types in some particulars
requiring special notice ; they are closely allied to the teeth
of the Chalicotheriwm of Kaup.

The three rear molars considerably exceed, in all their
dimensions, the corresponding teeth of A. commune; and
the two rear molars also differ from the corresponding teeth
of A. commune in the following respect, that their width is
greater than their length. This proportional compression
lengthwise belongs to the last two pre-molars of the Sewalik
fossil, and it holds also with the back molars of the Chalico-
therium. The outer surface presents, both vertically and
horizontally, the usual double chevron, or W-form of Ano-
plotherium, with the three salient vertical bulges swelling up
from the base to the crown; but with this difference from
Anoplotherium, that the surface of the re-entering angles is
more inclined inwards. The latter point is one of agree-
ment with Chalicotheriwm, in which the outer ridge of the
crown is so inflected as to be brought into the middle of the
plane of the tooth. The interspaces forming these re-enter-
ing angles are more unequal than in A. commune, the an-
terior one being much the broader. The posterior one in the
last molar is placed very obliquely, sloping backwards and
inwards. In these respects also the fossil agrees closely with
Chalicotherium. The vertical bulges, more especially the
rear one of the last molar, are slightly notched near the
apex into a lobule of the enamel, but much less so than in
Chaticotherium. In consequence of the progress of wear
being more advanced in the two other back molars, they
show no indications of this notch.

From the great inflection of the outer surface, the longi-
tudinal ridge of the crown is strongly zig-zagged. The apex
of the anterior re-entering angle gives off a transverse ridge,
which is much inclined downwards, and joins on with the
base of the isolated conical cusp (a, a’, a’’) in the anterior
and inner corner of the tooth, a cusp characteristic of Ano-
plotherium. In the Sewalik species, as in Chalicotheriwm,
this cusp is much larger, more pointed, surrounded by deeper
hollows, and more in relief than it is in A. commune. It is
even more developed than in Chalicotherium. The apex of
the posterior re-entering angle gives off a like transverse
ridge which sweeps round into the posterior side, and forms
in the germ a sort of three-sided pyramid, connected by a

ae

ANOPLOTHERIUM AND GIRAFFE. 193
low ridge with the cusp. The anterior border of the crown
is formed of a similar low ridge, sweeping round to the inner
side of the cusp, upon which it terminates near the middle
of the cusp. This ridge is less developed than in A. com-
mune}

The penultimate and antepenultimate are so like the last
molar that the authors deem it sufficient to refer to the
figures. The penultimate is the largest of the three, and
the antepenultimate considerably the smallest. There is in
all the three molars a strong development of the cusp ;
though, from the different stages of wearing, it shows dif-
ferently in the several teeth. In the back tooth it is intact
and has a sharp edge; in the penultimate the point is just
worn off into a slight oblique facet; in the antepenultimate
it is ground low down into a circular depressed disc, sur-
rounded by a ring of enamel.

The other teeth in the specimen (Plate XVII. fig. 1) are the
last three false molars. What was the entire number of this
series, whether it extended to four, as in A. commune, or was
limited to three, the specimen affords no certain indication.
If there was a fourth tooth (which is most probable) it must
have been in a rudimentary or reduced state, as in the rhi-
noceros, and must have been disconnected from the rest of
the series by being placed somewhat forwards in a diasteme ;
for no indication is obtained, from the appearance of the
anterior tooth, or from the remains of any alveolus, that
there was another tooth close in front of the sixth. These
three premolars, taken in succession from rear to front,
diminish rapidly in size; and in the aggregate are much
Shorter than the same three teeth in A. commune, the joint
length in the Sewalik fossil being 1:8 inch, whereas in the
smaller jaw of A. commune it is 2°3 inches. In the latter, as
is the case in the Ruminants, the anterior premolars are
narrow and elongated; in the Sewalik fossil they are short
and wide. This general condensation of the premolars adds
to the probability of the existence of a vacant diasteme. All
the premolars exhibit, in a well-developed form, the charac-
teristic cusp. The posterior two have their outer surface flat
or slightly convex; and they contract inwards towards the
cusp in a sub-cuneiform shape, the cusp and inner side being
bounded by a low basal ridge.

' MS. Note by Dr. Falconer.— This
low anterior ridge corresponds with the
anterior inner crescent of Anop. com-
mune. In the latter species, while the

_ other three crescents are as perfectly

developed as in the ruminants, the an-
terior inner lobe has only the front horn

VOL. I.

of the crescent, and never shows more
than a narrow strip of ivory by wear.
The reduction, which begins in Anop.
commune, is carried still further in the
Sewalik species; a rudiment of the lobe
only appearing, while the conical tuber-
cle is proportionally increased.’—[ Ep. ]

Oo

194 - FAUNA ANTIQUA SIVALENSIS.

The antepenultimate premolar of the Sewalik fossil is
somewhat different from the two others, being much smaller,
and contracting upwards into a trenchantedge. The cusp is
connected by a transverse ridge with the main ridge of the
crown, and the basal ridge is reduced to a small mammilla in
front of the cusp.

Plate XVII. fig. 2 represents the outside of the right upper
jaw, comprising the four back molars, and is an exact coun-
terpart, so far as it goes, of the left upper jaw.

There is little else shown by this specimen than what re-
gards the teeth. The muzzle appears to have fined off
rather abruptly in front of the malar protuberances, and
the orbit to have been advanced more forward on the face,
and to have been more depressed below the brow than in
A. commune.!. The upper orifice of the sub-orbital canal is
seen opening behind the anterior angle of the orbit, the floor
of which appears to have extended behind the post-orbital
processes.

The dimensions, as compared with those of A. commune,
and Chalicotheriwm Goldfussi, are as follows :—

Ch. Goldfussi | A. Sivalense | A. commune
Inches Inches Inches
Length of the series of 6 molars é . 71 55 55
Ditto 3 true molars 5 37 3-2
Ditto 3 premolars . 2-1 18 2°3
Length of the last true molar 1:85 13 1:2
Width of ditto . : - : : 1:94 15 1:2
Length of the penultimate true molar 7 1°4 1-2
Width of ditto . : : : : 1:9 1:55 11
Length of the antepenultimate true molar . 1°3 Tel its
Width of ditto . . : 5 1:4 cl iti
Length of the last premolar 0°8 0°75 0°65
Width of ditto . 2 : 3 1:15 0°88 0°75
Length of the penultimate premolar . 0°6 0°70 0-90
Width of ditto . ‘eel : : 0-9 0°80 0°62
Length of the antepenultimate premolar 0:75 0°55 0°75
Width of ditto . : , . 0°5 0°55 0°50
Height of the last molar . 0°87 1h 0°65

These measurements show the Sewalik species to have been
larger than A. commune, and smaller than Chalicotheriwm
Goldfussi. One of the most striking points in which it
differs from the two latter terms of comparison is in the
dimensions of its back molar, which, with the same amount
of wear, is about half an inch higher than in A. commune,
and in this respect considerably exceeds even the longer and
wider tooth of the Chalicotherwum.

1 See Ossemens Fossiles, tom. iii. tab. 57. fig. 1.

ANOPLOTHERIUM AND GIRAFFE. 193

Length Width Height
A, Sivalense . : i f 5 13 15 :
A. commune . ' E é 6 12 1:2 0°65
Ch. Goldfusst : : ; . 1:8 1:94 0°87

On the whole the Sewalik species appears to be most closely
allied to the Chalicotherium Goldfussi. The existence of a
vacant diasteme in front of the anterior tooth would constitute
a difference from the Anoplotherian type of some importance.
The characters generally show a return from the ruminant
tendencies of the Cuvierian species back to a more pachyder-
matous type, and a closer affinity with the rhinoceros,
between which and A. commune it may ultimately prove to
be an intermediate form. Until the evidence for separation
is conclusive the authors suggest leaving it with the genus
Anoplotherium. The A. commune was determined by Cuvier
to be of the size of a small ass; the A. Sivalense would rank
in dimensions between a horse and the small Sumatran
rhinoceros.

Remarks on Chalicotheriwm.—Kaup appears to have founded
this genus, as distinct from Anoplotherium, on real or sup-
posed differences, 1st, in the rear molars; 2nd, in the in-
cisors; 3rd, in the canines. The difference in the rear
molars consists in the size of the lobule of the enamel, into
which the vertical bulges near the apex are notched ; this
character indicates, as he conceives, an affinity with the
Tapir and Lophiodon. But this lobule, even if constant, does
not appear to the authors of sufficient importance to consti-
tute the basis of a generic distinction. The general form of
the rear molars of both the upper and lower jaws is only an
enlarged and less rectangular representation of those of Ano-
plotherium. Moreover, in the direction of the ridges of the
crown, and in the insulation of the conical cusp, the accordance
between Chalicotherium and Anoplotherium is complete.
As to the second distinction, drawn from the supposed form
of the incisors, the detached tooth which he figures and de-
scribes as a lower incisor (Oss. Foss. liv. ii. p. 30, Pl. VIL),
judging from the figures and from a cast which the authors
_ have examined, very closely resembles, both in form and in
the development of the crown, the penultimate premolar of
the A. Swalense. The channeled sides and the bifid extremity
of the fang, indicating two confluent fang roots, and the com-
plicated form of the crown with three mammille on the inside,
appear to the authors strongly to militate against regarding the
tooth as an incisor. They therefore consider this tooth as an
upper premolar (and probably as the penultimate) one of the
right side.

As to the third distinction, drawn from the canine teeth,
judging from a cast of the detached fragment which Kaup

02

196 FAUNA ANTIQUA SIVALENSIS.

describes and figures as the canine of Ch. Goldfussi, the
authors consider that’ determination as problematical. It
seems to them to bear a resemblance in form rather to the
lower incisor of an animal allied to rhinoceros. They advance
these doubts with the utmost deference to the distinguished
author.

Remarks on the Genus Anoplotherium.—The true Anoplo-
theria of Cuvier (of which A. commune may be regarded as
the type), together with the A. Sivalense and the Chalicothe-
rium (Anoplotherium ?) Goldfussi are allied, by their den-
tition, to Rhinoceros. The Dichobunes, A. leporinum, A.
murinum and A. obliquum, Cuvier arranges with considerable
doubt, and provisionally only, among the Anoplotheria. He
considers it not impossible that the two latter species were
small Ruminants. The A. cervinwm of Professor Owen (Geol.
Trans. 2nd ser. vol. vi. p. 45), obtained by Mr. Pratt from
Binstead in the Isle of Wight (Idem. vol. iii. p. 451), is ad-
mitted on all hands to be exceedingly like a musk deer.
Such heterogeneous materials are too much for the limits of
any one genus. Cuvier imagined the separation of the two
metacarpal bones to be a character limited to the Anoplo-
theria exclusively. He has also regarded the union of the
metacarpal bones as holding without exception in all the
ruminants ; and this law with respect to ruminants, though
empirical, he regards as equally certain with any conclusion
in physics or morals, and as a surer mark than all those of
Zadig (Disc. Prél. p. 49).

The authors, having had an opportunity of examining the
skeleton of an African ruminant, the Moschus aquaticus of
Ogilby, described in the ‘ Proceedings’ of the Zoological
Society by that gentleman from a living specimen, found it
wanting in the above supposed essential character of the ru-
minants, and possessing the above supposed distinctive cha-
racter of Anoplotherian Pachyderms. Its metacarpals are
distinct along their whole length ; its foreleg, from the carpus
downwards, is undistinguishable from that of the peccary;
and its succentorial toes are as much developed as in the last-
mentioned animal.

The deviation from the ordinary ruminant type, indicated
by the foot of this Moschus, is borne out by a series of modi-
fications in the construction of the head and in the bones of
the extremities and trunk, all tending in the direction of the
pachyderms.

The authors believe the present to be the first announce-
ment of the existence of such an anomaly in any living ru-

1 See appendix to memoir on Chalicotheriwm, No. 11.—[Ep.]

ANOPLOTHERIUM AND GIRAFFE. 197

minant; they had previously ascertained the occurrence of
the same structure in a fossil ruminant from the Sewalik
hills. As the Dorcatherium of Kaup breaks down the
empirical distinction between the ruminants and pachy-
derms, as regards the number of the teeth, so does the
Moschus aquaticus as regards the structure of the feet.

Giraffe.—In the 7th volume of the ‘ Journal of the Asiatic
Society of Bengal’ (pp. 658-660) is a communication dated
‘Northern Doab, July 15, 1838,’ and intituled, ‘Note on a
Fossil Ruminant Genus allied to Girafiide, in the Sewalik
hills, by Captain P. T. Cautley.’ The specimen referred to
in that paper was the third cervical vertebra of a ruminant,
which, for the reasons therein assigned, was supposed to have
been a giraffe. At that time the authors of the present com-
munication had not access either to drawings of the osteology
or to a skeleton of the existing giraffe; but the grounds for
referring the vertebra to that genus were, that it belonged to
a ruminant with a columnar neck, the type of the ruminants
being preserved, though very attenuated in its proportions ;
that the animal was very distinct from any of the camel
tribe; that it was in the giraffe that there existed such
a form most aberrant from the mean in respect of its great
elongation. That the bone belonged to a giraffe was put
forth at the time as only a probable inference, and chiefly
to serve as an index to future inquiries.

The authors, having since the former period obtained addi-
tional specimens, and had access to the fullest means of
comparison, are now able to place on the record of deter-
mined Sewalik fossils one very marked species of giraffe,
and also indications of a second species, which, so far as
the scanty materials go, appears to come near to that of
Africa.

The first specimen to which they refer is the identical ver-
tebra noticed by Captain Cautley in 1858. (See Plate XVI.
- fies. 1-4.) It is an almost perfect cervical vertebra. It were
needless to enter on the characters which prove it to have
belonged to a ruminant. Its elongated form shows that it
belonged to one with a columnar neck; that is to say, either
to one of the camel and Auchenia tribe, or to a giraffe, or to
some distinct and unknown type. The fossil differs from the
vertebra of a camel, Ist, in the position of the vertebral
foramina (a, a’); 2nd, in the obsolete form of the upper trans-
verse processes. According to the masterly analysis of the
Macrauchenia by Professor Owen, the Camelide and Macrau-
chenia differ from all other known mammalia in the following
peculiarity: that the transverse processes of the six inferior

cervical vertebree are without perforations for the vertebral

198 FAUNA ANTIQUA SIVALENSIS.

arteries, which enter the vertebral canal along with the
spinal cord, then penetrate the superior vertebral laminze,
and emerge on the canal again close under the anterior
oblique processes. This structure appears in the cervical
vertebree of the Sewalik fossil camel. In the vertebra now
under consideration, on the contrary, the foramina (a, a’)
maintain their ordinary position, that is, they perforate the
transverse processes, and appear on the surface of the body
of the vertebra.

Since the bone therefore does not belong to a camel, is it
the bone of a giraffe? There is preserved in the Museum of
the Zoological Society the skeleton of a young Nubian giraffe
which died at the Society’s gardens. When its third cervical
vertebra is placed in apposition with the fossil, the two are
found to agree in every general character, though they dis-
agree in some of their proportions, and in certain minor pecu-
liarities. In this young and immature giraffe the length of
the third cervical vertebra is 74 inches; what, then, is the
length of this bone in the adult Nubian giraffe ? The authors,
from their not having had under their examination this ver-
tebra from an adult animal, have been unable to ascertain
this point directly; but they are able to infer it, from the
length of a detached bone preserved in the Museum of the
Royal College of Surgeons of London, which is the second
cervical vertebra of a giraffe, nearly, but not quite, full grown.!
The length of this bone is 114 inches. Now in the skeleton
of the young giraffe belonging to the Zoological Society the
2nd and 3rd cervical vertebre are exactly of the same length.
The authors infer, therefore, that in an animal nearly full-
grown, such as was that to which the detached bone at the
College of Surgeons belonged, the length of the 3rd cervical
vertebra is 11} inches; and consequently, that the length of
the same bone in an animal which has reached full maturity
is about 12 inches.?

That the fossil vertebra belonged to an adult which had
long attained its full size is shown by the complete synostosis
of the upper and lower articulating surfaces, by the strong
relief of the ridges and the depth of the muscular depres-
sions. But the length of this bone is only a little more than
eight inches. As the other dimensions of the fossil and
recent vertebre that the authors placed in apposition are
nearly in proportion to their respective lengths, it follows

1 This appears from the detached | young giraffe in the museum of the
state of the upper and lower articulating | Zoological Society is 103 feet; that of a
heads of the bone. full-grown Nubian giraffe is 16 feet.

? The height of the skeleton of the

ANOPLOTHERIUM AND GIRAFFE. 199

that this fossil species of giraffe was one-third shorter in the
neck than an adult of the existing Nubian variety.

But it was not only in size that the two giraffes differed:
they differed also in their proportions. In the young oiratte
at the Zoological Society the vertebra, which is 73 inches
long, has a vertical diameter of 3:8 inches ; whereas in
the fossil species the vertebra, which is 8 inches long, in-
stead of having a vertical diameter exceeding 4 inches (as
it ought, if its breadth were proportional to its length), has
a vertical diameter of only 3°6 inches. This goes to prove
that in this fossil giraffe the neck was one-tenth more
slender in proportion to its length than the neck is in the
existing species.

The inferior surface of the body of the vertebra is more
eurved longitudinally in the fossil than it is in the recent
bone; the height of the arc in the former case being to the
height in the latter as 3 is to 2.

On the under surface of the fossil vertebra a very distinct
longitudinal ridge (b) runs down the middle, and this ridge
is wanting in the recent bone; but this difference, probably,
is chiefly owing to difference of age.

In the fossil vertebra the upper articulating head (c) is
very convex; for with a transverse diameter of 1:4 inch it
has a vertical height of 1 inch; laterally it is a good deal
compressed.

The posterior articulating surface (d) forms a perfectly cir-
cular cup, two inches in diameter ; and this diameter, in the
immature Nubian giraffe, is one-tenth greater, although
the vertebra is one-sixteenth shorter. This affords a further
proof of the comparative slenderness of neck in this fossil
species. (See Plate XVI. fig. 4.)

In regard to the apophyses, the inferior transverse pro-
cesses (7, 1) are sent off downwards and outwards from the
lower part of the anterior end, exactly as in the recent
species, and they are developed to nearly the same amount
of projection. ‘There is, however, this considerable difference,
that whereas in the recent species they do not run half-way
down the body of the vertebra, in the fossil they are decur-
rent along the whole of its length in well-marked laminar
ridges, which are confluent with the nearly obsolete ridges
of the upper transverse processes, the united mass near the
posterior end being dilated into two thick aleeform expan-
sions (¢, é).

In the fossil, as in the recent, bone, the superior transverse
processes are seen only in a rudimentary state; in the
former, however, they run forwards across the body with
less obliquity, and consequently make the canals for the

200 FAUNA ANTIQUA SIVALENSIS.

vertebral arteries twice as long as they are in the recent
bone. In the fossil the orifices (a, @) of these canals divide
the length of the vertebra into three nearly equal portions ;
whereas in the recent bone the orifices are both included
within its anterior half. .

The anterior oblique processes (f, f) have the same general
form and direction both in the fossil and recent species ;
but in the former they are considerably stouter and larger,
and their interspace is less. The articular surfaces are
convex, and are defined exactly as in the recent species.

The posterior oblique processes (g, g) of the fossil differ in
form very little from those of the recent bone ; in the fossil,
however, the articular surfaces are considerably larger ; and
the ridges in which they are continued along the side of the
upper vertebral arch are much less convergent than in the
recent bone; so that in the latter this part is somewhat
heart-shaped; whereas in the fossil it is nearly oblong, and
‘ looks squarer,’ so to speak.

The spinous process (4) in the fossil is the same thin
triangular lamina that is seen in the recent species ; and it
differs only in having its most prominent point lower down
on the arch.

The spinal canal is very much of the same form and
dimensions in both the fossil and the recent vertebra. At
this point some of the matrix remains attached to the fossil
bone, and prevents any very precise measurement.

As a minor point of agreement between the fossil and
recent bones, it may be noted that, in both, the foramen (k)
for the small nutritious artery on the inferior side of the
body of the vertebra is on the right. In the other cervical
vertebrae of the recent skeleton this solitary foramen is on
the left.

From the above comparisons it appears that the fossil
vertebra, while it is very distinct from that of a camel, fulfils
all the conditions required for a strict identification with
that of a giraffe; that its peculiarities are not of greater
than specific importance ; and consequently do not warrant
its being referred to a distinct and unknown type among the
ruminants.

The following are the dimensions, in detail, of the third
vertebra in the adult Sewalik fossil and in the immature
Nubian giraffe, 104 feet high, in the Museum of the Zoolo-
gical Society :—

DESCRIPTION OF PLATE XVI.

CAMELOPARDALIS.

Figs. 1 to 4. Third cervical vertebra of Camelopardalis Sivalensis,

one-half of the natural size. Copied from drawings by Mr.
Scharf in No. 98 of the Proceedings of the Geological Society,
and by Mr. Dinkel in an unpublished Plate of the Fauna An-
tiqua Sivalensis. The specimen is in the British Museum ; Cat.
No. 39,760. ad, orifices of the arterial canals; 6, longitu-
dinal ridge on the underside of the body ; ¢ ec, upper articulating
head; d, lower articulating surface; ¢ e, aleform expansions of
the transverse processes; f/f, superior oblique processes; g 9,
inferior oblique processes; hh, spinous process; 772, inferior
transverse processes; k, foramen of the nutritious artery. (See
pages 197 & 543.)

Figs. 5a and 56. Views in plan and profile of the last two upper molars

Fig. 6.

Fig. 7.
Fig. 8.
Fig. 9.

of Camelopardalis affinis, one-half of the natural size. This as
well as the succeeding figures, copied from No. 98 Proc. Geol.
Soc:, Pl. I.

Last upper molar, right side, of Camelopardalis affinis, one-half
of natural size.

Last lower molar, left side, of ditto, one-half of natural size.
Last lower false molar, left side, of ditto, one-halfof natural size.

Second upper false molar, right side, of ditto, one-half of natural
size.

Fig. 10. Rugose reticulated surface of the enamel of upper molars of

ditto, magnified to twice the natural size. (See page 202.)

VOL. I.

Pe seit ey iy

Vol. I. Plate 16.

Tae

nie

OSchars del JDinkel ith
l+o 4. Camelopardalis Sivaiensis. 5t010. C. affiuis. :

ANOPLOTHERIUM AND GIRAFFE. 201

Sewalik Nubian
Fossil. Recent.
Inches Inches
Length between the ends of the oblique processes 8-1 75
Length of the body of the vertebra between the articulating heads 78
Greatest width at the posterior end of the body, between the trans- 3+] 2:8
verse processes. ‘
Least width at the middle of the body, ‘between the upper tr ae 165 16
verse processes . .
Width between the outer margins ‘of the upper oblique processes « 265 2°55
Width of sinus between the upper oblique processes. ° 5 valet 1:2
Width between outer edges of posterior oblique processes. . 25 2°3
Least width of spinal arch between the ridges connecting the
upper and lower oblique processes ; :
Vertical diameter, posterior end of vertebra . : . 36 3°8

Vertical diameter, anterior end, between the inferior border arti- 26 2-7
culating head and upper margin spinal canal }
Antero-posterior diameter articulating head . 5 5 5 Al) 1°55
Transverse diameter articulating head at the middle. . . 14 15
Greatest diameter articulating head . 14 1:8
Vertical height articulating head . di = Also)
Length of articulating surface, lower oblique process : . 16 1:2
Width of ditto - : . 10 0-8
Length of articulating surface, upper oblique process : . 12 0°85
Width of ditto : . 08 0-7
Vertical diameter, spinal marrow, , posterior end . . 1:25
Vertical diameter, articulating cup, posterior end. . 2:0 2:2
Transverse diameter ditto ditto 5 . 2:0 2°3
Diameter upper transverse processes . : : . . 08 0-7

Hence the authors conclude that there belonged to the
Sewalik fauna a true well-marked species of giraffe closely
resembling the existing species in form, but one-third less
in height, and with a neck proportionately more slender ;
and for this small species they propose the name Camelopar-
dalis Swalensis.

Second Fossil Species of Giraffe.—The fossil specimens next
to be described have been in the possession of the authors
ever since 1836. ‘They are fragments from the upper and
lower jaws of another fossil species of giraffe, in which the
teeth are so exactly of the same size and form with those of
the existing species, and so perfectly resemble them in every
respect, that it requires the calipers to establish any difference
between them.

The largest specimen (Pl. XVI. figs. 5, 5 a) is a fragment
of a left upper jaw containing the two rear molars. The
back part of the maxilla, beyond the teeth, is attached,
and clearly proves that they belonged to a full grown animal.
These teeth were compared with the teeth, in the same
stage of wearing, contained in the head of an adult female
giraffe belonging to the Museum of the College of Surgeons,
and the fossil and recent teeth were found to agree together
in the most minute particulars. The following are the
corresponding dimensions of the fossil and recent teeth :—

202 FAUNA ANTIQUA SIVALENSIS.

Fossil. Recent.
Inches Inches

Joint length of the two back molars, upper jaw é c c » 25 2°55
Greatest width of last molar : a : ; .14 13
Ditto ditto of penultimate molar . 7 é : - . 146 1°35

The second specimen (Pl. XVI. fig. 6) is the rear molar of
the right upper jaw, corresponding exactly in size and form
with that of the left side, but if anything, rather more worn,
and belonging therefore probably to a different individual.
The agreement extends down to the small cone of enamel at
the base of the hollow between the barrels on the inside.
Its dimensions are :—

Length : : F é t . 1:2 inch.
Width. : F ‘ sill

The third specimen PL XVI. fig. 7) is a fragment of the
left lower jaw, containing the last molar. It has precisely
the form and proportions of the corresponding tooth in
the left lower jaw of the female head referred to, and the
same development of its third barrel or heel, which is
always found in this tooth in ruminants. Its dimensions
are :—

Length ' & A 5 A : . 1:7 inch.
Greatest width . ; ; , ee ak)

The fourth specimen (Pl. XVI. iS 8) is the last false molar
of the left lower jaw, detached. It agrees closely with the
corresponding tooth in the recent female head’ above re- ~
ferred to. This tooth is thicker in proportion to its length
in the giraffe than in other ruminants, and this constitutes
one of the most distinctive characters of the giraffe’s pre-
molars. The anterior semi-barrel appears a trifle longer
than the corresponding tooth of the recent animal; but this
is owing to a difference of wear, and is not borne out by
measurement. The dimensions are :—

Fossil. Recent.
Length . : : . . 1-0 inch. 1:0 inch,
Breadth F : : . 0-9 0°86

The authors are possessed of the same tooth of the right
lower jaw, detached; but have not thought it necessary to
figure it.

The fifth specimen (Pl. XVI. fig. 9) is the penultimate false
molar of the right upper jaw. It is of the same size and
form with the corresponding tooth in the recent female head,
with this difference, that it has three tubercles at the inside
of the base. On a sixth specimen of the first false molar
_ of the right upper jaw, which is not represented among the
figures, there are three similar tubercles similarly placed.

ANOPLOTHERIUM AND GIRAFFE. 203

It would require an extensive comparison of recent heads to
determine what value attaches to this peculiarity ; whether
the tubercles are constantly absent from the teeth of the
recent species, or appear occasionally as a variation on those
of individuals. The dimensions of the penultimate false
molar of the upper jaw are :—

Fossil. Recent.
Length . ; - 2 . 1:0 inch. 0°95 inch.
Breadth. : A 5 3 plat 1:12

There is a peculiar, finely reticular, striated and rugose
surface to the enamel of the teeth of certain quadrupeds, the
appearance of which the authors compare to that of a fine
net, forcibly extended, so as to bring the sides of the
meshes together. This texture they formerly described as
existing on the surface of the molars of the Sivatherium. It
is found also on the teeth of the recent giraffe, and is more or
less conspicuous on those of the hippopotamus. It is not
observed in the camel, the moose deer, or the larger bovine
ruminants; or if ever present, it is but faintly developed.
This texture is well marked on the enamel of the teeth of
this second species of giraffe. A magnified representation of
. itis given in Pl. XVI. fig. 10.

The series of teeth last described, excepting the fifth and
sixth specimens, are all but undistinguishable from those of
the Nubian giraffe; and the authors have sought in vain for
any distinctive character by which to discriminate them.
_ There is no good evidence to show that this fossil species
and the living are even different; but in putting the case
thus, the authors are far from advancing that the species are
identical. The materials are far too scanty to warrant a
conjecture to that extent.

Since the neck of the OC. Sivalensis was one-third too short
and slender to sustain the head that would have suited the
teeth last described, the authors consider it a necessary con-
sequence that these teeth belonged to a distinct species.
Had the difference been less considerable, they might have
hesitated regarding this conclusion; but the difference
between 8 inches and 12 inches in the length of the same
cervical vertebra of two adult animals of the same genus
admits, in their opinion, of no other construction than distinct-
ness of species. For the present, until sufficient materials
shall be obtained to determine the relationship between the
African giraffe and the second Sewalik species, in refe-
rence to their supposed resemblance, the authors propose to
mark the latter by the provisional name of Camelopardalis

affinis.

204 FAUNA ANTIQUA SIVALENSIS.

General Remarks.—In a former communication to the So-
ciety (Geol. Trans. 2nd ser. vol. v. p. 503) the authors no-
ticed the remarkable mixture of extinct and recent forms
which constituted the ancient fauna of Northern India. An
extinct testudinate form, Colossochelys Atlas, as enormous in
reference to other known Chelonians as the Saurians of the
lias and the oolite are to their existing analogues, is there
associated with one or more of the same species of crocodile
that now inhabit the rivers of India. The evidence respect-
ing one of these species of crocodile, resting as it does on
numerous remains of individuals of all ages, is considered by
the authors as nearly conclusive of the identity of the fossil
with its recent analogue. These reptiles occur together with
extinct species of such very modern types as the monkey, the
camel, the antelope, and (as has now been shown) the giraffe ;
and these are met by species of the extinct genera Siva-
therium and Anoplotherium. As regards the geographical
distribution of the true Anoplotheria, those hitherto dis-
covered have been confined, as the authors believe, to Europe ;
and as regards their geological distribution, to the older and
middle tertiaries. In India this genus continued down to
the period when existing Indian crocodiles and probably
some other recent forms had become inhabitants of that
region.

It might be expected that in a deposit containing Anoplo-
therium, Palzotherian remains also would sooner or later be
discovered. However, among the very large collection of
fossil bones from the tertiary sub-Himalayan range, made by
the authors during ten years in that part of India, they have
never found a single fragment of a head or tooth which they
were able to refer to Paleotherium. This is merely a
negative result, and only proves the rarity of that form.’

Although there occur among the Sewalik fossils abundant
remains of almost every large pachydermatous genus, such
as the elephant, mastodon, rhinoceros, hippopotamus, sus,

1 Mr. M‘Clelland, in his paper on
Hexaprotodon (Journ. Asiatic Society of
Bengal, vol. vii. p. 1046) casually men-
tions a species of Palzotherium as oc-
curring among the Sewalik fossils. But
he does not describe or figure the speci-
men. Messrs. Baker and Durand in
their remarks appended to their cata-
logue of the Dadoopoor collection (Jdem,
vol. y. p. 836), mention four specimens
containing teeth of the upper and lower
jaws belonging to what they provi-
sionally designate ‘Cuvierian genera :’
in regard to one of which, having the

upper and lower jaws in contact, they
state that, ‘although it affords some
analogies both to the Paleotherium
and Anoplotherium, its essential pecu-
liarities are sufficiently remarkable to
cause it to be separated from either
genus. Till these specimens are either
figured or described, the point must
remain undecided in regard to Palzo-
therium being represented in the Sewa-
lik fauna. [These specimens are de-
scribed in the subsequent paper on
§ Chalicotherium.’—Ep. |

Oe

ANOPLOTHERIUM AND GIRAFFE. 205

horse, &c., yet no remain has been found referrible to the
Tapir, a fact the more remarkable, inasmuch as one of the
only two existing species of that genus is now confined
to the larger Indian islands and a part of the adjoining
continent.

The finding of the giraffe as a fossil furnishes another
link to the rapidly increasing chain which (as the discoveries
of year after year evince) will sooner or later connect extinct
with existing forms in a continuous series. The antelope and
the bovine and antlered ruminants have numerous representa-
tives, both recent and fossil. The camel tribe comprises a
considerable fossil group, represented in India by the Camelus
Swvalensis, and is closely approached in America by the extinct
pachydermatous Macrauchenia. The giraffe has hitherto
been confined, like the human race, to a single species, and
has occupied an isolated Position in the order to which it
belongs. It is now as closely represented by its fossil ana-
logues as the camel; and it may be expected that, when the
ossiferous beds of Asia and Africa are better known, other
intermediate forms will be found, filling up the wide interval
which now separates the giraffe from the antlered ruminants,
its nearest allies in the order according to Cuvier and
Owen.'

The giraffe throws a new light on the original physical
characters of Northern India; for whatever may be urged in
regard to the possible range of its vegetable food, it is very
clear that, like the existing species, it must have inhabited
an open country, and had broad plains to roam over. Ina
densely forest-clad tract, like that which now skirts the foot
of the Himalayahs, it would soon have been exterminated by
the large feline feree, by the hyzenas, and large predaceous
bears which are known to have been members of the old
Sewalik fauna.

Postscript.—Since the above remarks were submitted to the
Society, M. Duvernoy’s paper, embodying two communica-
tions read to the Academy of Sciences on the 19th May and
27th November last, has appeared in the January number of
the ‘ Annales des Sciences Naturelles.’ These notices were
published in the ‘ Comptes Rendus,’ but were unknown to
the authors at the time. M. Duvernoy describes the lower
jaw of a fossil giraffe found in the bottom of a well, lying on
the surface of a yellow clay, along with fragments of pottery

1M. G. de St. Hilaire, in his zeal for| tomical proofs were all against this
the mutability of species, imagined that | inference; but if a shadow of doubt
he had detected in the Sivatherium the | remained, it must yield to the fact, that
primeval type which time and necessity | in the Sewalik fauna the Giraffe and the
had fined down into the giraffe. Ana- | Sivatherium were contemporaries.

206 FAUNA ANTIQUA SIVALENSIS.

and domestic utensils, in the court of an ancient donjon of
the 14th century in the town of Isoodun, Département de
V’Indre. Considerable doubt remains as to the bed and source
whence the fossil was derived. M. Duvernoy attributes the
jaw to a distinct species of giraffe, which he names Camelo-
pardalis Bitwrigum. Professor Owen, from the examination
of a cast, confirms the result, expressing his conviction ‘ that
in the more essential characters the Isoodun fossil closely
approaches the genus Giraffe, but differs strikingly from the
(single) existing species of the south and east of Africa, and
that the deviations tend towards the sub-genus Elk.’

M. Duvernoy also mentions the discovery by M. Nicolet of
a tooth in the molasse near Neufchatel, determined by M.
Agassiz to be the outer incisor of a fossil giraffe.—Duvernoy,
Annales des Sciences Naturelles, No. for January 1844.

APPENDIX.

L—Descrirwon sy Dr. Fatconer or Fosstz Remains oF GIRAFFE IN
THE Museum oF Asiatic Society or BENGAL.

A. From the Sewalik Hills.

No. 405. Camelopardalis Sivalensis.—Upper extremity of right
metacarpus, probably of Girajfe, so inferred from sudden contraction
below articular head. From the Sewalik hills near Nahun.

No. 560. Camelopardalis Sivalensis?—Fine fragment comprising
the lower jaw, right side, with three milk molars 7m s¢tu, and the germ of
the first true molar embedded in the jaw. The teeth are well preserved,
the two posterior premolars being only slightly touched by wear, the
third milk molar shows the three barrels and composite form charac-
teristic of that tooth; the surface of the enamel presents in a well-
marked manner the peculiar netted rugosity, which distinguishes the
teeth of the Sivatherium and Brahmatherium and Giraffe. The jaw evi-
dently belonged to a young animal. The body of the horizontal ramus
is thick, a portion of the diasteme remains, which is thick and low;
there are no materials of comparison to determine with certainty to
what genus the fossil belongs; but it is inferred from the characters of
the enamel and general form to be the lower jaw of a young Giraffe.

No. 561. Camelopardalis Sivalensis.—Fragment of lower jaw, left
side, horizontal ramus, with three milk molars of a still younger animal
than No. 560, and of smaller proportions, containing the three milk mo-
lars of which the most anterior is in germ, and just protruding from the
jaw; the germ of the first true molar is seen behind. Enamel sur-
face of this specimen also shows rugous netting ; but it is too imper-
fect for confident determination.

No. 690. Camelopardalis Stvalensis.—Fragment comprising the

ANOPLOTHERIUM AND GIRAFFE. 207

upper portion of the shaft of the left radius, the articulating head broken
off; the bone is much flattened, and its outer border forms a consider-
able curve, in consequence of the abrupt expansion of the articulating
head, and the sudden contraction of the shaft below it. It is not appa-
rent whether the articulating epiphyses had been synostosed; but the
bone is nearly equal in all its principal dimensions to the correspond-
ing one of the existing Girajfe, though it is considerably more flattened.
It has the greyish weathered appearance of a Sewalik fossil of the sand-
stone matrix. This is a valuable and rare specimen. Presented by Col.
Colvin.

B. From Perim Island.

No. 43. Camelopardalis Sivalensis ?—Humerus, right, in two frag-
ments; lower articulating surface perfect; the upper broken off imme-
diately below the head ; resembling in form exactly the humerus of the
giraffe, but a little larger.

No. 52. Camelopardalis.—Lower end of metacarpal bone, left side,
with articulating surfaces, of the size of existing giraffe.

No. 60. Camelopardalis.—Second or third dorsal vertebra of giraffe,
concave and convex articular surfaces present, apophyses wanting.

IJ.—On toe Numper or Existinc Species oF GIRAFFE.
MS. Memorandum from Note-Book.

Some naturalists have attempted to establish two species of giraffe,
founded chiefly on the geographical range of the Nubian and South
African varieties, and on external characters derived from the skin. But
any conclusions built on the former would be begging the question,
and dermal marks are insufficient to sustain a case of the kind, unless
borne out by peculiarities in the skeleton, or other structural differences.
Those who have seen the range of colour through which the tropical
antelopes run, from a light blue or pale ash in the young male through
every grade to a deep black in the old, such as is presented in the An-
tilope cervicapra and Damalis risia, will distrust any distinction
resting on the skin marks or development of the knee scope in the
giraffe. The distinctness of the suture of the bone which supports
the alleged third or intra-orbital horn, and the great relief of the
pedicle shown in the figures in Riippel’s atlas have been adduced
as proofs of the specific difference of the North African giraffe; but
Professor Owen states that he could detect no evidence of such a suture
on the original cranium. I am informed, also, by a distinguished
zoologist, and one of our best authorities on the Ruminantia that he
examined the head from which the figures were taken, in the Frankfort
Museum, along with Dr. Riippel, and found neither the suture nor the
elevated pedicle shown in the figures, both of which Dr. Riippel admitted
to be exaggerations introduced by the artist who drew them. No good
evidence, therefore, has been adduced in favour of there being two ex-
isting species of Giraffe; the proofs at present all tend to the opposite
conclusion. My authority, however, found a considerable difference
in the extent and figure of the lachrymal bone, a character, if con-
stant, of much importance.—[H. F. ]

208 FAUNA ANTIQUA SIVALENSIS.

X. ON CHALICOTHERIUM SIVALENSE.!

BY H. FALCONER, M.D.

On November 15, 1843, a paper by Major Cautley and my-
self was read to the Geological Society, entitled ‘On some
Fossil Remains of Anoplotherium and Giraffe from the Sewalik
Hills,’ &c., an abstract of which appeared in No. 98 of the
Society’s ‘Proceedings.’ In the concluding remarks upon
the species referred to Anoplotheriwm, we stated that our
fossils came nearest to the Chalicotheriwm of Kaup, and that
the generic determination then assigned to them by us was
to be considered more as provisional than positive, pending
the discovery of other more conclusive materials. This
opinion is expressed in the ‘Abstract’ in the following
terms :—

‘On the whole the Sewalik species appears to be most
closely allied to the Chalicotheriwm Groldfussi. The existence
of a vacant diasteme in front of the anterior tooth would
constitute a difference from the Anoplotherian type of some
importance. The characters generally show a return from
the ruminant tendencies of the Cuvierian species back to
a more pachydermatous type and a closer affinity with
Rhinoceros, between which and Anoplotherium commune it
may ultimately prove to be an intermediate form. Until the
evidence for separation is conclusive, the authors suggest
leaving it with the genus Anoplotherium.’ (‘ Proceedings,’
No. 98, p. 289; and antea, p. 195.)

Our reasons for hesitating to adopt Kaup’s generic name
of Chalicotherium arose from our entertaining doubts regard -
ing the validity of the grounds urged by that eminent
palzontologist,? for establishing the genus as distinct from
Anoplotherium. The characters adduced by Kaup were,
the ascertained form of the back molars of the upper
jaw, but more especially the supposed form of the incisors
and canines. We suggested that the complicated crown

1 This paper was written in October | the reader is referred to the description
1847, for the Geological Society, but was | of Plate lxxx. of the Fauna Ant. Siy.—
never presented, and is now for the first | [Ep.]

*time published. For the measurements 2 Oss. Foss. de Darmstadt, livrais, ii.
of the specimens described in the paper | p. 30, Pl. vii.

CHALICOTHERIUM SIVALENSE. 209

and the compound fang of the tooth regarded by him as an
incisor proved it rather to be one of the anterior premolars,
an inference which has proved to have been correct; and
that the supposed upper canine resembled more the incisor of
a species of Rhinoceros, to which it would appear Dr. Kaup
now refers it. The only other adduced character was the
form of the back molars, which, admitting the amount of
difference indicated, did not appear to us to be of sufficient
importance to constitute, alone, the basis of a generic dis-
tinction, as these molars upon the whole exhibited little
more than ‘an enlarged and less rectangular representation
of those of Anoplotherium,’ with which they entirely agreed
in the insulation of the conical cusp, so characteristic of that
genus.

Since the date of the memoir in question additional
materials have turned up to us, which fully establish the
validity of Kaup’s genus; while they prove at the same
time the Sewalik species of Chalicotheriwm, through a very
unexpected combination of characters in the construction of
the jaws, to have been widely different from Anoplotheriwm.
So unexpected, indeed, are those characters that Chalicotheriwm
must be regarded as one of the most remarkable and aberrant
pachyderms that has yet been met with, either in the fossil
or recent state.

The object of the present communication is to make

«known the nature of the new evidence respecting the Indian
fossil species, and to extend its comparison with the Huropean
species, upon the additional specimens discovered at Hppel-
sheim, which have been figured and described by Kaup in
the ‘Akten der Urwelt.? It would appear, by a manuscript
communication from Mons. Pomel, that remains of the same
genus, and probably of the Eppelsheim species, have been
discovered in the rich ossiferous beds near Sansans, in
the south of France, associated with Dinotheriwm, Mastodon,
and cther forms characteristic of the miocene deposits of
Hppelsheim.

The specimens now at our disposal put us in possession of
the whole of the dental characters of Chalicotheriwm Swalense.
The most important is a fragment comprising the anterior
half of an adult head, with the upper and lower jaws in
natural apposition, and exhibiting the greatest portion of the

_ dental series of both jaws (Plate XVII. figs. 3 and 4). The two

back molars, deficient in this specimen, are fortunately shown
in the most perfect state of preservation by the fragments re-
presented in our previous memoir (Plate XVII. figs. 1 and 2).
This beautiful specimen, originally in the Dadoopoor collec-

tion of Major Baker and Captain Durand, of the Bengal

VOL. I, P

210 FAUNA ANTIQUA SIVALENSIS.

Engineers, belongs now to the Museum of Marischal College,
Aberdeen, the authorities of which University with the most
prompt liberality forwarded it to London for comparison and
description, on the application of Major Cautley and myself
to that effect. It was found embedded in a yellow clay
matrix, being in the mineral condition, which we have else-
where designated as ‘soft fossil.’ (Geological Proceedings,
loc. cit. p. 235.) The head when discovered in situ appears
to have been in the most complete state of integrity, judging
from the undisturbed condition of the parts now remaining,
and to have been broken during the process of excavation by
the native collectors employed for the purpose. In con-
sequence of this unlucky accident, we are for the present
deprived of the important information which the specimen
would otherwise have yielded, regarding the construction and
relations of the cranium proper of Chalicotherium.

The specimen next in importance is a fragment comprising
the left half of the lower jaw, from the angle on to the com-
mencement of the symphysis, of an individual which was not
quite full grown. It supplies the character of nearly the
whole of the inferior molar series. It is in the mineral con-
dition which we have called ‘hard fossil’ (Proceedings, loc. cit.),
having been embedded ina sandstone matrix. This specimen
was forwarded during last May to Paris for examination and
comparison by M. Laurillard, in the hope that it might
be matched by some remains in the French palzontological,
collections ; and I am indebted to that excellent observer,
and to M. Pomel, through the kindness of Sir Roderick
Murchison, for opinions respecting it which I shall have
occasion to refer to in the sequel (Plate XVII. figs. 6 and 7).

The detailed description of theteeth given in our first memoir
was published in an abbreviated form in the ‘ Proceedings,’
the abstract not having been the production of the authors.
Although correct in the main points, the descriptive minutize
—a matter of prime importance where teeth are concerned—
were so abridged in the abstract, that I deem it necessary on
the present occasion to claim the indulgence of the Society
for redescribing the teeth, more especially with reference to
the new light which the specimens since acquired have re-
flected on the dental characters of Chalicotheriwm.

Incisors.—First with regard to the incisors. THERE WERE
NO INCISIVE TEETH to Chalicotheriwm Swalense in either jaw!
It was as toothless in this respect as the most typical among
the Hdentata. The evidence is happily of the most demon-
strative character. The intermaxillary bones are preserved
in the upper jaw of the Dadoopoor specimen (fig. 3), perfect
to their tips. They consist of narrow slender slips of bone _

As PALTRY fe cong
rf oe cae hae tah

Fig. 1.

Fig. 2.

DESCRIPTION OF PLATE XVII.
CHALICOTHERIUM SIVALENSE.

Left upper jaw containing three molars and three premolars of
Chalicotherium Sivalense, one-half of the natural size. Copied
from a drawing by Mr. Ford in the Fauna Antiqua Sivalensis,
Plate LXXX., fig. 3. a, a’, a’, the conical cusps. This spe-
cimen, which is in the British Museum, was also figured in the
Proc. Geol. Soc., No. 98, as Anoplotherium Sivalense. (See
pages 191, 216 & 524.)

Upper jaw, right side, with the four back molars, and part of the
orbit of Chalicotherium Sivalense, one-half of the natural size.
Copied from fig. 2 of Plate LXXX. of F. A. S., and fig. 2,
Plate IL., of Proc. Geol. Soc., No. 98. Specimen in British
Museum. (See pages 194, 218 & 523.)

Figs. 3 and 4. Anterior half of an adult head of Chalicothertum Siva-

Fig. 5.

lense, one-half of the natural size. Copied from drawings by
Mr. Ford in Plate LX XX. of the F. A.S. The upper and lower
jaws are in natural apposition, and the greater portion of the
dental series of both jaws is visible. The specimen is in the
Museum of Marischal College, Aberdeen. (See pages 209 &
523.)

Is anterior portion of lower jaw of same specimen as last, show-
ing the two canines. (See page 211.)

Figs. 6 and 7. Left half of lower jaw of Chalicotherium Sivalense, one-

half of the natural size. Copied from drawings by Mr. Ford in
Plate LXXX. of the F. A. S. Specimen in British Museum.
(See pages 210 & 524.)

VOL. 1.

Vol.L. Plate 17.

Chalicothertum Sivalense.

CHALICOTHERIUM SIVALENSE. 211

converging to a sharp point, in which it is apparent not
only that there were no incisors, but also that such teeth
could not at any period of the animal’s age have existed in a
developed form in the upper jaw. These intermaxillary
bones are even more rudimentary in amount of development
than occurs in the typical Ruminantia, with which order
they closely agree in form. The evidence respecting the
absence of lower incisors is equally conclusive. The anterior
portion of the lower jaw of the same specimen is perfect to
the alveolar edge. A detached canine is seen on either side,
but the intervening space, ordinarily occupied by incisors,
is without a vestige of such teeth; and the alveolar border
of the incisive region is contracted in correspondence with
the convergence of the intermaxillary bones above, and

sloped off to a fine edge, in which it is clear that no incisive

teeth could have been implanted. There is not a trace of
them, even in the most rudimentary form.

We have here then, in the absence of incisive teeth in both
jaws, a character of a very unexpected nature, which at
once distinguishes Chalicotheriwm from Anoplotheriwm, and
from every other known genus of the order. There is
nothing analogous to it among any of the Ungulata, whether
pachyderms or ruminants, hitherto described.

Canines.—The upper jaw of Chalicotheriwm Sivalense was
equally devoid of canines as of incisors. A vacant diasteme
of considerable length stretches without interruption from
the anterior premolar to the tip of the intermaxillary bone,
on either side. The animal was of sufficient age to have
developed permanent canines, if the species had been supplied
with such teeth; and there is no indication upon the diasteme
in the shape of an obliterated alveolar pit, that the fossil was
furnished with deciduous canines of the milk series in the
upper jaw. A slight vacuity has been picked out, in clearing
away the matrix, on the diasteme at the point of junction
between the maxillary and intermaxillary bones, which at
first sight might be taken for the alveolus of a very rudimen-
tary canine; but the appearance, partly caused by an accidental
fracture, is deceptive; the vacancy is seen to extend across
the palate, and indicates merely the very slight nature of the
connection between the maxillary and intermaxillary bones.
In this second character of dental suppression the Sewalik
Chalicotherium also differs from Anoplotherium.

But canines were present in the lower jaw. They are repre-
sented by figs. 3 and 5. The crown is thick, cuneiform, and
somewhat triangular in shape, thinning off into a blunt apex,
the anterior edge being short and nearly vertical, while the pos-

terior edge is longer and more sloping. It is implanted with
P22

212 FAUNA ANTIQUA SIVALENSIS.

an inclination forward, by a strong thick simple fang, equalling
the crown in diameter. It is nearly of the same size and form
as the anterior premolar, and does not rise above the level of
that tooth. The enamel of the apex is polished by attrition
against the pad of the upper gum, but unworn. In the mode of
implantation, general incisor-like form, and degree of develop-
ment, the lower canine of Chalicotheriwm agrees with the
corresponding tooth of Anoplotheriwm commune; but the
crown is more simple, being devoid of the basal-notched lobules
occurring in the latter genus. Its position in the jaw is also
more advanced than in Anoplotherium ; the diastemal interval
of nearly an inch exceeding the space which would have
been occupied by the first or suppressed premolar, had the
latter tocth been developed.

Was there any sexual difference in the canines of Chalico-
therium? Were they present in the male and wanting to the
female? Was the fossil from a male or female? These
are interesting points to determine; but with materials at
present limited to a single example it would be idle to
attempt solving them; for irregularities in the degree and
order of dental suppression are so numerous and variable
among different genera in the Ungulate tribes, that there is
not a clue to a probable inference on the subject. That the
canines in the fossil were full grown is proved by the
form of the crown and condition of the fang, independently
of the evidence furnished by the wear attained by the penul-
timate molar that the animal was adult.

Further, it may be urged that the teeth here described as
canines may be considered rather as representing the outer-
most incisor of either side. This is in some measure an
open question, but the massive deeply-implanted fang is
entirely that of a canine; and the teeth have the position
ordinarily occupied by the canines in allied genera. The
analogy of the upper jaw, so far as it is worth, is also against
their being regarded as incisors.

Kaup, in his first account of Chalicotheriwm, describes and
figures a large tooth, which was found detached, as probably a
canine of C. Goldfussi; but it is omitted among the figures
given with his more recent description in the ‘Akten der
Urwelt;? whence it would appear that he no longer holds
that opinion. In our first memoir it was regarded as a
lower incisor of one of the Eppelsheim species of rhinoceros.
The evidence now adduced regarding the Sewalik fossil
proves, a fortiori, that it could not have been a canine of a
species of Chalicotheriwm. Bronn describes it, in accordance
with Kaup’s first opinion, and thereon rests a conclusion as to
the affinities of the genus.

re Goer

CHALICOTHERIUM SIVALENSE. 213

Molars.—Next, with regard to the molar teeth. There
were six molars on either side in both jaws of Chalicotheriwm
Swalense. For those of the upper jaw I must revert to the
figures previously given in the ‘ Proceedings’ (Pl. XVII. figs.
land 2). The jaws being locked in the Dadoopoor specimen,
the crown surfaces of the teeth are not visible for description.
Fig. 1 represents the left side of the upper jaw, with the
whole of the molar series in sitw beautifully preserved, of an
adult animal. In the‘ Akten der Urwelt,’ Kaup has figured a
corresponding fragment of the left upper jaw of Chalicotheriwm
Goldfussi from Hppelsheim, containing also a series of six
molars. The teeth of the two species are exceedingly alike in
form and proportions, those of the European species being
larger; they both exhibit a strong resemblance to the molars
of Anoplotherivwm commune, as represented by Cuvier in Pl.
XLVI. fig. 2, vol. iii. of the ‘Ossemens Fossiles.’ The Sewalik
specimen and the latter correspond exactly in age, the last
molar in both being in the germ state, with the ridges scarcely
affected by wear. The comparison between them is in conse-
quence greatly facilitated. The teeth in the Eppelsheim
specimen are more worn.

Upper Molars.—The crowns of the true molars (2 m. 3 m.)
in Chalicotherium Sivalense form irregular cubes, with a
square deeply excavated grinding surface broken up into
four principal points or ‘hillocks,’ separated longitudinally
and transversely by wide open clefts. The outer surface, as
in Paleotherium and Anoplotherwum, is divided unequally into
two hollow interspaces by three vertical eminences, ascending
from the base about half-way upwards; but the anterior and
central of these elevations, instead of being well defined and
keel-shaped, as in the two latter genera, form in Chalico-
theriwm enormous rounded convex bulges, projecting much
above the level of the hollow spaces included between them,
while the posterior angle is depressed and overlapped by the
anterior bulge of the next succeeding tooth. In Rhinoceros
these eminences are obsolete, the outer surface of the upper
erinders being nearly flat; in Anoplotheriwm they are well
marked, as also in Paleotheriwm; in Chalicotherium they
attain the maximum of development presented by the Ungu-
late genera, fossil or recent. In Ch. Goldfussi of Kaup the
anterior of these bulges is the largest, and it is notched at
the apex, so as to resemble very closely the corresponding
lobe or anterior semicone of the upper grinder of Lophiodon
Tapiroides. They further agree in the horizontal outline,
and in the low elevation of the crown. But here the resem-
blance ceases; the form and position of the other principal
points and the transverse direction of the main ridges ar

214 FAUNA ANTIQUA SIVALENSIS.

entirely different in Lophiodon. In the Sewalik species the
middle bulge is the most salient, and most deeply notched at
the apex, where it forms an obtuse lobule. The two hollow
interspaces are strongly inclined inwards as they ascend and
terminate each in a three-sided pyramidal point, forming in
the vertical direction a double chevron or W outline, the re-
entering angles being at the apices of the bulges. In like
manner, from the inflexion of the pointed terminations of the
hollows and the projection of the bulges, the outer longitu-
dinal ridge of the crown follows a bold zig-zag direction, and
another double chevron is formed in the horizontal plane of
the grinding surface. All this takes place in Chalicotherium,
exactly as in Palwotheriwm and Anoplotheriwn, but ina greater
degree, the angular flexures being more marked. The an-
terior hollow space is considerably wider than the posterior,
which especially in the last molar slopes inwards towards the
back end of the tooth. Ch. Goldfusst, in these particulars,
agrees entirely with the Sewalik species, but besides the
lesser elevation of the crown the angular flexures of the
outer longitudinal ridge are still greater, so that their
apices occupy nearly the middle of the grinding surface.

The anterior outer principal point, which forms in the
germ a distinct three-sided pyramid, wears down into a
crescentic disc of ivory. Its inner edge descends nearly
vertically towards the great central hollow, and is not
connected by means of a uniting ridge with the internal
large conical point which is opposed to it. A wide angular
gap intervenes between them.

The posterior outer lobe forms a similar, but smaller, three-
sided low pyramid, the inner angle of which is connected
with a corresponding ridge from the posterior inner ‘ hillock’
opposed to it. A wide fissure having an antero-posterior
direction, and opening towards the posterior side of the
crown, intervenes. It is not as in Anoplotherium continuous
with the central longitudinal valley, but terminates at the
angle of junction of the posterior lobes, as in Palcotherium
and Rhinoceros, giving rise in advanced wear to an isolated
patch of enamel corresponding with the posterior annu-
lar ‘fossette,’ described by Cuvier in these genera. This fos-
sette is well shown by the third molar, which is much worn.

The inner side of the crown is rounded at the angles, and —

forms a nearly semicircular contour towards the palate. Like
the outer division, it is composed of two principal lobes or
‘hillocks.’ The posterior of these forms a low three-sided
pyramid smaller than the corresponding outer lobe which is
opposed to it. It is separated from the great conical tubercle
which constitutes the anterior eminence by a wide gap formed

CHALICOTHERIUM SIVALENSE. 215

by the opening of the bisecting transverse valley, which ex-
pands in the centre of the tooth into a deep and very open
irreeular fissure having a triangular bottom.

The conical tubercle is situated at the inner side, as in
Anoplotheriwm, and forms a large projecting lobe arising from
a broad disconnected circular base. The side towards the
palate is nearly vertical; the apex is compressed into a sharp
convex edge, sloping off towards the centre of the tooth. It
stands opposed to the anterior outer pyramid, an angular
gap intervening. A low ridge forming the inner margin of
the crown descends from its base, connecting it with the
posterior inner pyramid and subtending the opening of the
transverse bisecting valley. A similar ridge from the an-
terior side of the base connects it with a small front lobe to
be next noticed. These connecting ridges are wanting in
Anoplotherium; they continue intact during advanced wear, as
arcuate enamel cornua appended to the disc of the tubercle.
Tn consequence of its isolated position, the conical tubercle
remains long unconfounded with the confluent discs of the
other lobes; the apex is first abraded obliquely on the side
next the axis of the tooth; it then wears down to a depressed
dise of ivory encircled by a ring of enamel, and it remains in
this state when the other coronal eminences are ground
down to a common field; while in Anoplotheriwm it is lost at
a much earlier period.

Up to this point the accordance of Chalicotheriwm Sivalense
with Anoplotheriwum commune is complete, in the general re-
presentation of the coronal lobes of the back molars. But in
Anoplotherium, corresponding with the anterior outer hillock
there is an interior smaller lobe in front of it, separated by
a distinct furrow, the two forming, like the posterior lobes,
an opposed anterior pair. This lobe, which represents the
anterior inner crescent of the symmetrical molars of the
ruminants, is less developed in Anoplotheriwm than the other
lobes, and during wear it does not present more than an
oblong disc, corresponding with the anterior horn of the
ordinary crescent. This reduction in form is accompanied
by an excessive development of the fifth coronal eminence, or
‘conical tubercle,’ which in the allied genera ordinarily
forms only a slender pillar at the opening of the transverse
middle valley. The deviation from the typical symmetry of
the ruminants first visible in Anoplotheriwm is carried out still
further in Chalicotherium, the anterior inner lobe being: still
more reduced, while the conical tubercle is proportionally still
more developed. The exact determination of this point in
the fossil is of great interest, as the symmetry or otherwise
of this part of the molars usually bears an important relation

216 “FAUNA ANTIQUA SIVALENSIS.

to the construction of the extremities as regards the number
of toes, thereby indicating the affinities of the genus.

~The characters presented by the reduced imner lobe in
Chalicotherium Swalense are these :—From the apex of the
anterior outer vertical bulge a low ridge proceeds across,
bounding the tooth in front; it is concealed about the
middle of its course by the overlapping of the adjoining
tooth, but reappears near the inner third of the distance, in
the form of a low compressed tubercle, which, as above de-
scribed, is continued forwards in an enamel ridge, sweeping
round the base of the great conical lobe upon which it ter-
minates. This small tubercle represents the inner division
of the anterior pair in Anoplotherium. A well-marked ante-
rior transverse narrow valley, which remains distinct after
advanced wear, separates it from the corresponding and
opposed outer lobe, and also from the base of the conical
tubercle ; therefore, the anterior inner, or fourth symmetrical
lobe, which in Anoplotheriwm is reduced to one horn of the
ordinary crescent, is still more rudimentary in Chalicotherium,
attaining little beyond the development of an anterior ‘ talon,’
while the conical tubercle, representing the fifth or odd
coronal lobe, is proportionally enlarged at the expense
of the reduced segment. There is no basal collar or ‘ bour-
relet ’ upon this or any other part of the back molars of Cha-
licotheriwm Sivalense.

The difference of pattern yielded by the grinding in face
during progressive wear is finely exhibited by the different
conditions of the three back molars (Plate XVII. fig.1). The
last (m. 3) shows the germ form of the crown, the wear being
limited to the posterior inner face of the anterior outer pyra-
mid. The penultimate (m. 2) shows the two outer and the pos-
terior inner lobe, half worn and confluent into a common
field of ivory, while the conical tubercle is almost intact. The
posterior fissure forms an open gap separated from the cen-
tral valley by a belt of ivory, as in the molars of Paleothe-
rium. The termination of the great transverse valley forms
a triangular depression in the middle of the tooth, while its
opening remains in the state presented by the germ. The
anterior valley still presents the condition of a transverse
fissure. The first true molar is considerably more advanced
in wear; the united disc of the three pyramids is nearly flat ;
the posterior fissure is reduced to an oblong peninsular patch
of ‘ fossette’ of enamel; the conical tubercle is ground down
to an isolated annular depression, with a divergent enamel
ridge at either side; the transverse middle valley forms two
triangular depressions united by a narrow neck; and the an-
terior transverse valley is reduced to two enamel pits, the

CHALICOTHERIUM SIVALENSE. 217

one corresponding in size and position with the posterior
‘ fossette ’ and the other separating the conical tubercle from
the rudiment of the anterior inner lobe:

The upper back molars, more especially the two last, are
enormously large, in comparison with the other teeth or with
the dimensions of the head. If found isolated they would
seem suitable to an animal approaching the size of a Rhino-
ceros, whereas the anterior part of the lower jaw and the
muzzle do not reach the dimensions of the Indian Tapir.

Upper Premolars.—The upper premolars in Chalicotheriwm
are limited to the three last: the normal first being sup-
pressed. They differ more from the corresponding teeth of
Anoplotherium commune than do the true molars. In the
latter the premolars are compressed lengthwise, and their
outer surface retains the three-ridged division which is pre-
sented by the true molars, while the conical tubercle, at the
inner side, is lost in a salient edge, which goes all round the
circuit of their oblong crown.! In Chalicotherium Siwvalense,
on the contrary, the premolars are compressed in the trans-
verse direction, so as to make them somewhat cuneiform from
the outside inwards; their outer surface loses the vertical
bulges of the back molars and becomes nearly flat, as in Rhi-
noceros ; while they retain the conical tubercle at the in-
side, as a well-marked distinct lobe throughout. Taken in
succession, from rear to front, they diminish very rapidly in
size, the united length of the three teeth not exceeding
much the width of the penultimate trie molar; while in
Anoplotherium commune the proportion in length of the three
true molars to the three last premolars is nearly as 3 to 2.
The internal conical tubercle, as in the case of the back
molars, gives off in front and behind a low basal ridge,

_bounding the teeth on their inner side. The form of the
outer surface of these teeth is well shown on the right side
of the Dadoopoor specimen. The antepenultimate or first
(theoretically the second) has a cuneiform outline to the
outer edge of the crown, and is implanted by a double con-
nate fang. The second (p.m. 2) is of the same form but
broader. The last (p.m. 8) is still broader, and presents an
obscure indication vertically of a division into an anterior
and posterior lobe. The first true molar (m. 1) which follows,
is distinguished at once from the premolars by the saliency
of its projecting bulges. The form presented by the worn
erinding surface of the premolars is well exhibited by fig. 1

1 «Teur caractére est d’avoir dans le | tranchant.’—Cuvier, Oss. Foss., tom. iil,

genre une couronne oblongue, entourée | p. 19 (4to. edit.).
de toute parte d'un rebord saillant et

218 FAUNA ANTIQUA SIVALENSIS.

of Plate XVII. The inner conicle tubercle is retained on
to the anterior tooth, with but an inconsiderable modifica-
tion of the character which it shows in the back molar,
and separated by an intervening valley from the outer
lobe. There is no trace of a rudimentary first molar in any
of the fragments. In the Dadoopoor specimen the alveolar
border falls off in an abrupt step upon the diasteme in front
of the anteror tooth.

The molars and premolars of the upper jaw of C. Goldfussi
agree in form and proportions so nearly with those of C. Si-
valense that it is not necessary to do more than refer to
Kaup’s figures. The crown ridges and principal points are
less elevated, and the dividing vallies more open and shal-
lower. The tooth which Kaup figures and describes as the
middle lower incisor of the left side is here regarded as the
penultimate upper premolar of the right side.

Inferior Molars.—The molars in the lower jaw, like those
above, were limited to six, viz. three true molars and three
premolars.

The Dadoopoor specimen, Plate XVII. figs. 3 and 4, shows
on the left side the anterior four of these teeth entire, and the
front half of the fifth, in apposition with the upper jaw,
while the detached left ramus (figs. 6 and 7) presents the last
four molars in situ, together with the empty alveoli of the two
first. This fragment had belonged to a young animal which
had just acquired its permanent teeth, the last premolar being
fully protruded but unworn, while the last true molar is in the
germ state, and had not entirely pushed through the gum ; the
crown of its posterior half is broken off. The teeth and the
form of the jaw are so much alike in the two specimens,
that there cannot be a doubt of their having belonged to the
same species of animal.

The inferior molars of Chalicotheriwm Sivalense differ more
from those of Anoplotheriwm commune than the upper, and
the points in which they differ constitute characters of resem-
blance to Paleotheriwm, while at the same time they are
marked by some peculiarities which are not met with in either
of those genera. These teeth have the general form of those
of Paleotheriwm, consisting of two semi-cylindrical lobes ;
and they further agree with the latter genus, in the two last
premolars being double-barrelled, like the three true molars ;
while in Anoplotherium the premolars differ very materially in
form from the back teeth. The convexity of the lobes is less
uniform and cylindrical than in Paleotheriwm; as in Anoplo-
theriwm they contract upwards to a small pyramidal point at
the outer side, but without bulging into convexities below, as
in the latter genus; while at the same time they hardly ex-

ee,

CHALICOTHERIUM SIVALENSE. 219

hibit a vestige of the basal ‘bourrelet,’ or collar, which is so
constantly found upon the teeth of Paleotheriwm. The tren-
chant edge of each lobe, after forming the small outer point,
is bent round to the inner side, so as to form a compressed
curve, which wears down to a crescentic pattern. There are
two peculiarities connected with these edges which require
special notice. The anterior edge of the front lobe descends
uniformly and obliquely forwards, from the outer point to a
low termination at the inner side, whence it shelves a short
way backwards to join on with the base of the internal middle
eminente, as occurs in the lower molars of Macrauchema
Patagonica, while its posterior edge and both edges of the
rear lobe bend in a concave line across, to terminate each in
a small elevated point at the inside; so that the posterior
lobe has two inner points, while the anterior lobe has only one.
This is precisely the reverse of what takes place in Anoplo-
theriwm, in which the anterior lobe has two internal points,
while the rear lobe has only one. Further, the angle of
junction of the two crescents forms a double or bifid point,
as in the Palewotheriwm of Montpellier (P. Awrelianense)
whereas in the other Paleotheria and in Anoplotheriwm it
is always simple. This character was specially noticed by
M. Laurillard, in comparing the fragment with lower jaws
of Paleotheriwm, in the Paris Museum.! This bifid point is
seen in Plate XVII. figs. 6 and 7; it is also exhibited by the
first and second true molars of the Dadoopoor lower jaw, a
vertical furrow indicating the line of division. The penulti-
mate molar (m. 2) shows a well-marked posterior talon ridge,
which commencing at the base of the posterior inner point is
directed downwards and across to terminate at the base of
the outer surface of the posterior crescent: thus reversing
what is presented by the anterior edge of the front lobe. This
talon is feebly developed on the first true molar (m. 1), and
is entirely wanting to the last premolar (p.m. 3), being the
only character by which the latter is distinguished from the
true molars. Direct evidence as to the form of this talon in
the last molar (m. 3) is wanting, in consequence of the fracture
in the specimen of the posterior lobe; but enough of the out-
line remains to show that this tooth was devoid of the third
lobe or crescent, which is constantly found in all the species
of Anoplotheriwm and Palwotherium. This is a character of
generic importance in which Chalicotherium Siwalense agrees

1 *Qette mAchoire semble tenir des | pellier, puisque, l’angle de réunion des
Palzotheriums, des Anoplotheriums et | deux croissans est double, ou forme deux
des Rhinoceros. Les dents ont la forme | points en regardant les dents par la face
générale de celles des Palzotheriums et | interne,’ &e.—(MSS. Com. from M. Lau-
surtout de celles du Paleoth. de Mont- | rillard.)

220 FAUNA ANTIQUA SIVALENSIS.

with Macrauchenia Patagonica, according to the lower jaw so
determined by Owen,! and with Rhinoceros; but both of
these genera are without the posterior ‘talon’ to the lower
back molars, found in the Sewalik form. Chalicotherium
differs also from Rhinoceros in the equal height of the an-
terior and posterior lobes.

The inferior molars decrease very uniformly in size from
the last true molar to the penultimate or first (theoretically
the second) premolar, which has a simple cuneiform crown re-
sembling in size and shape the lower canine. It is implanted
by a double confluent fang. The second premolar (theore-
tically the third) exhibits a reduced condition of the double
crescent of the back molars, the internal point of junction
being bifid and elevated as in them. The last premolar
(p.m. 3) is double-barrelled and differs in no respect from the
true molars except in the absence of the posterior talon. A
question may be raised, whether the first premolar is not re-
presented by the tooth above described as an inferior canine.
But its advanced position in the jaw, close to the incisive
border and the thick procumbent simple fang by which it is
implanted, appear to be conclusive against a view of this
kind.

In consequence of the close resemblance between the upper
molars of the Sewalik and Eppelsheim species of Chalicothe-
rium, it might be expected that the lower molars would also
agree. Kaup describes and figures two teeth which he refers
to the lower jaw of Chalicotherium. I have compared casts
of these with the lower molar of Chalicotherium Siwalense,
and the generic accordance was found to be complete. The
one (Kaup, loc. cit., tab. vil. fig. 5) is a worn last molar of
the left side, broken behind (regarded by Kaup as penulti-
mate). In correspondence with the upper molar, the crown
is proportionally broader and less elevated than in the
Sewalik species ; the anterior edge of the front lobe is not
so much inclined, but the point in which it terminates shelves
backwards towards the base of the middle internal lobe, as
in the Indian fossil; the internal point at the confluence of
the two crescents is also considerably elevated, and the
outer termination of a posterior talon is discernible; the
other tooth is an antepenultimate true molar of the right
side. Hxcepting the greater width and lower elevation of
the crown it agrees closely with the corresponding tooth of
Chalicotheriwum Sivalense in form and proportional size. It
would appear that this accordance runs throughout the

1 Odontography, p. 602, pl. 135, fig. 7. The specimen is in the British Museum:
‘ Paleont. Coll.’ No. 19,950.

CHALICOTHERIUM SIVALENSE. 221

series of lower molars. M. Pomel, to whom the detached
lower jaw was submitted in Paris by Sir R. Murchison,
was immediately struck with the resemblance. His words
(extracted from a communication with which I have been
favoured by Mons. Pomel) are :—

‘Cette machoire inférieure me parait ressembler en tout
point, sous les rapports génériques, avec une piéce découverte
a Sansans et qu’on ne peut rapporter qu’a une machoire
supérieure trouvée dans le méme gite, que j’ai vue dans le
cabinet de M. de Blainville et qui est certainement celle du
Chalicotheriwm, Kaup. Les molaires ont exactement la méme
forme générale, et la derniére parait également avoir manqué
de troisiéme croissant. Les molaires antérieures sont de
méme plus semblables aux postérieures. Malheureusement,
le bord symphysaire a également été brisé.

‘Mais il me parait quil y a certainement une différence
entre les animaux.’

CHALICOTHERIUM.' Kaup.

Char. essent.—Dentes 26: viz. ine. $2; can. $7; mol. &. Pro

primoribus, laniariisque superioribus diastema: horum in-
feriores cuneiformes. Molarium superiorum 38 anteriores
transversim compressi, condensati; posteriores maximi,
quadrati, coron4 complicata quadricuspide; inferiores bilu-
nati, postrem6 conformi.—Oculidepressi. Nasus ? Palmee
et plante verosimiliter tridactyle.

Genus inter Pachydermata extinctum in Europe et Asiz
Australis fossariis obvium, dentium primorum defectu insig-
nitum: hine Anoplotherio, illince Rhinocerotidi affine: maxillee
superioris forma, Ruminantia, ceteroquin valde diversum,
simulans.

1. Chal. Goldfussi. (Kaup.) ‘Statura Rhinoc. Javanici.
Olim in Germania et Gallia degebat.

2. Chal. Sivalense. (Fale. & Caut.) Statura inter Rhinoc.
Javanici et Tapiri Indici medium. Olim degebat in India
Orientali.

1 From xdAt, eal; or pebble, and | which the European fossil was dis-
Onptov, the name being probably suggested | covered by Kaup. Cuvier’s Anthraco-
by the gravelly beds of Eppelsheim, in | theriwm is a similar case.

222 FAUNA ANTIQUA SIVALENSIS.

APPENDIX TO MEMOIR ON CHALICOTHERIUM.
Extracts From Dr. Fatconer’s NoTE-Books.

Darmstadt Museum: September 4, 1856,

Examined a beautiful specimen, hitherto unfigured, of the left upper
jaw of Chalicotherium Goldfussi, nearly in every way the counterpart
of our Sewalik specimen figured in the Geological Proceedings (PI. XVII.
figs. 1 and 2). It comprises the same width of palate, the same height
of mutilated maxillary, and the same number of teeth; but older, i.e.
in more advanced detrition; the correspondence in every general re-
spect is very remarkable. The specimen contains six molar teeth in
continuous line, i.e. 3 premolars + 3 molars.

The antepenultimate premolar (first) is exceedingly like the Sewalik
figure, being small, nearly triangular, with a main (outer) pointed cusp,
which is worn down as in the Sewalik specimen, with the small insular
detached interior cusp (placed at the posterior end developed into a
lobe), continued forwards in a basal bourrelet, which terminates
anteriorly in a small lobule, as in the Sewalik figure. This tooth is
proportionally much less worn than the three next, which are ground
down so as to involve the insular cusp in one common disc (i.e. very
much worn). The penultimate true molar is less worn, having the cup
of the insular disc untouched. The cusp of the last molar is perfectly
intact, but the outer main ridges (the chevrons) are well worn.

The two last molars are, as in the Sewalik specimen, nearly of a size;
the teeth then diminish most rapidly in front with the anisodon cha-
racters.

The others diminish very rapidly from the second true molar. The
three anterior premolars very much compressed, i.e. broader than long ;
the first nearly triangular.

Inches Metre

Length of series of 6 molars . r C . . . 3 (lt 18
Ditto 3 true molars . A . . 5 . 50 124
Ditto 3 premolars : . 2°18 055
Length of last true molar. a : - . 5 lies 048
} Width of ditto . 3 0 ; : c - Hd 048
Length of penultimate molar. . : - , lr 042
( Width of ditto 2 A : Fs - pele 047
eee of antepenultimate . . . . . Bale: 032
Width of ditto : . 145 035
Length of last premolar . 08 020
Width of ditto pp a, 028
Length of 2nd premolar (OR 016
1 Wad of ditto - : : - : . 0-9 022
Length of 1st premolar . ° A - 5 . : eOzH 017
Width of _— ditto : . : ° A . c 2) Ont 017

N.B. Inches taken roughly with tape.

Memo.—The anterior (antepenultimate, or first) premolar is distinctly
two-fanged, and the fangs are more divergent on the inside than on the
ouler.

Aa)

CHALICOTHERIUM SIVALENSE. 223

Saw no cast, nor the original of Kaup’s canine, fig. 4 of Plate VII.
of the Darmstadt fossil; but there is a triangular cusped tooth with a
long fang which must have had a very oblique, i.e. sloping insertion in
the jaw, which may be a canine of Chalicotherium belonging probably to
left side of upper jaw, or to right side lower. It is much larger than our

_ lower ‘ canine,’ more triangular in section, with an edge behind, and

rather blunt in front; apparently unfigured. This supposed canine is
larger than the anterior premolar; in our Sewalik (Baker’s) head, the
lower canine and anterior premolar are nearly of a size.

Memo.—Kaup, in conversation, tells me the supposed canine proved
to be an incisor of Rhinoceros Goldfuss?.

Chalicotherium.—Munich, June 15, 1861.—Most interesting of all
the Pikermi collection are a set of specimens of a very large species of
the same genus as our Sewalik Chalicotherium. This is the Nestorithe-
rium, Kaup (Beitrage, viertes Heft. 1859), which is figured and described
by Wagner (1857) under the name of Rhinoceros pachygnathus.

The principal specimen, fig. 15, Plate VII., execrably drawn, consists
of the greater part of a cranium from the sinciput on to the middle
of the diasteme, and the greater part of the nasals; but the whole of
the sphenoidal and occipital portion wanting. The entire chaffron, with
greater part of the nasals, is present, but so crushed down upon the
maxilla of the left side, containing the molars, that not a trace even
of an orbit can be detected. It looks as if the specimen never had had
an orbit! Frontal portion behind the last molars broad and concave,
longitudinally and across, as in an Arab horse. Cerebral box (bo/te)
broad and without ridges over the temporals, something as in the genus
Equus. Nasals also seem as like those of a horse, but the lower extre-
mities broken. ‘The animal very old (from the teeth), and no trace of
suture between nasals and frontal; but a line of groove marking the
internasal and interfrontal suture. The maxillary with molars, on right
side, entirely wanting. No determinable trace of a sub-orbital foramen,
or of incisive bones.

Dentition of left side—-All the molars are in situ, viz. three true
molars, three p.molars = 6 together, with 1 2in. in length of a sharp
diastemal ridge in front of p.m. 2. The animal was very old, all the
teeth being ground down to a flat surface, except the last; and of the
last the isolated cone is ground down low; but the posterior angular
fissure-like valley is intact. The teeth in general form, and anisodontine
character, exactly like the Sewalik specimen which has the orbit, right
side; but there is this difference, that in the Pikermi specimen the
molars have a very strongly pronounced internal basal bourrelet show-
ing in the premolars, as in hippopotamus, and in the last molar, a very
bold, broad, sharp, bounding keel.

The dimensions are as follows :—

Length of pm.1. ,
Width of ditto

Length of pm.2 . ; ’
Width of ditto ‘ ° c -

Inches
Extreme length of fragment . - . . 18:
Length of line of 6 molars, » O4
Ditto of 3 true molars n . 63
Ditto of 3 premolars . ° . :

a )
mR O
Or

bo

ee Rigee Snag”

224 FAUNA ANTIQUA SIVALENSIS.

Inches
Length of pm.3 . . . . . ‘ - 1:22
Width of ditto : 2 ape 5)
Length of true m. 1 (outer side broken) . 1:65
Length of true m. 2 (outer side). . » 2S
Width of ditto (greatest at middle) . ao Al)
Length of true m. 3 (outer side 2°65) 27
Width of ditto (greatest in front) 2:
Length of diastemal edge remaining 1:2

Memo.—The teeth, both upper and lower, show distinctly on their
comparatively thin enamel the parallel fibrous lines, as in Rhinoceros.

The greater part of the palate is present, and the sinus has its bottom
in a line with the extremity of the last molar.

The anterior premolar (p.m. 1) is sharp and wedge-shaped in front,
wide behind. No trace of dise of pressure in front ; basal bourrelet
and isolated cusp shown by disc; crown ground down low.

P.m. 2, worn down to an uniform disc, with slight remains of two
fissures.

P.m. 3, much worn, and like 2; bourrelet well marked.

N:B. The outer surface of the three premolars quite flat, without
chevrons.

T.m. 1, antepenultimate true molar, much worn ; disc of isolated cone
confluent with the rest of the surface ; outer surface broken off, and the
W-shaped chevrons lost.

Penultimate true molar worn down low (lower even than in the Sewalik
true mol.1); the disc of its conical cusp confluent in front with main disc,
but distinct behind. The outer surface shows very markedly the W
pattern and bulges of the Sewalik form. The posterior angular fissure
ground down, leaving only a slight crescentic depression.

The last true molar shows the disc of the cone distinct, or nearly so
(a fracture and chasm in the specimen conceals this partly). It shows
the chevrons and bulges very boldly ; and the posterior angular valley,
with its bounding ridges, quite intact and shaped exactly as in the Se-
walik species.

In a general way, the Pikermi species differs from the Sewalik one in
having the last molar longer than the penultimate, the reverse of which
occurs in the Sewalik. Also in the latter the true molars are broader in
reference to their length than in the Pikermi.

The diastemal ridge is very sharp, and there are no indications of
canines ; and none of the incisive slips of the Sewalik form.

Of the lower jaw there are two specimens: one young, with both
rami and symphysis, and milk dentition; the other very old, with both
rami, six molars on either side much worn, the whole of the ‘symphysis,
the diastemal edge on left and right, and the greater portion of the eden-
tulous incisive margin.

The following are the principal dimensions :—

Inches
Extreme length of specimen . “ . . 15
Length of line of 6 molars. F “ 5 = 859
Ditto of 3 true molars - ‘ - = uO:
Ditto of 3 premolars - OS
Length of diasteme to incisive ‘border left side. 2°85
Width between diastemal ridges at base . 1:3

ee

BIMGG ©

rf

~—-=. "= ©

ree

nae

CHALICOTHERIUM SIVALENSE. 225
Inches
Length ofp.m.2 . ‘ ; c 2 5 . O07
Ditto of p.m. 3 : r 0 : ; A ee by

1:0
Ditto of p.m. 4 : : c : : Bmeweleat
Ditto of true m. 1 (middle) 4 : : Smt)
Ditto of true m. 2 (outer) 0 : 271
Ditto of true m. 3 . : ; 2°3

P.m. 2 is the least worn of all—compressed; no disc of pressure in
front.

P.m. 3 has crown mutilated both sides.

P.m. 4 shows the two crescents, as in rhinoceros: crown well worn.

T.m. 1, or antepenultimate, is very much worn: crescents confluent.

T.m. 2, (penultimate), also well worn; crescents confluent and yielding
cordate discs.

True m. 3, less worn, discs nearly confluent, somewhat reniform in
character.

All these molars show a distinct basal bourrelet outside at base.

The diastemal ridge nearly entire on left side, very sharp, and shows
not the slightest indication either of a detached first premolar or of a
canine. F urther, the left half of the incisive border present and en-
tire, without the slightest trace of either incisive teeth or their alveoli.

The symphysis commences in a line with the posterior third of p.m. 3,
and the total length of symphysis to incisive border is about four
inches, or a little more.

The diastemal ridges rise into a sharp edge, leaving between a
deepish gutter or spout, which expands a little as in the Sewalik species
towards the incisive border—but only a little.

The most remarkable character of all is, that the lower surface of the
symphysis is contracted forwards into a sharp keel as in a boat, which
terminates suddenly at 2:7 inches in front of the anterior premolar, and
then the short incisive margin projects forwards to its slight expansion,
without a trace of this keel! showing a very remarkable edentate cha-
racter. Nothing of this keel seen in the Sewalik species, and it is very
different from Lartet’s specimen, called Chalic. Larteti by Kaup.

Certainly no lower incisives, as in Lartet’s jaw.

ON THE INCONVENIENCE OF RETAINING THE DICHOBUNES IN
. ANOPLOTHERIUM,

Great inconvenience arises from retaining the Dichobunes in Anoplo-
therium, and quoting the species under the name of that genus, the
true Anoplotheria being allied to rhinoceros in their dentition with
the natatory habits of the otter in the case of A. commune, while some
of the species ranked under Dichobune are nearly indistinguishable from
the musk deer. Such heterogeneous materials are too much for the
limits of any genus. Cuvier considered his arrangement of them as
merely provisional—‘ Les deux espéces suivantes (A. murinum and A. 0b-
liquum) sont encore plus douteuses, car il ne serait pas impossible qu’ elles
appartinssent 4 de petits ruminans, et toutefois jusqu’ & ce qu’on en
ait la preuve rigoureuse, on puit les laisser dans les Anoplotheriums, au
moins pour la nomenclature.” Cuvier even regarded A. leporinum as a
provisional name only (Ossemens Fossiles, tome iii. p. 70). The incon-
venience alluded to has been felt in the case of Mr. Pratt’s fossil, from

VOL. I. Q

226 FAUNA ANTIQUA SIVALENSIS.

Binstead (Geol. Trans., Series II. vol. iii. p. 451), which is the A. cer-
vinum of Professor Owen (idem, vol. vi. p. 45), admitted on all hands to
be prodigiously like a musk deer, while Anop. (Chal.) Sivalense and
the Chalicotherium Goldfussi press in the direction of rhinoceros.
Should the two separate cannon bones have proved constant in all the
species now ranked under Anoplotherium and been limited, as Cuvier
imagined, to them, the character might have been better used to define
a group of genera than as the means of defining a genus. But we have
ascertained the singularly interesting fact that the united cannon bones
so eloquently handled by the illustrious Cuvier in the ‘ Discours Prélimi-
naire ’ as invariably present in the ruminants are subject to exceptions,
and no longer ‘ as certain as any conclusion in physics or morals ;’ nor
a mark ‘surer than all those of Zadig’ (Disc. Prélim. p. 49). We have
had an opportunity of examining the skeleton of the African Moschus
aquaticus (Ogilby) described from a living specimen by Mr. Ogilby in the
Zoological ‘ Proceedings,’ and we find that the metacarpals are distinct
along their whole length! with the succentorial toes as much developed
as in a Peccary. The fore leg, from the carpus downwards, is indis-
tinguishable from that of a Peccary! We had previously ascertained
the same structure in the foot of a fossil ruminant from the Sewalik
hills, but believe that the present is the first announcement of the fact
in regard to an existing form of this order (see antea, page 196). The
deviation from the ordinary ruminant type indicated by the foot is borne
out by a series of modifications in the head, bones of the extremity and
trunk, all tending in the direction of the Pachydermata. Inasmuch as
the Dorcatherium of Kaup breaks down the empirical distinction as to
the teeth between the Ruminantia and Pachydermata, so does the
Moschus aquaticus in regard to the feet.—[H. F. 1843. ]

CAMEL.

to
iy)
bor |

XI. ON THE FOSSIL CAMEL OF THE SEWALIK
HILLS.!

BY HUGH FALCONER, M.D., AND CAPTAIN P. T. CAUTLEY.

Amonest the most interesting of the fossil remains of
Mammalia, which have been found in the Sewalik strata, the
Camel may, undoubtedly, take up a high position. Inde-
pendently of the speculations which the remains of this genus
would lead to, relatively to the form and features of the
country previously to their entombment, the circumstance of
the Camel having been up to this period a desideratum in
fossil zoology adds very considerably to the interest of the
present discovery.

The only? remain which we find noted is in Cuvier’s ‘ Osse-
men’s Fossiles,’ where a reference is made to the Merycotheriwm
Sibericum of M. Bojanus, which Cuvier decides to be an un-
doubted species of Dromedary. This remain consists of three
teeth brought by a merchant from Siberia; the place or
stratum in which it was found is unknown, and Cuvier’s re-
mark—‘ Si les trois dents que M. Bojanus vient de publier
sont effectivement fossiles ’—throws an uncertainty even on
its antiquity.

In the identification of the Sewalik fossil there can be no
doubt; and although we should have preferred delaying this
paper until we had a perfect skull, we may, perhaps, be ex-
cused for entering upon the description, since the portions of
the skeleton we already possess, including parts of the skull,
are sufficiently marked to remove all doubts as to the animal
to which they belonged.

The Camel is placed in systematic arrangements at the

1 Reprinted from the ‘Asiatic Re-|p. 507. Besides the Merycotherium,

searches,’ vol. xix. p. 115. 1836. Subse-
quently to the publication of this paper,
Many additional remains of the Sewalik
camel, including the vertebre and
the bones of the anterior and posterior
extremities, were figured in the ‘Fauna
Antiq. Sivalensis’ (Plates lxxxvi. to xe.),
to the description of which the reader
is referred.—[ Ep. |

2 Ossemens Fossiles, tom. v. part ii.

Cuvier also notices in the same article
a fossil femur, of which he says ‘qui
ressemble aussi beaucoup, dans ce qui
enreste, 4 celui d’un chameau.’ A draw-
ing of the specimen, which was found
near Montpellier, was sent by M. Marcel
de Serre to Cuvier. Our information
does not extend later than the third
edition of the Ossemen Fossiles, in 1826.

Q 2

228 FAUNA ANTIQUA SIVALENSIS.

head of the Ruminantia. In common with the rest of the
family it has a compound stomach. Its molars have the form
characteristic of Ruminants, and its skeleton generally is
constructed on the same plan. But the skull differs very
materially in form from that of the horned Ruminants; and
we see in the less complete anchylosis of the metacarpal and
metatarsal bones, and in the greater division of the carpus
and tarsus, an approach to a higher family among the Mam-
mala. The anomalous character of the pseudo-canines, the
presence of incisors in the upper jaw, the thickness of skin
and horny soles of the feet, show a strong affinity with the
Pachydermata. This affinity is greatest with the Solidungula,
which the camels approach by their more divided carpus and
tarsus; while the former approximate to the camels by their
soldered metacarpals and metatarsals.

In drawing a comparison between the skull of a Camel and
that of a horned Ruminant, the peculiarities of the former
are exhibited in the great width and massiveness of the cranial
portion, contrasted with that of the muzzle, which is slender;
the position of the orbit is more central, and its edges more
prominent, owing to the elongation of the cranium and to
the greater development of the temporal fossa.

We cannot do better, however, than follow Cuvier in his
observations on the same subject, who, in drawing his com-
parison with the separate bones of the head, remarks—‘ In
the true Camels the occipital crest is more elevated, and the
temporal fossee more hollowed than in the Lamas, this de-
velopment being nearly as great in the Camel as in the
Carnivora; the occipito-temporal suture is considerably in
front of this crest; the nasal bones are narrower at their
bases, and a much larger space intervenes between the small
membranous portion situated at the angle of the nasals and
the lachrymal bone, a very small portion of which is exposed,
and which does not even extend to the internal sub-orbital
foramen in the orbit.’ These remarks are applicable to the
Ruminants generally, as well as to the Lamas, although the
passage from which we have made the quotation is intended
for the latter. We may add two other points in which a
marked difference exists between the skull of the Camel and
the horned Ruminants : first, in a greater depth of maxillary
bone, and a consequent elevation of the nasals; second, in
the external nasal aperture being provided with three pairs
of bones, the nasals and intermaxillaries being separated ;
this intervening space is a general feature in the Camel,
although subject to great variations in extent. We have seen
specimens with the nasals and intermaxillaries separated by
a space of two inches, and others again with only one quarter

CAMEL. 229

of an inch of the maxillary bone appearing on the nasal
aperture. In the horned Ruminantia this peculiarity is only
observable in the Yak and Auroch.

We have considered the above observations applicable in
pointing out the differences that exist in the osteology of
the Camel’s head and that of other Ruminants, previously to
- entering upon the teeth; as with these two points fully ex-
plained, the identification of our fossil is placed beyond all
doubt, and it will only remain then for us to show the
differences which it exhibits.

Tt has been before noted that the anomalous character of
the teeth is one of the points connecting the Camel with a
higher family. The molars, however, are, as is normal in
Ruminants, in number twenty-four; six on each side in the
upper, and the same in the lower jaw. The first molar, which,
from having the simple and pointed form of a canine tooth,
has by some naturalists been termed a second canine, is one
of the chief peculiarities of the Camel. It is situated at some
distance from the remaining molars, which, in number five,
are in a continuous series ; in the lower jaw the second molar,
or that which may be considered the first of the series, is de-
seribed by Cuvier as falling out at an early age; and, not
being replaced, it leaves a series of four teeth only.

We have examined a number of skulls of the common or
Arabian Camel, and have only found one example of the ex-
istence of this second molar in the lower jaw; the series in
all other cases consisting of four, with a wide intervening
space between the first and third. That it is a part of the
true series is undoubted, from its existence in the specimen
above alluded to. It is exceedingly small and rudimentary.
Its position is central on the space between the -first and
third of the whole series.

The skull in which we have observed this small and
deciduous tooth is from a fine specimen of the Camel pro-
cured at Hissar. The animal to which it belonged was full-
erown, we should say somewhat passed the adult state,
judging from the bones of the cranium being anchylosed, and
a consequent absence of sutures. There is a greater develop-
ment of all the distinctive characters in this specimen, in
depth of maxillaries, comparative dimensions of the cranial
and facial portions, &c., arising from the superior growth to
which the animal has attained, and from the superior class of
animal from which it was selected; and the space occupied
by the maxillary bone in the external nasal opening is also
smaller than we have observed in any other specimen.

In the upper jaw the line of molars consists of one sharp
pointed tooth similar to a canine, and situated at some dis-

230 FAUNA ANTIQUA SIVALENSIS.

tance in front of the others, which are in a continuous series ;
the two first being single-barrelled, and the three last, or the
three true molars, double-barrelled.

In the lower jaw a sharp pointed tooth, corresponding with
that in the upper jaw, is situated in the same way; and takes
the place of the first in the series. The intervening space
between this and the continuous line contains the second or
small deciduous molar. The remaining four teeth are placed
together, the first of these (or third of the whole series) con-
sisting of a similar cylinder, and the three permanent molars,
as is usual with the Ruminants, being double-barrelled with
the exception of the last, which is triple-barrelled. Finally,
as in the Sheep and Antelopes, the cylinders are well defined,
and without any approach to accessory pillars.

There are two canines in the upper jaw, as we find in
another section of the Ruminantia.

The presence of incisive teeth in the upper jaw is pecultar
to the Camel as a ruminant. Of these teeth there are two,
corresponding in position to the outer incisors, and similar in
form to the canines. In taking a lateral view of the skull
this similarity of form in the incisor, canine, and first molar,
gives the appearance of three canines in the upper jaw.

In the lower jaw there are, as is normal in the family,
eight incisors, differing in form from those of other genera—
the outer ones taking the simple and pointed form as
described above, and the six intermediate ones being more
regular in proportions than is usual in Ruminants, and
having on each side a nick or hollow on the grinding surface.

The teeth of the Camel, then, are as follows :—

Upper Jaw.—2 Incisors.

2 Canines.
12 Molars, two of which are pointed and have
been termed second canines.

Lower Jaw.—8 Incisors, two of which are pointed.

12 Molars, two of which are pointed, and two
deciduous at an early age.

The chief peculiarities of the skull are—

. Narrowness of muzzle.

. Advanced position and prominence of orbits.

. Elevation of sagittal and occipital crests and develop-
ment of temporal fossze.

Narrowness of nasal bones at the posterior extremity.

. Extreme depth of maxillary bone, producing an arched

appearance in the nose of the animal.

. Form of sphenoid and basillary portion.

. Number of bones on the external nasal aperture.

NO OO CWroe

CAMEL. 231

This summary brings us at once to the comparison of our
fossil species with the existing Camelide.

In pursuance of the rule that we have proposed to follow
in naming the new species so as to ally them at once to the
mountain series, whence their remains have been obtained,
we propose calling the largest, and that nearly approaching
the Indian species, Camelus Sivalensis; for the second or a
smaller species, the description of which we shall enter upon
more fully hereafter, and which may perhaps have been more
closely allied to the Lama, we propose the name of Camelus
antiquus.'

CAMELUS SIVALENSIS.

Of the Camelus Sivalensis we draw our description from the
remains both of the skull and of the bony structure of the
animal generally. We have at present only portions of the
skull to guide us. The remains of the lower jaw, however,
are perfect, including the coronoid processes. The arti-

culating ends of bones are in sufficient abundance, and in a

sufficient state of preservation to enable us to form a very
tolerable idea of the size and height to which the animal
must have attained.

To commence with a comparison between the fossil skull
and that of the Dromedary or Common Camel in use in the
Bengal Provinces. From the imperfection of our fossil
fragments, and the sutures not being distinctly traceable in
most cases, we must be satisfied with a view limited more to
the general character than to the detailed boundaries of the
bones; yet it is fortunate, that in some cases where these
boundaries are especially required as a distinctive character,
as in the naso-frontal and naso-maxillary suture, our frag-
ments, imperfect as they are, have been provided with them.

The form of the skull, position of sutures (as far as our
fossil fragments exhibit), and the teeth, both in number and
character, very closely resemble the existing species above
referred to. We draw our comparison from a fragment con-
sisting. of the posterior portion of the nasals and maxillary
bones, with the frontal to the posterior border of the orbits.
This fragment would alone establish the genuine position of the
animal, and in the absence of a perfect skull we could not have
possessed a specimen more applicable to our present purpose.
This fragment, in fact, contains three of the most prominent
points in which the Camel differs from all other Ruminants.
Here we have the contrasted breadth of the frontal and facial
bones, the extreme narrowness of the posterior extremity of

1 Only one species of Camel, C. Siva- | ‘Fauna Antiqua Sivalensis.’—[Ep.]
lensis, was subsequently figured in the

232 FAUNA ANTIQUA SIVALENSIS.

the nasals, and the great distance between that point and the -
anterior border of the orbit, distinctly shown. In the fossil
there is a strong resemblance in all these points to the species
now existing; the swelling of the frontal is as highly de-
veloped, and the deep superciliary notch as well defined.
The narrowness of the nasal bones on their approach to the
frontals is well marked, as also that space occupied by the
membranous portion in rear of the nasal bones; and the
superciliary foramina correspond in size and position, being
placed as remote from the orbit as in the living animal. In
viewing this fragment laterally we observe that the orbit has
an excess of length on its antero-posterior diameter, the orbit
of the existing Camel being either a perfect circle, or having
the excess of length in its vertical dimension (Plate XVIII.
fig. 1, or F.A.S. Pl. LXXXVI. 4a). The mutilated state of
this fragment (see fig. 4, Pl. LXXXVI., F.A.S., giving an
upper representation of the fossil) does not admit of remark
or comparison further than that the animal to which it
belonged was far advanced in age, and had arrived at that
period and state of dentition when the obliteration of the
crescentic lines was complete, and when the grinding surface
consisted of ivory with an imperfect margin of enamel. The
second or third false molars, or those with a single cylinder,
are here in position; exhibiting a remarkable affinity to those
in the existing Camel both in form and in the contrasted
obliquity of wear (to front and rear respectively in the first and
second teeth), which is such a peculiarfeature in the old animal.

Fig. 2 of Plate XVIII. (or F.A.S. Plate LXXXVL. fig. 3) is
a representation of another fragment ; both jaws being locked
together, but the anterior and posterior extremities, with the
upper surface of the skull, wanting. The animal from which
this remain originated was young—its ultimate permanent
tooth not completely developed, and the third milk molar
still in position. The general character is that of the present
Camel, the form of maxillaries, thickness of lower jaw and
external appearance of teeth, corresponding as closely as two
skulls of one species would do in the animal now existing.
The position of the sub-orbital foramen, however, is rather
higher up on the maxillary, and the diminution in depth or
tapering of the lower jaw is not so considerable as we observe
to be the case in the existing Camel.

Figs. 3 and 4 of Plate XVIII. (or F.A.S. Pl. LXXXVI.
figs. 5 and 5 a) give us a still further insight into the form
of the head of the fossil Camel. The lower jaw (with the
exception of the upper portion of the ascending branches
including the condyle and coronoid processes) is quite perfect ;
and the lines of molars of the upper jaw are also in position.

DESCRIPTION OF PLATE XVIII.

CAMELUS SIVALENSIS.

Vig. 1. Lateral view of cranium of Camelus Sivalensis, one-fourth of the

natural size. The specimen is in the British Museum, and is
copied from a drawing, by Mr. Bone, in the Fauna Antiqua Siva-
lensis. Plate LXXXVL, fig. 4a. (See pages 232 & 533.)

. Is of another specimen in the British Museum, showing both

jaws of Camelus Sivalensis locked together, and one-fourth of
the natural size. The figure is copied from Plate LXXXVI.,
fig. 3, of the F. A. 8. (See pages 232 & 532.)

Figs. 8 and 4. Are horizontal and profile views of the lower jaw of

Camelus Sivalensis, one-fourth of the natural size, and copied
from figs. 5 and 5 a of Plate LXXXVI. of the F. A. S. This
specimen is also in the British Museum. (See pages 232 &
523.)

Fig. 5. Fragment of upper jaw of Camelus Stivalensis, containing the

second and third true molars, three-eighths of the natural size.
In the British Museum. Copied from Plate LXXXVIL, fig. 3,
of the F. A. 8. (See pages 234 & 533.)

VOL, I. c

Vol.I. Plate 18.

Ss

GRBane dal. 5 Bidet Yith

Gamelus Snvalensis.

Mn,

“ate

A,
:
‘
é

1) ies

tear lems gt

CAMEL. 233

Here we may express a regret as to the want of careful super-
intendence in excavating and removing these fossils from the
stratum. It is possible, in the present case, that the whole
of the upper portion of this specimen might have been
obtained had proper care been taken in removing the cir-
cumjacent matrix. It will be seen that the upper portion

has been as it were cut off on a line with the alveoli of the

upper molars, leaving not only them but also the upper
canines in position.

Fig. 5' represents the lower jaw of the existing Camel, and
placed in juxta-position with the fossil will convey a tolerable
idea of the form and character of both. The measurements
of the lower jaws are annexed.

Existing from Existing from

Dimensions, Lower Jaw ave Hisser Pe ee
In. Mét. In. Mét. In. Mét.
1, Extreme length from alveolus
of incisors to rear of ramus . | 16°8 | -428 |15-0 *3882 (15:2 “387
2. Ditto expanse of alee tothe rear | 81 | 25 6-4 "162 | 6:2 157
3. Ditto length of symphysis -| o4 | 116 | 5°325) -185 | 4:95 | -125
4, Width of jaw over first or
pointed molar. : -| 21 | 052 | 1:95 | -050 | 2-0 050
5. Length of molars in series 59 | 715 o-4 1387 =| 5:2 "132
6. Distance between Ist or
pointed molar and the 8rd
or Ist of the series : . | 2°95] 075 | 2°575) -065 | 26 066
7. Extreme depth of ramus at the
ascending angle . : . | 4°25] -108 — — | 40 101

This fossil is the remain of a very old animal; the canines
and pointed teeth are worn down to a flat surface, and the
molars which can be partially examined, from the circum-
stance of the two jaws having been fossilized obliquely on each
other, appear to have lost all their enamel excepting the
exterior border. We before stated that, although the cranium
and facial bones were entirely removed, the lines of molars
on both jaws were in position; the animal had, partly from
age and partly from accident, lost the first molar from the
right side of the upper jaw. This tooth must have dropped
out at an earlier period than nature had ever intended, and
the coronal surface of the corresponding tooth in the lower
jaw, for want of the wear which it ought to have had under
the natural course of detrition, is distinctly marked with its
crescentic lines of enamel. The fall of the tooth has further
led to a tendency on the part of the rear teeth to incline for-
wards and partially occupy its place; this inclination had so far

* Refers to Plate xx. yol. xix. of ‘ Asiatic Researches.’—[Ep.]

234 FAUNA ANTIQUA SIVALENSIS.

advanced, as to prevent the posterior cylinder of the tooth in
the lower jaw from undergoing any change, the front and fore-
most cylinder only of this penultimate tooth having undergone
this singular alteration and arrived at this diseased form.

In referring to the table of measurements above given, the
excess in size of the fossil is the most prominent feature, but
the relative dimensions of the different parts, with one excep-
tion, bear a close affinity to those of the existing Camel. In
the jaw of the latter there appears to be a greater depth at
the alveolus of the ultimate molar at the commencement of
the ascending branch, which may possibly result in some degree
from age and the more perfect development of the teeth; but
in other respects the resemblance is striking. The exception
to which we refer is in the second measurement, showing the
breadth between the rami, or ascending branches to the rear,
a difference of some importance, as it involves in the structure
of the cranial portion of the skull an increase of width, and a
greater distance between the articulating or glenoid surfaces
for the condyles of the lower jaw. We are borne out in the
correctness of this inference, by the remains of our second
species of Camel, which fortunately is very perfect in the
cranium, and where the breadth and form of this region in
comparison with that of the existing Camel are very different.
To this, however, we shall refer in its proper place, satisfying
ourselves with the conclusion that the Camelus Siwalensis and
the Camelus antiquus corresponded in this respect, and that the
former differed from the existing Camel in the form and excess
of width of the cranium. In comparing the teeth of the fossil
represented in Plate XVITI. figs. 3, 4, and 5, with those of the
Camelus dromedarius we observe no difference, excepting that
which may have arisen from their difference in age. That of the
fossil must have been considerable, as may be observed in the
wear and flattened surface of the fourth or pointed incisor. °
The intermediate incisive teeth have been slightly disarranged
in the matrix; and it will be observed in the drawing that
one of these teeth has been displaced, and is now situated
above, embedded in the rock. The grinding surfaces of all
the incisors are much worn, and all marks of the lateral nick
completely obliterated. On the opposite side to that repre-
sented in the plate, the canine tooth of the upper jaw is
situated in position with its point downwards, embedded in
the mass of matrix, which, it may be observed, reposes on
the anterior parts of the fossil. This canine is much worn
both on the point and on the anterior side; and its resem-
blance to that of the Camel of the present day is sufficiently
close to make any further comparison unnecessary.

Figs. 6 and 6a (Pl. LXXXVII. F.A.8.) are portions of the

CAMEL. 235

lower jaw of the skull, a fragment of which is represented
in fig. 1 of Plate XVIII. The mass from which these re-
mains were recovered was carefully broken up by ourselves,
and the broken pieces united afterwards. A great portion
of the cranium appears to have been disintegrated and so
mixed up with the matrix as to make all attempts at a
separation ineffectual. The anterior part of the lower jaw
has suffered in this way, but the extreme good fortune of
rescuing that portion represented in the figures above
alluded to, consisting of the rami of both the right and
left side, with the condyles and coronoid processes entire, is
ample compensation for the loss, more especially as the inci-
sive extremity was already in our possession, and we were
only in want of the articulating and coronoid processes to
complete the jaw. The difference in form of the fossil will
be observed on a reference to fig. 5,! which is a representa-
tion of the lower jaw of the existing Camel. In giving the
measurements, we place in juxta-position those of the Hissar
and Suharanpoor Camels before referred to.

Existing

Fossil Existing Hissar Suharanpoor

In. Met. In. Mét. In. Met.

Total depth from top of coronoid
process to lower border of jaw | 8°3 e2ite Ore 233 | 7-9 “20
Ditto from surface of articulating
condyle to lower border of jaw | 5°45 | -1388 | 81 205 | 6-4 162
Ditto from heel to lower border of
jaw 0 : : 6 . | 36 7091 | 6:0 152 | 4:7 “119
Depth from top of coronoid process
to upper margin of articulating

condyle : . * 5 |) 2A |) We" |) eal 028 | 1:5 038
Breadth of condyle on transverse
measurement 5 - B alas "044 | 165 | 042 | 16 041

The fossil ramus has more the form of that of the Ox than
of the Camel; the slenderness of its proportions resembles
that of the Cervide more than of the Camelide, to which it
belongs; and were it not for the heel or step on the posterior
ascending margin, which as the generic mark establishes its
position, we should have been at some loss in recognizing as
the remain of a Camel, a fragment bearing in its external ap-
pearance so strong a resemblance to the Ox, Deer, or Antelope.
Independently of the heel, the Camel now existing is rather
peculiarly formed in this part, in comparison with other
Ruminants. The Buffalo is that which has the nearest
approach to it. In the existing Camel the ascending branch

1 See note, p. 283.—[Ep.]

236 FAUNA ANTIQUA SIVALENSIS.

rises at nearly a right angle from the line of jaw; it has con-
siderable breadth on its antero-posterior dimension, and its
coronoid process is short, straight, and massive. In the fossil
Camel the ascending branch is as oblique as in the Ox; it has
no excess of breadth on its antero-posterior dimension, and
its coronoid process is long, slightly curved back, and slender.
Here are points of difference sufficiently striking, but there is
a still further difference in the form of the condyle, that of
the fossil having a much longer transverse diameter than in
the existing Camel; its proportions are much more slender,
and the depression on its upper margin much deeper. We
may remark, however, that the slenderness of the fossil con-
dyle is only comparative with reference to that of the same
genus now existing, and that it bears no resemblance whatever
to the condyle of either the Buffalo, Ox, or other Ruminants.
The condyles of the two former are much slighter, and the
upper articulating surface much narrower than in the Camel.

On the peculiarity of form above described as appertaining
to the ramus of the lower jaw, we are naturally struck by the
close resemblance it bears to that of the horned Ruminants,
and its marked variation from the same bone belonging to the
Camel of the present day ; and we should be inclined to refer to
the extreme length of the coronoid process as a point tending,
in all probability, to unravel the mystery, were we not struck
with the discrepancies that appear even amongst animals of
the same species, in the length and dimensions of this process.

The length and breadth of the coronoid process appear to
be distinguishing features in all animals where there is a
ereat depth of the temporal fossa, and great elevation of the
zygomatic arch; and also in those animals possessing the
power of great lateral motion of the condyle in the glenoid
cavity. Amongst the former may be included all the Carni-
vora and predatory animals; amongst the latter, the Rumi-
nantia, to which only we shall at present refer, although
there appears to be considerable obscurity as to the changes
that modification of form of this process entails upon the
physical economy of the animal.

The Capride, including Antelopes, appear to have the
coronoid process more developed than the Bos; the Bos more
than the Camel. We observe that this process in the common
Goat, C. hircus, is long and broad, and in the A. Chikarra
long but narrow; in the A. tetracornis it is short. In two
specimens of the male and the female of the NylGhau (A. picta,
Pallas), we observe that the female has an exceedingly long
coronoid process, much curved to the rear; whereas that of
the male is short, straight, and pointed. We could give
additional instances, were it necessary, for a want of any rule

CAMEL. 237

of uniformity; in fact, the inferences of the value of this
process in establishing any peculiarity in the organization of
the masticatory faculties, appear to us, as we before noted, to
be clothed in considerable obscurity.

With these remarks on the osteology of the head, we will
—after the following table, showing the comparative sizes of
the ultimate tooth in the upper jaw of a number of fossils,
compared with that of the Camel now existing—proceed to
the rest of the bony structure, which, from the strong resem-
blance that exists between the fossil and modern types, will
occupy but a small space.

Comparative Dimensions of the Ultimate or Third Permanent Molar
(see Plate XVIII. fig. 5) in the upper jaw of the Fossil and of the

evisting Camel.

Existing

Hissar 1st Fossil 2nd Fossil 3rd Fossil

Dimensions j
In. Met. | In. | Mét. In. Mét. In. | Met.

Length on its antero-pos-

terior measurement. . | 1:75) :044; 1°90) 049} 2:00| -050) 2°00] -050
Breadth on its greatest la-

teral bulge. : - | 1:00) -025, 1°35) -034 a 7032 | 1:20] -030

Of the vertebral column we possess a very perfect atlas,
with part of the axis attached to it. The form is precisely
that of the corresponding vertebra of the existing Camel,
with the exception of an increased development to the ridge
on the lower side, and a consequent increase of depression of
the lateral surfaces in which the arterial foramina are
situated. The shortness of the wings and the conical form
of the atlas correspond with that of the existing Camel.

Of the axis we have only a mutilated specimen; the pos-
terior portion is altogether wanting, but the remainder is
similar to that of the present Camel.

We possess a number of the lower extremities of both the
humerus, and the radius and the ulna—some of them in
-connected joints, others separate ; but we have not been able
to detect, amongst the numerous fragments in our possession,
the connecting pieces by which the bone in its whole length
could be established. We see no difference whatever in
these fragments, and in their articulating ends, from those of
the existing animal. The anchylosis of the radius and ulna
is as complete; the surfaces for the articulation of the
seaphoid and cuneiform bones as flat and unmarked by a
hollow; and the lower extremity of the ulna is as destitute
of an external process for embracing the cuneiform bone of
the carpus, as in the existing Camel.

238 FAUNA ANTIQUA SIVALENSIS.

The carpal bones are equally correspondent in form and
number—the small bone in rear of the os magnum and con-
nected bone having the same marked and spherical-headed
articulation with the scaphoid.

We now come to the metacarpal bone, of which we have a
very perfect specimen attached to the end of the radius and
ulna by the intermediate carpal bones. We see no difference
in form, and no peculiarity requiring remark. We derive
data, however, from this, for establishing the comparative
size to which the Sewalik Camel arrived. Our dimensions are
placed in juxta-position with those of a full-grown and
common-sized Camel of the present day :

Fossil Existing
In. Met. In. Met.
Extreme length of metacarpal bone from its arti-
culation with the carpals to articulation with
the phalanges. - : : : . | 18°85] 479 | 15°85} -403

The extent of non-anchylosis, and the form of the pulleys
for articulation with the phalanges, and the phalanges them-

selves, appear to correspond, and to be equally characteristic |

in both the fossil and the living Camel.

With the posterior extremities we are not so well provided,
and shall content ourselves with a reference to the femur
only, of the lower end of which bone we possess a variety of
specimens. Of the largest of these we annex the dimensions:

Fossil Existing
In. Met. In. Met.
Extreme transverse breadth in condyles. 5 | eee "124 | 4:7 118
Ditto breadth from front to rear - : . | o°4 137 | 5-1 130
Ditto breadth between condyles . . : - | 0°675] -0165| 1:075| :027
|

With the exception of the proximity of the condyles to
each other in the fossil, there is no marked difference; all
the hollows and protuberances on the bone of the existing
animal have their corresponding ones in the fossil. The
femur, in its length, also appears to have had as great a
curve forward as we observe in that of the Camel of the
present day.

This brings us to a conclusion on the comparative differ-
ences between the Camelus Sivalensis and the Camelus drome-
darius now existing. Although the fossil fragments from

iiecinatatinaite

CAMEL. 239

which this comparison has been derived are not either so
perfect or so numerous as we could have wished, they are
still sufficiently so for every purpose of comparison; and in
some cases we have been even struck with the remarkable
perfection of the fossil, considering its soft and, in many
cases, imperfectly indurated quality, added to the intimate
combination with, as well as adhesion to, the matrix, which
consists of a light-coloured clay with a small admixture of
sand.

In recapitulation of our above remarks, therefore, we will
note that, independent of the peculiarities described as
existing in the cranium of the Camelus Sivalensis, upon which
peculiarities we rest its specific character, there must have
been others in its external form. These differences, how-
ever, could not have extended far; its general character must
have borne a close affinity to that of the same animal of the
present day; and although we have proofs of its size having
exceeded our existing Camel in a proportion equalling at least
one-seventh of its height, we are unfortunately ignorant of
the effects that domestication may have caused in the
deterioration or otherwise of the Camelus dromedarius,
especially in a country and amongst a race of people who
pay little attention to its improvement, so long as the natural
increase is sufficient to supply their wants and add to their
comfort. The Camelus dromedarius, from which our com-
parison has been drawn, must not be confounded with the
Camelus Bactrianus, or Camel used by the Arabs.

For the Camel in all its perfection we must seek the shores
of the Caspian, to the hordes and wandering tribes who from
generation to generation have looked upon this animal as
the only means by which they could exist—as the only means
by which communication could be maintained over oceans of
sand and miles of desert. If any care be given to the breed
of the Camel in its domestic state, we should expect to find it
in this quarter ; but among the people of India, who use the
animal merely as a beast of burthen, and carry on the breed
much in the same way as they do with their other domesti-
cated animals, we have no reason to expect any improvement.
In the Government stud we have no doubt that all feasible
means are exerted to improve the breed, or at least to prevent
deterioration, by maintaining a stock from the largest and
finest grown animals. It will be noted that one of the skulls
referred to in this paper is from the stud, and the person to
whom we are indebted for its use as a means of comparison
described it as having belonged to a very large male Camel ;
but here, also, we see no great difference in size, although
there are differences in the greater development of the bones

240 FAUNA ANTIQUA SIVALENSIS.

of the head and face. The constant influx of Camels in the
whole sweep of the Indus and its branches, from Ludiana to
Shikarpur, or even to the Indian Ocean, most undoubtedly
keeps up the supply, but does not add anything to the
improvement of the species. Indeed, we are inclined to con-
sider that the Camel has deteriorated in size from that to
which it attained in its wild and natural character; and
should our inference be correct, the dimensions obtained
from the comparative measurement of the bones of our fossil
species may lead to a very tolerable idea of the size to which
the Camel reached, when unshackled by the trammels of
man, and leading its existence in the wilds of its own native
region.

We regret our inability, from want of specimens, of adding
to this comparative statement the dimensions and peculiarities
of form of the Bactrian or true Camel with two humps,
Camelus Bactrianus of authors. The Camelus dromedarius, or
the Dromedary with one hump, is the animal from which we
have drawn our description. In Stark’s ‘Natural History ’
the former is stated to be the longer of the two; Camelus
Bactrianus being described as ‘about 10 feet long,’ and
Camelus dromedarius as ‘nearly 8 feet long. We are not
aware of the limits upon which the above measurements are
drawn ; but in taking those of a perfect vertebral column,
from the atlas to the last caudal vertebra of the common-
sized Camelus dromedarius, we obtain a measurement of
9 feet 10 inches. Including the head, the total length of the
Camelus dromedarius is 11 feet 4 inches; and this must be
considered as under the full measurement, from the absence
of inter-vertebral cartilages, which connect the vertebre in
the living animal. Stark’s specific character evidently leaves
an impression of a superiority of size in the Camelus Bactri-
anus. We learn from Elphinstone, in his ‘ History of Caubul.,’
that the height of the latter animal is considerably less; that
it is shorter and stouter; well adapted for rocky and hilly
countries; and, from its shortness of limb, less liable to
accident than its tall and slenderly-formed congener.

On the Camels in Afghanistan, the author above-mentioned
remarks : ‘The Dromedary is found in all the plain country,
but most in sandy and dry parts; this is the tall long-legged
animal common in India. The Bactrian Camel (which I
understand is called Uzhree in Toorkee) is much more rare,
and I believe is brought from the Kuzzauk country, beyond
the Jaxartes. He is lower by a third at least than the
other, is very stout, and covered with shaggy black hair,
and has two distinct humps, instead of the one hump as
the Dromedary. The Boghdi Camel, in the south-west of

CAMEL. 241

Khorasan, is shaped like the last mentioned, but is as tall as
the Dromedary. LEHyven this last varies, the Dromedaries of

_ Khorasan being lower and stouter than those of India.’!

Again: ‘Many Dromedaries are bred here, or at least by the
tribes whose residence is partly in Damaun. They are much
darker in colour than the common Camel, have shorter and
stronger limbs, and are far better calculated for work among
hills.’ ?

It would appear from Elphinstone’s remarks that there are
three species of Camel:

Ist. That which has obtained the specific denomination of
Camelus dromedarius; tall, slender, with one hump, and
common to India.

2nd. Camelus Bactrianus, or the Bactrian Camel, which is
one-third less in height than the former one, stout, covered
with black hair, and with two distinct humps.

3rd. The Boghdi Camel, resembling in shape the Bactrian,
but with the height of the Dromedary or Camelus dromedarwus.

The Khorasan Dromedary may be considered as a variety
of the first species, or Camelus dromedarius, with less height
but stouter proportions.

In Griffith’s ‘ Translation of the Régne Animal,’ notice is
drawn to a third species, distinct from the Bactrian and
Arabian Camels, in the possession of Ruguere; of this third
species, however, the characters are not given. Hamilton
Smith, in Griffith’s ‘ Cuvier,’* divides the genus into the two
species, Bactrianus and Dromedarius, considering these as

_the parent stocks from which a number of breeds and

varieties have sprung, ‘all nevertheless depending on the
very trivial distinctions of colour, size, and form ;’ but the
specific characters of these parent stocks differ very materially
from those derived from Elphinstone’s work before alluded
to, most especially with reference to the Camelus Bactrianus,
described by that author as one-third lower in height than
the Arabian. Camel. Hamilton Smith‘ says, in describing the
Camelus Bactrianus: ‘ His height may be considered superior
to the Arabian, and the bulk of his body more considerable.
The large breed of this species attains seven feet and a half
from the top of the hunches to the ground, the legs are pro-
portionately short, and the body long.’ The height of the

_ Arabian Camel, according to the same authority, does not
exceed seven feet. ‘Those of Turkey are the strongest and

the best suited for burthen; those of Arabia and Barbary

- the lightest ; and those of India, where there are breeds for

1 Elphinstone’s ‘ Caubul,’ second edit. 3 Synopsis Mammalia, vol. v. p. 298.
1819, vol. i. p. 280. 4 Vol. iv. p. 48.
2 [bid. vol. ii. p. 72.
VOL. I. R

242 FAUNA ANTIQUA SIVALENSIS.

both purposes constantly supplied by fresh importations from
the north-west, are probably inferior in their class to those
more in the vicinity of their original climate.’

We have been desirous of ascertaining the excess of height
to which the existing Camel arrives, to form a correct
opinion of the proportionate size of our fossil species; and
for this purpose have consulted those authorities from whom
it was most likely to draw accurate information. The accounts
are sufficiently conflicting ; but that of Griffith, as an authority
_ ona point of natural history, may be considered as the best
and the most properly to be depended upon. Assuming,
therefore, that the comparative heights and proportions of
the Camelus Bactrianus and Camelus dromedarius, as shown
by Griffith, are the true ones, and that the latter is of a
smaller size than the Bactrian Camel, we are still borne out
in our conclusions with regard to the excess of dimensions of
the fossil or Camelus Sivalensis, and that this excess applies
generally towards all the species of Camel now existing.

Norruern Doan: July 15, 1836.

APPENDIX TO MEMOIR ON CAMEL.

T.—On tHe Fossiz Remains or CAMELIDE OF THE SEWALIKS. By
Capt. P. T. Cautiey.!

‘But the most interesting discovery was that of a camel, of which
the skull and jaw were found. It is to be observed that no decisive
proof of any of the Camelidw—either Camel, Dromedary, or Lama—had
ever been hitherto found among fossil bones, although Cuvier had
proved certain teeth brought from Siberia to be undoubtedly of this
family, if they were really fossil, which he doubted. This discovery in
India was therefore extremely interesting, as supplying a wanting
genus. But for this very reason, it became the more necessary to
authenticate the position of this supposed camel’s remains the more
clearly, especially as there were abundance of existing camels in the
country, which there could not be in Siberia. The Indian account is
somewhat deficient in this respect, leaving us in doubt whether the
bones, admitted to bear a very close resemblance to the living species,
were found in a stratum, or loose and detached.’—Dissertations on
Subjects of Science connected with Natural Theology. By Henry Lord
Brougham, F.R.S., &e. Vol. i. pp. 213-14. 1839.

It is only within the last few months that the most interesting
volumes from which the above is an extract have reached this remote
part of India. Long as the extract is, however, its introduction may
be permitted, as affording us the opportunity of removing all doubts of
the existence of the camel among the Fossil Fauna of the Sewaliks, by

' Reprinted from the ‘Journ. As, Soc.,’ vol. ix. p. 620. 1840.

ii

res)

CAMEL. 243

a few supplementary remarks, which a re-perusal of the original paper
published in the ‘ Transactions of the Asiatic Society of Bengal,’ with
reference to the paragraph above quoted, renders necessary.

To those who have interested themselves in the discovery of the
fossil remains, which has been made in the Sewaliks, it need hardly be
necessary to allude to the two very distinct states in which the mine-
ralization has taken place: that in which the fossil is impregnated
more or less with iron in the form of a hydrate, and that where the cal-
careous elements of the bone are nearly or entirely unaltered, and the
medullary hollows filled with matrix; the former universally existing
in those remains extracted from the sandstone rock, the latter, from the
subordinate beds or substrata, either consisting of clay, or of an admix-
ture of clay, sand, and shingle. ‘The difference in external appearance
is remarkable, the sandstone fossil bemg to a common observer an
organic substance converted into stone, whereas that which is found in
the clay strata not only conveys an idea of a lesser antiquity, but looks
like a substance merely in a progressive state of petrifaction.

As the beds of clay, &c., are inferior in position to the extensive
sandstone strata, the palm of antiquity rests with the fossils of the clay ;
these very imperfect and half-fossilized looking remains being evidently
of older date than those of the sandstone.

With very few exceptions, the only remains that have been dis-
covered, scattered on the face of the mountains or in the ravines and
water-courses which drain them, are those from the sandstone strata.
Those from the lower beds appear to be of a quality too little indurated
to withstand the effects of weather and exposure. The greater pro-
portion of the latter, amongst which are some of our most interesting
genera, viz. Simia, Anoplotheria, Camelide, &c., were exhumed—re-
moved out of the parent strata in which they were originally embedded.
The remains of Ruminants and Rhinoceroses broyght to light in this way
were singularly striking: numerous crania of both families—in many cases
not having shed their milk-teeth—being found closely and compactly
embedded together, the stratum of rock being a perfect Golgotha, not
of the skeletons of old and worn-out animals, but of those that were cut
off when young or in the prime of their existence.

In the osteology of the camel there are certain distinctive marks
which at once separate it from the true Ruminantia, laying aside the
peculiarities of the cervical vertebre, in the absence of perforations
for the vertebral arteries in their transverse processes, which, with the
atlas excepted, is universal in the family, and separates it not only from
the Ruminants, but from all other existing Mammalia. There are two
very simple points of difference which can never be mistaken by the
most careless observer, the first being the want of anchylosis in the
lower extremities of the metatarsal and metacarpal bones, that of the
camel exhibiting itself in a cleft or separation of the two bones to a
distance of two or three inches from the articulating surface, whilst the
same bones of the Ruminants are perfectly undivided; and, secondly,
the marked distinction existing in the carpal bones of the camel, in
the separation of the scaphoid and cuboid, these two bones being joined
together in the true Ruminantia.

Of these metatarsal and metacarpal bones, we have forwarded speci-
mens both to the British Museum and to the Geological Society of

R 2

244 FAUNA ANTIQUA SIVALENSIS.

London, extracted from the lower beds of the Sewalik strata, as well as
from the sandstone rock. Numerous other specimens of the same
family have also been sent to England, the more perfect remains of
crania being still in our possession, although ultimately intended for
the British Museum.

The most valuable remains of Camelide which have as yet been dis-
covered in these hills, and which were figured in the ‘Transactions of
the Bengal Asiatic Society,’ were dug out in my presence. The stratum —
in which they were found consisted of a sandy clay, inclined at an
angle to the horizon of about 20°; the position being about half a mile
north-east of the village of Moginund, which lies at the foot of the
range, and the elevation about 400 or 500 feet above that village.
These fossils were removed by a working party over whom I was
standing, and taken to my camp immediately afterwards. There can
be no demurrer on their being fossil remains, for even had they not
been exhumed before me, their state of fossilization is a proof of their
not having belonged to the existing family; and the position in which
I found them was such that, laying aside their being a part of an in-
clined stratum of rock, no camel of the present day, at least, could
have reached such an awkward locality, the excavation having taken
place at the head of a deep ravine, terminating in a slip, in a wild pre-
cipitous region, far away from the habitation of man, and far removed
from even the grazing ground of village cattle.

In the paper above referred to, certain specific differences are noted
between the fossil and existing camel, which, a fortiori, establish the
discovery of the animal in the former state. As these appear to have
been overlooked by Lord Brougham, I will, in referring your readers
to the memoir in question, note that the most remarkable points of
dissimilitude were in that portion of the cranium connected with the
lower jaw, the breadth between the articulating or glenoid surfaces
for the condyles of the latter being much greater than that in the
animal now existing—a peculiarity not confined to one solitary spe-
cimen, but common to others, amongst which was a very perfect
cranium of a second species, for which we proposed the name of C.
antiquus, procured from the sandstone strata. With the marked dif-
ference above alluded to, it was natural to expect some modification in
form to the condyles and rami of the lower jaw. In this we were not
disappointed. The obliquity of the ascending branches similar to that
of the ox, their form, and the excess of transverse diameter of the con-
dyle, were points of great difference between the fossil and living
animal, and in total correspondence with the peculiarities of the cranium.
It will be observed that the difference of structure in the skull is by no
means of trifling importance, and, as far as the subject of this paper is
concemed, is evidence that the bones found by us could never have
been the remains of the animals now existing in India.!

That the camel lived at the same time with the Sivatherium, Anoplo-
therium, Simia, Hippopotamus, Rhinoceros, and with the very proto-
type of the Crocodiles and Garials now abounding in the great rivers

1 At the lower extremity of the meta- | separation of the points of articulation
tarsals and metacarpals the cleft appears | is somewhat greater, a remark drawn
to be somewhat less in the fossil than | from an inspection of a great number of
in the existing camel; in the latter the | fossil remains of this part of the animal.

CAMEL. 245

and estuaries of modern India, there can be no doubt, as far as the re-
searches in the Sewalik hills have exhibited proofs.

As a fossil discovery, the camel is of great interest. Its position
with regard to the Pachydermata and Ruminants is a link of a now
broken chain. The Sivatherium was one, and Mr. Owen’s Macrau-
chenia was another, to explain the mystery and add two links to a
broken series. That future discovery will tend still further to prove
the wisdom of design is an inference borne out by every succeeding
step in paleontological research.

Whether the camel has existed in an originally wild state in any
period within the historical era, is a question that has been argued at
considerable length. The animal in a state of domestication is spoken
of during the early period of the Scriptural writings, and by subsequent
authors at all periods of history. It is mentioned by Strabo and Dio-
dorus Siculus as having been found in a wild state in Arabia about
the commencement of the Christian era.

Pallas, who argues on the evidence of the Tartars, that the wild camel
is found in Central Asia, is met by Cuvier in the well-known fact of
the Calmucks being in the habit of giving liberty to all sorts of animals
on religious principles. The natives of Hindostan, who act in the same
way, and are guided by similar motives, have, in their affection for the
cow and ox, given rise to a race of wild cattle perfectly distinct from
those of the forest. In the districts of Akbarpoor and Dostpoor, in the
province of Oude, large herds of black oxen are or were to be found,
in the wild and uncultivated tracts—a fact to which I can bear testi-
mony from my own personal observation, having, in 1821, come in
contact with a very large herd of these beasts, of which we were only
fortunate enough to kill one, their excessive shyness and wildness
preventing us from a near approach at any second opportunity. The
wild horses of Southern America are another proof of the tendency of
animals to congregate in herds, and assume the character of originally
wild animals, although, properly, the offspring of domesticated cattle
set at liberty. The proof, however, after all, is merely in the possi-
bility of domesticated animals being able to return again to a state of
nature, and assume the functions of their primitive designation.

The object of this paper is merely to establish the fact of the camel
having been found in a fossil state in the Sewalik hills, the identifi-
cation being more complete, perhaps, than that of any other of the
numerous genera and species which these hills have made us ac-
quainted with. Judging from the number of the remains of this
family in our collections, the camel could not have existed in great
abundance, and their proportion to the true Ruminants must have been
comparatively small.

Nortuern Doan: Sept. 8, 1840.

Il.—Description sy Dr. Fatconer or Fossiz Remains oF CAMEL IN
Museum or Astatic Society oF BENGAL.

From Sewalik Hills.

No. 595. Camelus Sivalensis.—Fragment of 7th cervical vertebra,
comprising the body, with the superior transverse processes attached ;

246 FAUNA ANTIQUA SIVALENSIS. |

the oblique processes and neural arch broken off. Both articular sur-
faces shown ; the inferior not completely synostosed, showing the animal to
have been adult, but not aged. The body, inferiorly, is more curved than
in the corresponding bone of the dromedary, and the median ridge less
prominent. Also, the length of the body, posteriorly, from the offset of
the transverse process to the border of the articulating cup, is conside-
rably longer than in the dromedary. The upper articulating head is
more globular, and the body proportionally narrower and longer.
Length of body ‘ ¢ : : - 5°5 inches.
Width of ditto behind the transverse process . c Booey ass

The two articulating costal surfaces are seen. Nothing shown as to
the course of the vertebral foramen. From Moginund deposit.

No. 596. Camelus Sivalensis.—Lower jaw, fragment of right side,
containing three well-worn molars in situ, a good deal covered with
matrix. Horizontal ramus broken off on a line with the teeth in front
and behind. From Sewalik hills.

No. 624. Camelus (?)—Upper end of right metacarpal with a wide
articulating head suddenly contracting into the shaft. Section of shaft
nearly triangular. From Moginund deposit.

No. 755. Camelus Sivalensis—Fragment of lower jaw, right side,
containing a single molar; this being the only specimen, besides
No. 596 in the collection. From Moginund clay.

zn tah per Say

SIVATHERIUM GIGANTEUM. 247

XII. SIVATHERIUM GIGANTEUM.

A New Fossi, Ruminant GENUS, FROM THE VALLEY OF
Murxkunpa, IN THE SewaLIik BrRaNcH OF THE SUB-

HimaLAyAN Movunrarns.!

BY HUGH FALCONER, M.D., AND CAPTAIN P. T. CAUTLEY.

Tux fossil which we are about to describe forms a new ac-
cession to extinct zoology. This circumstance alone would
give much interest to it. But, in addition, the large size sur-
passing the rhinoceros, the family of mammalia to which it
belongs, and the forms of structure which it exhibits, render
the Sivatherium one of the most remarkable of the past
tenants of the globe that have hitherto been detected in the
more recent strata.

Of the numerous fossil mammiferous genera discovered
and established by Cuvier, all were confined to the Pachy-
dermata. The species belonging to other families have all
their living representatives on the earth. Among the rumi-
nantia, no remarkable deviation from existing types has
hitherto been discovered, the fossil being closely allied to
living species. The isolated position, however, of the giraffe
and the Camelide made it probable that certain genera have
become extinct which formed the connecting links between
them and the other genera of the family, and further be-
tween the Ruminantia and the Pachydermata. In the Siva-
therium? we have a ruminant of this description connecting

The Sewalik or Sub-

1 This memoir is reprinted from the
‘ Asiatic Researches,’ vol. xix. p. 1. 1836.
Different views of the cranium of the
Sivatherium are given in the two last

published plates of the ‘Fauna Antiqua |
Four other |

Sivalensis’ (xci. and xcii).
plates, not published, illustrate the
remaining parts of the skeleton. The
bones of the cranium, including the

lower jaw, are also figured by Royle in |

his ‘Illustrations of the Botany of the
Himalayahs,’ vol. ii. Plate vi—| Ep. ]

2 We have named the fossil Sivathe-
rium, from Siva, the Hindoo god, and

Onptov, bellua.
Himalayan range of hills is considered
in the Hindoo mythology as the Lutiah
or edge of the roof of Srva’s dwelling in
the Himalayah, and hence they are called
the Siva-ala, or Sib-ala, which by an easy
transition of sound became the Sewalik
of the English. The fossil has been dis-
covered in a tract which may be included
in the Sewalik range, and we have given
the name of Sivatherium to it, to com-
memorate this remarkable formation
so rich in new animals. Another deri-
vation of the name of the hills, as ex-

248 FAUNA ANTIQUA SIVALENSIS.

the family with the Pachydermata, and at the same time
so marked by individual peculiarities as to be without an
analogue in its order.

The fossil remain of the Sivatherium, from which our
description is taken, is a remarkably perfect head. When
discovered it was fortunately so completely enveloped by a
mass of stone, that although it had long been exposed to be
acted upon as a boulder in a water-course, all the more im-
portant parts of structure had been preserved. The block
might have been passed over, but for an edging of the teeth
in relief from it, which gave promise of something additional
concealed. After much labour, the hard crystalline covering
of stone was so successfully removed that the huge head
now stands out with a couple of horns between the orbits,
broken only near their tips, and the nasal bones projected in
a free arch, high above the chaffron. All the molars on both
sides of the jaw are present and singularly perfect. The
only mutilation is at the vertex of the cranium, where the
plain of the occipital meets that of the brow, and at the
muzzle, which is truncated a little way in front of the first
molar. The only parts which are still concealed are a por-
tion of the occipital, the zygomatic fossee on both sides, and
the base of the cranium over the sphenoid bone.

The form of the head is so singular and grotesque that the
first glance at it strikes one with surprise. The prominent
features are :—first, the great size approaching that of the
elephant; second, the immense development and width of
the cranium behind the orbits; third, the two divergent
osseous cores for horns starting out from the brow between
the orbits ; fourth, the form and direction of the nasal bones,
rising with great prominence out of the chaffron, and over-
hanging the external nostrils in a pointed arch; fifth, the
great massiveness, width, and shortness of the face forward
from the orbits ; sixth, the great angle at which the grind-
ing plane of the molars deviates upwards from that of the
base of the skull.

Viewed in profile, the form and direction of the horns and

plained by the Mahant, or high priest at

also called Gangaja, gaja being, in
Deyra, is as follows :—

Sewalik, a corruption of Siva-wala, a

between the Jumna and Ganges, from
having been the residence of Iswara
Srva and his son Ganzs, who, under the
form of an elephant, had charge of the
westerly portion from the village of
Dudhli to the Jumna, which portion is

| Hindoo, an elephant. That portion east-
| ward from Dudhli, or between that yil-
name given to the tract of mountains |

lage and Hurdwar, is called Deodhar,
from its being the especial residence of
Deota, or Iswara Stva; the whole tract
however between the Jumna and Ganges
is called Stva-ala, or the habitation of
Siva: unde der. Sewalik.

DESCRIPTION OF PLATE XIX.
SIVATHERIUM GIGANTEUM.

Cranium of Sivatheriwm gigantewm ; one-sixth of the natural
size ; copied from the drawing, by Mr. Ford, in Plate XCI. of
the Fauna Antiqua Sivalensis. In the British Museum. (See
pages 249 & 537.)

VOL. I.

. ' :
“i ‘
g \

G.H Ford del. JDimkal lth.

Sivathernam giganteum.

Vol. I. Plate 19.

;
\
i ;
if <
i

nab
Soatiaas

W.West amp.

SIVATHERIUM GIGANTEUM. 249

the rise and sweep in the bones of the nose give a character
to the head widely differing from that of any other animal.
The nose looks something like that of the rhinoceros, but
the resemblance is deceptive, and only owing to the muzzle
being truncated. Seen from the front, the head is somewhat
‘wedge-shaped, the greatest width being at the vertex, and
thence gradually compressed towards the muzzle, with con-
traction only at two points behind the orbits and under the
malars. The zygomatic arches are almost concealed and
nowise prominent; the brow is broad and flat, and swelling
latterly into two convexities; the orbits are wide apart, and
have the appearance of being thrown far forward, from the
great production of the frontal upwards. There are no crests
or ridges; the surface of the cranium is smooth, the lines are
in curves, with no angularity. From the vertex to the root
of the nose the plane of the brow is in a straight line with
a slight rise between the horns. The accompanying draw-
ings will at once give a better idea of the form than any
description. (Plates XIX. and XX.)

Now, in detail of individual parts, and to commence with
the most important and characteristic, the teeth :—

There are six molars on either side of the upper jaw. The
third of the series, or last milk molar, has given place to the
corresponding permanent tooth, the detrition of which and of
the last molar is well advanced, and indicates the animal to
have been more than adult.

The teeth are in every respect those of a ruminant, with
some slight individual peculiarities (See Plate XX. fig. 1).

The three posterior or double molars are composed of two
portions or semi-cylinders, each of which encloses, when par-
tially worn down, a double crescent of enamel, the convexity
of which is turned inwards. The last molar, as is normal in
ruminants, has no additional complication, like that in the
corresponding tooth of the lower jaw. The plane of grinding
slopes from the outer margin inwards. The general form is ex-
actly that of an ox or camel, on a large scale. The ridges of
enamel are unequally in relief, and the hollows between them
unequally scooped. Hach semi-cylinder has its outer surface
in horizontal section, formed of three salient knuckles, with
two intermediate sinuses; and its inner surface, of a simple
arch or curve. But there are certain peculiarities by which
the teeth differ from those of other ruminants.

In correspondence with the shortness of jaw, the width of the
teeth is much greater in proportion to the length than is usual
in the family: the width of the third and fourth molars being to
the length as 2°24 and 2:2 to 1°55 and 1°68 inches, respectively ;

250 FAUNA ANTIQUA SIVALENSIS.

and the average width of the whole series being to the length
as 2°13 to 1:76 inches. Their form is less prismatic; the
base of the shaft swelling out into a bulge or collar, from
which the inner surface slopes outward as it rises, so that
the coronal becomes somewhat contracted ; in the third molar

the width at the coronal is 1:93, at the bulge of the shaft 2°24. -

The ridges and hollows on the outer surface descend less upon
the shaft, and disappear upon the bulge. There are no acces-
sory pillars on the furrow of junction at the inner side. The
crescentic plates of enamel have a character which distin-
cuishes them from all known ruminants; the mner crescent,
instead of sweeping in a nearly simple curve, runs zig-zag-
wise in large sinuous flexures, somewhat resembling the form
in Hlasmotherium.

The three double molars differ from each other only in their
relative states of wearing. The antepenultimate being most
worn, has the crescentic plates less curved, more approximate,
and less distinct; the penultimate and last molars are less
worn, and have the markings more distinct.

The three anterior or simple molars have the usual form
which holds in Ruminantia, a single semi-cylinder with but
one pair of crescents. The first one is much worn and
partly mutilated; the second is more entire, having been
a shorter time in use, and finely exhibits the flexuous
curves in the sweep of the enamel of the inner crescent; the
last one has the simple form of the permanent tooth which
replaces the last milk molar, it also shows the wavy form of
the enamel.

Regarding the position of the teeth in the jaw, the last four
molars, viz. the three permanent and the last of replacement,
run in a straight line, and on the opposite sides are parallel
and equi-distant; the two anterior ones are suddenly directed
inwards, so as to be a good deal approximated. If the two
first molars were not thus inflected, the opposite lines of teeth
would form exactly two sides of a square, the length of the
line of teeth and the intervals between the outer surfaces of
the four last molars being almost equal, viz. 9°8 and 9:9
inches respectively.

The plane of detrition of the whole series of molars from
rear to front is not horizontal, but in a slight curve, amd
directed upwards at a considerable angle with the base of the
skull, so that when the head is placed so as to rest upon the
occipital condyles and the last molars, a plane through these
points is cut by a chord along the curve of detrition of the
whole series of molars at an angle of about 45 degrees. This
is one of the marked characters about the head.

a gt

DESCRIPTION OF PLATE XX.
SIVATHERIUM GIGANTEUM.

Two other views of the cranium of Sivatherium gigantewm,
represented in Plate XIX.; one-sixth of the natural size.
Fig. 1 shows the palate with the molar series on either side ;
fig. 2 shows well the offsets of the anterior and posterior
horns. (See pages 249 & 537.)

VOL. I.

to. a

Vol.I. Plate 20.

GH. Ford del. J Dinkel ith. : W.West imp .

Sivatherium giganteum.

SIVATILERIUM GIGANTEUM. 251

Dimensions of the Teeth.

Length Breadth

Inches Inches
Last molar right side . ‘ 0 : ; : : — 2°35
Penultimate ditto ; ! A A 5 : ; 2°20 2°38
Antepenultimate ditto. i : ‘ i é S 1:68 2:20
Last simple molar. F : : ‘ A : 1°55 2°24
Second ditto ‘ j : F ; ; ; 5 1:70 1:95
First ditto . ‘ 5 A 3 A : : : 1:70 1:90
Outer Inner

surfaces surfaces
Interval between the surfaces of last molar 9-9 55
Ditto ditto ditto third molar 9°8 i)
Ditto ditto ditto second molar 8:4 45
Ditto ditto ditto first molar 64 2)

Space occupied by the line of molars, 9°8 inches.

Bones of the Head and Face.—From the age of the animal
to which the head had belonged, the bones had become an-
chylosed at their commissures, so that every trace of suture
has disappeared, and their limits and connections are not
distinguishable.

The frontal is broad and flat, and slightly concave at its
upper half. It expands laterally into two considerable swel-
lings at the vertex, and sweeps down to join the temporals in
an ample curve, and with no angularity. It becomes nar-
rower forwards to behind the orbits, and then expands again
in sending off an apophysis to join with the malar bone and
complete the posterior circuit of the orbit. The width of the
bone where narrowest, behind the orbit, is very great, being
16:2 inches. Partly between and partly to the rear of the
orbits there arise, by a broad base passing insensibly into the
frontal, two short thick conical processes. They taper rapidly
to a point, a little way below which they are mutilated in the
fossil. They start so erect from the brow that their axis is
perpendicular to their basement, and they diverge at a con-
siderable angle. From their base upwards they are free from
any rugosities, their surface being smooth and even. They
are evidently the osseous cores of two infra-orbital horns
(Plate XIX. and Plate XX. fig. 2). From their position and
size they form one of the most remarkable features in the
head. ‘The connections of the frontal are nowhere distin-
guishable, no mark of a suture remaining. At the upper end
of the bone the skull is fractured and the structure of the
bone is exposed. The internal and outer plates are seen to

252 FAUNA ANTIQUA SIVALENSIS.

be widely separated, and the interval to be occupied by large
cells, formed by an expansion of the diploe into plates, as in
the elephant. The interval exceeds 24 inches in the occipital.
On the left side of the frontal, the swelling at the vertex has
its upper lamina of bone removed, and the cast of the cells
exhibits a surface of almond-shaped or oblong eminences with
smooth hollows between.

The temporal is greatly concealed by a quantity of the
stony matrix, which has not been removed from the temporal
fossa. No trace of the squamous suture remains to mark its
limits and connection with the frontal. The inferior pro-
cesses of the bone about the auditory foramen have been
destroyed or are concealed by stone. The zygomatic process
is long and runs forward to join the corresponding apophysis
of the jugal bone, with little prominence or convexity. A
line produced along it would pass in front, through the tube-
rosities of the maxillaries, and to the rear along the upper
margin of the occipital condyles. The process is stout and
thick. The temporal fossa is very long and rather shallow.
It does not rise up high on the side of the cranium; it is
over-arched by the vee sides of the frontal bone.
The position and form of the articulating surface with the
lower jaw are concealed by stone, which has not been
removed.

There is nothing in the fossil to enable us to determine
the form and limits of the parietal bones: the cranium being
chiefly mutilated in the region which they occupy. But they
appear to have had the same form and character as in the
ox; to have been intimately united with the occipitals, and to
have joined with the frontal at the upper angle of the skull.

The form and characters of the occipital are very marked.
It occupies a large space, having width proportioned to that
of the frontal, and considerable height. It is expanded late-
rally into two ale, which commence at the upper margin of
the foramen magnum and proceed upwards and outwards.
These alee are smooth, and are hollowed out downwards and
outwards from near the condyles towards the mastoid region
of the temporal. Their inner or axine margins proceed in a
ridge arising from the border of the occipital foramen,
diverging from each other nearly at right angles, and enclose
a large triangular fossa, into which they descend abruptly.
This fossa is chiefly occupied by stone in the fossil, but it
does not appear shallow, and seems a modification of the
same structure as in the elephant. There is no appearance
of an occipital crest or protuberance. The bone is mutilated
at the sides towards the junction with the temporals. Both
here and at its upper fractured margin its structure is seen

SIVATHERIUM GIGANTEUM. 253

to be formed of large cells, with the diploe expanded into plates,
and the outer and inner laminze wide apart. This character
is very marked at its upper margin, where its cells appear to
join on with those of the frontal. The condyles are very
large and fortunately very perfect in the fossil (Plate XX. fig. 1) ;
the longest diameter of each is 4°4 inches, and the distance
measured across the foramen magnum from their outer angles
is 7:4 inches—dimensions exceeding those of the elephant.
Their form is exactly as in the Ruminantia, viz. their outer
surface composed of two convexities meeting at a rounded
angle: one in the line of the long axis, stretching obliquely
backwards from the anterior border of the foramen magnum ;
the other forwards and upwards from the posterior margin,
their line of commissure being in the direction of the transverse
diameter of the foramen. The latter is also of large size, its
antero-posterior diameter being 2°3 inches and the transverse
diameter 2°6 inches. The large dimensions of the foramen and
condyles must entail a corresponding development in the verte-
bre, and modify the form of the neck and anterior extremities.

The sphenoidal bone, and all the parts along the base of
the skull, from the occipital foramen to the palace, are either
removed or so concealed by stone as to give no characters for
description.

'The part of the brow from which the nasal bones com-
mence is not distinguishable. The suture connecting them
with the frontal is completely obliterated, and it is not
seen whether they run up into a sinus in that bone or
how they join on with it. Between the horns there is a rise
in the brow, which sinks again a little forward. A short way
in advance of a line connecting the anterior angles of the
orbits there is another rise in the brow. From this point,
which may be considered their base, the nasal bones com-
mence ascending from the plane of the brow at a considerable
angle. They are broad and well arched at their base and
proceed forward with a convex outline, getting rapidly nar-
rower, to terminate in. a point curved downwards, which
overhangs the external nostrils. For a considerable part of
their length they are joined to the maxillaries; but forwards
from the point where they commence narrowing, their lower
edge is free and separated from the maxillaries by a wide
- sinus, so that viewed in lateral profile their form very much
resembles the upper mandible of a hawk, detached from the
lower. Unluckily in the fossil, the anterior margins of the
maxillaries are mutilated, so that the exact length of the
nasal bone that was free from connection with them cannot
be determined. As the fossil stands, about four inches of the
lower edge of the nasals, measured along the curve, are free.

254 FAUNA ANTIQUA SIVALENSIS.

The same mutilation prevents its being seen how near the
incisives approached the nasals, with which they do not
appear to have been joined. This point is one of great im-
portance, from the structure it implies in the soft parts about
the nose. The height and form of the nasal bones are the
most remarkable feature in the head; viewed from above
they are seen to taper rapidly from a broad base to a sharp
point, and the vertical height of their most convex part
above the brow at their base is 34 inches.

The form of the maxillaries is strongly marked in two
respects: first, in their shortness compared with their great
width and depth; second, in the upward direction of the line
of alveoli from the last molar forwards, giving the appearance
(with the licence of language intended to convey an idea of
resemblance without implying more) as if the face had been
_ pushed upwards to correspond with the rise in the nasals, or
fixed on at an angle with the base of the cranium. The ten-
dency to shortness of the jaw was observed in the dimensions
of the teeth, the molars being compressed and their width
exceeding their length to an extent not usual in the Rumi-
nantia. The width apart between the maxillaries was no-
ticed before, the interval between the outer surfaces of the
alveoli equalling the space in length occupied by the line of
molars. The cheek tuberosities are very large and promi-
nent, their diameter at the base being 2 inches, and the
width of the jaw over them being 12°2 inches, whereas at the
alveoli it is but 9°8 inches. They are situated over the third
and fourth molars; and proceeding up from them towards
the malar there is an indistinct ridge on the bone. The
infra-orbital foramen is of large size, its vertical diameter
being 1:2 inch; it is placed over the first molar as in the ox
and deer tribe. The muzzle portion of the bone is broken off
at about 2°8 inches from the first molar, from the alveolar
margin of which to the surface of the diastema there is an
abrupt sink of 1:7 inch. The muzzle is here contracted to
5:8 inches, and forwards at the truncated part to about 4:1.
The palatine arch is convex from rear to front, and concave
across. No trace of the palatine foramina remains, nor of the
suture with the proper palatine bones. The spheno-pala-
tine apophyses and all back to the foramen magnum! are either
removed or concealed in stone. In front, the mutilation of
the bone, at the muzzle, does not allow to be seen how the
incisive bones were connected with the maxillaries ; but it ap-
pears that they did not reach so high on the maxillaries as the

“1 With the exception of a portion of the basilary region, which resembles that
of the Ruminants.

Pe ee ee eee

SIVATHERIUM GIGANTEUM. 255

union of the latter with the nasals. The same cause has ren-
dered obscure the connections of the maxillaries with the
nasals and the depth and size of the nasal echancrure or
sinus.

The jugal bone is deep, massive, and rather prominent.
Its lower border falls off abruptly in a hollow descending
on the maxillaries ; the upper enters largely into the forma-
tion of the orbit. The posterior orbital process unites with
a corresponding apophysis of the frontal, to complete the
circuit of the orbit behind. The zygomatic apophysis is
stout and thick, and rather flat. No part of the arch, either
in the temporal or jugal portions, is prominent; the interval
between the most salient points being greatly less than the
hind part of the cranium, and slightly less than the width
between the bodies of the jugals.

The extent and form of the lachrymals cannot be made
out, as there is no trace of a suture remaining. Upon the
fossil, the surface of the lachrymal region passes smoothly
into that of the adjoining bones. There is no perforation
of the lower and anterior margin of the orbit by lachrymal
foramina, nor any hollow below it indicating an infra-orbital
or lachrymal sinus. It may be also added, what was omitted
before, that there is no trace of a superciliary foramen upon
the frontal.

The orbits are placed far forwards, in consequence of the
great production of the cranium upwards and the shortness
of the bones of the face. Their position is also rather low,
their centre being about 3°6 inches below the plane of the
brow. From a little injury done in chiselling off the stone,
the form in circle of the different orbits does not exactly
correspond. In the one of the left side, which is the more
perfect, the long axis makes a small angle with that of the
plane of the brow. The antero-posterior diameter is 3:3
inches, and the vertical 2°7 inches. There is no prominence
or inequality in the rim of the orbits, as in the Ruminantia.
The plane of the rim is very oblique; the interval between
the upper or frontal margins of the two orbits being 12:2
inches, and that of the lower or molar margin 16:2 inches.

256 FAUNA ANTIQUA SIVALENSIS.

Dimensions of the Skull of the Sivatherium giganteum.' f
fas Métres
From the anterior margin of the foramen magnum to the alveo-
lus of first molar : : . | 18°85} -478
From ditto to the truncated extremity of the feeds - 2 . | 20°6 | °5268
From ditto to the posterior margin of the last molar . 10°3 | °262
From the tip of the nasals to the cere fractured margin of the
cranium. 3 > F . | 18:0 | 4668
From ditto ditto to ditto along the curve 19-0 | 4822
From ditto ditto along the curve to where the nasal arch begins
to rise from the brow . : PE ANB 188
From the latter point to the fractured margin of the cranium , | 11-2 | -284
From the tip of the nasals to a chord across tle tips of the
horns. : ‘ wip SOPs -2L0
From the anterior angle, right orbit, to the first molar 9-9 | -261
From the posterior ditto ditto to the fractured margin of the
cranium. 1271 | 3075
Width of cranium at the vertex (mutilation at left side restored)
about . : - : . : . | 22°0 | *659
Ditto between the orbits, upper - borders : 5 : - . | 12°2 | -3095
Ditto ditto, lower borders . ‘ . | 16-2 | -4108
Ditto behind the orbits at the contraction of the frontal : . | 146 | 3705
Ditto between the middle of the zygomatic arches. : . | 16-4 | 4168
Width between the bodies of the malar bones . 16°62 | 422
Ditto base of the skull behind the mastoid processes (amutilated
on both sides) . . | 19°5 | 496
Ditto between the check tuberosities of the maxillaries : 12°2 | *38095
Ditto of muzzle portion of the maxillaries in front of the first
molar . ; : Pap ests (es)
Ditto of ditto where truncated (partly restored). ; . |e 2 Hee
Ditto between the outer surfaces of the horns at their base «| L236 32
Ditto ditto ditto fractured tips of ditto . : . | 13°65 | °347
Perpendicular from a chord across tips of ditto to the brow .| 4:2 | +165
Depth from the convexity of the occipital condyles to middle of
frontal behindthe horns. 11°9 | °302

Ditto from the body of the sphenoidal to ditto between the horns | 9:94] -252
Ditto from middle of the palate between the third and fourth
molars to ditto at root of the nasals_. 7:52 | °192
Ditto from posterior surface, last molar, to extremity of thenasals | 130 | -331 |
Ditto from grinding surface penultimate molar to root of the

nasals. - 10°3 | -262
Ditto from the convexity n near the tip of the nasals to the pala-

tine surface in front of the first molar - 5°53 | 14
Depth from middle of the ala of the occipital to the swell at

vertex of frontal . : 8-98 | -228
Ditto from inferior margin of the orbit to erinding ‘surface 5th

molar . 73 | °186
Ditto from the crinding surface, 1st molar, to edge of the palate -

in front of it : . | 26 | -066
Space from the anterior angle of orbit to tip of the nasals - | 10-2 | -2695
Antero-posterior diameter left orbit . A : : c .| 33 | °084
Vertical ditto ditto. ‘ - ‘ A 2°7 | -0685
Antero-posterior diameter of the foramen magnum : : . | 23 | 058
Transverse ditto ditto. : : .| 2°6 | -066
Long diameter of each condyle - : 2 c 5 é -| 44 | :112 ;
Short or transverse ditto of ditto. 2-4 | -0603 .
Interval between the external angles of ditto measured across :

the foramen : a ’ : ‘ : : ; ; 74 | *188

1 To facilitate comparison with the | ‘ Ossemens Fossiles,’ the dimensions are
large animals described in Cuvier’s | also given in French measure.

=n

SIVATHERIUM GIGANTEUM. 257

Among a quantity of bones collected in the neighbourhood
of the spot in which the skull was found there is a fragment
of the lower jaw of a very large ruminant, which we have
no doubt belonged to the Sivatherium; and it is even not
improbable that it came from the same individual with the
head described. It consists of the hind portion of the right
jaw, broken off at the anterior third of the last molar. The
coronoid apophysis, the condyle, with the corresponding part
of the ramus, and a portion of the angle, are also removed.
The two posterior thirds only of the last molar remain;
the grinding surface is partly mutilated, but is sufficiently
distinct to show the crescentic plates of enamel and proves
that the tooth belonged to a ruminant. The outline of the
jaw in vertical section is a compressed ellipse, and the
outer surface is more convex than the inner. The bone thins
off, on the inner side towards the angle of the jaw, into a
large and well-marked muscular hollow; and running up
from the latter, upon the ramus towards the foramen of the
artery, there is a well defined furrow, as in the Ruminantia.
The surface of the tooth is covered with very small rugo-
sities and strieze, as in the upper molars of the head. It had
been composed of three semi-cylinders, as is normal in the
family, and the advanced state of its wearing proves the ani-
mal from which it proceeded to have been more than adult.

The form and relative proportions of the jaw agree very
closely with those of the corresponding parts of a Buffalo.
The dimensions compared with those of the Buffalo and Camel
are thus :—

Sivatherium Buffalo Camel
Inches Inches Inches
Depth of the jaw from the alveolus of last
molar 5 3 é F : y 4:95 2°65 2°70
Greatest thickness of ditto 5 6 4 2°3 1:05 1-4
Width of middle of last molar . . ; 1°35 0°64 0:76
Length of posterior 2nd and 8rd of ditto . 2°15 0:95 115

No known ruminant, fossil or existing, has a jaw of such
a large size; the average dimensions above given being more
than double those of a Buffalo, which measured in length of

_ head 19-2 inches (‘489 métres), and exceeding those of the

corresponding parts of the Rhinoceros. We have, therefore,
no hesitation in referring the fragment to the Sivatherium
giganteum. (See Plate XXII. fig. 1.)
The above comprises all that we know regarding the
osteology of the head from an actual examination of the
VOL. I. s

258 FAUNA ANTIQUA SIVALENSIS.

parts. We have not been so fortunate hitherto as to meet
with any other remain, comprising the anterior part of the
muzzle either of the upper or lower jaw.' We shall now
proceed to deduce the form of the deficient parts, and the
structure of the head generally, to the extent that may
be legitimately inferred from the data of which we are in
possession.

Notwithstanding the singularly perfect condition of the
head, for an organic remain of such enormous size, we can-
not but regret the mutilation at the muzzle and vertex, as it
throws a doubt upon some very interesting points of structure
in the Sivatherium: 1st, the presence or absence of incisive
and canine teeth in the upper jaw, and their number and
character if present; 2nd, the number and extent of the
bones which enter into the basis of the external nostrils;
and 3rd, the presence or absence of two horns on the vertex,
besides the two infra- orbital ones.

Regarding the first poimt, we have nothing sufficient to
guide us with certainty to a conclusion, as there are rumi-
nants both with and without incisives and canines in the
upper jaw; and the Sivatherium differs most materially in
structure from both sections. But there are two conditions
of analogy which render it probable that there were no inci-
sives. First: in all ruminants which have the molars in
a contiguous and normal series, and which have horns on
the brow, there are no incisive teeth. In the Camel and its
congeners, where the anterior molar is unsymmetrical and
separated from the rest of the series by an interval, incisives
are present in the upper jaw. The Sivatherium had horns,
and its molars were in a contiguous series; it is therefore pro-
bable that it had no incisives. Regarding the canines there
is no clue to a conjecture, as there are species in the same
genus of ruminants both with and without them. Second:
the extent and connections of the incisive bones are points
of great interest, from the kind of development which they
imply in the soft parts appended to them.

‘ In a note received from Captain | party of workpeople employed by a gen-
Cautley, while this paper is in the press, | tleman with whom I was unacquainted,

that gentleman mentions the discovery
of a portion of the skeleton of a Siva-
therium in another part of the hills: see
Journal As. Soe., vol. iv. ‘ During my re-
cent {rip to the Sewaliks near the Pinjore
valley, the field of Messrs. Baker and
Durand’s labours, I regretted much my
inability to obtain the dimensions of one
of the most superb fossils I suppose
that ever was found. It was unfortu-
nately discovered and excavated by a

and although I saw the fossil when in
the rock, I was prevented from getting
the measurements afterwards. This
specimen consisted of the femur and
tibia,with the tarsal, metatarsal, and pha-
langes of our Sivatherium” It is much
to be regretted that such an opportunity
should haye been lost of adding to the
information already acquired of this new
and gigantic Ruminant.—[J. Prrysep,
Sec. As, Soc. ]

——

a i a

SIVATHERIUM GIGANTEUM. 259

In most of the horned ruminantia, the incisives run up
by a narrow apophysis along the anterior margins of the
maxillary bones, and join on to a portion of the sides of the
nasals; so that the bony basis of the external nostrils is
formed of but two pairs of bones, the nasals and the inci-
sives. In the Camel, the apophyses of the incisives terminate
upon the maxillaries without reaching the nasals, and there
are three pairs of bones to the external nostrils: the nasals,
maxillaries, and incisives. But neither in the horned rumi-
nants, nor in the Camel and its congeners, do the bones of
the nose rise out of the plane of the brow with any remark-
able degree of saliency, nor are their lower margins free to
any great extent towards the apex. They are long slips of
bone, with nearly parallel edges, running between the upper
borders of the maxillaries, and joined to the ascending pro-
cess of the incisive bone, near their extremity, or connected
only with the maxillaries; but in neither case projecting
so as to form any considerable re-entering angle, or sinus,
with these bones.

In our fossil, the form and connections of the nasal bones
are very different. Instead of running forward in the same
plane with the brow, they rise from it at a rounded angle of
about 130 degrees, an amount of saliency without example
among ruminants, and exceeding what holds in the Rhino-
ceros, Tapir, and Palzotherium, the only herbivorous animals
with this sort of structure. Instead of being in nearly
parallel slips, they are broad and well arched at their base,
and converge rapidly to a sharp tip which is hooked down-
wards, over-arching the external nostrils. Along a consider-
able portion of their length they are unconnected with the
adjoining bones, their lower margins being free and so wide
apart from the maxillaries as to leave a gap or sinus of
considerable length and depth in the bony parietes of the
nostrils. The exact extent to which they are free is un-
luckily not shown in the fossil, as the anterior margin of the
maxillaries is mutilated on both sides, and the connection
with the incisives destroyed. But as the nasal bones shoot
forward beyond the mutilated edge of the maxillaries, this
circumstance, together with their well-defined outline and
symmetry on both sides of the fossil, and their rapid converg-
ence to a point with some convexity, leaves not a doubt that
they were free to a great extent, and unconnected with the
incisives.

Now to determine the conditions of the fleshy parts, which
the structure in the bony parietes of the nostrils entails.

The analogies are to be sought for in the Ruminantia and
Pachydermata.

82

260 FAUNA ANTIQUA SIVALENSIS.

The remarkable saliency of the bones of the nose ‘in the
Sivatherium has no parallel, in known ruminants, to guide
us; and the connection of the nasals with the incisives, or
the reverse, does not imply any important difference in struc-
ture in the family. In the Bovine section, the Ox and the
Buffalo have the nasals and incisives connected; whereas
they are separate in the Yak,' and Aurochs. In the Camel
they are also separate, and this animal has greater mobility
in the upper lip than is found in other ruminants.

In the Pachydermata both these conditions of structure
are present and wanting in different genera; and their pre-
sence or absence is accompanied with very important differ-
ences in the form of the corresponding soft parts. It is,
therefore, in this family that we are to look for an explana-
tion of what is found in the Sivatherium.

In the Elephant and Mastodon, the Tapir, Rhinoceros, and
Paleeotherium, there are three pairs of bones to the external
nostrils, viz. : the nasals, the maxillaries, and incisives.?— In
all these animals the upper lip is highly developed, so as to be
prehensile as in the Rhinoceros, or extended into a trunk as
in the Elephantand Tapir; the amount of development being
accompanied with corresponding difference in the position
and form of the nasal bones. In the Rhinoceros they are
long and thick, extending to the point of the muzzle, and of
ereat strength to support the horns of the animal; and the
upper lip is broad, thick and very mobile, but little elongated.
In the Elephant they are very short, and the incisives enor-
mously developed for the insertion of the tusks, and the trunk
is of great length. In the Tapir they are short and free ex-
cept at the base, and projected high above the maxillaries ;
and the structure is accompanied by a well-developed trunk.
In the other Pachydermatous genera there are but two pairs
of bones to the external nostrils, the nasals and incisives ;
the latter running up so as to join on with the former; and
the nasals, instead of being short and salient, with a sinus
laterally between them and the maxillaries, are long and run
forward united to the maxillaries, more or less resembling the
nearly parallel slips of the Ruminantia. Of these genera
the horse has the upper lip endowed with considerable mo-
bility ; and the lower end of the nasals is at the same time
free to a small extent. In all the other genera there is
nothing resembling a prehensile organ in the upper lip.

In the Sivatherium the same kind of structure holds as is
found in the Pachydermata with trunks. Of these it most

1 Cuvier, Ossemens Fossiles, tome iv. 2 Cuvier, Ossemens Fossiles, tome
p. 131. ni. p. 29.

SIVATHERIUM GIGANTEUM. 261

nearly resembles the Tapir. It differs chiefly in the bones of
the nose being larger and more salient from the chaffron, and
in there being less width and depth to the naso-maxillary
sinus than the Tapir exhibits. But as the essential points
of structure are alike in both, there is no doubt that the Siva-
therium was invested with a trunk like the Tapir.

This conclusion is further borne out by other analogies,
although more indirect than that afforded by the nasal bones.

Ist.—The large size of the infra-orbital foramen. In
the fossil the exact dimensions are indistinct, from the margin
having been injured in the chiselling off of the matrix of stone;
the vertical diameter we make out to be 1:2 inch, which, per-
haps, may be somewhat greater than the truth, but anything
approaching this size would indicate a large nerve for trans-
mission, and a highly developed condition of the upper lip.

2nd.—The external plate of the bones of the cranium is
widely separated from the inner, by an expansion of the
diploe into vertical plates, forming large cells, as in the cra-
nium of the Elephant, and the occipital is expanded laterally
into alee, with a considerable hollow between, as in the Hle-
phant. Both these conditions are modifications of structure,
adapted for supplying an extensive surface for muscular at-
tachment, and imply a thick fleshy neck, with limited range
of motion; and, in more remote sequence, go to prove the
necessity of a trunk.

3rd.—The very large size of the occipital condyles, which
are greater, both in proportion and in actual measurement,
than those of the Elephant, the interval between their outer
angles, taken across the occipital foramen, being 7:4 inches.
The atlas and the rest of the series of cervical vertebrae must
have been of proportionate diameter to receive and sustain
the condyles, and surrounded by a large mass of flesh. Both
these circumstances would tend greatly to limit the range
of motion of the head and neck. But to suit the herbivorous
habits of the animal, it must have had some other mode of
reaching its food, or the vertebree must have been elongated
in a ratio to their diameter, sufficient to admit of free motion
to the neck. In the latter case the neck must have been of
ereat length, and to support it and the load of muscles about
it, an immense development would be required in the spinal
apophysis of the dorsal vertebrae, and in the whole anterior
extremity, with an unwieldy form of the body generally. It
is, therefore, more probable that the vertebree were condensed
as in the Elephant, and the neck short and thick, admitting
of limited motion to the head, circumstances indirectly cor-
roborating the existence of a trunk.

A4th.—The face is short, broad, and massive, to an extent

262 FAUNA ANTIQUA SIVALENSIS.

not found in the Ruminantia, but somewhat resembling that
of the Elephant, and suitable for the attachment of a trunk.

Next with regard to the horns :—

There can be no doubt that the two thick short and conical
processes between the orbits were the cores of horns, resem-
bling those of the Bovine and Antelopine sections of the Ru-
minantia. They are smooth, and run evenly into the brow
without any burr. The horny sheaths which they bore must
have been straight, thick, and not much elongated. None of
the bicorned Ruminantia have horns placed in the same way,
exactly between and over the orbits ; they have them more or
less to the rear. The only ruminant which has horns similar
in position is the four-horned Antelope! of Hindostan, which
differs only in having its anterior pair of horns a little more
in advance of the orbits than occurs in the Sivatherium.
The correspondence of the two at once suggests the question,
‘Had the Sivatherium also two additional horns on the vertex?’
The cranium in the fossil is mutilated across at the vertex,
so as to deprive us of direct evidence on the point, but the
following reasons render the supposition at least probable :—

1st.—As above stated in the bi-cavicorned Ruminantia,
the osseous cores are placed more orless to the rear of the orbits.

2nd.—In such known species as have four horns, the
supplementary pair is between the orbits, and the normal
pair well back upon the frontal.

3rd.—In the Bovine section of Ruminantia the frontal
is contracted behind the orbits, and upwards from the con-
traction it is expanded again into two swellings at the lateral
angles of the vertex, which run into the bases of the osseous
cores of the horns. This conformation does not exist in such
of the Ruminantia as want horns or as have them approxi-
mated on the brow. It is present in the Sivatherium.

On either supposition the infra-orbital horns are a re-
markable feature in the fossil, and if they were a solitary pair
on the head, the structure, from their position, would perhaps
be more singular than if there had been two additional horns
behind.?

Now to estimate the length of the deficient portion of the
muzzle, and the entire length of the head :—

In most of the Ruminantia, where the molars are in a con-
tiguous uninterrupted series the interval from the first molar
to the anterior border of the incisive bones is nearly equal to
the space occupied by the molars, in some greater, in some a
little less, and generally the latter. In other Ruminantia,
such as the Camelide, where the anterior molars are unsym-

1 The Tetracerus or Antilope quadri- ? See Appendix, No. I. and Plate xxi,
cornis and Chikarra of authors. —[Eb.]

ee

SIVATHERIUM GIGANTEUM. 263

metrical with the others, and separated from them by being
placed in the middle of the diasteme, this ratio does not hold,
the space from the first molar to the margin of the incisives
being less than the line of molars. In the Sivatherium the
molars are in a contiguous series, and if on this analogy
we deduce the length of the muzzle we get nearly ten inches
for the space from the first molar to the point of the incisives,
and 28°85 inches for the whole length of the head from the
border of the occipital foramen to the margin of the incisives ;
' these dimensions may be a little excessive, but we believe
them not to be far out, as the muzzle would still be short for
the width of the face in a ruminant.

The orbits next come to be considered. The size and posi-
tion of the eye form a distinguishing feature between the
Ruminantia and the Pachydermata. In the former it is large
and full, in the latter smaller and sunken, and the expression
of the face is more heavy in consequence. In the Sivathe-
rium the orbit is considerably smaller in proportion to the
size of the head than in existing ruminants. It is also
placed more forward in the face, and lower under the level of
the brow. The rim is not raised and prominent as in the
Ruminantia and the plane of it is oblique, the interval be-
tween the orbits at their upper margin being 12:2 inches, and
at the lower 16-2 inches. The longitudinal diameter exceeds
the vertical in the ratio of five to four nearly, the long axis
being nearly in a line from the naso-maxillary sinus across
the limb of the zygomatic arch. From the above we infer
that the eye was smaller and less prominent than in existing
ruminants; and that the expression of the face was heavier
and more ignoble, although less so than in the Pachydermata,
excepting the horse; also that the direction of vision was
-considerably forwards as well as lateral, and that it was cut
off towards the rear.

This closes what we have been led to infer regarding the
organs of the head. With respect to the rest of the skeleton
we have nothing to offer, as we are not at present possessed
of any other remains which we can with certainty refer to
the Sivatherium.' Among a quantity of bones? collected
from the same neighbourhood with the fossil head, there are
three singularly perfect specimens of the lower portions of the
extremities of a large ruminant, belonging to three legs

' See Appendix, Nos. If. and Tii— | racters we are inclined to suspect that it
{Ep. | is allied to some other gigantic species
2 We note here a very perfect cervical | of Ruminant, of the existence of which
vertebra of a ruminant in our posses- | we have already a tolerable certainty.
sion, which must have belonged to an} Of the existence of the Elk, and a
animal of proportions equal to that of | species of Camelide, Lieut. Baker, of the
the Sivatherium ; but from certain cha- | Engineers, has shown us ample proof.

264 FAUNA ANTIQUA SIVALENSIS.

of one individual. They greatly exceed the size of any
known ruminant, and, excepting the Sivatheriwm gigan-
tewm, there is no other ascertained animal of the order
in our collection, of proportionate size to them. We
forbear from further noticing them at present, as they appear
small in comparison for our fossil; and, besides, there are in-
dications in our collection, in teeth and other remains, of
other large ruminants different from the one we have described.

The forms of the vertebrae, and more especially of the carpi
and tarsi, are points of great interest to be ascertained ; as
we may expect modifications of the usual type adapted to the
large size of the animal. From its bulk and armed head,
few animals could be strong enough to contend with it, and
we may expect that its extremities were constructed more to
give support than for rapidity of motion. But, in the rich
harvest which we still hope to reap in the valleys of the
Murkunda, it is probable thatspecimens to illustrate the greater
part of the osteology of the Sivatherium will at no very
distant period be found.

The structure of the teeth suggests an idea regarding the
peculiarities of the herbivorous habits of the animal. In the
description it was noticed that the inner central plate of
enamel ran in a flexuous sweep, somewhat resembling what
is seen in the Elasmotherium, an arrangement evidently in-
tended to increase the grinding power of the teeth. It may
hence be inferred that the food of the Sivatherium was less
herbaceous than that of existing horned ruminants, and de-
rived from leaves and twigs ; or that, as in the horse, the food
was more completely masticated, the digestive organs less com-
plicated, the body less bulky, and the necessity of regurgitation
from the stomach less marked than in the present Ruminantia.

The following dimensions, contrasted with those of the
Elephant and Rhinoceros, will afford a tolerably accurate idea
of the size of the Sivatherium. They are characteristic
although not numerous :—

pan
“| - 1-horne
Elephant Sivatherium Rikinoceros
af Inches Inches Inches

From margin of foramen magnum to the

first molar. a : 2 5 . 23°10 18°85 24:9
Greatest width of the cranium. 7 x 26:0 22-0 12°05
Ditto ditto of face between the malar

bones. 5 ‘ i : 5 , 18:5 16°62 9°20
Greatest depth of the skull . 5 4 17°80 11:9 11:05
Long diameter of the foramen magnum . 2°55 2°6 2°6
Short ditto ditto ditto , 2-4 2°3 15
Average of the above és ‘ 4 : 15:06 12°38 10°22

Sn a ee

SIVATHERIUM GIGANTEUM. 265

If the view which we have taken of the fossil be correct,
the Sivatherium was a very remarkable animal, and it fills
up an important blank in the interval between the Ruminantia
and Pachydermata. That it was a ruminant, the teeth and
horns most clearly establish ; and the structure which we have
inferred of the upper lip, the osteology of the face, and the
size and position of the orbit, approximate it to the Pachy-
dermata. The circumstance of anything approaching a pro-
boscis is so abnormal for a ruminant, that at the first view
it might raise a doubt regarding the correctness of the
ordinal position assigned to the fossil; but when we inquire
further, the difficulty ceases.

In the Pachydermata there are genera with a truuk, and
others without a trace of it. This organ is, therefore, not
essential to the constitution of the order, but accidental to
the size of the head, or habits of the animal in certain genera.
Thus, in the Elephant nature has given a short neck to sup-
port the huge head, the enormous tusks, and the large grinding
apparatus of the animal; and by such an arrangement, the
construction of the rest of the frame is saved from the disturb-
ance which a long neck would have entailed. But as the
lever of the head became shortened, some other method of
reaching its food became necessary; and a trunk was appended
to the mouth. We have only to apply analogous conditions
to a ruminant, and a trunk is equally required. In fact, the
Camel exhibits a rudimentary form of this organ, under dif-
ferent circumstances. The upper lip is cleft; each of the
divisions is separately moveable and extensible, so as to be
an excellent organ of touch.

The fossil was discovered near the Murkunda river, in
one of the small valleys which stretch between the Kyarda-dun
and the Valley of Pinjore in the Sewalik or Sub-Himalayan
belt of hills, associated with bones of the fossil Hlephant,
Mastodon, Rhinoceros, Hippopotamus, &c. So far as our
researches yet go, the Sivatherium was not numerous.
Compared with the Mastodon and Hippopotamus (H. Siva-
lensis, Nobis, a new species characterized by having six incisors
in either jaw) it was very rare.

Norruern Doss: September 15, 1835.

266 FAUNA ANTIQUA SIVALENSIS.

APPENDIX.

1.—MS, Note sy Dr. FALconeR on A SPECIMEN DISCOVERED
By Cou. Convin.!

In a preceding article, in describing the head of the Siva-
therium gigantewm, the mutilation of the cranium at the
vertex was noticed as one of the deficiencies in the specimen
from which the description was taken. Although, in conse-
quence, deprived of direct evidence on the point, we were led
to infer (p. 262), partly from analogy and partly from the
form of the portion of the frontal over-arching the temporal
fossa, that the Sivatherium had two horns at the back of the
head, besides the two infra-orbital horns presented by the
specimen—that the animal was, in short, a huge four-horned
ruminant. A superb specimen lately discovered by Col. Colvin
has put the matter at rest and entirely confirmed our infer-
ence. Besides, these rear horns, by their form, constitute
one of the most remarkable characters about the fossil.

The fossil consists of the posterior half ef the head broken
obliquely off behind the orbits. The right anterior horn is
present, the left one has been removed by the oblique direc-
tion of the fracture, which stretches back so as to include
part of the base of the left rear horn. The bases or pedicles
of both the back horns are present, and the right one to some
extent. The occipital is almost entire, and its ridges and de-
pressions are well marked, and the condyles entire.

The frontal, as in the bovine genera, runs back so as to
meet the occipital plane, sloping only a little behind the rear
horns. As in the Bovines, the parietals are not distinguish-
able from the occipitals with which they are anchylosed. The
general plane of the occipital descends vertically as in the
Bovines, from its intersection with the frontals. The condyles
project obliquely outwards from the plane of the occipital
more than in the Bovide, Cervide, or Capride. The occipital
is much broader for its height than in any of these, the one
dimension being nearly double that of the other. The middle
of the plane of the bone is occupied by a deep hollow as de-
scribed at page 252, triangular downwards. The occipital
crest runs in a double arch, as in the Antilope cervicapra,
the commissure of the arcs descending by a peak into the

1 Now for the first time published.—[Ep. ]

i
Chi eet
Th eis
ears

' 2 sre
i
in

Fig. 1.

Fig. 2.

Fig. 3.

Fig. 4.

Fig. 5.

DESCRIPTION OF PLATE XXI.
SIVATHERIUM GIGANTEUM.

Left side of lower jaw with four molars, about one-fifth of the
natural size. Copied from a drawing in Royle’s ‘ Illustrations of
the Botany of the Himalayahs,’ vol. ii., Plate VL, fig. 1d. This
specimen differs from that referred to at page 257, but it was
described and figured by Colonel Colvin in the Journ. As. Soc.,
vol. vi. p. 152. The specimen was found near the sources of
the Sombe river, north of Dadoopoor and east of Nahun, and is
now in the British Museum. Cat. No. 40,667.

This is the important specimen discovered by Col. Colvin, and .
now in the Museum of the University of Edinburgh, which
proved the correctness of Dr. Falconer’s inference as to the
existence of a posterior pair of horns in Sivatherium. The spe-
cimen is drawn one-sixth of the natural size. A description of
it by Dr. Falconer will be found at page 266. (See also pages
268 & 538.)

Is a fragment from the middle of the posterior horn. It was
originally in Sir Proby Cautley’s collection, and is now in the
British Museum. Cat. No. 39,525. It was found to corres-
pend to the posterior horn-core of the specimen represented in
fig. 2. The figure is one-sixth of the natural size, and is copied
from a drawing, by Mr. Ford, in an unpublished Plate of the
Fauna Antiqua Sivalensis. (See pages 268 & 539.)

Restoration by Mr. Dinkel of the skull of Stvatherium gigan-
tewm, showing the anterior and posterior pairs of horns, one-
eighteenth of the natural size.

Fragment of sternum, about one-seventh of the natural size.
The figure is copied from a drawing, by Mr. Dinkel, in an un-
published plate of the Fauna Antiqua Sivalensis. (See pages
270 & 540.)

VOL. I.

-umequed 13 UMTTOYYBATS

TOTAL Tp pag HD

SIVATHERIUM GIGANTEUM. 267

hollow, and the suture, as in the Bovide, appears to have
been lower than the crest.

The mastoid regions on both sides are of great width, equal-
ling that of the space across the two condyles. The frontal,
as before mentioned, is contracted between the rear and front
horns, and is depressed slightly in the interval between them,
from which it swells on either side in a bulge which runs into
the base of the horns. The base of the left rear horn is only
present in part. The right one shoots out laterally and back-
wards, and slightly upwards, as in the Bovide, and from the
rear portion of the frontal, where it meets the plane of the
occipital exactly as in the Bos. It overhangs the temporal
fossa, and the fossa has the same form exactly as in the Bovide.
The pedicle for the first 44 inches of its rise from the swell of
the brow is contracted ; it then expands with a hollow and a
flattish edge both posteriorly and anteriorly, its vertical
diameter being 3°3 while its longitude is 5°68 at the contrac-
tion, but at the expansion it is 7?. The horn now presents
the extraordinary appearance of trifurcating, one main branch
composing the greatest portion of the diameter of the horn,
being given out from the centre and proceeding laterally and
upwards. It has a central hollow core of about 2 inches
diameter and walls of about an inch in thickness. The posterior
branch, which is broken off at its base, is sent off at right
angles to the central one, and nearly in the same plane. It
has no hollow core, but is cancellated throughout, like the
articulating heads of the bones of the extremities. It appears
to have proceeded directly backwards, and to have been much
smaller than the central one. The hollow core of the central
branch is incomplete at its outer and lower side, having no
cincture of bone here, but joins on with an oblong produc-
tion of hollow proceeding downwards and outwards. This
lower branch at its anterior and lower side is surrounded by
a cancellated paries of bone, which is sharp and flattened for-
wards and nearly two inches in thickness. This appears to
have been the centre of a third branch or off-set from the
common pedicle proceeding directly outwards. The anterior
or front horn is broken off close to the base, which is hollow,
and with an appearance of large cells from the parietes of it.

Now what was the character of the horns? Were they
cores of hollow horns as in the Bovide ?—or branched antlers
as in the Cervide ?—or were the front the former, and the rear
the latter? About the front ones there can be no doubt.
They are conical, rise rapidly to a point, are smooth, have
no burr, are hollow at their base, and are formed of large
cells throughout: no ruminant had ever antlered horns
of this sort. They must, therefore, have been cavicorned

268 FAUNA ANTIQUA SIVALENSIS.

cores. Besides, no ruminant with antlers was ever seen with
four bases to the horns. With regard to the rear ones, their
structure is most perplexing, the main branch is hollow as in
the Bovide, they have no burr or appearance of articula-
tion; but at the same time they give undoubted proofs of
having had two branches, the distinct bases of which are seen,
and there is every reason to believe they had a third. No
cavicorned core is known to be branched in this way, after
the manner of the solid antlered horns of the Cervide, but at
the same time they have no burr as all the Cervide have.

They are smooth, they are not solid, as all the Cervide are,

but hollow ; at least, the central and “outer ones are so. The
horns in the Cervide always come off from the forehead much
in advance of the occipital, with long parietals between. In
the Bovide they come off exactly overhanging the occipital:
so do these. In the specimen the plane of the occipital is
exactly as in the Bovide ; there are no distinct parietals, the
frontals run up to the occipital crest, and there give off their
cores. Therefore, both from structure and analogy, the rear
horns of the Sivatherium were at least three-branched, and
at the same time cavicorned! The dimensions of Col. Colvin’s
specimen are as follows :—

Measurements.

Inches
Longitudinal diameter of pedicle of right back horn where contracted 58
Vertical ditto ditto ditto 7 StS
Greatest width mutilated of occiput. ee et
Height of occipital from upper border foramen to top of cores. . 5°25
Breadth of the condyles (somewhat broken or compressed) . 6:3
Interval between the surfaces of temporal fossee behind the pedi
cles of horns. 13-4
Length from the middle of a line drawn across the anterior margin
of the base of front horns . : ; : . 135

[The specimen to which the slions semana refer was ob-
tained by Colonel Colvin from the lower hills below and west
of Nahun. It was figured and briefly described by him in
the ‘Journal of the Asiatic Society’ for February 1837. It
was also figured by Dr. Falconer in an unpublished plate of
the ‘Fauna Antiqua Sivalensis’ (Plate A., fig. 2). In
Captain Cautley’s collection a large flat horn was found,
which corresponded to Colonel Colvin’s specimen: this is
figured i in the same plate (Plate A., fig. 5). Colonel Colvin’s
specimen is now in the Museum of the Edinburgh Uni-
versity; the flat horn is in the British Museum (Cat.
No. 39,524). These specimens are also figured in Plate X XT.
figs. 2 and 3 of this work, where a restoration of the skull
by Mr. Dinkel (fig. 4) is also given, which will explain the
structure and relations of the four horns better than any
description.— ED. |

SIVATHERIUM GIGANTEUM. 269

II.—MS. Notes on Bones oF SIVATHERIUM NOT DESCRIBED
In Memorr.

Atlas.—The form of the atlas, from two fragments which
exist of it, appears to have been very different from that of all
living ruminants. The body of the vertebra and the are
surrounding the medullary canal are not peculiar. The an-
terior articulating surface resembles that of the eland; the
two cavities are continued to the median line, without any
mesial ridge as in the eland. The posterior articulating sur-
face resembles in every respect that of the buffalo. Below,
the body of the vertebra is not distinct, as in the buffalo and
eland, where on each side it forms a right angle, but the body
passes into the transverse apophysis by a slightly inclined
plane, as in the giraffe. Asa result of this, the foramen in
the transverse process is not concealed by the body of the
vertebra. But the remarkable feature of the atlas consists
in the form of the transverse process. In the ruminants
this process is depressed; it is very large posteriorly, and
diminishes gradually in front, where it presents a straight
edge. In the Sivatherium the transverse process is much
less developed. At its middle it is very narrow, but it ex-
pands a little at the two extremities of the bone, in such a
way that the outer part of the process is concave towards the
body. The four angles of the bone form almost a square,
perhaps slightly wider in front than behind, but the extremi-
ties of the transverse processes are too worn to determine this
point with certainty. Length of body, 3-3 inches; minimum
breadth a little behind the middle of the body, 7:8; extreme
breadth of articulating surface with axis, 6-7. The corres-
ponding measurements in buffalo are 2:1; 10°4; and 5.

[The specimens of the atlas referred to are in the British
Museum. Cat. Nos. 39,526 and 39,527. They are both
figured in an unpublished plate of F.A.S. (B. 1 and 2).—Ep.]

Azxis.—In form the axis is that of a ruminant. It resembles
that of the eland and of the ox in the situation of the foramen,
by which the vertebral artery enters the spinal canal anteriorly,
and also in the form of the odontoid process and of the arti-
culating surface of the atlas; below, at the middle of this
surface, there is not the cleft seen in the eland, nor the echan-
crure observed in the buffalo. The spinous process projects a
little more backwards. The posterior articulating surfaces
project less backwards and are more distant from one another
than in the eland. In this respect they resemble those of
buffalo. Length of body with odontoid process, 7:6 inches ;
width of articulating surface with atlas, 5-7.

270 FAUNA ANTIQUA SIVALENSIS.

[This specimen is in British Museum, No. 39,528, and is
figured in an unpublished plate of F.A.8. (B. 3).—Eb.]

Sternum.—The sternum agrees very nearly in form with
that of Bos wrus (British Museum skeleton). It is very deep
and narrow on the upper or pectoral surface and rather broad
on the inferior surface. It also widens behind and has the
same pointed termination. It shows the five posterior hollow
discs for the reception of the cartilages. The anterior keel-
shaped portion is broken off. The B. wrus shows seven discs
for rib cartilages, viz. the six anterior for separate cartilages,
and the last, or seventh, for the confluent cartilages of the
posterior ribs. The sternum of buffalo is very broad and flat,
and so is that of the deer; that of Bos wrus alone resembling
Sivatherium, but from B. wrus, as well as from all other
ruminants, the Sivatherium sternum differs in its complete
ossification.

[The sternum referred to is in British Museum, and is
ficured in an unpublished plate of F.A.S. (C. 1.) See also
Plate XXI. fig. 5.—Ep.]

Scapula.—aA portion of the left scapula most resembles that
of the camel. The spine is very near the anterior margin,
and sufficiently perpendicular to the surface of the scapula,
but the acromion process is less prominent than in the camel.

[See description of F.A.S. Plate C., figs. 2 and 3.—Ep.]

Humerus.—The deltoid crest is as prominent as in the
horse. The tuberosities at the upper end rise less above the
bicipital canal than in most ruminants. The greater tuber-
osity is deeply indented by a furrow through which the second
head of the biceps passes. In the camel the external furrow
or canal which is the deepest is also the narrowest, whereas
in the Sivatherium the external furrow is broader than the
internal. The lower end of the bone is broad, and has three
well-marked ridges, of which the inner is broad and flat. Pos-
teriorly the condyles are prominent and have a deep cavity
between them, rather less narrow than in the camel. There
is no hole init. The tuberosities above the lower articula-
ting surfaces are very prominent. The extreme transverse
diameter of the lower end of a very large specimen is 7°8,
whereas the transverse diameter at the middle of the articu-
lating surfaces is only 6°5.

[See description of F.A.S. Plate C., figs. 4 and 5.—Ep.]

Bone of Forearm.—The forearm presents a sort of transition
from the Ruminants to the Pachyderms. In the latter the
ulna is generally separate from the radius, but in the rumi-
nants the two bones are united throughout their length, and
are only distinctly visible at their extremities. In the Siva-
therium, as in the horse, a furrow runs throughout the whole

SIVATHERIUM GIGANTEUM. 271

length of the bone, and at either end there is a fissure. The
nature of the two bones is observed on the outer surface,
where the ulna forms a prominent ridge along the posterior
and outer angle of the bone. In the lower third, the radius
retires from this ridge, and a longitudinal furrow is observed.
The olecranon projects considerably behind the upper end of
the radius. The latter presents three distinct articular pits
for the corresponding prominences on the lower surface of
the humerus; the innermost is the largest and the middle
one the deepest. The anterior tuberosity below the articu-
lation is very prominent. In the camel the ulna is invisible
at the diaphysis, and the upper surface presents only two pits,
the two outer ones being confluent. The lower end presents
more marked elevations and depressions than in the camel.
The cavity which receives the semilunar bone is much larger,
and assumes an oval form in front. On the contrary, the
depression for the cuneiform bone is smaller than in the
camel. Both the depressions are very deep.

[Formeasurements, see F.A.S. Plate C., figs. 7, 8, & 9.—Ep. ]

The Carpus is extremely like that of the buffalo, presenting
the same number of bones similarly shaped, and with the
same relative dimensions, except that they are a little higher.
The cuneiform also has its posterior surface, not vertical as
in the buffalo, but inclining downwards and backwards. The
articulating surface for the pisiform being on this surface,
inclines in a like manner. The pisiform has not been found.

[See description of Plate C., figs. 9, 10, 11, 12, and 13, in
F.A.S.—EDp.]

Metacarpus.—In proportion as we descend the bones of the
leg, the differences from the camel increase. The cannon
bone does not at all resemble that of the camel and approaches
nearer that of the buffalo. The relative proportion of its
length and breadth is intermediate between these two
animals; in the buffalo the length being to the breadth as
41 to 1, in the Sivatherium as 5 to 1, and in the camel
as 83 to 1. The anterior surface over its upper third pre-
sents a more prominent ridge than in the buffalo. The
posterior surface is more scooped out than in the buffalo.
The nutritive foramen is situated on this surface at exactly
the same spot as in the buffalo, viz. a little above the lower
end. This lower end is also the same as in buffalo.

[See description of Plate C., fig. 15, in F.A.S.—Ep. ]

Femur.—The superior end seen in front resembles that of
the camel; it is flattened and without depression, but the
trochanter rises more upwards than in the camel. From
above, this extremity looks more like the buffalo’s, on account
of the transverse length of the articulating head and the

272 FAUNA ANTIQUA SIVALENSIS.

shortness of the neck. The articulating head is semi-cylin-
drical, instead of being spherical, as in the camel. The
transverse diameter of the head is more than twice as long
as the neck; in the camel they are about equal. The great
trochanter is much larger than in the camel, but not so large
as in the buffalo. At the inferior extremity the inner side
is longer than the other. This end much resembles the
corresponding part of horse. On the articulating surface there
is a small ridge between the lower part of the pulley and the
inter-condyloid fossa, which is not found in the camel. The
oval cavity situated between the external condyle and the ex-
ternal border of the pulley is well circumscribed in the camel ;
but its direction, instead of being exactly transverse, as in
camel, slants from behind forwards and outwards. At the lower
part of the shaft, on the posterior external angle, is a depression
about the same as in the buffalo. This depression does not
exist in the camel, and is much more marked in the horse.

[See description of Plate D., figs. 1, 2, 3, and 4, in F.A.S.
—Ep.]

Tibia.—The two depressions on the upper end are deeper
than in the camel, and the anterior tuberosity rises higher
towards the articulation. The lower end is rectangular, the
transverse diameter being half as large again as the antero-
posterior. The notch on the outer side is small and more in
front than in the camel, the articulating surface for the
fibula is also less developed. The posterior and inner angle
of the shaft near the lower end has a well-marked groove,
about two inches in length; in the camel this furrow is
scarcely visible.

[See description of Plate D., figs. 5, 6, 7, 8,and 9, in F.A.S,
—Ep.

ea te calcaneum is very similar to that of the buffalo.
The heel is more compressed laterally. The superior angle
is more acute. The astragalus is intermediate in form between
that of the camel and that of the buffalo, but much nearer
the latter. The anterior part of its lower surface does not
present the deep transverse depression seen in the camel.
The tarsus has but four separate bones, as is the case with
ruminants, except the camel, the scaphoid being united to the
cuboid. The superior articulating surface of this compound
bone has three cavities, and differs from the buffalo in the
circumstance that the middle depression is not separated by
a well-marked depression, from a fourth small articulating
surface on the vertical surface. The cuneiform has not been

found.

[See description of Plate D., figs. 10, 11, 12, and 13, in -

F.A.S.—Ep.]

~)
\.

Ree tog Yea an,

SIVATHERIUM GIGANTEUM. 273

Metatarsus.—This bone, like the metacarpus, is of inter-
mediate dimensions between the camel and buffalo, but is more
like the latter. In the middle of the anterior end is a deep
and well-defined furrow much more marked than in the
buffalo; the upper part of the posterior surface has also a
much deeper mesial depression than in buffalo. Extreme
length, 16:4 inches; transverse diameter of upper end, 3:9;
of lower end, 4°05; of middle of shaft, 2:1.

Iii.—Descrirtion sy Dr. Fatconer or Fossit Remains or Siva-
THERIUM IN THE Museum or THE AstatTic Society oF BENGAL.

No. 327.—Fragment of lower jaw of Sivatherium giganteum, left
side, of a very young animal, showing one molar embedded in germ in
the alveolus, and part of one in front emerged from the alveolus. These
are probably the last milk molar and the first true. The specimen
includes behind the base of the ascending ramus and coronoid. The
tooth in germ exhibits well the peculiar rugous reticulation of the
enamel, so characteristic of the genus. The section in front shows the
molar pressing down nearly to the inferior border of ramus with a very
limited dental canal. From the Sewalik hills near Nahun, presented
by Conductor Dawe.

No. 328.—Detached molar, lower jaw, with crown worn, covered .
with a little sandstone matrix. From the Sewalik hills. Presented by
Colonel Colvin?

Vo. 329.—Superb specimen of the horn, showing the basal snag and
nearly the whole of the broad leafy expansion, broken off only towards
the apex; the convex surface shows four deep wavy ramified channels
for blood-vessels of very large size and converging near the base; the
horn is convex on one surface and concave on the other. It is
very thick at the inner border, and becomes thinner outwards and °
upwards; the general form of the specimen is an irregular trapezium.
The figure by Lieut. Baker shows the form and characters of the
antler very well. From the Sewalik hills, Ganawur Khal, near the
Haripal branch of the Sombe River. See Journ. As. Soc. iv. 506, Pl.
XLIV. fig. 3, also v. 184, D. 3, 4. Presented by Conductor Dawe.
This antler was found connected by matrix to the cervical vertebra
(fig. 2 of Pl. XLIV. vol. iv. Journ. As. Soc.), and both were described
by Lieut. Baker as belonging to a fossil elk.

No. 330.—Fragment composing half the molar of a Sivatherium.
Presented by Colonel Colvin.

Wo. 331.—Fine specimen of second cervical vertebra nearly entire and
deficient only at the anterior and upper part of the spinous process. The
bone had belonged to an old animal, as the epiphyses are completely
synostosed and the ridges and depressions strongly marked. The vertebra
is remarkable for the very great width of the posterior surface measured
at the extremities of the aliform expansion of the transverse processes,
this dimension considerably exceeding the entire height. The body of
the bone is broad at the upper articulating surface ; the odontoid process

VOL. I. T

274 FAUNA ANTIQUA SIVALENSIS.

is thick, and there is a very rapid contraction immediately below the
vertebral foramen, which is situated high up on either side; the
median ridge along the under side of the body is in great relief and
thick and prominent at its extremity. The form of the body differs from
that of all other ruminants in the great divergence and spread of the
posterior continuation of the inferior transverse processes, which ter-
minate in a thick projecting mass, instead of being inflected downwards
and inwards as in the bovine ruminantia; the spinous process also
differs remarkably, inasmuch as it appears to have been a short thick
mass instead of a leafy expansion, as in most other ruminants. The
inferior articulating surface is partly concealed by matrix, but it appears
to be oblique and deep; the posterior oblique processes are broad and
flat. The length of the superior vertebral lamine is short as compared
with other ruminants. The projecting edge of the odontoid process is
slightly broken off. In the aggregate of its characters, the axis of the
Sivatherium is remarkable for shortness and thickness.

Dimensions.

Inches

Length of vertebra measured from termination of posterior inferior
surface to edge of odontoid. : 3 : c

From ditto to margin of anterior articular surface . : A eatoeil
Width of articulation with atlas. : : : : a . 56
Ditto of constriction of body . : ; 5 : : . 36
Ditto between the posterior terminations of the transverse processes,

partly restored on the right side : c : : ¢ . 95
Vertical diameter of vertebral canal ; a é A é Bo. AD)
Transverse ditto ditto . i : B ‘| 5 F 5 2"

Found with specimen No. 329, and presented by Conductor Dawe,
See Journ. As. Soc. iv. 506, Pl. XLIV. fig. 1.

No. 332.—Specimen of cervical vertebra, probably the fourth,
more mutilated than the axis. The anterior and posterior articu-
lating surfaces of the body present, as also the two posterior
oblique articulating surfaces; the anterior oblique articulating pro-
cess on the right side is partly shown, on the left broken off at its
base, as also the spinous process; on the left side a part of the superior
and inferior transverse processes (Owen) is shown; on the right they
are broken off close to their base. A good deal of matrix remains on
the fossil, concealing some of the parts. The anterior articulating head
of the body is ovate in its outline as in the ox, but more vertical in its
direction. The posterior articulating cup is very deep. The posterior
oblique articulating processes are very broad. Vertebral foramina
large with a short canal. The bone is too much’ mutilated to afford
dimensions, but corresponds in dimensions with the axis, and was found
in the same place adhering to the large antler No. 329. All ‘three
probably belong to the same animal.—Journ. As. Soc. vol. iv. p. 506,
Pl. XLIV. fig. 2.

No. 333.—Sixth cervical vertebra, showing the anterior and posterior
articulating surfaces, the former nearly entire; the bases of the anterior
and posterior oblique processes present, but the articulating surfaces
broken off, as also the spinous and transverse processes. Vertebral
foramen is present, on left side, of enormous size, with an exceedingly
short canal,

.

SIVATHERIUM GIGANTEUM. 275

The articular head of the body is very convex and broad and less
pointed downwards than in the ox; its general direction being also
more vertical. The length of the body is very short for the “other
dimensions.

Dimensions.

ne

Vertical diameter of articular head 7 . : ae)

Transverse ditto ditto . : : : : : i My 2s

Vertical ditto of vertebral foramen . 1

Transyerse ditto ditto i

From near Nahun. This probably « came Meet the same mines as the
three last—Journ. As. Soc. vol. vi. p. 899.

No. 334.—Mutilated specimen of vertebra (either, last cervical or
first dorsal) presenting the greater part of the body with the articular
surfaces entire, but the neural arch and apophyses broken off; less
mutilated on the right side than on the left, and it is there seen that there
was no vertebral foramen: the posterior articulating surface shows,
on its right margin, the smooth depression of a costal articulation: the
anterior articulating surface is very large, globular, and projecting ; the
posterior shows a deep cup, which is ‘oblique in its vertical direction,
indicating that the bone is probably the last cervical. Too much
mutilated to furnish any dimensions. From the Sewalik hills.
Presented by Conductor Dawe.—Journ. As. Soc. vi. 899.

No. 835.—Lower end of humerus left side, showing the whole of the
articular head with only a short portion of the shaft. Shaft continuous
with the inner condyle, the outer condyle projecting outwards with
great obliquity as in the rhinoceros. ‘The outer condyle shows
markedly the projecting ridge bisecting its articulating surface, as is
normal in the ruminants.

Dimensions.
Inches
Width of articular surface at lower border. 3 E Beis)
Ditto greatest width at the upper part of outer condyle F : oe a2o)
Breadth of articular surface inner condyle. . : ; ane)
Ditto outer ditto . : : ‘: : H ‘ ‘ é oe HEE

No. 836.—Fragment comprising the upper half of the anti-brachium
right side, showing a considerable portion of the shaft and the whole of
the articular surface of the radius, which is entire; the olecranon
broken off; a small splint of the ulna adhering to the shaft. The
articular surface of radius shows the depressions for the corresponding
surface of the humerus; the inner is large and nearly circular, inclining
obliquely inwards to a lower level than the outer. The outer side of
the bone is marked by a salient tuberosity. Shaft contracts rapidly
from the articular surface, and, in front, immediately below the outer
articular surface, there is a deep rough pit.

Dimensions.

Inches
Extreme width of articular head . i 3 3 . hie
Transverse diameter of articular surface . 8
Ditto of inner surface . 6 Bi
Ditto of outer ditto i ; : : 23)
Antero-posterior diameter of i inner wurkice i F . ; . 2°85
Ditto of outer ditto “ ¢ ? F : F é 33

T 2

276 FAUNA ANTIQUA SIVALENSIS.

From the Sewalik hills. Presented by Colonel Colvin.—Journ. As.
Soc. vol. v. 182.

No. 837.—Nearly counterpart of No. 336, but of the left side and of
smaller size. It shows the articular surface still more perfectly, and
more especially the part in apposition with olecranon. Reference and
source same as last.

No. 338.—Fine specimen of cervical vertebra, probably the fifth,
showing the whole of the body and the greater portion of the apophyses.
It agrees very closely in general form and dimensions with No. 332,
the various processes being more entire, but more concealed by matrix ;
the body is short, the anterior articular head formsa broad oval, broader
and more abruptly pointed than No. 332, The posterior articular cup
is nearly circular, deep, and with a raised rim. It is very oblique in
the vertical direction.

The superior and inferior transverse processes are separated from
each other by a broad fossa as in Bos. The oblique processes show
broad articular surfaces and the base of the spinous process presents a
stout projection with an oval section. A good deal of sandstone matrix
fills the intervals between the different apophyses, the spinal canal and
the vertebral foramina.

Dimensions.
Inches
Length of body from articular head to aaa border of articulat-
ig cup .
Verte 3 diameter of articular head .
Transverse ditto at base.
Vertical ditto of articulating cup
Transverse ditto ditto : : : ; ‘ : .
From the Sewalik hills. Presented by Conductor Dawe. This
specimen, in conjunction with Nos. 332, 333, and 334, shows the four
last cervical vertebre, the atlas and third being alone wanting in ‘the
collection.

No. 339.—Lower end of humerus, left side, showing both condyles
with a portion of the shaft attached; the inner condyle as in No. 335,
thé dividing ridge upon the articular surface of the outer condyle is very
prominent, and the articular surface is in fine preservation. The
exterior surface of the outer condyle forms an irregular depressed disc.
This specimen is in two pieces.

From the Sewalik hills. Presented by Colonel Colvin. See Journ.
As. Soc. v. 182. This specimen is of smaller size than No. 335, but
the articular surface is better preserved. It agrees in size with the
articular surface of the radius, No. 337, so as nearly to match.

No. 340.—Fine fragment comprising the inferior half of the right anti-
brachium, the fibular portion being broken off, but showing the whole
of the articular surface of the radius, which agrees very closely with
that of the bovine ruminants; the bone is enormously large when com-
pared with the radius of the recent Gour.

Ss
mwweo

Dimensions.

Length of fragment : : : - : - : 2 - 95
Width of articulating head. 5
Antero-posterior diameter of head at the inner saa BCs
Transverse diameter of shaft where broken . : 2. tae a01er
Vertical ditto. 2

SIVATHERIUM GIGANTEUM. 277

The medullary cavity is filled with a solid mass of crystals of
carbonate of lime.

This and the six following specimens, from the Sewalik hills, were
presented by Colonel Colvin.—Journ. As. Soc. v. 182.

No. 341.—Lower end of radius left side, resembling exactly the
former, but still more perfect in its articular surface, which is as com-
plete as in a recent bone.

The dimensions are smaller than in the preceding specimen.

¥ Inches
Width of articulating head. A F c . - A 6) Gril
Height of styloid apophysis of ulna. 13

No. 342.—A similar specimen of same side, alike in every respect,
except that the portion of the shaft is shorter.

No. 343.—Another of the same side, agreeing in every respect,
except that the shaft of the bone is considerably longer, comprising the
lower two-thirds. Dimensions smaller than the two last. Probably
belonged to a female. The shaft of the bone, instead of being broad
and flat, is compressed laterally, so as to form an angle upwards with
considerable depth, showing a somewhat triangular outline in section,

No. 344,—Fragment of lower end of left radius, the outer portion of
the lower articulating surface broken off.

No. 845.—Specimen comprising the inferior third of the right radius,
with the carpal bones and upper part of the metacarpus connected
together by matrix; the two principal bones being bent, so as to be
parallel to each other. The bones are so enveloped by matrix as to
present no distinct character ; the core of the metacarpal is divided in
section by a vertical plate of bone showing its composite nature.

Dimensions.
Inches
Width of articulating head of radius partly broken off . : . do
Ditto of head of metacarpus . 0 : S c . 48

No. 346.—Fine specimen consisting of the left metacarpus perfectly
entire, cleared of matrix. The bone corresponds exactly, in the form of
its articular terminations, with the same bone in the Gour, but is
proportionally shorter, thicker, and broader in its dimensions. The
articular epiphyses are synostosed, indicating the animal to have been
adult, but probably a female, as it is considerably smaller than
No. 345.

Dimensions.
Inches
Length of bone. - : ° : : : : : . 13°
Transverse diameter of carpal articulating surface . . 2 . 43
Antero-posterior ditto inner side. - a é A a . W4
Ditto outer side . F ‘ ; é - A ; 0 a ales}
Transverse diameter of inferior articular surface . : ;. - 43

The interval between the divisions of the articular ends below is
very narrow and the salient ridge between the pulley surfaces very
salient, as in Bos.

No. 347.—Fine specimen comprising the upper half of the tibia,
showing the whole of the articular surface nearly complete ; the form

278 FAUNA ANTIQUA SIVALENSIS.

agrees closely in every respect with that of Bos, the posterior outer
angle being deflected downwards exactly as in Bes: the longitudinal
ridge along the shaft proceeding from the tubercle is thick and massive,
with a deep fossa on its outer side.

Dimensions.
Inches
Length of fragment - 6 : : ° 2 OZ
Transverse diameter of articular head ‘ : c - 68
Antero-posterior diameter of inner articular surface < 3 . 3d
Ditto of outer 38

Purchased at auction Age i head of Blephas Fee

No. 348.—Fragment comprising nearly the whole length of the shaft
of the right tibia with the articular surface below, the upper end broken
off a little below the tubercle; the articular surface corresponds closely
with that of the Gowr, the bone being proportionally much thicker and
shorter.

Dimensions.
Inches
Length of fragment : : : . ; : . 15°5
Transverse diameter of articular head . F 5 A i Seiie
Antero-posterior ditto of outer cup . : 3 . 3d

This and the remaining specimens from the Sawalil: hills. Presented
by Colonel Colvin. —Journ. As. Soc. vol. v. p- 183. ;

No. 349.—Lower end of left femur, presenting the condyles and
rotular pulley with a short portion of the shaft. A good deal of
obliquity of direction of the rotular pulley from within outwards ;
borders of pulley more or less injured; condyles nearly of equal size;
form and depression closely resembling what are seen in Bos.

Inches
Length of fragment : ; : : A : . 85
Transverse diameter of articular head. : : - 2 i
Interval between condyles 14
Length of rotular pulley 37
Antero- -posterior diameter ofi nner condyle 5 5 : - easel
Transverse ditto. : . : ; : . 24
Antero-posterior diameter of outer condyle 3°7
Transverse ditto ditto 21

The form of shaft of bone not srell ett:

No. 350.—Astragalus, right, nearly entire, partly covered with
matrix, but the impressions and articular surfaces sufficient to show
that it resembles in general form the astragalus of Bos, the proportion
being thicker, broader, and shorter.

Inches
Length of bone : 4 A d . , ; f . 48
Width of lower sone 5 . 5 * ‘ : : = ese
Ditto upper ditto . ; - ° C 0 ; : 5 » 34

No. 851.—Compound bone of tarsus, including scaphoid, cuboid, and
two cuneiform forms, of left side.

General form and articular surface like those of the ordinary
ruminants.

Transverse diameter : . : ; 5 A é : . 14:5
Antero-posterior ditto. : é c ; : : , » 4:6

SIVATHERIUM GIGANTEUM. 279

No. 352.—Left calcaneum, apophyses broken off, but presenting the
whole of the articular surfaces as distinctly as in the recent bone.

Inches
Length of fragment : : 9 c ; : : é . O38
Ditto astragalus articulation . é . : ; : ¢ . 33
Height of bone. : : 5 ose

No. 353.—Right caleaneum, presenting the whole of the long apophy-
sis and the base of the articulation with the astragalus; lateral portion
broken off. The bone is short, thick, and stout.

Inches
Length of apophysis to articular surface . : ‘ 4 : 5 OF
Height of ditto . : , : C : c . c . 28

No. 354.—Lower end of right tibia, presenting the whole of the
articular surface with an extraneous piece of bone adhering to the side.

No. 355.—Proximal metacarpal or metatarsal phalanx of outer toe,
left side; a thick and massive bone as compared with the ordinary
ruminants. The proximal articulation very deep; distal articulation
very unequal and oblique.

Inches
Length . : : ; 5 : 0 : : : a1
Height of proximal articular surface. ; ; : : « 25
Width of ditto : : : Hs : . : : C a) 20

No. 864.—Middle phalanx of toe, of large size, but very short.

280 FAUNA ANTIQUA SIVALENSIS.

XIII. THE FOSSIL BOVIDA, CERVIDA, AND
ANTILOPIDZ OF INDIA.

[Tue fossil Bovide, Cervide, and Antilopide of India, have
still to be described.

Dr. Falconer distinguished at least six new species of
Bovide, among the Sewalik fossils, which he designated in a
manuscript synopsis, Hemibos triquitriceras, Amphibos acuti-
cornis, Amphibos elatus, Amphibos antilopinus, Bos Swalensis,
and Bos occipitalis. The two first are figured in unpublished
plates of the ‘Fauna Antiqua Sivalensis’ (Plates H. and L.),
by referring to a description of which, further on in this
volume, the numbers of the specimens in the Catalogue of the
British Museum will be found. Outline sketches of Bos Siva-
lensis are among the unpublished drawings for the Fauna
deposited in the British Museum; and the specimen de-
scribed under No. 562 of the Catalogue of the Museum of
the Asiatic Society of Bengal was probably Bos occipitalis.
There are no means of identifying the two remaining species.
References to the Sewalik Bovide, with figures by Messrs.
Baker and Durand, will also be found in the ‘ Journal of the
Asiatic Society’ for October, 1835, vol. iv. p. 569. Dr.
Falconer also described two species of fossil Bovide from the
valley of the Nerbudda in his Catalogue of the Museum of
the Asiatic Society of Bengal, viz. Bos Paleindicus and Bos
Namadicus. The former is figured in an unpublished plate
of the ‘ Fauna Antiqua Sivalensis’ (G.), in the description of
which the catalogue numbers of the specimens in the British
Museum are given, and from which the figures in Plate XXII.
have been copied. Brief descriptions and figures of the fossil
buffaloes of the Nerbudda valley by Dr. Spilsbury will also
be found in the ‘Journal of the Asiatic Society,’ vol. iii.
- p. 399, vol. vill. p. 952, vol. ix. p. 551, vol. x. p. 626, and
vol. xiii. p. 765.

The Sewalik Collection in the British Museum also con-
tains numerous remains of Cervide. In a manuscript synopsis
of Sewalik fossils by Dr. Falconer, two unnamed species of
Cervus, a third, Cervus Paleindicus and a species of Dorcathe-
rium (moschinum), are entered under the head of Cervide,

DESCRIPTION OF PLATE XXII.

Bos Paua#inpicus AND Bos NAMADICUS.

Fig. 1. Shows the cranium and horns of Bos Paleindicus from the -

Nerbudda, about one-eighth of the natural size. The drawing
is a restoration from several specimens in the British Museum
(Cat. Nos. 389,759 and 39,715). These specimens were ascer-
tained to belong to the same individual, and the restoration has
been executed by Mr. Dinkel in accordance with an outline
drawing found among Dr. Falconer’s papers. The separate
specimens are also figured in an unpublished Plate of the Fauna
Antiqua Sivalensis. (See pages 280, 546, & 554.)

. Lateral view of same skull as shown in fig. 1, one-fifth of the

natural size.

. Section of horn of same specimen, one-fifth of the natural size.

. Fragment of cranium showing forehead, portion of right horn,

and left horn-core of Bos Namadicus, one-tenth of the natural
size. The specimen has been drawn by Mr. Dinkel from the
original in the British Museum (Cat. No. 39,760), and is also
figured in an unpublished Plate of the F. A. 8. (See pages
287 & 545.)

. Profile view of another cranium of Bos Namadicus, one-fifth of

the natural size, drawn by Mr. Dinkel from the original in the
British Museum. Cat. No. 39,758. (See pages 280 & 545.)

i. 6. Section of horn of Bos Namadicus, one-fifth of the natural size.

(See pages 286 & 545.)

VOL, I.

a a aa aa

SsnoTpute eq sog e=issunn Saiineecacameeanee ae mos

tans esteem omer

BOVIDA, CERVIDA, AND ANTILOPIDA. 281

Dr. F. also refers to numerous remains of Cervus in his Cata-
logue of the Museum of the Asiatic Society of Bengal; and a
description with figures of vertebrz and antlers of the ‘fossil
Elk of the Himalayah,’ by Lieut. Baker, but subsequently
referred to Sivatherium by Dr. Falconer (see page 273), will
be found in the ‘Journal of the Asiatic Society of Bengal’
for September, 1835, vol. iv. p. 506. Another fossil species
of Cervus from the valley of the Nerbudda was designated by
Dr. Falconer, Cervus Namadicus, but has not been described.
Dr. Spilsbury, in his account of fossils found in the valley of
the Nerbudda, refers to fragments of upper and lower jaws
and teeth of deer. (See Journ. As. Soc. vol. x. p. 626.)

The Sewalik fossils also included several species of Ante-
~ lopes which were designated by Dr. Falconer A. Paleindica,
A. gyricormis and A. picta, inva manuscript synopsis. Brief
descriptions of some of these Antelopes with figures, by
Messrs. Baker and Durand, will be found in the ‘Journal of
the Asiatic Society,’ vol. iv. p. 569, and vol. xii. p. 769. The
- Antilope Paleindica is also described by Dr. Falconer in the
Catalogue of the Museum of the Asiatic Society of Bengal,
and the specimen thus designated is said to agree closely with
the Sewalik fossil Antelope, described and figured by Capt.
Baker, in the ‘ Journal of the Asiatic Society’ for September,
1843. Capt. Baker’s specimen is now in the British Museum
(No. 39,594), and is represented in Plate XXIII. from original
drawings by Mr. Dinkel. Among the Sewalik specimens in
the British Museum there are also remains of an Antelope
with twisted horns, probably the A. gyricornis of Falconer.
—Ep.|

I.—Description By Dr. Fatconer oF REMAINS OF LARGE Bovine
RUMINANTS FROM THE SEWALIK Hitts, in THE Museum oF THE
Asiatic Society oF BENGAL.

Wo. 562. Bos.!'—Fine skull of a bovine ruminant, nearly perfect,
from the occiput on to the diastema, showing the zygomatic arches,
temporal fosse and the whole of the spheno-palatine region, together
with two lines of molars, in situ; the crowns of those on the left side
broken off; the three posterior molars on the left side nearly entire; a
horn-core is present on the left side, absent on the right through a fracture,
which has carried it off below the base. The cranial part of the skull
differs remarkably from all known bovine ruminants in this respect,
that the occipital bone appears to terminate at the occipital crest, or
close to it, and that no part of the parietals enters into the occipital
plane. The horns are pyriform in section, with a very sharp edge
behind and a broad surface in front ; they are closely approximated on

) Bos occipitalis, Fale. ?—[Ep.]

282 FAUNA ANTIQUA SIVALENSIS.

the brow, and start outwards and upwards, but curve forwards towards
their tip. The plane of the frontal is flat to between the commence-
ment of the horns, and then descends in a sudden curve between the
horn-cores, to meet the plane of the occipital at an obtuse angle. The
occipital crest is very prominent ; orbital rim also prominent, the lachry-
mal bones present rough tuberosities at the orbital margin, as in the
bovine group ; there is also no lachrymal fissure; the two supra-orbital
foramina very large. The nasals are received into an angular fissure
of the frontals, their apices ascending nearly to a line with the anterior
border of the orbit. The orbits differ in a remarkable manner from
those of the ordinary bovine ruminants, in having their greatest
diameter in the vertical direction, instead of antero- -posteriorly. Between
the supra-orbital foramina there is a raised portion of the surface of
the frontal of a horse-shoe shape, about a line in thickness, and with a
rugous fimbriated margin about two inches in breadth, the sinus at the
posterior part passing gradually into the surface of the frontal. This
rugous disc is unknown in other ruminants. The maxillaries contract
abruptly in front of the orbits; their tuberosities are very prominent.

Teeth—The three posterior molars on the left side are well worn,
showing the animal to have been fully adult, and a large accessory
pillar, narrow and compressed, but of considerable depth, is seen in the
sinus between the two barrels of each molar on its inner side, as in the
bovine ruminants. The palate is broad.

Dimensions.

Inches
Teeth from occipital condyles to commencement of diastema . :

Ditto from summit of occipital crests to apex of nasals

Width at posterior margin of orbits

Ditto at anterior ditto

Constriction of maxillaries in front of orbits

Ditto of frontals behind orbits

From base of occipital to summit of occipital crest .

Width of occipital region measured at the pars petrosa

Height from surface of palate to frontal .

Width between outer border of condyles. . : ; :

Distance between occipital crest and base of horn . 5 : 3

Length of temporal fossa

Interval between base of horns :

Height from palate to broken extremity of nasals :

Length of contraction between base of horn and pce margin
of orbit . 3

Vertical diameter of left orbit

Antero-posterior diameter of left orbit

Vertical diameter right orbit .

Antero-posterior ditto :

Length of line of molars (5) right side

Greatest width of palate :

Length of left horn-core fragment .

Antero-posterior diameter at base .

Thickness of ditto at ditto

Antero-posterior diameter near broken tp

Transverse ditto

a
Cas
aD

PITA OTOND AS
On or Co Ore OO ~I

onto

(I oo pel EL
He wo

bo
~

The dimensions of the oe epi aifiee a little j in conecaneees of
crushing on the right side. The principal distinctive marks are: 1.
The occipital not rising above the occipital crest. 2. The great
narrowness of the parietal region between base of horn and occipital
crest. 3. The close approximation of the horn-cores. 4. The short

tit

—

BOVIDA, CERVIDA, AND ANTILOPIDA. 283

interval between the base of the horn and the posterior border of orbit.
5. Projection of orbits above plane of frontal. From Colonel Colvin’s
collection.

No. 571. Bos ?—Cranium of a good sized bovine ruminant, in
the cranial portion nearly entire, together with the maxillaries on to
the diastema; nasal bones wanting. Specimen shows core of left horn
a good deal abraded, the right broken off, but the specimen is so
much enveloped by matrix as to conceal all the principal parts; the
occipital, parietal, and frontal regions resemble 562, of which this may
have been a female.

No. 572. Bos ?—Fine fragment comprising lower jaw, right
side, showing the last five molars in situ. The teeth correspond in
pattern of crown and accessory pillars with the bovine ruminants;
they are well worn, and the posterior talon of the last is broken off.
The compressed accessory pillar between the barrels agree closely with
the corresponding teeth of upper jaw of No. 562, of which species this
was probably the lower jaw.

Tnches
Estimated length of last five molars : F : : 3 . 65
Length of fragment : ; : 5 : : : ; ee On
Height to alveolus, behind . - ; : : ; 5 2 aS
Ditto, in front : j 4 : : a 5 5 les

No. 574. Bos ?—Fragment comprising tiie terminal portion of a
curved and conical horn-core, probably bovine, which shows no indica-
tion of a ridge on its upper edge.

Inches
Length of fragment 5 C 3 : ci . : . 58
Antero- -posterior diameter at base 0 - : e c 4 aaa
Transverse ditto . 4 ; : . : q Meg

No. 575. Bos ?— Portion of ae of a fee horn-core of a bovine
ruminant, presenting a different section from the horn-core of 562.
Inches
Long diameter. : : ‘ ; - : c : . 4:2
Short ditto . ; - : : : : z 2 6 . 38
No. 576. Bos ?—Fraement of horn-core, presenting the charac-
ter of 562, with a sharp upper edge.
No. 577. Bos ?—Fragment of horn-core in two pieces, present-

ing nearly a square section like 575, but of smaller size; fitting a
detached frontal portion of cranium.
Inches
Length of fragment of horn . ; 2 : ‘ . : 5 ea)
Subsequently the frontal portion of this specimen was discovered.
It is the specimen referred to by Mr. E. Blyth as identical, or nearly so,
with Capra Sakeen (Bl.), vide Journal As. Soc. xi. p. 103.

No. 578. Bos ?—Fragment of a horn-core presenting nearly a
square section, with a sharp edge on one side, differing in form from
562.

No. 579. Bos ?— Another portion of horn-core of similar form to
last, with sharp edge on the concave side, belonging to upper part of
core. Section nearly triangular.

No. 580. Bos ?—Part of cranium of a middling-sized ruminant,

284 FAUNA ANTIQUA SIVALENSIS.

showing only occipital and part of sphenoid, too defective for further
determination.

No. 607. Bos ?—Skull nearly complete from the occipital con-
dyles on to the diastema, with two lines of molars, both orbits and the
right zygomatic arch; both horn-cores broken off, and the upper
lamina of the frontal bone denuded, so as to afford no characters in
that part. The skull is nearly of the same size as 574. The molars
are far advanced in wear, and show the compressed accessory pillar
between the barrels, exactly as in 572. Rime of orbits more or less
damaged, but the left shows the great vertical height seen in 572.
From Sewalik hills, sandstone matrix.

Il.—Descrietion By Dr. Fatconer oF Fossiz SprctEs oF Bos
From THE NEeRBUDDA, IN THE Museum oF THE AsIATIC SocIETY
or BENGAL.

No. 18. Bos Paleindicus.—Cranium with the cores of 2 horns
attached, broken off in front across the orbits, where it is much muti-
lated; occiput, right condyle, and left styloids, with the occipito-sphenoid
region and zygomatic fossa on either side nearly entire. Core of right
horn in 4 pieces, extending nearly to the apex. Left core in 3 pieces,
of less extent. Dimensions—Length of right core 2 feet 9 in.; breadth
at base 64 in.; thickness 4j in. Length of left core 1 foot 11 in.;
breadth of skull at base of occiput 13 in.; ditto from occipital condyles
to plane of frontal 10 in. The horn-cores spread out more horizontally,
and with a less inclination upwards than in the existing wild buffalo,
slightly concave anteriorly and convex behind. A cord stretched
between the tips would subtend the plane of the cranium behind the
frontal. In all these respects it differs from the existing wild buffalo,
and so far as the horizontal offset is concerned it approximates to the
Gayal, from which, however, it differs in the flattened form of the
horns and in every other respect. The posterior border of the cores
encroaches much upon the temporal fossa, which is narrow. These
characters are so constant, as shown by still better specimens in the
British Museum, that there can be little doubt that the species is dis-
tinct from the existing wild buffalo. The specimen is in a soft friable
condition, and covered with calcareous matrix ; probably belonged to
an adult male of large size. This fine specimen has remained unfigured
in the Journal. It would appear to have been the intention of Mr.
J. Prinsep to have described and drawn it; this he postponed, ex-
pecting similar specimens from Conductor Dawe. His subsequent
fatal illness prevented this design from ever being carried out (vide
Jour. As. Soc., vi. 489). Found by Dr. Spilsbury, near Sagauni (or
Sejouni).

No.19. Bos Paleindicus——Mutilated cranium broken off across the
orbit with the base of the left core attached, the right core with the
parietals and vertex broken off. Occiput nearly entire with the con-
dyles, occipito-sphenoid region and right styloid and mastoid entire,
together with the left temporal fossa. Width of core at base 65 in.;
thickness 44. Width between the outer marem of occipital condyles

a Y

BOVIDA, CERVIDA, AND ANTILOPIDA. 285

54. Long diameter of occipital condyle 25:75 in.; transverse 1:75.
Foramen magnum, antero-posterior diameter, 1°7 in. ; transverse dia-
meter 1:75 in. This would appear to be an adult male specimen of
same species as No. 18. From Hoshingabad (vide Jour. As. Soc., vol.
viii. 950, figs. 3 and 9).

No. 20. Bos Paleindicus——Specimen consisting of a nearly entire
cranium broken off anteriorly at the diastema, and bearing the cores of
both horns, the right broken off a little above the base, the left 9 inches
long. All the molar teeth on both sides protruded and worn, the occi-
pital portion agrees in form with the large specimen No. 18, of which it
would appear to be an adult female, judging from the less prominent
development generally and the smaller size of the horns. Length of
specimen 174 in.; height from sphenoid to surface of frontal 8 in. ; width
of skull at constriction between the horn-cores and orbits 8} in.; width
of brow between middle of orbits 74 in. Length of brow from vertex
to upper margin of orbit 74 in. ; length of line of molars, left side, 6} in. ;
greatest width of left horn 5 in.; thickness 3 in.

The left horn-core, right condyle, right styloid, left zygomatic arch,
and glenoid surface, palate, occiput, frontal, and both orbits, also the
two maxillaries, are nearly entire. Nasals broken off. The specimen is
encrusted with a hard calcareous gritty matrix. (From Hoshingabad.)
Memo.—This cranium is described by Mr. James Prinsep, and a series
of 8 different dimensions as compared with a skull of a buffalo (not
stated whether wild or tame) prepared by Dr. Evans (Journ. As. Soc.,
vol. iii. p. 399).!

No. 21. Bos Paleindicus.—Specimen comprising left maxilla nearly
complete in height and containing 6 molars protruded from the jaw : all
the rest of the skull wanting. Length of line of molars 6 in.; height
from edge of molars to upper margin of the fragment 93 in. Probably
adult.

No. 22. Bos Palwindicus.—Lower jaw, left side, comprising nearly
the whole length of horizontal ramus, broken off in front near symphy-
sis, and behind at the offset of the ascending ramus in a line with the
molars, containing 6 adult molar teeth, all protruded and worn, but not
of an old animal. The grinding pattern is a little obscured by matrix,
which encrusts the specimen; but it would appear to correspond with
Bos Paleindicus. Length of fragment 13°25 in.; length of line of
molars 74 in.; height of jaw at the last molars 4 in.; at the anterior
edge of the 1st premolar 2} in. Length of diastemal portion 24 in. ;
greatest thickness 1? in.

No. 30. Bos Namadicus.—Enormous core of bovine ruminant, en-
crusted with matrix, much curved in the outline and nearly circular in
section. Girth, 5 inches above base, 1 foot 7 in. Length along convex

1 Respecting this specimen Mr. J. | the forehead, as well as the angle of the
Prinsep writes in Journ. As.Soc., vol. iii. | occiput, both tend to rank it with this
. 399 :—‘ The direction of the horns in | genus; or at least certainly to separate
the Hoshingabad fossil skull give it at | it widely from the Aurochs and the do-
first sight the appearance of a buffalo’s | mestie ox, as described by Baron Cuvier.’
head: and the convexity and breadth of ' —[Ep ]

286 FAUNA ANTIQUA SIVALENSIS.

curve 3 feet 8 in. Diameter where tip is broken off 8 in. This horn-
core differs in size, form, and curvature from any known ruminant. It
is much more circular than that of the Gour or Gayal, in which respect
it differs still more than the Bos Paleindicus, and it would appear to
indicate a distinct fossil species now extinct.

The core is hollow throughout, having a thin shell. Surface encrusted
thickly with matrix, which requires removal. Consists of 2 fragments,
which join; a portion of the tip is wanting.

No. 81. Bos Namadicus.—Fragment of horn-core, left side, detached,
presenting the same characters as No. 30, namely, a curved outline
and nearly circular section; it is less covered with matrix, and at the
upper and convex edge shows a ridge similar to Bos grunniens, which it
resembles also in circular section. The specimen is partly covered
with hard sandy matrix and is harder and stronger than the other
bovine remains above described. Length 11 inches; girth near base
143 in.; at small end of specimen 13 in. (From Jahnsee Ghat on the
Nerbudda.)

No. 32. Bos Paleindicus.—Detached fragment of horn-core with a
slight portion of the frontal attached to its base; the left side; hollow and
in the soft condition of the other Nerbudda fossils.

Length 11} in.; girth near base 193. Long diameter 7 in.; short
diameter 4} in.

No. 33. Bos Namadicus.—Mutilated cranium, presenting the occi-
pital and sphenoidal region nearly entire ; left occipital condyle and
right mastoid complete; occipital and frontal surfaces perfect, from
superior margin of foramen magnum to the commencement of the
nasals ; cores of both horns broken off—on the left side within the base
of the pedicle of the core—and on the right side the fracture includes
the base of the core, the margin of the frontal and orbit, all of which
are removed. A portion of the upper border of left orbit remaining.
The frontal plane longitudinally is slightly convex with a shallow con-
cavity upwards, between the commencement of the horn-cores. The
vertex projects posteriorly to a great extent, so as to overarch the plane
of the occipital condyle 22 inches. Occipital region concave from
above downwards; in these two respects differing very notably from
the skulls of the Gour and Gayal. Judging from the section of the
core on the left side, it was more or less cylindrical in form; in this
respect also differing from these two species.

Length of specimen from vertex to broken margin in front, 13 inches;
height from posterior edge of body of sphenoid to vertex, about 84 inches ;
width of skull at base of occiput, 95 inches; height of occipital surface
from inferior margin of foramen magnum to posterior border of vertex,
7 inches; long diameter of occipital condyle, 22 inches; short ditto,
1-6 inch; width of frontal at constriction behind orbits, 8} inches.
Dr. Spilsbury states that this cranium and the specimen No. 31 were
found in the same place; they aremarked No.1and No. 2. Inconjunction
with the huge horn-core No. 30, they would appear to ‘afford conclu-
sive evidence of a fossil species of Bos from the Nerbudda, distinct
from Bos Paleindicus and Gour and Gayal, or from any other described

BOVIDA, CERVIDA, AND ANTILOPIDA. 287

existing form. From the complete synostosis of the two frontals, it is
inferred that the animal was an aged adult; and the smaller size of the
horn, as compared with specimen No. 30, would seem to indicate that
it was a female.!

The forehead of this specimen is flat and slightly concave above; it
is square, taking the base between the orbit; its height is about equal
to its breadth. The horns are attached to the extremity of the highest
salient line of the head; the plane of the occiput forms an acute angle
with the forehead (it is over- -arched), and the plane of the occiput is
nearly quadrangular, instead of semi-circular, all these being distinc-
tive characters of the Taurine Bovide (ox), as contrasted with the
Bisons and Aurochs. (Vide Cuvier, Ménagerie du Muséum, article
‘Zébu,’ p. 4.)

No. 34. Bos Namadicus.—Specimen of the posterior part of the
cranium, nearly perfect on the left side from the vertex to the middle
of the orbit; mutilated on the right side; the inter-frontal suture, and
that uniting the two lateral occipitals with the superior occipital are
widely open, showing that the animal was very young, which is further
borne out by the dimensions. The left horn-core is attached to the
extent of about 5 inches and is given off horizontally, with a slight
inclination upwards. The occipital condyles, the styloids, and tym-
panic bulle are entire on both sides, together with the auditory
foramina and the whole of the basi-sphenoid region. The right side of
the vertex and the upper part of the frontal, together with the right
orbit, are gone. Part of the left orbit remains. The plane of the
frontal exhibits the same general form as in specimen No. 35, and the
vertex projects posteriorly, overarching the occipital exactly as in that
specimen. Length of fragment from vertex to anterior fracture,
64 inches; width of forehead at constriction behind the orbits, 65
inches; long diameter, 2°6 inches; short diameter, 2'3 inches; ex-
treme distance between the outer border of the two occipital condyles,
3°6 inches. The core is cylindrical in section, like specimen No. 51,
being a little flattened in front and more convex behind. This speci-
men would appear to afford further proofs of the inference derived from
31 and 33, as to the distinctness specifically of Bos Namadicus.

No. 35. Bos ?—Specimen consisting of facial part of the cranium
from the orbits, exhibiting the maxillaries on both sides with the entire
line of molars 7m situ, the posterior part of the diastema and a portion
of the nasals; the intermaxillaries broken off. This specimen is exactly
in the same mineral condition, with a ferruginous tinge, as the cranium
No. 20.

Dimensions.—Length of the molar series, right side, 53 inches; dis-
tance between the outer surfaces of the molars at the middle of the
series, 43 inches; width of palate taken at the rear of the last molar,
3 inches; ditto in the middle, 3°25 inches; ditto between anterior
molars, 3 inches. The molars on the right side are quite entire, they
are all protruded and well worn. Species indeterminable without
more detailed comparison.

1 Vide Dr. Spilsbury in Journ. As. Soe. xiii. 765, Pl. ii. fig. 3 ¢.

288 FAUNA ANTIQUA SIVALENSIS.

No. 36. Bos ?—Fragment comprising facial part of cranium,
like 35, but crushed laterally, presenting the two maxillaries with the
whole series of molars in each. Nasals broken off by a fresh fracture
near commencement of diastema. The palate is broken from the
crushing, and the opposite lines of molars are at unequal heights.
Length of left molar series, 5-9. The molars are all protruded and in
the same state of wearing as in No. 35. The specimen is encrusted
with an argillaceous matrix.

No. 41. Bos Namadicus.—Fine specimen of cranium, nearly entire,
exhibiting the whole of the chevron region from near the vertex on the
right side, as far as the extremity of the nasals, which are broken off
at their tips; also all the maxillary portion of the diastema, with a
portion of the right incisive; the two orbits nearly entire; the two zygo-
matic arches and temporal fosse; the four back molars on the right
side, and the five front molars on the left side, all protruded and worn
so as to show that the animal was fully adult but not aged. Also the
whole of the lower surface of the cranium, from the diastema back
along the palate and spheno-palate region on to the occipital foramen.
The basi-sphenoid protuberances, occipital condyles, foramen magnum,
tympanic bulle, and glenoid cavity perfectly entire, with the auditory
foramen, right side, partially shown. Also the posterior and basal por-
tion of the occipital entire.

Deficient only at the vertex, which is broken off obliquely from the
right angle downwards and forwards to the left, the fracture having re-
moved the left side of the occipital plane, the whole of the occipital
crest, and the cores of the two horns. On the left side the fracture
encroaches upon the plane of the frontal, but on the right the base of
the core of horn is zn situ, broken off vertically in a line with the
supra-auditory apophysis at the root of the jugum. ‘Truncated in
front about the middle of the incisives; the supra- and infra-orbital
foramen present on both sides. In all other respects the specimen is
singularly perfect, with the exception of the inferior border of the
right orbit, which is abraded. The form of the cranium in this speci-
men differs very notably from that of the Gour and Gayal and other
bisontine forms, recent or fossil, in its much greater elongation as
compared with the width, in the lesser prominence of the orbital
margin, and in the more slender proportions of the whole of the facial
and muzzle regions. In these respects it exhibits a considerable re-
semblance to the taurine division of Bos.as represented by Bos primi-
genius. 'This is well shown in the form of the construction between
the base of the horn-cores and orbits, which is in a long curve. From
Jahnsee Ghat on the Nerbudda.!

' Extreme length of specimen, 18 inches; from the broken margin of
vertex to commencement of the diastema, 16 inches; from ditto to
upper border of orbit, 6} inches; antero-posterior diameter of orbit,
3°3 inches; transverse ditto, 2°5 inches; length from anterior border
of orbit to commencement of diastema, 7-3 inches ; length from broken
border of vertex to the naso-lachrymal termination of the frontal, 10-5.
Length of truncated portion of muzzle in front of molars, 2:2 inches ;

1 Vide Jour. As. Soc, Beng. viii. 951.

:

BOVIDA, CERVID.Z, AND ANTILOPID.®. 289

length of line of molars, left side, 61 inches; width of frontal at con-
striction behind the orbits, 7-1 inches; ditto between the outer borders
of the orbits, 9 inches; height from surface of occipital condyle to
plane of frontal at vertex, 6-4 inches; length of nasals to commence-
ment of incisive bone, 7-2 inches; width of face immediately in front
ef orbits, 6-2 inches; ditto at the maxillary tuberosities, 5:9 inches;
ditto at commencement of diastema, 3-4 inches.

Interval between outer margin of zygomatic arches, 8:2 inches; in-
terval between external surface of maxillaries at the rear of the back
molar, 4:7 inches; ditto, at commencement of penultimate molar,
9°3 inches; interval between inside of back molars, 3-1 inches; ditto,
between the penultimate molars, 3°2 inches; ditto, in front between the
two premolars, 2°9.; width between the diastemal ridges of the maxilla-
ries, 15 inch; from anterior border of orbit to infra-orbital foramen,
6-2 inches.

The interfrontal suture is distinctly shown, also the transfrontal and
the other sutures of the face.

From the description and dimensions above given, the cranium is
considered to indicate a distinct species of Bos and to agree with the
specimens Nos. 32, 34, and 35, and to be a female. The general form
presents many analogies with the cranium of Bos primigenius figured in
Owen’s Brit. Fossil Mammalia, fig. 208, page 498, and with the crania
figured in the Ossemens Fossiles, Pl. CLXXIL, figs. 1 and 2.

No. 47. Bos Namadicus?—Fragment of cranium, very much muti-
lated, showing only occiput with occipital condyles, foramen magnum,
and right mastoid process, with a nest of calcareous crystals occupying
the cranial cavity and the cancelli of the hollow vertex. This must
have belonged to an animal of very large size. Query? Did the huge
horn-core No. 31 belong to this cranium?

TI.—Descrirtion sy Dr. Farconer or Fosstt REMAINS oF ANTELOPE,
' FROM THE SEWALIK HILus, In THE Museum OF THE ASIATIC
Society oF BENGAL.

No. 569. Antilope.—Fragment of skull, comprising the whole of the
cranial portion nearly entire with the occipital and parietal regions,
right orbit, and the bases of two horn-cores, that on the left side 14 in.
long. ‘The animal was of a size between the Nylghau and the Antilope
cervicapra; the frontal on either side shows a prominent ridge from
the base of the horn towards the occiput as in the Nylghau; the fron-
tals are too much mutilated to show whether there are ridges in front
of the horn. The horn-cores are cylindrical and appear to start from
the brow as in Antilope cervicapra; tympanic bulle large.

No. 573. Antilope—Fragment of cranium of an antelope, comprising
the left frontal detached, with a thick cylindrical horn-core attached,
and showing a part of the cup of the left orbit. It corresponds very
closely in size and appearance with 569, and probably belonged to the
same species which is considerably larger than Antilope cervicapra.

VOL. I. U

290 FAUNA ANTIQUA SIVALENSIS.

No appearance of a spiral twist in the core of the horn as in that group
of antelopes.

Inches

Length of horn-core : ; : : : ; E oA

Antero-posterior diameter. : d : 4 : F ty Beale

Transverse ditto iff
From near Nahun.

lores oA!

No. 581. Antilope Paleindica.—Fine cranium of an antelope, com-
prising two horn-cores, both orbits and the facial portion as far as the
diasteme with the line of molars on either side; the specimen is at once
distinguished by the close approximation of the horn-cores, by the
sudden inclination which the posterior part of the frontal makes with the
nasal plane in a line with the middle of the orbits, and by the deep
lachrymal fosse in front of the orbits. The specimen agrees most
closely in form and proportion with that figured at vol. xii. Journ. As. Soc.
p- 770, figs. 1 and 2, described by Captain Baker. The horns diverge
slightly and at the same time curve backwards and outwards. The
species differs very remarkably in form from any Asiatic antelope
known, and Captain Baker compares it to the Sassaybe or Acronotus
lunatus. It is considerably larger than Antilope cervicapra—more
nearly approaching the size of the Nylghau. Occipital and sphenoidal
regions entirely wanting. Molar teeth devoid of any accessory pillar
between the barrels.

Inches

Length of fragment from behind the horn-cores to commencement

of diasteme_ . : : . - : : 0 10:
Length of right horn-core fragment : : - c : fateh
Interval between cores at base - : : . : é . 05
Ditto at two inches height. : d : é : ; a) LUE
Antero-posterior diameter of orbit . : , - : . - 23
Length of line of molars : : : : : : - . 39
Height from alveolar border to middle of nasals_ . é : - 42

The supra-orbital fosse are very deep and marked, as in the
Antilope cervicapra and as shown in Captain Baker's figure, although
otherwise described in his paper. The supra-orbital foramina are very
much smaller than in that species. (Vide Plate XXIII.)

No. 582. Antilope Palewindica.—Another specimen very like the
last, but more perfect, contained in a mass of matrix enveloping a great
number of bones of different animals in all directions, among others
apparently the metatarsal extremity of a Sivatherium. The right horn-
core is of considerably greater length than in No. 581, and diverges
outwards exactly as in Captain Baker’s drawing; orbits, zygomatic
arches and molars on both sides are present, but covered with matrix.

IV.—Descriptions By CapTaIN Baker oF Fosst, ANTELOPE OF
SewaLik Hints. (Antitope PaLzinpica or FaLconer.)

1. From Journ. As. Soc. for Oct., 1835, vol. iv. p. 569.

Figs. 40, 41, and 42 are different views of a skull of an animal allied
to the Antelope; the length and narrowness of the face, the height of the
nose, and the peculiar setting on of the horns are all more conspicuously
exemplified in another specimen of a similar skull, which Colonel

ow

DESCRIPTION OF PLATE XXIII.
ANTILOPE PALAINDICA.

Figs. 1, 2, and 3. Three different views of cranium of
Antilope Paleindica, drawn by Mr. Dinkel from the original
in the British Museum (Cat. No. 39,594). About one-fourth
_ of the natural size. (See pages 290 & 555.)

VOL. I.

Vol. L. Plate 23.

ae Antilope Palseindica. W.West imp.

BOVIDA, CERVID®, AND ANTILOPID. 291

Colvin purposes presenting to the Asiatic Society. Our specimen,
however, has the advantage of possessing the cranium and occiput
entire.

2. From Journ. As. Soc. for 1848, vol. xit. p. 770.

The Fossil differs from the Indian Antelope, in the greater elongation
of its face, the straightness of its profile, the close juxta-position of its
horns at the base, the absence or small development of the infra-orbital
sinus, and the small size of the supra-orbital foramina. In all these
respects it resembles one or other of the African genera, from the de-
scriptions of which, by Captain Harris, I have extracted the following :—

‘ Acronotus caama, or Hartebeest.—Head remarkably heavy, narrow,
and long. Horns seated upon the summit of a beetling ridge above the
frontals; very close together and almost touching at the base. No
sub-orbital sinus.’

‘ Acronotus lunatus, or Sassaybe-—Head long, narrow, and shapeless,
wearing a bubaline appearance ; facial line straight ; eyes high in the
cranium ; indistinct lachrymal perforation.’

As far, therefore, as can be judged from a description, which, like
the above, has no particular reference to the osteology of these ani-
'mals, they appear to have a considerable resemblance to our Fossil.
[This specimen of Captain Baker’s is now in the British Museum
(No, 39,594), and is represented in Plate XXIII.—Ep., |

uw 2

to
So
bho

FAUNA ANTIQUA SIVALENSIS.

XIV. ON THE FOSSIL QUADRUMANA OF THE
SEWALIK HILLS.

I.—NoTICE ON THE REMAINS OF A Fosstn MoNKEY FROM THE
TERTIARY STRATA OF THE SEWALIK HILLS IN THE
Norru or Hinpostan.!

BY CAPTAIN P. T. CAUTLEY, F.G.S., AND H. FALCONER, M.D.

(Dated November 24th, 1836: read June 14th, 1837.)

Tur most highly organized mammifers hitherto described in
a fossil state, so far as our information extends, have belonged
to the Cheiroptera; and the instances of these on record are very
few.2. That quadrumanous remains should be wanting is by
no means surprising, without the necessity of supposing that
they did not exist. The countries of which the ancient races
have been most completely investigated, had a climate un-
suited to be the habitat of the tribe, as we now know it, when
the more recent or superficial deposits were in progress of
formation. If we refer to the remote epochs when the climate
was suitable, and when genera now associated with the
Monkeys were abundant, it is easy to conceive that the latter
might have existed in numbers, without their remains being
entombed. It requires, in all instances, many unconnected
circumstances for the preservation of organic bodies, and
their subsequent disclosure. Amongst the most important
of these are the habits and organization of the animals them-
selves. As in the case of birds, it might be predicated that
this lucky concurrence of circumstances would be rare with
quadrumanous remains. The very perfection in the organi-
zation of the Monkey entails, as a consequence, that his solid
frame should seldom continue to indicate the previous exist-
ence of the individual. His admirable agility and social
habits protect him against most aggressions. A flood might
suffocate in their dens, over a large tract of country, the
burrowing tribes; it might sweep from under the feet of
the monkey, hundreds of its herbivorous and predaceous
fellow-tenants of the forest, and bury them in the near shingle
or far distant estuary, or drown and deposit them in the stag-
1 This memoir is reprinted from the | Museum of the Society.—[Ep.
Transactions of the Geol. Soc. of Lon- ? Brewster's Edinburgh Journal of

don, vol. y. 2nd series, p. 499. The astra- |, Science.
galus described was presented to the

QUADRUMANA. 293

nant swamp—while he would remain secure. The tree on
which he was perched might totter, and yield to the under-
mining current, and he still escape and feed on his wonted
fruits, undisturbed by the destruction around. When the
debt of nature comes to be paid, his carcase falls to the
ground, and immediately becomes the prey of the numerous
predaceous scavengers of torrid regions, the Hyena, the
Chacal, and the Wolf. So speedily does this occur, that in
India, where Monkeys occupy, in large societies, mango groves
around villages, unmolested and cherished by man, the traces
of casualties among them are so rarely seen, that the simple
Hindoo believes that they bury their dead by night.

When the ancient races of India began to open upon us in
the new forms and the exuberant variety which the fossils of
the Sewalik hills exhibit, we were early led to anticipate that:
some trace of quadrumanous animals would soon be met with
to perfect a series, which would be incomplete without them.
Several months ago we became possessed of a solitary speci-
men, which put the matter, in our own minds, beyond all doubt.
We deferred making it public, however, in the hope of soon
finding specimens of the cranium and teeth; being unwilling
to rest the announcement on anything less characteristic.
That chance has since fallen to our fellow-labourers in the
pursuit, Messrs. Baker and Durand, of the Bengal Engineers,
who have lately discovered a specimen, consisting of a con-
siderable portion of the face, and the whole series of molars
of one side, of a quadrumanous animal belonging to a much
larger species than the bone we found.

Our fossil is the specimen which accompanies this commu-
nication. It is the astragalus of a right hind leg. It is
completely mineralized, having a specific gravity of about 2°8,
and it appears to be impregnated with hydrate of iron.
Although but a solitary bone of the foot, the relations of
structure are so fixed that the identity of the fossil is as cer-
tain as if the entire skeleton were before us. The very shallow
excavation of the superior surface (a in figs. 6 and 7) for the
pulley-like articulation with the tibia; the form and extent of
the lateral articulating surfaces (B, c, figs. 6,7, 8,and 9) for the
external and internal malleoli; the considerable elongation of
the apophysis for the head and neck of the bone (¢, H) ; the
slight obliquity with which it is sent off from the body ; and
the diagonal direction and form of the principal articulating
surface (D) with the calcaneum, are characters which, taken
in conjunction, incontestably prove that the fossil is a quad-
rumanous astragalus. It would be needless, therefore, to
dwell on the points of difference between it and the astragali
of those orders of Mammalia which have an allied form. It
is only requisite to ascertain how it agrees with the corres-

294 FAUNA ANTIQUA SIVALENSIS.

ponding bone in existing species of Quadrumana. It closely
resembles, in size and general form, the astragalus of the
Semnopithecus entellus, which we send along with the fossil
for comparison.

The principal dimensions are as follows :—

Dimensions.
Sewalik Fossil Monkey Semnopithecus entellus

Extreme length of astragalus. : + 1:3) “mechs: . 1°35 inch.

Extreme width of body of astragalus . ood an ke wr ckO3) 5

Length of body . 3 3 , ‘ : 08 aa 8 », OBO J;

Greatest diameter of navicular head . UG one as, eee srn0:G0) ~,;

Thickness of ditto ; i ¢ ‘ =, NOMA = Taaties Shion ~ 35
Fic. 6. Fic. 7.

FOSSIL, RECENT.

The chief peculiarities of the fossil astragalus, compared with
that of the Entellus, are these:—The upper articulating surface
for the tibia (A) is more convex than in the Entellus, and less
square in the outline, the lateral margins (a, b) approximating

Fic. 8. Fie. 9.

FOSSIL. RECENT,

as they run backwards; the outer one being also more ele-
vated. The peroneal articulation (B) is precisely of the same
form and extent as in the Entellus; and the rough fossa be-
tween it and the large calcanean surface (D) also corresponds.

QUADRUMANA. 295

The articular surface (c) for the inner malleolus somewhat
differs: in the fossil it is long, shallow, and rather pyriform
in outline, while in the Hntellus it is cup-shaped, deeper, and
more extensive. The other pits and inequalities of the inner
side correspond; but the entire surface slopes off more ob-
liquely in the fossil. The great calcanean surface (D) has the
same diagonal direction, with reference to the upper surface,
as in the Hntellus; it has also the same form, but it is more
vaulted, and has less stretch and width. Its inner margin is
bounded by the shallow, pulley-shaped fossa (£) for the tendon
of the flexor pollicis longus muscle, entirely as in the Entellus ;
and the rough pit between it and the anterior caleanean sur-
face (F) is alike in both. The head-and-neck apophysis is
sent off as in the Hntellus. The upper surface of the neck
(¢) is narrower and less sloped. The scaphoid surface of the
head (#) is altogether less extensive. The head itself is not
so thick and massive, and its long direction slopes more ob-
liquely upwards than in the Entellus; its inferior articular
surface is less, and there is a wide, rectagonal, rough gutter
or fossa running half way across so as to make two sur-
faces. In the Hntellus the fossa is obsolete and only indicated
by a minute foramen, so that these articular surfaces run
into one. This is the greatest difference observable in the
fossil. The rough fossa at the outer side of the neck is alike
in both.

With these inconsiderable peculiarities, the fossil agrees so
closely in size and general form with the astragalus of the
Entellus, that it probably belonged to the same sub-genus ;
- still the points of difference are sufficient to leave no doubt,
that the fossil must be assigned to a distinct species. In
equalling the Hntellus, it would belong to the larger Quadru-
mana. ‘This is all the information the specimen conveys,
regarding the animal from which it came; but we may hope
to meet with remains, which will develope its entire osteology,
more especially that of the cranium and face. The fossil was
found by a party of Hindoo collectors employed by us on the
fossil tract of the Sewalik hills; and was brought to us mixed
up with a promiscuous collection of the remains of the Hip-
popotamus, Mastodon, Ruminants, &c., like the specimens
which have been sent to the Society. We have not therefore
the means of knowing the exact locality where, and the cir-
cumstances under which, it was found.

_ The discovery is interesting in itself as supplying a deficient
link in the series of the former tenants of the globe, but
ereatly more so in connection with the races with which the
fossil was associated. We have excavated from, or found in
the débris of, different beds of the same formation which
yielded the fossil astragalus, the remains of a species of

296 FAUNA ANTIQUA SIVALENSIS.
Anoplotherium,! the Crocodilus biporcatus and OC. (Leptor-
hynchus) Gangeticus,? respectively the Magar and Gavial, two
species which at the present day inhabit the quiet waters of
the Ganges. Here then are two most instructive facts:
Quadrumana co-existed with a member of the oldest ascer-
tained pachydermatous genus of Europe; and two reptiles
now the contemporaries of man in the Hast, lived, and may
have laved, in the same waters along with a species of one of
the mammiferous genera which characterize the Eocene period
of the West; affording another illustration of constancy in the
order of nature, of an identity of condition in the earth of the
olden time with what it exhibits now, and of the invariableness
of organized forms. The two decurrent ridges on the face
which specifically distinguish the C. biporcatus of the present
day are as marked and distinct on the individuals which
existed perhaps centuries of centuries ago; and an ankle
bone of the Sewalik fossil Monkey so closely resembles that of
a living species, that it is difficult to explain the difference.
The Sewalik fossils abound in monuments of this sort.
There is a mixture of the new and of the old, of the past and
of the present, of familiar with surprising forms, together
with a numerical richness, such as no other explored region
has exhibited within so comparatively limited a space. The
Camel,’ the Antelope, and Anoplotherium, have been found,
intermixed with each other in the same bed. 'There are re-
mains of the Elephant, Mastodon, Hippopotamus,’ Anthra-
cotherium, Rhinoceros,° Hog, and Horse; the Tapir alone of
the large existing Pachydermata being without a representa-
tive. In the Sivatherium® is seen a huge Ruminant exceeding
in size the largest Rhinoceros; it is also armed with four
enormous sheathed horns, divided and foliated like the
Dicranocerine Antelopes, and able to contend for mastery
with the Mastodon. Contrasted with him in the same family

is the puny Musk Deer, scarcely larger than a Hare.

1 Anoplotherium Sivalense, a new spe-
cies, of a size somewhat larger than the
A. commune of the Paris basin. The
species is known to us by two upper
jaws in our possession with the series
of molars complete. We therefore quote
it unhesitatingly. (See antea, pp. 190
and 208.—Ep.)

2 Known to us by specimens compris-
ing the whole of the cranium and
muzzle. They do not differ more from
the existing individuals than these do
from one another in varieties dependent
on age and sex. Asiatic Researches,
vol. xix. Part II., Art. II. (See Appen-
dix to Memoir on Crocodile.—Ep.)

3 Camelus Sivalensis.(Nob.), Asiatic
Researches, vol. xix. Part II., Art. X.,

There

a species of the size of the existing
Camel. (See antea, p. 227.—Ep.)

4 Asiat. Res., Art. III. Hippopotamus
Stvalensis (Nob.), & H. dissimilis (Nob.).
(See antea, p. 1380.—Ep.)

5 Journal of the Asiatic Society, vol.
iv. p. 706, and vol. vy. p. 486. (See
antea, p. 157.—EpD.)

6 Asiat. Research ut supra, Art. I.
Sivatherium gigantewm (Nob.). Since
the memoir was printed, Col. Colvin,
Bengal Engineers, has got a specimen
of the cranium with the bases of the four
horns attached, and we have in our
possession an almost entire rear horn,
which has given the characters noted
above. (See antea, p. 268.—EHp.)

QUADRUMANA. 297
are the Cat! and the Dog tribe, the Hyena, Bear,’ and
Ratel,? and other Carnivora. In the feathered races there
are Gralle, greatly surpassing in size the gigantic Crane of
Bengal (Ciconia argala). Among the Reptilia, besides the
- Magar and Gavial, there were other Crocodiles‘ of enormous
bulk, approaching the largest Saurians ; and the Testudinata,
which have hitherto held but a humble rank beside their
Saurian co-ordinals, here show their giant representatives.
In addition to numerous species of Emys and Trionyx not
bigger than the small Terrapins of the sluggish brooks of
Hindostan, we possess humeri and femora of this tribe (with
corresponding fragments of the bucklers) as large as the equi-
valent bones of the Indian Rhinoceros. As the Pterodactyle
more than realized the most extravagant idea of the Winged
Dragon, so does this huge Tortoise come up to the lofty con-
ceptions of Hindoo mythology ; and could we but recall the
monsters to life, it were not difficult to imagine an Elephant
supported on its back.

Sunarunpoor: Nov. 24, 1836.

1 Asiat. Res., Art XI. Felis cristata

(Nob.). Smaller than the Tiger. (See
posted, p. 315.—Ep.)

2 Ibid. Art. XII. Ursus Sivalensis
(Nob). Size of the U. speleus. (See

posted, p. 321.—Ep.)
3 Messrs. Baker and Durand, Journ.
Asiatic Society, vol. v. p. 581.

4 CO. Leptorhynchus crassidens (Nob.),
an immense species far exceeding exist-
ing ones, and forming a passage from
the Gayials into the true Crocodiles. It
has the cylindrical muzzle and synos-
torized lower jaw of the former with the
blunt thick teeth of the latter.

+ 10.

FROM A SKETCH IN PENCIL IN ONE OF DR. FALCONER’S NOTE-BOOFS BY THE
LATE PROFESSOR EDWARD FORBES,

298 FAUNA ANTIQUA SIVALENSIS.

Il.—Susp-Himatayan Fossin REMAINS OF THE DADOOPOOR
CoLLECTION.!

BY LIEUTS. W. E. BAKER AND H. M. DURAND.

QUADRUMANA.

Lyell, when combating the inconclusive evidence advanced
in support of the theory of the progressive development of
organic life, notices the absence of remains of quadrumanous
species in a fossil state, and the hypothesis which this cir-
cumstance has by some geologists been considered to counte-
nance. He, however, draws attention to the fact, that the
animals which are found in sub-aqueous deposits are in
general such as frequent marshes, rivers, or the borders of
lakes, and that such as live in trees are very rarely discovered ;
he adds, moreover, that considerable progress must be made
in ascertaining the contemporary pachydermata, before it can
be anticipated that skeletons of the quadrumanous tribes
should occur. Considering the great number of relics assign-
able to the Pachydermata, Ruminantia, and Fere, which the
Sub-Himalayan field has produced, it is not therefore surpris-
ing that at length the half jaw of a quadrumanous animal
should be brought to light. The circumstance being inter-
esting in several respects, we have not deferred its com-
munication until further research should put us in possession
of more perfect specimens; the chances are against the pro-
bability of more being brought in for some time. In the
interval it may be as well at once to add to the Sub-Himalayan
list of fossils one species belonging to the order of the Quad-
TUMaAnNG.

The specimen in question was found in the hills near to the
Sutlej,; and it appears from the attached matrix to have been
derived from a stratum very similar in composition to the one
described as occurring at the Moginund deposit. The frag-
ment consists of the right half of an upper jaw; the molars
as to number are complete; but the first has lost some of its
exterior enamel: and the fifth has likewise had a portion of
the enamel from its hind side chipped off. The second and
third molars are a good deal worn, and the state of the fourth
and fifth such as to indicate that the animal was perfectly
adult. The canine is small, but much mutilated, its insertion
into the jaw and its section being all that is distinct.

’ Reprinted from the ‘Journal of the Asiatie Society’ for Noy. 1836. yol. v.
p. 739.—[ Ep. | ;

7 oe

QUADRUMANA. 299

From the inspection of the molar teeth, the order to which
the animal belonged is sufficiently evident; but there is
enough of the orbit remaining to afford additional and very
satisfactory proof. The lower part of the orbit and the start
of the zygomatic arch being very distinct, would alone remove
all doubt from the subject, the orbits of the Quadrwmana
being peculiar and not easily to be confounded with those of
other animals.

On comparison with the delineations of the dentition of
this order of animals given by F. Cuvier, the fossil bears some
resemblance to the genus Semnopithecus; the section of the
canine and the form and size of the false molars are very
similar to the example taken by F. Cuvier from a head of
the species Mauwrus, found in Java. Had the drawing
been taken from the Entellus, a species which inhabits
India, the comparison would in this instance have been more
satisfactory ; but the Mauwrus being chosen as the type, and no
mention made of other difference except length of canines,
the various species may be supposed to present no material
departure from the type in form of molars. The third molar
in the fossil is so much worn as not to admit of being com-
pared with drawings from unworn teeth; the fourth is like
that of the Mawrus, but the fifth does not resemble the ana-
logous molars of any of the existing species as represented by
F. Cuvier, for the fossil tooth possesses a small interstitial
point of enamel at the inner side, which does not appear to
have place in any of those delineated. The incisors are
absent, but the inter-maxillary is clearly distinguishable.

Were it not for the size of the canine and the fifth molar,
the specimen presents some resemblance to the genus Macacus,
given as the type of the genera Macacus and Cynocephalus.
The smallness of the canine and the large size of the molars
cause the fossil to approach more nearly to the Semnopithecus
than to the Macacus ; the difference is, however, great between
the two, for the Hntellus is said to attain the length of three
and a half feet, whereas the length of the fossil animal, if the
space occupied by the molars and their size be deemed suffi-
cient ground for a conjecture, must have been equal to that
of the Pithecus satyrus: the space taken up by the molars is
2°15 inches. This circumstance, and the differences before
pointed out, clearly separate the fossil from the species be-
longing to the genera Cynocephalus or Semnopithecus. The
Specimen is imperfect, but it indicates the existence of a
gigantic species of quadrumanous animals contemporaneously
with the Pachydermata of the Sub-Himalayahs, and thus sup-
plies what has hitherto been a desideratum in palzontology
—proof of the existence, in a fossil state, of the type of

300 FAUNA ANTIQUA SIVALENSIS.

organization most nearly resembling that of man (Vide Plate
XXIV. figs. 1 and 2).!

TII.—On AppitTionaL Fosstu SPECIES OF THE ORDER QUAD-
RUMANA FROM THE SEWALIK HILLs.?

BY H. FALCONER, M.D., AND CAPTAIN P. T. CAUTLEY.

In the November number of the Journal, vol. v. p. 739,
Messrs. Baker and Durand have announced, in the discovery
of a quadrumanous animal, one of the most interesting results
that has followed on the researches into the fossil remains of
the Sewalik hills. The specimen which they have figured
and described comprises the right half of the upper jaw, with
the series of molars complete; and they infer that it belonged
to a very large species. In the course of last rains we de-
tected in our collection an astragalus, which we referred to a
quadrumanous animal. The specimen is an entire bone, free
from any matrix and in a fine state of preservation from
having been partly mineralized with hydrate of iron. It cor-
responds exactly in size with the astragalus of the Semnopi-
thecus entellus or Langoor, and the details of form are so
much alike in both, that measurement by the callipers was
required to ascertain the points of difference. We have for-
warded the specimen with a notice to the Geological Society
of London, after keeping it. some months in reserve, having
been diffident about resting the first announcement of fossil
Quadrumana on anything less decisive than the cranium or
teeth.

This astragalus, in conjunction with Messrs. Baker and
Durand’s specimen, satisfied us of the existence of at least
two distinct fossil Quadrumana in the Sewalik hills. We
have lately become possessed of several fragments, more or
less perfect, belonging to the lower jaws of two species, both
smaller than Messrs. Baker and Durand’s fossil. These we
shall now proceed to notice.

The principal specimen is represented in Plate XXIV.
figs. 3 and 4. It consists of both sides of the lower jaw; a
great portion of the right half is entire with the whole series
of molars; the left half is broken off to the rear of the ante-

1 The fossil described in this paper; Botany of the Himalayahs.’ Plate vi.

js now in the British Museum, and from
it original drawings have been made by
Mr. Dinkel for this reprint. The draw-
ings accompanying the memoir, as it ap-
peared in the ‘Journal of the Asiatic
Society,’ were little more than rough out-
line sketches. The fossil is also figured
by Royle in his ‘Ulustrations of the

fig. 2.—[ Ep. |

2 Reprinted from the ‘Journal of the
Asiatic Society’ for May 1837, vol. vi.
p. 354. The drawings accompanying
the original paper being merely rough
outline sketches, fresh drawings have
been made by Mr. Dinkel from the spe-~
cimens in the British Museum.—[ Ep. |

” -

DESCRIPTION OF PLATE XXIV.
(JUADRUMANA.

The figures in this Plate represent all the fossil remains of
Quadrumana from the Sewalik Hills, in the British Museum,
drawn of the natural size from the originals by Mr. Dinkel.

Fig. 1. Upper jaw, right side, of the gigantic quadrumanous animal
described by Messrs. Baker and Durand at page 298. (Cat.
No. 387,157.)

Fig. 2. Palate view of same specimen, showing the crowns of six
molars.

Fig. 3. Lower jaw of a smaller species closely resembling but larger
than Semnopithecus entellus, described by Dr. Falconer at page
300. ‘The whole series of molars is present on the right side.
(Cat. No. 15,709.)

Figs. 5 and 6. Fragment of body of right side of the lower jaw of
another species of Monkey, containing the four rear molars.
Described by Dr. Falconer at page 802. (Cat. No. 15,710.)

Figs. 7 and 8. Another fragment (Cat. No. 15,711) of right side of *
lower jaw of a quadrumanous animal containing the last molar,
which agrees exactly in size with the corresponding tooth in
figs. 5 and 6, (See page 303.)

VOL. I.

"ge Ue wntp eng

Tie ISOM\ M

nO ae

y ST

QUADRUMANA., 301

penultimate molar. The two middle incisors are present, and
also the left canine broken across at its upper third. The right
canine and the lateral incisors had dropt out, leaving only the
alveoli. The molars of the left side are destroyed down to the
level of the jaw. The right ramus is wanting in more than
half its width, together with the articulating and coronoid
processes, and a portion of the margin at the angle of the jaw
is gone. The specimen is a black fossil, and strongly ferru-
ginous ; the specific gravity is about 2°70. It was encased
in a matrix of hard sandstone, part of which is still left
adhering to it.

The jaw had belonged to an extremely old animal. The
last molar is worn down so as to have lost every trace of its
points, and the three teeth in advance of it have been reduced
to hollowed-out discs, encircled by the external plate of
enamel. The muscular hollow on the ramus for the insertion
of the temporal muscle is very marked, being *35 inches deep
with a width of 55.

The dimensions contrasted with those of the Langoor or
Semnopithecus entellus and the common Indian monkey or
Pithecus rhesus, are as follows :—

Ratio of the
Pethecus |Sewalik fossil
rhesus to the En-

tellus

Fossil Se- | Semnopi-
walik thecus
Monkey | entellus

Dimensions of the Lower Jaw

5 Inches Inches Inches
1. Extreme length from the anterior

margin of the ramus to the
middle incisors : : 5 3°6 2°85 2°5 4 3:2
2. Extreme length of jaw (calculated
in the fossil) . : p c
3. Height of jaw under the 2nd
molar measured to the margin

53 4: 36 4 3:02

of the alveoli . 1:35 1:05 85 4 3:1
4. Ditto at the rear molars . 1-2 11 95 4 36
5. Depth of symphysis . : , 19 1-4 uel 4 3
6. Space occupied by the molars 2°3 ee) 15 4 33
7. Interval between the 1st molars . 9 75, 65 4 32
8. Antero-posterior diameter of the

canine . e : : ¢ sf) “4. 3 4 32
9. Width of jaw behind the chin

under the 2nd molar : 1:15 1:05 95, 4 37

As in all other tribes of animals in which the species are
very numerous, and closely allied in organization, it is next
to impossible to distinguish an individual species in the
Quadrumana from a solitary bone. In the fossil, too, the
effects of age have worn off those marks in the teeth, by
which an approximation to the subgenus might be made. It
very closely resembles the Semnopithecus entellus in form,

302 FAUNA ANTIQUA SIVALENSIS.

and comparative dimensions generally. The differences ob-
servable are slight. The symphysis is proportionally a little
deeper than in Hntellus, and the height of the body of the
jaw somewhat greater. The chin, however, is considerably
more compressed laterally under the second molar than in the
Entellus, and the first molar is more elongated and salient. So
much of the canine as remains has exactly the same form as
in the Entellus, and its proportional size is fully as great. As
shown by the dimensions, the jaw is much larger than in the
full-grown Hntellus; in the former the length would have
been about 5:3 inches, while in the latter it is exactly 4 inches.
The fossil was a species of smaller size than the animal to
which the specimen described by Messrs. Baker and Durand
belonged, but less so than it exceeds the Hntellus.

Our limited means for comparison, restricted to two living
species, besides the imperfection of the fossil and the few
characters which it supplies, do not admit of afirming whether
it belongs to an existing or extinct species; but the analogy
of the ascertained number of extinct species among the
Sewalik fossil mammalia makes it more probable that this
monkey is an extinct one than otherwise. There is no doubt
about its differing specifically from the two Indian species
with which we have compared it.

The next specimen is shown in Plate XXIV. figs. 5 and 6.
It is a fragment of the body of the right side of the lower jaw
containing the four rear molars. The teeth are beautifully
perfect. It had belonged to an adult although not an aged
animal, the last molar having the points a little worn, while
the anterior teeth are considerably so. The dimensions
taken along with age at once prove that it belonged to a
different and smaller species than the fossil first noticed.

The dimensions are as follows :—

Smaller fossil | Larger fossil Semnopithe-| Pithecus

Dimensions of the Lower Jaw a lee a aecieneaites enagne
Inches Inches Inches
1. Length of space oceupied
by the four rear molars 1:48 1-7 1:48 1:25
2. Height of jaw at the third
molar . . 5 : 95 ikal 3)

The length of jaw, therefore, estimated from the space oc-
cupied by the teeth, would be 4 inches, while in the larger
fossil it is 5°3 inches; a difference much too great to be de-
pendent merely on varieties of one species. Besides, we have
another fragment, also belonging to the right side of the

3
.
:
;
¢

QUADRUMANA. 305

lower jaw, and containing the last molar, which agrees ex-
actly in size with the corresponding tooth in the figured
specimen. ‘This goes to prove the size to have been constant.
The fossil, although corresponding precisely in the space oc-
cupied by the four rear molars with the Entellus, has less
height of jaw. There is further a difference in the teeth. In
the Entellus the heel of the rear molar is a simple flattened
obliquely surfaced tubercle, rather sharp at the inside. In the
fossil, the heel in both fragments is bifid at the inside. The
same structure is observable in the heel of the rear molar of
the common Indian monkey, P. rhesus. It is therefore pro-
bable that the fossil was a Pithecus also. It was considerably
larger, however, than the common monkey, and the jaw is
more flattened, deeper, and its lower edge much sharper, than
in the latter. This difference in size and form indicates the
species to have been different.

It would appear, therefore, that there are three known
species of fossil Quadrumana from the Sewalik hills: the first
a very large species, discovered by Messrs. Baker and Durand ;
the second a large species also, but smaller than the first,
and considerably larger than the Hntellus; the third, of the
size of the Hntellus, and probably a Pithecus; and further,
that two of the three at least, and most probably the third
also, belonged to the types of the existing monkeys of the old
Continent, in having but five molars, and not to the Sapajous
of America.

There are at present upwards of 150 described species of
existing Quadrumana; and as the three fossil ones all be-
longed to the larger sized monkeys, it is probable that there
are several more Sewalik species to be discovered. We have
some specimens of detached teeth, of large size, which we
conjecture to be quadrumanous; but their detached state
make this conjecture extremely doubtful.

Besides the interest attaching to the first discovery in the
fossil state of animals so nearly approaching man in their
organization as the Quadrumana, the fact is more especially
interesting in the Sewalik species from the fossils with which
they are associated. The same beds, or different beds of the
same formation, from which the Quadrumana came, have
yielded species of the camel and antelope, and the Anoplo-
therium posterogenium (Nob.):' the first two belonging to
genera which are now coexistent with man, and the last to a
genus characteristic of the oldest tertiary beds in Europe.
The facts yielded by the Reptilian orders are still more in-
teresting. 'Two of the fossil crocodiles of the Sewaliks are

1 Chalicotherium Sivalense.-—[Ep.]

304 FAUNA ANTIQUA SIVALENSIS.

identical, without even ranging into varieties, with the Cro-
codilus biporcatus! and Leptorynchus Gangeticus which now
inhabit in countless numbers the rivers of India; while the
Testudinata are represented by the Megalochelys Sivalensis
(Nob.),? a tortoise of enormous dimensions which holds in its
order the same rank that the Iguanodon and Megalosaurus do
among the Saurians. This huge reptile (the Megalochelys) —
certainly the most remarkable of all the animals which the
Sewaliks have yielded—from its size carries the imagination
back to the era of gigantic Saurians. We have leg bones
derived from it, with corresponding fragments of the shell,
larger than the bones in the Indian unicorned Rhinoceros !

There is, therefore, in the Sewalik fossils, a mixture in the
same formation of the types of all ages, from the existing up
to that of the chalk; and all coexistent with Quadrumana.

P.S. Since the above remarks were put together, we have
been led to analyze the character presented by a specimen in
our collection which we had conjectured to be quadrumanous.
The examination proves it to be so incontestably. The speci-
men is represented in fig. 11, A. and B.

Fic. 11.

A and B represent the canine of natural size; at C it has been reduced and
placed in position with the lower jaw of the Sumatra Orang-Outang. (The drawing
has been copied from one in the ‘ Journ. As. Soc.,’ vol. vi. Pl. xviii— Eb.)

It is the extra-alveolar portion of the left canine of the upper
jaw of avery large species. The identification rests upon
two vertical facets of wear, one on the anterior surface, the
other on the inner and posterior side, and the proof is this.
The anterior facet b has been caused by the habitual abrasion
of the upper canine against the rear surface of the lower one,

1 Crocodilus bombifrons. See appendix to Memoir on Crocodiles.—[ Ep. ]
2 Colossochelys Atlas —I[Ep.]

el Sa hat Kod ice a

QUADRUMANA. 305

which overlaps it, when the jaws are closed or in action.
This facet would prove nothing by itself, as it is common to
all aged animals in the Carnivora and other tribes in which
the upper and lower canines have their surfaces in contact.
The second facet, c, must have been caused by the wear of the
inner and rear surface of the canine against the outer surface
of the first molar of the lower jaw. But to admit of such
contact, this molar must have been contiguous with the
lower canine without any blank space intervening ; for if there
was not this contiguity the upper canine could not touch the
lower first molar, and consequently not wear against it.
Now, this continuity of the series of molars and canines
without a diasteme or blank interval is only found, through-
out the whole animal kingdom,'! in man, the Quadrumana,
and the Anoplotheriwm. ‘The fossil canine must therefore
have belonged to one of these. It were needless to point out
its difference from the human canine, which does not rise
above the level of the molars. In all the species of Anoplo-
thervum described by Cuvier, the canines, while in a contiguous
series with the molars, do not project higher than these,
being rudimentary asin man. Of the Sewalik species, Ano-
plotherium posterogenium (Nob.), we have not yet seen the
canines ;? but it is very improbable, and perhaps impossible,
that the fossil could belong to it. For if this species had a
salient canine, it must have been separated from the molars
by an interval as in the other Pachydermata; otherwise the
jaws would get locked by the canines and molars, and the
lateral motion required by the structure of the teeth and the
animal’s herbivorous habit would be impracticable; and if
there was this interval, the upper canine could not have the
posterior facet of wear. The fossil canine must therefore
have belonged to a quadrumanous animal. This inference is
further borne out by the detrition of the fossil exactly cor-
responding with that of the canines of old monkeys.
The dimensions are :—

Length of the fragment of canine. : : j - 1-75 inch.
Antero-posterior diameter at the base 5 : ‘ . roy Mie
Transverse ditto A 3 3 F F F ‘ . ‘hae hos
Width of the anterior facet of wear . : 5 : "Gil Gite:

The two diameters are greater than those of the canine of
the Sumatra Orang-outang described by Dr. Clarke Abel? as
having been 7} feet high. The Cynocephali have large and
stout canines, more so comparatively than the other Quadru-
mand. But to what section of the tribe our fossil belonged,

1 Cuvier, Oss. Foss., tome iii. p. 15. 3 Asiatic Researches, vol. xv. p. 498,
* Tt had no canines, see antea, p. 211.
—[Eb.]
VOL. I. x

306 FAUNA ANTIQUA SIVALENSIS.

we have not a conjecture to offer. We may remark, how-
ever, that the tooth is not channeled on three sides at the
base, as in the Entellus. Does the fossil belong to the same
species as the jaw discovered by Messrs. Baker and Durand,
or to a larger one?

Appended Note by Mr. Prinsep, Ed. Journ. As. Society.—
We have sketched Dr. Falconer’s highly curious fossil tooth
in position with the lower jaw of the Sumatran Orang-outang
from the Society’s Museum (fig. 11). There is a third facet
of wear at the lower extremity d which, on reference, we
find Dr. Falconer attributes like ¢ to attrition against
the first molar, being observable, he says, in many aged
animals. The worn surfaces ¢ and d are uniformly polished,
and have evidently originated from attrition against a tooth ;
but with regard to the principal facet b, we confess we have
a degree of scepticism, which can only be removed by a cer-
tainty that the fossil had been seen extracted from the matrix.
In the first place, the great extent of the worn surface and
its perfect flatness could hardly be caused by attrition against
the lower canine which should produce a curvature measured
by the length of the jaw as radius. In the next place, the
enamel of the tooth is less worn than the interior and softer
part of the fossil; and thirdly, on examination with a mag-
nifier, numerous scratches are visible in divers directions:
all these indicating that the facet may have been produced
on the fossil, by grinding it on a file or some hard flat surface.
On showing the fossil to Madhusudana, the medical pandit
of the Hindoo College, he at once pronounced that the tooth
had been ground down to be used in medicine, being a sove-
reign specific in the native pharmacopeia. This circumstance
need not necessarily affect the question, for it is probable
that the native druggist would commence his rubbing on the
natural plane, if any presented itself to his choice; but Dr.
Falconer and Capt. Cautley, to whom we have returned the
fossil with a communication of our doubts, assure us in reply
that the fossil tooth was brought in along with a large col-
lection, so that there is every improbability of its having been
in possession of a native druggist. At any rate, it is not on
the front wear that they so much rest their argument of its
origin, as on the posterior abrasion which could only happen
in the jaw of a quadrumanous animal. In fact, they have
recent Quadrumana showing precisely similar wear on a small
scale, and no other head will do so. We find only one ex-
ception in the Society’s Museum, viz. the Tapir, whose right
upper incisor (or non-salient canine) falling between the two
lower ones is worn nearly in the fashion of the fossil, but it
is less elongated.

QUADRUMANA. 307

[The following is an extract from a letter addressed by Mr.
Prinsep to Dr. Falconer on April 7th, 1837:—‘TI have not
only read your reasoning with attention, but have conferred
with the best authorities here, and with the best of all, viz.
the Orang-outang himself in our Museum, anent the quadru-
manous canine tooth. The result is a firm conviction that
you are right. Indeed the tooth fits so well on to the Orang
that it might be thought (and has been!) to belong to this
very individual.’

Five years afterwards, also, in 1842, Dr. Falconer instituted
a close comparison between this tooth and the corresponding
tooth of three skulls of the Orang-outang, contained in the
Museum of the Asiatic Society in Calcutta, and found their
agreement so close that he conjectured that the extinct
Sewalik form had been a large ape allied to Pithecus satyrus.
See vol. 11. ‘Memoir on Antiquity of Man.’—Ep. |

TV.—On AppiTIoNAL QUADRUMANOUS REMAINS FROM THE
TERTIARY DeEposits OF THE SEWALIK Hiuus.!

BY H. FALCONER, M.D.

Last November we forwarded to Mr. Lonsdale for his ac-
ceptance a fossil astragalus of a quadrumanous animal,
which we requested he would do us the favour of submitting
to the Geological Society, along with a descriptive notice
which accompanied the specimen. We intimated that Lieut.
Baker, of the Bengal Engineers, had discovered a large por-
tion of the face, comprising the series of molars, of a much
larger species, also from the Sewalik hills; establishing the
contemporaneous existence of at least two species of Quad-
rumana, along with both extinct and recent mammiferous
genera, such as the Camel, Anoplotherium and Sivatherium,
and also some existing species of Crocodiles. We expressed
the hope that we should soon come to the possession of more
quadrumanous remains from the tract which has hitherto
yielded us so rich a harvest. This anticipation has already
been realized.

Tn the ‘ Journal of the Asiatic Society’ for May, 1837, we
have figured and described several additional remains con-
taining teeth, and one of them a nearly complete lower
jaw, of fossil Quadrumana from the Sewalik hills. In a
postscript to the paper we have noticed a remarkable frag-
ment consisting of a detached canine, which we refer

1 This paper was written in 1837, but | Geological Society, and_is now for the
was never sent, as intended, to the | first. time published.—[Ep.]
x2

308 FAUNA ANTIQUA SIVALENSIS.

to the same order. The specimen is herewith submitted
to the Society. It belonged to the left side of the upper
jaw. The evidence which would make it quadrumanous
is briefly this: There are two surfaces of wear, the one a flat
smooth disc on the anterior side, and the other a flexuous
smooth surface on the posterior and inner side, with a slight
oblique abrasion of enamel at the worn apex; the anterior
wear occasioned by the play of the lower canine upon the
upper, and the rear one, by the play of the posterior and
inner surface of the upper canine against the outer surface
of the first molar of the lower jaw. To produce this posterior
wear two conditions in the jaw were necessary: first, the
anterior molar of the lower jaw must have been quite close
to its adjoining canine; and second, the upper canine must
have projected above the level of the other teeth, so as to
overlap the lower first molar; otherwise the wear could not take
place. Now the first of these conditions is common to man,
Quadrumana and the Anoplotherium, and with a very partial
exception restricted among Mammalia to them; the second,
among these three genera, is confined to the Quadrumana.

In some of the Carnivora, more especially in the Bear,
in which the number of molars is very variable, sometimes
the anterior false molar of the lower jaw is placed close to
the canine, as in the Ursus Tibetanus, and then the upper
canine may play slightly against it. But this tooth, under
these circumstances, is very rudimentary, simall, and single-
fanged, and the utmost amount of its effect is to produce
a slight vertical streak on the enamel of the canine, at its
rear surface. The broad flat disc of wear which the fossil
has on its anterior side we believe never occurs in the canines
of Carnivora, while it does in monkeys. The exception does
not, therefore, appear to invalidate the evidence that the
fossil is quadrumanous.

It has been objected to us by a scientific friend, Mr. James
Prinsep, to whom the specimen was sent to be examined and
figured for the ‘Journal of the Asiatic Society,’ that the an-
terior surface is too extensive to have been naturally produced;
that it bears the marks of artificial abrasion; and that, as we
ourselves did not see it dug out of the matrix, it was probably
ground down, as such teeth are highly prized as a medicine
and charm, in certain parts of India.

On these objections we have to remark that the anterior
wear is quite as great, and precisely similar in aged male
monkeys, as is seen in the canine of a male Entellus, which
accompanies the fossil; that the edge of enamel is elevated
slightly above the flat dise of ivory, which could not be the
case had the surface been artificially ground down; that the

QUADRUMANA. 309

streaks on the surface were probably caused by rubbing on
a table in an attempt to remove a slight portion of the matrix,
which still adheres to the middle of the disc; that in the
neighbourhood of the Sewalik hills such fossil bones and
teeth have no reputation either as medicines or charms, for
amongst the enormous collections of bones and teeth that we
have made in the tract we have not met with a single in-
stance of a specimen having been ground down, or manipu-
lated in any way for a charm or medicine; and that, although
we did not see the specimen exhumed, nor do we even know
the exact spot from which it came, yet we have no reason to
suspect that it was ever in any other hands till got by our
collectors. It was brought in mixed up in a large collection
of other fossil bones.

Tf our inference be correct, it proves the existence of fossil
Quadrumana in the Sewalik hills as large as the gigantic
Orang-outang of Sumatra, described by Dr. Clarke Abel. The
evidence usually supplied by organic remains is conclusive,
from these remains being either forms of structwre, such as
bones, or casts of forms, such as coprolites, or the prints of
birds’ feet. In this case it is neither, but a secondary result
or alteration of a form of structure, which generally proves
nothing more than the age of an animal. But in the fossil
it appears to us to be sufficiently good to be conclusive.

The inference is of considerable geological interest, and we
hope, by sending the specimen to the Geological Society, that
it may either be confirmed or refuted.

V.—NorTe on A CoRRECTION OF PUBLISHED STATEMENTS
RESPECTING Fossin QUADRUMANA.! »

BY H. FALCONER, M.D., F.R.S., F.GS.

Of the sciences which unite to build up the superstructure
_ of geology there is none more rapid in its progress, or which
contributes more to the enlargement of the pile, than Palzon-
tology. Year after year brings forth with unerring certainty
its complement of extinct forms, new to the system, or extends
the area of what was but partially known before; while, at
irregular intervals facts of such importance spring unex-
pectedly to light, that the science makes a bound in advance,
and we are carried irresistibly onward by the impulse. Too
often in such cases, elated by the result and fascinated by
the prospects opened before us, we neglect to apply to the

1 The paper of which this is a portion | years before the author's death, but was
was written in 1862, less than three | never published.—[ Ep. ]

310 FAUNA ANTIQUA SIVALENSIS.

sudden stride the close and severe scrutiny which marks the
ordinary progress of the science. We push on, taking it for
granted that the steps which we have leaped over are secure,
and that all is sound in our wake. Buta lesson of wisdom,
sooner or later, is certain to overtake us. Occasions arise
which awaken us to a sense of insecurity, our confidence is
shaken, and we are compelled to cast a searching glance
behind us. The inexorable law which tracks every hurried
and unsound result of human labour, material or intellectual,
exacts reparation. What we have built up in haste we are
compelled to throw down and reconstruct with care and
attention. When assured that all is made right, we start
again on our onward progress to achieve renewed success,
sobered by the lesson of the past, and provided with safe-
guards for the future.

The same kind of retrospect which from time to time we
cast on the. material facts, justice demands of us to apply
also to the history of discovery in the science. Facts which
are now fused in the common mass may have exercised a
powerful influence when first brought to light. The impartial
historian will regard them in this light, and not merely as
they now appear. He will also be scrupulously careful to
award to the first observers fairly what is their due; for,
apart from the abstract consideration of justice, the only
guarantee which we have that our own labours shall be
respected in the future is the fairness with which we our-
selves deal with the labours of our contemporaries and of
those who have gone before us.

These trite reflections are called up on the present occa-
sion by what has lately come to light in regard to the
discovery of fossil Quadrumana, and by the investigations
now in progress respecting the early appearance of the human
race upon the surface of the globe. Having had a humble
share in both, and historical injustice having been, as I
consider, done to me, though perhaps unwittingly, as regards
the date of the former, I think the time appropriate for indi-
cating an historical correction concerning the exact deter-
mination of supposed fossil Quadrumana.

When Cuvier died, monkeys and man were alike unknown
in the fossil state, and the negative evidence furnished the
ereat French anatomist with an argument, which he wielded
with powerful effect and to the general acceptance of man-
kind, in favour of the very modern appearance of both upon
the earth. But in 1836 and 1837 the successive discovery
of fossil Quadrumana in miocene strata in India and Europe
excited much interest, and communicated a hopeful forecast
to the aspirations of Paleontology. Now, the subject is so

QUADRUMANA. 311

far advanced that the great problem of the day is, ‘ How far
back in time are we justified in carrying the human race as
contemporary with the extinct mammalia of the quaternary
period?’ The two subjects are intimately connected by
many aspects, and the tendency of inquiry at the present
time is to indicate that they will be still more closely con-
nected by future research. The problem will doubtless be

- worked out with philosophic caution and scrupulous accuracy ;

but a lesson of prudence may be learnt from the quadrumanous
past in reference to the spurious or unreliable evidence which
the stimulus of a newly started and exciting subject is certain
to bring forth.

I will now call attention to the historical correction.

In 1836, fossil Quadrumana were discovered in the Sewalik
hills in India.

In February, 1837, Sir Charles Lyell, then president, in
his anniversary address to the Geological Society, used the
following terms with reference to the award of the Wollaston
medals: ‘ Dr. Royle has permitted me to read part of their
(Captain Cautley’s and Dr. Falconer’s) correspondence with
him, when they were exploring the Sewalik mountains, and
I can bear witness to their extraordinary energy and perse-
verance,’ &c. (op. cit. p. 35). Dr. Royle was then one of the
secretaries of the Geological Society ; and on the 3rd May of
the same year (1837), the following appears on the proceed-
ings of the Society :—Read: ‘ An extract of a letter, dated
Suharunpoor, 18th November, 1836, from Captain Cautley to
Dr. Royle, was next read, permitting the announcement,
which had long been communicated to the latter, of the finding
of the remains of a quadrumanous animal in the Sewalik
hills, or Sub-Himalayan range of mountains. An astragalus
was first found, but latterly a nearly perfect head, with one
side of the molars and one orbit nearly complete.’—Proceed-
ings, p. 044),

On the 14th June, 1837, a memoir by Captain Cautley and
Dr. Falconer, entitled, ‘On Remains of a Fossil Monkey, from
the tertiary strata of the Sewalik hills,’ was read to the
Geological Society. It bears date, Suharunpoor, 24th Nov.,
1836, and it appeared in the 5th volume, second series, of the
‘Geological Transactions,’ carrying that date. The original
fossil astragalus which it describes, together with the cor-
responding bone of the living Semnopithecus entellus, was
transmitted along with the memoir; and both are now
extant in the Museum of the Society.'

1 Sir Proby Cautley informs me that | was despatched, Dr. Falconer always
for many months before this paper | carried this astragalus in his pocket,

312 FAUNA ANTIQUA SIVALENSIS.

In explanation of the interval which elapsed between the
date of communication by the authors and the date when
the memoir was read to the Geological Society, it is to be
remembered that the now quick means of transmission to
and from India, by the overland route, did not at the time
exist, and that the memoir and the objects were forwarded
by the only channel then open, being the long voyage round
the Cape.

Assuming the dates set forth to be authentic—and about
this it is believed there can be ng question—indubitable
evidence is borne on two distinct publications of the Geolo-
gical Society, that Captain Cautley and myself had established
the existence of a fossil quadrumanous species in the strata of
the Sewalik hills, upon materials discovered, identified, and
described by ourselves, at least as far back as the 18th Nov.,
1836. In fact, the specimen which belonged to my collection
was identified by me in the shape in which it came before
the Society, in the spring of 1836; but being unwilling to
rest so important an announcement upon other materials
than the skull or teeth, Dr. Royle, to whom the discovery
was communicated, was requested to withhold it from pub-
lication. The fossil astragalus now before the meeting was,
T believe, the first quadrumanous fossil remain determined
by the eye of science. In affirming this, I do not mean to
prefer any claim anterior to November, 1836; but this strong
conviction in my mind will explain to the Society why I
should not feel disposed tamely to brook the injustice, that
the unreliable record of history should put me a full year
back in the order of discovery.

The next incident of the case is an important paper by
Captains Baker and Durand, on fossil quadrumanous remains
from the Sewalik hills, which appeared in the number for the
month of November, 1836, of the ‘Journal of the Asiatic
Society of Bengal.’ There is no record borne on the Pro-
ceedings of the Asiatic Society when the memoir was com-
municated or received, nor is any date borne on the memoir
itself. But it is manifest that the number of the journal in
question was not published, or even printed, until after the 7th
December, 1836, since it contains the record of the Pro-

and was thoroughly convinced of its
quadrumanous nature, although with
his characteristic caution he waited for
further evidence before making the dis-
covery public. The credit of the first
discovery of fossil Quadrumana has
generally been awarded to French
naturalists (see Supplementary Notes
to Dr. Buckland’s Bridgewater Treatise) ;

but the date of Falconer's first paper
was two months antecedent to the pre-
sentation, on the 16th January 1837, to
the French Academy of Sciences of a
Memoir by M. Lartet, respecting the
discovery of the lower jaw of an ape, in
the tertiary freshwater formation of
Simorre, in the south of France| Ep.]

QUADRUMANA. 313

ceedings of the meeting of the Asiatic Society held on that
date. Messrs. Baker and Durand, although later in point of
actual discovery, were before Captain Cautley and myself in
publication, they having wisely availed themselves of the
local scientific journal, while we forwarded our memoir to
the Geological Society. Our friends, therefore, established
a just claim to priority of publication, which we have never
dreamt of questioning. It is not with them that we are at
issue, but with the inexact historians of the case.

The 5th vol., second series, of the ‘Geological Transactions,’
contains a memoir by Captain Cautley, bearing upon the
fossil remains of the Sewalik hills; it contains comments in
a foot-note by the referee, and the index of the volume points
out that they were contributed by Professor Owen.

Let us now examine the reputed history. In the first
part of the ‘ British Fossil Mammalia,’ published in 1844,
the author gives an account of the occurrence of Quadrumana
in the fossil state. He assigns the first discovered case to
Lieutenants Baker and Durand, in 1836, and then goes on
to state, that in the ‘year following’ Captain Cautley and
myself discovered a considerable portion of a lower jaw con-
taining teeth. ‘ Fragments of two other lower jaws and an
entire astragalus were subsequently discovered by these gen-
tlemen.’ Professor Owen here omits any notice of our
memoir of the 24th November, 1836, which appeared in the
same 5th vol. of the ‘ Geological Transactions,’ along with the
other, which passed through his hands as referee, and bears
annotations by him. He jumbles all our quadrumanous
findings together, and the most important being the astra-
galus, which was first in date, he puts last.

Sir Charles Lyell gives an historical note in the ‘ Principles
of Geology,’ on the same subject, in these words :—‘ The
first quadrumanous fossils discovered in India were observed
in 1836 in the Sewalik hills, a lower range of the Himalayah
Mountains, by Lieutenants Baker and Durand, by whom
their osteological characters were determined (Journal of
Asiatic Soc. of Bengal, vol. v. p. 739), and in the year follow-
ing other fossils of the same class were brought to light and
described by Captain Cautley and Dr. Falconer.’ On this
occasion the author quotes our memoir in the ‘ Journal of the
Asiatic Society of Bengal,’ vol. vi. p. 354, for May, 1837;
but he makes no mention of the Astragalus Memoir of
November, 1836, and the note bears internal evidence that
in framing it he followed Professor Owen as his guide,
instead of consulting the original authorities. Beyond this
he is blameless, he not being a professed expert in mamma-~
lian Paleontology. But the combined high authority and

314 FAUNA ANTIQUA SIVALENSIS.

wide currency of his great work have contributed largely to
propagate an injurious error.

With the countenance of two such names as Professor
Owen and Sir Charles Lyell to the inaccurate statement, it
is not to be wondered at that it should have become incor-
porated in suecum et sanguinem with the reputed history of
the science, as a fact, by systematic writers. Among others,
I may cite Pictet, who, in his ‘ Paleontologie,’ faithfully
follows the author of the ‘ British Fossil Mammalia’ in
assigning to Captain Cautley and myself ‘lannée suivante,’
i.e. 1837.

The facts of the case are so plain, that I do not think it
necessary to urge them beyond a parallel illustration :—Sup-
pose a registry to be established in London for the record of
all births of British parentage within the Imperial dominions,
home and colonial; let two births, of distinct parentage,
take place, say in Calcutta, in the month of November, 1861;
~ let one of them be gazetted on the spot in a local periodical,
and let the other appear in the ‘ Times’ as soon as possible,
with authentic particulars, in the month of January, 1862.
What should be said of the accuracy of the registrar who
inserted in his record that the one birth had taken place a
year after the other?

Highteen years have passed since the statement first ap-
peared, in 1844, in the ‘British Fossil Mammalia,’ during
which period I have remained silent; and it may be thought
that the right of redress has lapsed, through default, by the
effluxion of time. I admit that every correction of the kind
should be vindicated at once, in the interest of truth, irre-
spectively of other considerations; yet personal indifference
to so palpable an inaccuracy, and constitutional aversion from
the hispid walks of controversy, led me to disregard it. The
matter would probably never have been noticed by me, had
not immunity from correction led to further error in Pro-
fessor Owen’s work, entitled ‘ Paleontology,’ the second
edition of which appeared in 1861, and in which the following
paragraph occurs in the part devoted to the fossil Quadru-
mana: ‘Genus Semnopithecus.—To this genus belong the
petrified jaws, teeth, and astragalus, from the older pliocene
or miocene rocks of the Sub-Himalayan hills, near Sutlej,
India, discovered in 1836 by Durand and Baker.’ Thus,
what we were the first to discover is not merely abstracted
from us, but we are struck out of the record. It was fuil
time, therefore, that I should step forward to vindicate the
cause of truth.

* * *

FELIS CRISTATA. 315

XV. ON FELIS CRISTATA, A NEW FOSSIL TIGER
FROM THE SEWALIK HILLS.!

BY HUGH FALCONER, M.D., AND CAPTAIN P, T. CAUTLEY.

To the large fossil species of the genus Felis hitherto de-
scribed we are now enabled to add another, from the tertiary
strata of the Sewalik hills, differing alike from the F’. spelea
and F’. antiqua of the ‘Ossemens Fossiles,’ and, so far as
our means for comparison enable us to judge, from every
known member of the genus. ;

The specimen from which we take our description is one
of the most perfect that has up to this time been exhumed
from the fossil tract. It was found at the foot of a sandstone
cliff, partly encased in a hard stone matrix. It consists of a
nearly entire head, deficient only in the temporal apophysis
of the left zygomatic arch, and in a small portion of the
sagittal crest. The incisors had dropped out, but the
alveoli are sufficiently distinct to indicate their number
and relative size. The cheek teeth are nearly entire, but
the canines are broken off at their bases. The fragility of
the specimen has deterred us from removing a portion of hard
stony matrix which fills up the right zygomatic fossa, and
conceals the base of the skull from the anterior margin of the
occipital foramen on to the posterior border of the palatine
bones.

Plate XXV. fig. 2, represents the head in profile, and
shows the peculiarities of the fossil in the most striking light.

Ist. The relative shortness of the facial portion of the
head, from the post-orbital apophyses of the frontal to the
border of the incisives, and the length of the cranial portion
from the same point to the occipital crest; the dimensions
being as 124 to 153. In this respect it differs from all exist-
ing species, in which, as exhibited by Cuvier,? the facial
portion of the head exceeds the cranial, generally to a con-
siderable extent.

1 This paper is reprinted from the | Sivalensis’ (see description of Plate K.),
‘Asiatic Researches,’ yol. xix. p. 135, | and from this the accompanying draw-
1886. Several specimens of the skull | ings have been copied.—[Eb. |
of Felis cristata are figured in an un- 2 Ossemens Fossiles, vol. iv. p. 147.
published plate of the ‘Fauna Antiqua

316 FAUNA ANTIQUA SIVALENSIS.

2nd. The outline of the upper surface of the cranium and
face. The facial line runs with a gentle curve, slightly
convex and nowise serpentine, to the rear of the post-orbital
apophyses of the frontal. The cranial line meets it without
angularity, and runs back horizontally to the occipital crest:
so that when the head is seated on the occipital condyles and
carnassier teeth, the plane of the base of the cranium is parallel
to that of the vertex. In this respect it differs remarkably
from all known large species of Felis, in which the cranial
line descends more or less either in a curve or slope, from the
' post-orbital apophyses to the occipital crest.

- 8rd. The saliency of the sagittal crest, which greatly ex-
ceeds that of all known Feline. (See Plate XXYV. fig. 1.)
4th. The height of the occipital, which is relatively greater
than in any other known species of the genus (figs. 2 and 4).
5th. The elevated position of the zygomatic arches, and
the strongly arched outline of their inferior margin.

Viewed from above (fig. 1), the contrasted proportions of
the cranium and face are well exhibited. The muzzle is
short. The canine region of the maxillaries swells greatly
out in the bulge of the alveoli, and between it and the malars,
the infra-orbital hollow is more abrupt and deeper than
generally holds in the large Feline. The nasals are short
and broad. The brow is wide. The infra-orbital region
of the frontal is marked by a deep longitudinal hollow. The
post-orbital processes of the frontal and malars are blunt
and little projecting. The post-orbital ridges of the frontal
meet at a very acute angle, leaving between them a well-
defined and narrow furrow. The length and prominence of
the sagittal crest and the height of the occipital crest are
strongly apparent. The parietals are seen to bulge out little
towards their upper margin, but considerably towards the
temporals.

Our means for comparison of the fossil with most of the
large Feline are restricted to the figures in the ‘Ossemens
Fossiles,’ which, however, are so perfect and characteristic as
to admit of the chief marks of distinction being very readily
seized. Of all the iarge species, the specimen most closely
resembles the Tiger, although considerably smaller in size and
perfectly distinct otherwise. The chief points of resemblance in
both are the great development of the sagittal and occipital
crests, the considerable surface of the occipital, the mode-
rate convexity of the brow and face, and the elevated position
of the zygomatic arches and the outline of their inferior
margin. ‘To exhibit the peculiarities of the fossil we shall
now give its dimensions in juxta-position with those of two
of the largest-sized Tigers, killed in the forests near Suharun-

Fig.

HH oo 9

DESCRIPTION OF PLATE XXV.

FELIS CRISTATA, DREPANODON SIVALENSIS, AND
DREPANODON LATIDENS.

. Upper surface of cranium of Felis cristata, about two-sevenths

of the natural size. The figure has been copied from a drawing
by Mr. Ford in an unpublished Plate of the Fauna Antiqua
Sivalensis. The specimen is in the British Museum (No. 37,133),
and resembles that figured in the ‘ Asiatic Researches,’ vol. xix.
Plate XXI. (See pages 315 & 548.)

. Profile view of same specimen.
. Palate view of same specimen.

. Posterior view of same specimen, showing occipital condyles and

foramen magnum.

. Fragment of upper jaw, right side, of Drepanodon Sivalensis,

three-fourths of the natural size, and containing a portion of the
canine, the posterior edge of which is finely serrated. The
specimen is in the British Museum (No. 16,350), and the figure
is copied from a drawing by Mr. Ford in an unpublished Plate
of the Fauna Antiqua Sivalensis. (See page 550, and vol. ii.
p- 456.)

. Fragment of lower jaw with three premolars of Drepanodon

Sivalensis, three-fourths of the natural size The specimen is
in the British Museum (No. 16,557), and is copied from a
drawing by Mr. Ford in an unpublished Plate of the F. A. S.
(See page 550.)

Figs. 7 and 8. Represent two views of a canine of the English Dre-

panodon (Machairodus) latidens. They have been reproduced
from figs. 8 and 2 of Plate F of an unpublished work by the
late Mr. McEnery on the fossil remains of Kent’s Hole’ Cave,
Torquay. The drawings are two-thirds of the natural size.
McEnery’s Plate has the following inscription:—‘ Teeth of
Ursus cultridens found in the cave of Kent’s Hole, near Torquay,
Devon, by the Rev. Mr. McEnery, January, 1826, in diluvial
mud, mixed with teeth and gnawed bones of Rhinoceros, Ele-
phant, Horse, Ox, Elk, and Deer, and bones of Hyznas, Bears,
Wolves, Foxes, &c.’ (See vol. ii. p. 460.)

VOL, I.

4)
Ml

a

‘SwBPNEL (| “QL,

Sy

FELIS CRISTATA. 317

poor, and of a younger animal almost exactly of the size of
the fossil.! The amount of wearing in the cheek teeth and
the condition of the sutures prove “that the fossil was full-
grown, although not aged.

Fossil Tiger Taree ee Taree, Tiber

English | French |English |} French |English | French
Inches | Métre | Inches | Métre | Inches | Métre

1. Extreme length from occipital
crest toalveoliof the incisors | 10:9 | :278 |14:1 | 358 |13:1 333
2. Ditto from anterior margin

occipital foramen to ditto .| 9:0 | °228 |11:7 | °298 |10°8 274

3. Extreme breadth of head at
zygomatic arches . 8:0 | :208 | 9°55 | *242 |10°0 253

4. From posteriorborderof palate
to the alveoli of the incisors | — _ 63 | 7180 | 5:9 149

5. From anterior margin of na-
sals to occipital crest . 92 | 233 |12°4 | -314 |11°85 | -300

6. From border of post- -orbital
process to ditto . ; 60 | 153 | 7:5 | °191 | 7-0 179

7. From ditto to the alveoli of
incisors. 4:9 | -124 6°6 168 | 61 156

8. Extreme height from base of

occipital foramen to occipital
crest . F 41 | +103 | 42 | °106 | 4:2 106

9. Width between outer surfaces
of occipital condyles . .| 22 | 055 | 2°3 | 058 | 2°45 | -062
10. Ditto of occipital foramen . | — — 1:15 | 029 | 1:25 | -031
11. Height of ditto . 3 = _ 0°8 020 | 08 020

12. Height of cranium from mid-
dle of frontal to the palatine
arch . _ — 4:6 ‘117 ~+| 4:25 | :108
13. Breadth on imaxillaries he
tween the canines and molars 3:0 | ‘076 37 093 | 3°5 088
14. Extreme length of 2nd and

3rd molars. F 2°2 | :055 2°32 | -059 | 2°2 "056
15. Breadth of maxillaries to rear

of carnassier : 4:6 | 114 54 1388 | 56 "142
16. Depth of infra-orbital " fora-

men . 4 5 . 0-75} 018 0:7 017 | 07 017
17. Width of dite 5 6 0°3 | :007 0-4 7010 | 0°375)| -094
18. Length of the nasal bones .| 28 O71 | 4:2 | °106 | 4:1 104
19. Length of carnassier tooth . 1:5 | :088 14 ‘0385 | 1°375| 034
20. Width of frontal between the

post-orbital apophysis .| 381 | 078 | 4:2 | 106 | 3:65 | :092
21. Length from frontal angle to

base of the nasals ; 2:9 | :073 3°6 090 | 2-9 073

We shall now make a comparison of the peculiarities of
the individual bones of the head with those of the Tiger.
This will make the points of difference in the fossil more
apparent.

The dentition of the fossil is as is normal in the genus.
There are six alveoli for the incisors, the two outer of which

-1 The dimensions of this younger animal were omitted from the paper as
originally published.—[ Eb. ]

318 FAUNA ANTIQUA SIVALENSIS.

are greatly larger than the intermediate, which are equal in
size. The canines are broken off at their base ; and the sec-
tion is seen to be the same as in the Tiger. The cheek teeth
had consisted of two false molars, a carnassier and a tuber-
culous tooth on either side. The tuberculous and the first
false molar had dropped out. The socket of the tuberculous
tooth is distinct on one side. Those of the first false molars
are more ambiguous. In all the specimens of the F. spelea
observed by M. Goldfuss,' this false molar was invariably
wanting, and he was induced to consider it as a specific dis-
tinction of the fossil. But Cuvier attributes its absence to
dropping out at an early period.

In our fossil we were at first led to think that there was
no small false molar, from the very contracted space between
the canine and the large false molar not admitting of room
for it. In the fossil the space is 0°3 inch, whereas in the
Tiger it is 0°7 inch. But on carefully clearing the interval
we have detected on the left side an alveolar cavity. In this
respect, therefore, the fossil does not differ from the ex-
isting large species. The great false molar and the carnas-
sier tooth resemble in form those of the Tiger exactly. But
they differ considerably in two respects, which we consider as
distinctive marks of the fossil. First, The length of the two
teeth in the fossil is exactly equal to that of the full-grown
Tiger No. 2, although it measures 10-9 inches in length of
head, while the Tiger is 13°1 inches ; Second, The large false
molar is directed inwards, so that its long axis makes a con-
siderable angle with that of the carnassier. This position of
the false molar holds in a slight degree in the genus Felis
generally, but it is véry marked in the fossil.

The obscurity of the sutures and the extent to which the
fossil is still enveloped in stone do not admit of our deter-
mining precisely the limits of many of the bones of the
cranium and face.

The frontal is considerably shorter than in the Tiger of the
same size, No. 3, and broader; so that it has more square-
ness of form, at the same time that the ridges stretching
from the post-orbital processes, by their prominence and
greater convexity inwards, give it an appearance of more
sharpness to the rear. These processes are more obtuse and
less salient than in the Tiger, and the outline of the frontal

1 Quoted in Cuvier, Oss. Foss., vol. iv. | all the teeth much worn, both the tuber-
p. 462. culous teeth and the small false molar

2 The presence, or absence, both of the | of one side are present. Upon the
tuberculous and the first false molar | opposite side all trace of alveolus has
appears to be very uncertain in aged | disappeared, whereas in younger animals
Feline. In a very old Tiger, No.1 of | with unworn teeth, we find the turber-
the measurements, with the canines and ' culous and first molars less complete.

FELIS CRISTATA. 319

portion of the orbit is less curved. The separate halves of
the bone are convex across, leaving a deep longitudinal hol-
low between.

The parietals are longer than in the Tiger. They are sunk
towards their commissure, giving greater appearance of
saliency to the sagittal crest. They bulge out at their junc-
tion with the temporals in conjunction with these bones.
The sagittal crest, as noticed above, from its great promi-
nence, is one of the most distinctive characters about the
fossil. It is nearly twice as much raised as in the largest-
sized Tiger. Its anterior extremity for a short way divides
into two, from running in continuity with the post-orbital
ridges of the frontal.

The occipital is large in all its dimensions. It greatly
exceeds that of the Tiger of the same size in height, and
equals that of the large Tiger, No. 2. Its margins expand
greatly laterally, in conjunction with the ridge ascending
from the petrous portion of the temporal bones.

The temporal on the right side is mostly concealed by stony
matrix. On the left it is broken at the zygomatic process.
The petrous portion is comparatively larger than in the Tiger.
The zygomatic arch is elevated and its lower margin is
arched more decidedly than in the Tiger. (In this respect
it resembles the ‘ Black Jaguar.”') The posterior angle is
more acute.

The corresponding process of the malar differs chiefly from
that of the Tiger in the post-orbital apophyses being blunt
and very slightly prominent.

The nasal bones are considerably shorter than in the
Tiger, and they taper less upwards towards their frontal
insertion.

The mawillaries chiefly differ from those of the Tiger, in
their greater prominence along the alveoli of the canines,
and the more decided hollow which occupies the infra-orbital
region. From the elevated direction of the zygomatic arches,
the posterior margin of the maxillaries descending from the
malars is higher than in the Tiger. The infra-orbital fora-
mina agree in form and position with those of the Tiger.

The ascending apophysis of the incisive bones runs higher
up between the nasals and maxillaries, giving thus a stronger
insertion to the bone.

We do not observe any very appreciable difference in the
palate, except that the bulge of the canines, and the inward
direction of the large false molars, appear to contract it in
width between these teeth.

1 Oss. Foss., vol. iv. Plate xxxiv. fig. 7.

320 FAUNA ANTIQUA SIVALENSIS.

The orbit is smaller than in the Tiger; the post-orbital
processes of the malars and frontal are more apart, and the
osseous ring consequently less complete.

Regarding the relations of the fossil with respect to other
species—

The fossil Lion of Gaylenreath, F’. spelwa, differs from it
in the great size, equalling that of the Lion in the outline of
the head, breadth of forehead, depth of zygomatic arches,
position of the infra-orbital foramina, and inconsiderable
sagittal crest.

The fossil F. antiqua differs in being greatly smaller.
The existing Lion is much larger, differing also in its recti-
linear profile, shortness of head, and want of prominence in
the sagittal crest and occipital.

The points of distinction and resemblance with the Tiger
have been noticed in detail.

The Panther somewhat resembles the fossil in the moderate
convexity of the head, but differs in size and in the little
prominence of the sagittal crest. The Jaguar has the same
elevated direction of the zygomatic arch as the fossil, but
differs greatly in the outline and form of the head.

The Cougouar has one character peculiar to it and the fossil
in the genus, in the face being shorter than the cranium ;
but it differs in size and form.

The other species of the genus differ at once in size.

We have named the fossil the Felis cristata, from its most
prominent character, the elevation of the sagittal crest.

The position in the genus will probably be after the Tiger.
Its size is intermediate between that species and the Jaguar.

We are indebted to Walter Ewer, Esq., of the Civil Service,
for an examination of the fossil, which was found by collec-
tors employed by that gentleman, under the direction of
Captain Cautley.

Norrseen Doas: April 15, 1836.

URSUS (HYAINARCTOS) SIVALENSIS.

XVI. ON URSUS SIVALENSIS, A NEW FOSSIL
SPECIES FROM THE SEWALIK HILLS.!

BY CAPTAIN P. T. CAUTLEY, AND HUGH FALCONER, M.D.

WE are now enabled to record another new form in fossil z0o-
logy drawn from the rich deposits of the Sewalik hills. Ina
preceding article we have noticed a new feline extinct species,
of dimensions approaching those of the existing Tiger; in
the present one, we shall endeavour to characterize another
member of the same family, of the genus Ursus, essentially
distinct from existing or extinct species in some prominent
points of its osteology, and remarkable also for large size,
like some other of its associated fossil contemporaries.

Our knowledge of the species is derived from two fossil speci-
mens. The one, consisting of the right half of the lower jaw
mutilated at the symphysis and ascending portion of the
ramus, exhibited in Pl. X XVI. figs. 3 & 4, gave us the first idea
of a new animal. The other, figs. 1 & 2, a subsequent acqui-
sition, is a superb specimen of the head, which, although a
good deal fractured, is at the same time so well preserved in its
principal features as to cause little difficulty in determining
the specific characters. The three rear molars are perfect on
one side and but little damaged on the other. Both canines
are present, and that of the right side is entire. The alveoli
of the false molars and incisors are distinct, although the
teeth are wanting. The only considerable deficiencies are in
the posterior and lower parts of the occiput, both zygomatic
arches, and in the lower end of the nasals, where a fissure
extends across the face on both sides towards the orbits.

The chief peculiarities of the fossil are to be found in the
teeth, which are constructed more after the type of the higher
Carnivora? than any other described species of the genus.

1 This paper was published in 1836

_ in the ‘Asiatic Researches,’ vol. xix. p.

193, unaccompanied by illustrations.
The illustrations in Pl. xxvi. are copied
from an unpublished plate (O.) of the
‘Fauna Antiqua Sivalensis, executed
about twelve years later, and in which
the fossil is designated Hyenarctos
Sivalensis. In this plate are figured the
femur, radius, ulna, and other remains

VOL. I. XG

of the fossil in addition to those des-
cribed in this memoir. In the same
plate are figured the portion of an upper
jaw with four molars and a tibia of a
smaller species of Bear from the Ner-
budda, named Ursus Namadicus (see
Pl. xxvi. fig. 5).—[ Ep. |

2 Underlined in pencil by Dr. F.—
[Ep.]

Bae Ce ers: A ial

322 FAUNA ANTIQUA SIVALENSIS.

Before entering upon these it will be convenient, for com-
parison, shortly to refer to the dental system of the Bears
generally.

The number of the teeth varies more in the different species
of the Bears than in any other genus of the Carnivora. The
incisors, canines, and the three posterior molars are constant
in both jaws in all the species. The false molars are subject to
ereat variation. They are usually entirely wanting! in the
upper jaw of the Ursus speleus, or large fossil Bear of Europe,
while in other species they amount to three on either side.
The same variation holds in the lower jaw. In the Ursus labi-
atus? there are four false molars on each side, while in the Ursus
speleus there is generally but one present: so that while the
number of teeth in the former extends in both jaws to 42, it
is commonly reduced in the latter to 30. This irregularity
in the number of false molars exists to a certain extent in
different individuals of the same species. In one skull of
the Ursus Tibetanus in our collection, which belonged to
an old animal, there are three false molars in the upper jaw
on each side ; while in the skull of a younger individual with
unworn teeth there are but two. In another skull there are
no false molars on the left side, while there is one on the
right side, there being at the same time three false molars
on either side of the lower jaw.

The characters depending on the form of the teeth are
very constant in the genus. The carnassier or antepenulti-
mate molar, in the upper jaw, has but two lobes or points
along its length; and the tubercle of its inner side is placed
opposite the rear lobe. The two rear molars are oblong, and
the last one has the additional development of a crenulated
spur or heel. The higher Carnivora differ widely in these
characters from the Bears; their carnassier teeth being three-
lobed, and the tubercle of the inside being placed forwards,
as in the Cats and Hyzna, while the rear or tubercular
teeth are reduced in number and are rudimentary.

We shall now proceed to the teeth of the fossil, which in
many respects deviate from the type of the genus, and
approximate that of the more perfect Carnivora. Along with
the dimensions we annex those of the Ursus speleus and
Ursus Tibetanus for comparison.’

1 Cuvier, Ossemens Fossiles, tom. iv. | have to regret this: omission, as one of
p. 351. the most marked characters about our
2 This is constant in several skulls | fossil is the great width of the molars;
in our collection. and we have no other source to refer to
3 In the ‘Ossemens Fossiles, Cuvier | regarding the Ursus speleus besides the
does not give the dimensions in width | ‘Ossemens Fossiles.’
of the teeth of the Ursus speleus. We

URSUS (HYAINARCTOS) SIVALENSIS. 328

Ursus Sivalensis | Ursus spelzeus | Ursus Tibetanus

Molars, Upper Jaw
In. Mét. In, Met. In. Met.

Length of last molar . : 11 028 | 184 | 048 | 11 028
Width of ditto : F 1:2 | 03 — — | 06 015
Length of penultimate ditto pele 2a | 038 1-23 | :031 | 0°8 02

Width of ditto. . : pleolede alee Oo — — | 0°55 | 013
Length of antepenultimate ditto . | 1:3. | 032 | 0°83 | 021 | 05 | -012
Width of ditto : 0°85 | °022 = 0°35 | -008

Molars, Lower Jaw
Length of last molar .

Width of ditto “ 5 = — — — | 0:48 | :012
Length of penultimate ditto . | 1:15 | °029 | 1:30 | :032 | 0°8 02
Width of ditto . ° : : | 0°75 | <019 — — 0°45 O11
Length of antepenultimate . 1°35 | °0335)| 1:32 | 033 | 0°8 02
Width of ditto ; F = — — — | 0°38 | :095
Length of last false molar . 6 || OS 023 — — | 04 ‘01
Width of ditto . : : | 05 | °012 — — | 025 | -006

The incisors as indicated by the alveoli were six, and the
external one of each side larger than the others, as is usual
in the family. The canines are of great size. The right one
is entire: its point is worn off, indicating the animal to have
been more than adult, and there are also stripes of wear bothon
its posterior and inner sides (fig. 2). It is 1-4 inch in length in
antero-posterior diameter at the base, and 1 inch transversely.
The socket of the first false molar! is close behind the canine,
that of the second is near the anterior one, and the tooth appears
to have been two-fanged.?, There can have been no other false
molars besides these two, the sockets of which are close
together, and occupy the interval between the canine and
carnassier, which is inconsiderable for the size of head,
being but 12 inch in length.

The three rear molars present marked peculiarities. The
antepenultimate or carnassier is of very large size ; it slightly
exceeds both of the rear teeth in length, and is about half an
inch longer than the corresponding tooth of the Ursus speleus.
Instead of having but two points like the rest of the Bears, it
has three,? the anterior lobe being well developed as in the
higher Carnivora ;! and the tubercle of the inside, instead of
being to the rear as in the other species, is advanced forwards
opposite the middle lobe. It has altogether a great analogy

1 Note in pencil by Dr. F—‘ This pre-; ? Underlinedin pencil by Dr. F.—[Ep.]
molar in all the true Bears is described 3 Underlined in pencil by Dr. F.—| Ep. |
as haying but a single fang, and in this 4 Note in pencil by Dr. F.—‘ Such as
view the two alveoli would imply two | the Hyzna.’—Ep.]
premolars, here reckoned as one. —[Eb. |

¥ 2

324 FAUNA ANTIQUA SIVALENSIS.

with the corresponding tooth of the Hyzna. The teeth of
the opposite sides are unequally worn.

The two rear or tubercular molars are also marked in their
form. Instead of being oblong, as in all the other species,
with their length greater by a third than the breadth, they
are square in our fossil. The penultimate, if anything, is
longer than the rear one, the reverse of which holds in the
rest of the genus. It has two tubercles at its outer side, as in
other species; at the inside it is somewhat shortened in
length, and the cleft between the tubercles is nearly obsolete,
so as to give the appearance of one large tubercle. In this
respect there is a remote analogy with the corresponding tooth
of the Dog, and a deviation from the usual type of the Bears.
The last tubercular presents as great a contrast in form with
that of the other species as the carnassier tooth does. At its
outer side there are two tubercles to the crown smaller than
in the penultimate, as is normal in the genus, and at its
inner side a ridge indistinctly divided by three slight furrows.
There is no heel to the tooth; the crown is square, and the
only part which can be considered as representing a heel or
spur ! is a flattish disc at the inside, alternate with the pos-
terior outer tubercle, and partly opposed to the rear portion
of it. None of the rest of the Bears have the last tubercular
in the upper jaw square, or without a crenulated spur added
on to the rear of it.

Our specimen of the lower jaw (figs. 3 & 4) is deficient in the
incisors and in the protruded portion of the canine. It is
broken off just where the latter emerges from its socket.
The section of the embedded portion of the canine gives 1°6
inch of vertical dimension and *95 transversely. The molars
are six in number. The two anterior false molars and the
last tubercular have dropped out, but the sockets remain
unobliterated. The anterior false molar was close behind
the canine, and there is not space for another to have been
inserted between. The second was close to the first and
almost in contact with the third false molar. This latter,
like the carnassier of the upper jaw, is of large size, compared
with the same tooth of the other species, and distinctly three-
lobed; this is another peculiarity, and further supports the
analogy shown by the upper carnassier with the Hyzna, to
the second false molar of which it bears a considerable resem-
blance. The anterior and posterior lobes are small, the
middle point being chiefly developed. The antepenultimate
or carnassier is so defined? as to give no indication of form to
notice, except its length. The penultimate or first tubercular

1 Note by Dr. F.—‘ Posterior or talon | 2 Underlined in pencil by Dr. F.—
lobe.’ —[Ep.] [Ep.]

Fig.

Fig.

3.

DESCRIPTION OF PLATE XXVI.

Ursus (Hymnarctos) SIVALENSIS AND URSUS
NAMADICUS.

. Profile view of cranium of Hyenarctos Sivalensis, one-fourth of

the natural size. The specimen is in the British Museum (No.
39,721), and the figure has been reproduced from an unpub-
lished Plate of the Fauna Antiqua Sivalensis. (See pages 321
& 551.)

. Shows the dental series on the right side of the same specimen

as that shown in fig. 1, three-fourths of the natural size. The
three rear molars and the canine are perfect, and the alveoli of
two false molars are also seen.

Greater part of the body of the lower jaw of Hycnarctos
Sivalensis, one-fourth of the natural size and seen in profile. |
The specimen is numbered 39,722 in the Brit. Mus. Catalogue,
and has been copied from a drawing by Mr. Ford in an un-
published Plate of the F. A. 5S. (See pages 324 & 551.)

Alveolar margin with dental series of same fragment of jaw,
three-fourths of the natural size.

Portion of upper jaw, with four molars of a smaller species of
Bear from the Nerbudda, Ursus Namadicus, one-half of the
natural size. The specimen is in the British Museum, No.
39,720, and the drawing has been reproduced from an un-
published Plate of the Fauna Antiqua Sivalensis.

VOL. I. d

Vol. I. Plate 26.

| GH Ford del. JDizikel ith

bs
1-4.. Ursus (Hyeemarctos) Sivalensis. 5. Ursus Namadicus.

URSUS (HYHNARCTOS) SIVALENSIS. 325

molar is oblong. It is broader for its length than generally
holds in the genus, and the crown is less complicated with
tubercles. Of the rear tubercular the socket alone remains,
the tooth having fallen out. It is situated with considerable
obliquity to the rest of the series in the root of the ascend-
ing portion of the ramus. The alveolus is inconsiderable,
and the tooth appears to have been comparatively small.

The teeth in tee fossil appear to have been thus: incisors,

343 ‘id 22 3+3
374: canines, {: false molars, 55: cheek teeth, 5_4: in
all, 38.

The size and form of the head bear out the specific distine-
tion established by the teeth. No Bear, fossil or recent,
attains the enormous size of our fossil, except the Ursus
speleus, and the absence of any bulge in the forehead above
the orbits at once distinguishes it from the latter. The
mutilation of the cranium at the occiput prevents an exact
comparison of the length with that of the Ursus speleus. In
the tables of the ‘Ossemens Fossiles,”! an adult specimen of the
latter measures 17-9 inches from the incisors to the occipital
erest. The fossil cranium, although mutilated at the occiput,
measures 17 inches; with the deficient portion restored, it
would probably measure 19 inches. The facial half of the
head, from the post-orbital processes to the incisors, mea-
sures 9°3 inches; and in almost all the Bears the cranial
portion is longer than the facial. Supposing this proportion
to hold in our fossil, the head would be more than 19 inches,
and would exceed that of the Ursus spelceus.

The form of the cranium in profile is shown in fig. 1. The
most striking feature is the almost rectilinear outline, and
absence of any notable curvature. From along the nasals to
between the infra-orbital processes is almost a straight line.
There is but a trifling degree of convexity from that back-
wards; and the sagittal crest rises in a very prominent ridge
above the parietals. No species of Bear has so straight a
cranium. The Ursus speleus is chiefly characterized by a
bulge of the forehead above the root of the nasals. The only
species which at all approaches the fossil in profile is the white
Polar Bear, Ursus maritimus.? But besides the great differ-
ence of size, the latter has nothing of the salient sagittal
crest, which is so prominent in the fossil; all the other Bears
have more or less convexity of profile.

Exclusive, therefore, of the teeth, the size and cranial out-
line would suffice to establish the fossil as a distinct species.
The other peculiarities of the head are three. The frontal

1 Tom. iy. p. 359, 2 Ossemens Fossiles, tom iy. Pl, xxi. fig. 4.

326 FAUNA ANTIQUA SIVALENSIS.

is very broad, although an accidental indented fracture on
the brow takes off from the measurement. The orbits have
considerable obliquity, and large size, the depth from the post-
orbital process to the malar margin being 3:1 inches. Their
anterior margin only advances to above the posterior surface
of the rear molar. The temporal ridges are but slightly
marked, and meet at an open angle as in the Ursus speleus.
From their commissure backwards there is scarcely any sink,
the sagittal crest starting with great prominence from the
parietals. The crest is not complete in its whole length,
being broken off obliquely towards the occipital. The parie-
tals swell out backwards and downwards. ‘The cranial cavity
appears to have expanded much laterally towards the occi-
put, which is broken off. The temporal fossze are of great
depth and extent; the zygomatic arches on both sides are
wanting; judging from the depth of the temporal fosse they
must have been of great expanse. The malar apophysis
comes off low and is 24 inches in height. The nasals are
partly removed by a fissure in the fossil extending across the
face. They appear to have been rather long, and the external
opening of the nostrils to have been much shorter and less
oblique than in the Ursus speleus. The muzzle is broad and
obtuse, being about one-fourth of the length of the head,
and a little wider than the inter-orbital portion of the frontal.
In this respect it resembles the Ursus labiatus. The palate
is strongly arched both longitudinally and transversely; the
ereatest depth from a line across the worn molars to its sur-
face being 2°3 inches. The horizontal plate of the palatines
hardly extends an inch beyond the rear molar, whereas in a
head of the Ursus labiatus, measuring a foot in length, it
extends more than two inches. The palatine sinus is also
proportionably narrow for the size of head.

In place of a single sub-orbital foramen there are three
distinct foramina, nearly of the same size, placed over each
other and a few lines apart (Pl. XX VI. fig. 1). They are con-
siderably advanced on the jaw, the uppermost being 1:6 inch
from the margin of the orbit and placed over the carnassier.
It is difficult to say whether this is common to the species, or
merely an individual peculiarity. Nothing of the sort is seen
upon the heads figured in the ‘Ossemens Fossiles.’

We only know the lower jaw by the fragment represented
in fig. 3. It consists of the greater part of the body of the
right side, broken off where the canine protrudes. It is also
deficient in the articulating and coronoid processes. There
is, therefore, little to remark about the form. The lower
edge has a good deal of curvature backwards, and the outer

URSUS (HYAINARCTOS) SIVALENSIS. 327

surface is deeply indented by a muscular hollow towards the
angle. The dimensions of the fragment are thus :—

Extreme length of the fragment . 4 ; 5 > . 10°3 inches.
Height of jaw over first falsemolar_ . : 3 : ae 2ROm ys
Ditto between the two rear molars 3 § ; é RBH 55
Greatest thickness at rear molar . 6 ; ; Al ORE) cp

We have not yet found out or identified any bones of the
trunk or extremities. The species does not appear to have
been abundant, as no other specimens of the head or teeth
have been discovered, so far as we know, among the immense
collections of fossil bones got from the Sewalik hills.

The dimensions of the cranium, contrasted with some of

the measurements of the Ursus speleus for comparison, are
thus :—

Ursus Sivalensis| Ursus spelzeus
Dimensions of Cranium na ea eee ee

In. Met. In, Mét.

From the incisives to the occipital crest! . — — | 17:9 | 457
Width of cranium between post-orbital pro-

cesses . : . | 546 | 7139 4:7 | °121

From the incisives to: a line between ditto : i) het = 4) Ri 8:6 | 245

From occipital crest to the same point! -| — — ‘| 11-2 | :285

Width of brow between the orbits . oi) Ag 1 20 —_ —

Ditto of muzzle over the canines 48 | 122 — —
Length from the alveoli of the incisors to the

posterior margin of the palate. 73° | +186 _ —
Width of palate in the interval between the car-

nassier molars, 2 c : . | 3°35 | 085 — —

Interval between the canines . ° - eH deg 7) _ ae

To conclude. It follows, that there existed along with the
Mastodon, Sivatherium, fossil Camel, &c., of the Sewalik
deposits, a large distinct species of Bear, equalling if not
exceeding the largest known of the genus. Its teeth de-
viated widely from the type of the genus, approximating it
more to the higher Carnivora than to any other species ; the
carnassier teeth of the upper jaw connecting it with the
Hyena, while the tuberculars have a more remote analogy
with those of the Dog. The size and extent of the temporal
fossee, and the prominence of the sagittal crest, taken in
conjunction with the teeth, show that it had a more strictly
carnivorous than a frugiverous habit.

We have designated it Ursus Sivalensis.

1 These two measurements are in-| length is 17 inches, or 434 métre, the
complete in the fossil, from the mutila- | second 7:7 inches or ‘197 métre.
tion at the occiput, The first or extreme

328 FAUNA ANTIQUA SIVALENSIS.

APPENDIX.

I.—Manuscript Memoranpum sy Dr. Fatconer, 1842.

In a communication read before the Academy of Sciences, on the 26th
July last (1841), contained in the ‘ Comptes Rendus’ of that date, M. de
Blainville has given a revision of the animals allied to the Badger,
which have been separated from the genus Ursus of Linneus. He takes
up the whole of the species, existing and fossil, comprising them in a
group to which he applies the name of Subursus, which he ranges in
three Sections:—WMeles, or Badgers; Procyon, or Racoons; and Ratel
and Caudivolvulus, including Arctitis and Cercoleptes, the Binturong and
Kinkajou. At the head of this group, preceding the Badger, but
excluded from any of the three sections, he places the Ursus Sivalensis
of Fale. and Caut. described in the 19th vol. of the Asiatic Researches,
raising it to the rank of a distinct genus, to which he provisionally
assigns the name of Sivalarctos. It is strange that M. de Blainville
should have adopted this term, while convinced against its being a Bear.
Tf he thought it nearer the Badger, Sivataxus or Sivameles, should such
combinations be admissible, would have been more appropriate. But
we cannot assent to his conclusion. ‘The Sewalik species still appears to us
to be nearer the true Bear than any other of the Plantigrade Series.
M. de Blainville’s remarks on the subject are thus: ‘ Dans ce groupe,’
(Subursus ou les Petits Ours), ‘réduit aujourd’hui en Europe au seul
Blaireau, et dont quelques ossements ont été recueillis dans le diluvium
des cavernes, j’ai eu l’avantage de trouver 4 /état fossil, outre les
traces de plusieurs autres especes, les indices certains de cinq formes
animales assez différentes pour constituer cing genres distincts dans la
maniére admise par les zoologistes modernes, c’est-a-dire, d’apres le
systéme dentaire (p. 165).

‘Le premier, que je désigne sous le nom de Sivalours, est établi,
d’aprés la description détaillée et comparative, avec mesures linéaires,
d’une téte presque entiére pourvue de ses dents, publiée par MM. Hugh
Falconer et Cautley, sous le nom d’Ursus Sivalensis, et qui differe
notablement par la forme, la proportion et le nombre des dents
molaires, en général plus carnassiéres, de toutes les espéces d’Ours
aujourd’hui connues. Cette téte provient des terrains tertiaires des
monts Sous-Himalayahs dans Inde” . . . . . Nous pouvons
encore rapporter probablement & ce groupe le singulier fossile que
MM. Cautley et Falconer ont décrit sous le nom d’ Ursus Sivalensis,
qui n’appartient certainement pas 4 ce genre, que nous avons désigné
provisoirement sous le nom. de Sivalarctos, et dont les ossements ont
été trouvés dans cet immense dépdt tertiaire de conglomérat qui
remplit toutes les pentes et les excavations des monts Sivaliens ou Sous-
Himalayahs (p. 179).’ In another place he calls the fossil ‘cette téte
d’Ours prétendu des monts Sivaliens.’

We have here M. de Blainville’s opinion given in two degrees of
affirmation: in the first instance that the so-called Ursus Sivalensis
differs in its teeth from all known Bears, and constitutes a distinct
genus of the Subursine group; in the second, that it certainly does not

ae

URSUS (HYANARCTOS) SIVALENSIS. 329

belong to the genus Ursus, but probably to the Subursine group. We
beg leave to take exception to both inferences.!

IJ.—Extract rrom Dr. Fatconer’s Nore-Boox.

(British Museum, Nov. 21, 1857.)

Madé a rough comparison of a huge ursine head from South America,
(Bravard, Buenos Ayres) referred to by Lartet, with the Hyenarctos of
the Sewalik hills.

Upper Dentition.—All the teeth on their alveoli present on both
sides, and the number exactly identical with that of the Hycenarctos
Sivalensis.

1. Asmall premolar, touching the canine, is present on the right side.

2. A two-fanged premolar, the alveoli only remaining on both sides.

3. A tooth resembling in general form the third of Hycenarctos, but
more worn and broader for the length. Instead of presenting three ex-
terior cusps, the flat summit, which is much worn, only shows the discs,
which are confluent, of two lobes, the anterior small cusp, which is so well
defined in Hycnarctos, being wanting; but there is the great agree-
ment of a mesial inner cusp placed opposite the transverse axis, as
in Hyenarctos. Its worn disc is confluent with that of the outer cusps.

4, A square molar, consisting of two outer and two inner cusps, the
outer pair having the discs confluent, and also the inner. It is more
moderate in form than the Sewalik tooth. This fourth resembles in
form the last of Hyenarctos.

5. A last molar, as in Hyenarctos, which, instead of being quadrate,
has the inner posterior discous surface produced behind, so as to make
the tooth oblong instead of square. The outer line, as in the Sewalik
fossil, consists of two tubercles.

Lower Dentition.—In number, the lower teeth are also the same.

1. A small one-fanged premolar touching the canine (not seen from
fracture in Hycenarctos).

2%. A two-fanged premolar, shown by the alyecd which are
more approximated than in the Sewalik fossil, and are straight fore and
aft (upper being diagonally oblique); doubtful if one or two.

3, A premolar, which is rounder, broader, and thicker, as in the
upper jaw, than the corresponding tooth of the lower jaw in Hycenarctos ;
in the latter, it is three-cusped, but in the American fossil, bi-cusped ;
front cusp wanting, middle cusp very large.

4. A carnassier which, as in the Sewalik form, is the most elongated
of all. In the Sewalik specimen the crown is hammered off, so as not
to admit of comparison.

5. A tooth homologous in form to the penultimate of Hyenarctos,
but thicker.

6. A last or tubercular, which is worn down to a roundish disc, as
in the Bears, is the smallest of the three, and not much larger than the
third premolar.

This tooth in Hycnarctos is only shown by its alveolus.

1 In an unpublished plate, however, | signated Ursus (sub-genus Hyenarctos)
of the ‘Fauna Antiqua Sivalensis,’ ex- | Stvalensis.—[Ep.]
ecuted about 1848, the species is de-

330 FAUNA ANTIQUA SIVALENSIS.

Upper canines very large and thick: incisors six; outer with a broad
top; middle very much compressed across; they are very much and
flatly worn; so also are the canines, but unequally on the opposite sides,
the left being ground nearly to the level of the incisors; the converse
with like characters in lower jaw, left canine being most worn.

The form of the skull differs much from Hyenarctos; it is much
broader, deeper, shorter, and thicker. 'The Sewalik form is more elon-
gated in every direction.

This difference is strongly shown in lower jaw. The Sewalik is thin
and high; the American fossil is thick and massive. The American was
a larger form, and was probably more heavy and less lithe.

From the very decided flat grinding of the teeth, the American fossil
was probably a vegetable feeder; the Indian was evidently an animal
feeder.

ENHYDRIODON. 331

XVII. ON ENHYDRIODON (AMYXODON), A FOSSIL
GENUS ALLIED TO LUTRA, FROM THE
TERTIARY STRATA OF THE SEWALIK
HILLS.’

BY H. FALCONER, M.D.

Tint comparatively lately naturalists were only acquainted
with the aquatic Mustelide through the true Otters ; and the
few species known were so much alike, and so well defined,
that Cuvier ranked them merely as a sub-genus of Mustela.
Subsequent researches brought to light a remarkable devia-
tion from the ordinary type, in the dentition of Enhydra.
The genus which we are now to describe exhibits a still
wider departure from the Otter form in its teeth, the
carnassier being so singularly modified that, when detached,
it is with difficulty recognized to be that carnivorous tooth.
At the same time the other characters of the head are so
clear and marked that not a shadow of doubt remains in
regard to the animal having been a true lutrine Mustelida.
The genus is further remarkable in the tribe for the size
which it attained, one of the species having been nearly as
large as the Indian panther.

The specimens on which the determination rests are three
heads (see Pl. XX VII.), which taken together are so com-
plete in regard to the teeth and general form of the cranium,
that little remains to be desired respecting the more import-
ant characters, except the form of the articular surface for
the lower jaw, which is not shown in any of the specimens.
There is also a fragment of the lower jaw, containing the
two rear molars, but it wants the ascending ramus and
condyle.

1 This paper was written in December
1843, but was never published. The
three heads referred to in the memoir
are now in the British Museum, and
were figured in an unpublished plate of
the ‘Fauna Antiqua Sivalensis,’ (P.)
On this plate the species is named
Enhydriodon ferox. The specimens
were afterwards named by Dr. Falconer
on labels in the British Museum,
Enhydriodon Siwvalensis. On the same

plate of the Fauna a skull and a lower
jaw of another new species of Otter
are figured, under the designation of Lzw-
tra Paleindica. These specimens are
also in the British Museum, Cat. Nos.
37,157 and 37,152 (see description of
Plate P.). The annexed drawings in
Pl. xxvii. have been copied from the
unpublished plate of the Fauna, above
referred to,—[ Eb. ]

332 FAUNA ANTIQUA SIVALENSIS.

In order to follow the relations of the teeth of Enhydriodon'
it is necessary to review the dental system of Lutra and

Enhydra.

The formula in Lutra is: incis. a can. Ss ; mol. 2+?
ae In the Enhydra (as shown in the solitary skull
preserved in the Museum of the Zoological Society) :
incis. aEHO)e can. a = mol. a = 32. The dental sys-
tem of the Otters is built very much on the plan of the
Martins (Fred. Cuvier, in ‘ Ossemens Fossiles,’ art. ‘ Dentition
of the Carnassiers,’ vol. iv.) with which it agrees numerically,
except in having a false molar less on either side of the
lower jaw. But in the signification of the teeth they differ
materially in the increased development of the tubercular of
the upper jaw, and in the greater degeneration, so to speak,
of the carnassier. The typical tubercle on the inside of the
latter, instead of being limited to a small knob connected
with the body by a narrow base, constitutes nearly half the
surface of the coronal, and is expanded into a wide disc,
bounded on its inner side by a sharp raised edge, occupying:
the whole length of the inside of the tooth. The outline of
the carnassier is in consequence nearly triangular. The body
is still distinctly tricuspid, as in the higher Carnivora, but the
anterior cusp is reduced to little more than a well-marked
serrature or lobe of the basal ridge. The tubercular has a
development proportionate to that of the tubercle of the
carnassier. It is somewhat trapezoidal in the outline of its
coronal, which is oblong in the transverse direction, that is,
considerably broader than long; it is divided lengthwise by
a deep hollow into two somewhat unequal halves, the outer
and smaller of which is subdivided by a shallow transverse
channel into two flattish surfaces bounded by a raised edge ;
while the inner is expanded into a flat disc, bounded by an
edge, as in the tubercle of the carnassier, but it is of greater
extent and more complicated in form. This arises from the
anterior border of the coronal being raised up into a promi-
nent trenchant ridge divided into two denticles, and distinct
from the bounding basal ridge which sweeps round it.

In the lower jaw the tubercular is of comparatively small
size, nearly square, and its coronal is divided by a transverse
low ridge into two flattened nearly equal surfaces. The car-
nassier may be considered as made up of two parts, separated

1 So named from évvdpis the Greek | modern Enhydra, which has a similar
word for Otter, and dd0vs. The | derivation.
name has no special reference to the

ENHYDRIODON. 333

by a deep transverse hollow, the anterior of which is formed
of three sub-equal pointed cusps disposed in a triangle, the
inner one representing the tubercle of the corresponding
upper tooth; the posterior portion consists of a dilated
flattened tubercle, sloping inwards, and bounded by a sharp
edge, which is raised at the outer side into an obsolete
posterior cusp, less distinctly marked at the inner side.

The false molars, canines, and incisors of the Otter agree
so closely with those of other Mustelide as not to require
special notice. In regard to the ‘wear’ of the molars gene-
rally, we may remark that from age and use they get blunted
and irregularly ground down; but the cusps never wear off
into the truncated depressed transverse discs exhibited in
the teeth of Hyzenas, a remarkable approach to which takes
place in those of Enhydriodon, to be noticed more particularly
in the sequel.

Tn the Enhydra the teeth undergo a singular modification,
both in number and configuration. It has normally the
Same amount of molars in the lower jaw as the true Otters,
and one false molar less in the upper jaw, so that the

dental formula, if regularly followed out in the canines and
incisors would be =
there are only four incisors in the lower jaw, the outer and
largest one on each side being wanting. The canines are
only remarkable for their small size in relation to the other
teeth and the bulk of the head.

But it is in the form of the molars of Enhydra that the
ereat deviation from the type of the Otters is observed. They
are large, robust, and amorphous: the coronal being com-
posed of a very thick layer of enamel. Everything like the
pointed cusps and trenchant edges of the Otters’ teeth has
disappeared, and given place to a flattened, sinuous, and
obtusely tuberculated surface, like the worn tubercular teeth
of the Bears. This is not a result of age, but is shown in
the young animal when the sutures of the head are still
open. Notwithstanding this disguise, the amorphous-like
inequalities can be traced to a modified representation of the
cusps and lobes of the teeth in the Otter. The upper
carnassier is of a triangular form, divided by a longitudinal
furrow, the outer portion being formed of two very obtuse
rounded tubercles, the anterior of which is the largest,
representing the central cusp of the carnassier of the Otter ;
but there is no trace of the anterior lobe of the latter. The
internal portion is still larger than in the Otter, and consists
of an expanded disc raised into a third obtuse tubercle, and
bounded by a blunt basal ridge. The upper tubercular is of

=34; but from a peculiar suppression

334 FAUNA ANTIQUA SIVALENSIS.

an irregular ovate form, and has the same obliquely trans-
verse direction as in the Otter. The surface is so flat and
sinuous that it is difficult to recognize its constituent
elements; but it is divided, as in Lutra, by a shallow
furrow; the outer portion consisting, as in the latter, of two
unequal tubercles; the inner, which is much larger, is
elevated into three obtuse tubercles, arranged in a triangle,
and surrounded by a blunt ridge.

In the lower jaw the tubercular is comparatively small, as
in the Otter. It is oblong across, and umbilicated, with four
indistinct surfaces surrounding the depression. The car-
nassier is of a broad ovate form; the flattened surface of
the crown being raised into four inequalities, representing
the three cusps and heel of the corresponding tooth in the
Otter. The false molars in both jaws are only remarkable
for their robust and obtuse form.

These minutize may appear tedious, but we have found it
necessary to enter into them so much at large, as the den-
tition of the fossil genus occupies in certain respects a mean
position between that of Lutra and Enhydra. In _deter-
mining fossil animals no assistance is derived from the soft
parts, or external characters; the means for their identifica-
tion are often limited, as in the present instance, to an
isolated part of the skeleton; and the anatomist has it
forced upon him to go rigidly into the comparison of every
character, in order to be able to apply with any certainty the
laws which regulate correlative organization to the recon-
struction of the form, and the determination of its affinities.

Now, in regard to the teeth of Hnhydriodon. The speci-
mens, fortunately, supply information regarding all those of
the upper jaw, even to the deciduous precanine false molar.
The incisors (Pl. XXVII. fig. 2) are of the normal number,
three on each side, the two interior of which are shown by
their transverse section to have been very compressed, their
length being three times their width. This compression, so
much greater than what is seen in either Lutra or Enhydra,
or in any other described Mustelida, is palpably connected with
the enormous development of the outer incisor on either
side, which relatively exceeds that of any known Carnivora.
This lateral incisor evidently served as a subsidiary canine,!
and the only analogous case which occurs to us is found in a

1 From this canine character of the | fossil mammiferous genera. It is men-

outer incisor, we were induced, in the
first instance, to name the genus
Amyxodon, or, par excellence, Lama-
rian, but subsequently adopted that of
Enhydriodon as significant of its affini-
ties, a useful consideration in naming

tioned under the name Amyxodon in a
catalogue of Sewalik fossils in the
Journal of the Asiatic Society of Bengal,
vol. iv., 1836, p. 707, and in Dr. Royle’s
‘Tllustrations of the Botany of the
Himalayahs,’ vol. i. p. 31.

— ee TT

DESCRIPTION OF PLATE XXVIII.
ENHYDRIODON SIVALENSIS AND LutTrRA PALAINDICA.

All of the figures in this Plate have been reproduced from
drawings by Mr. Ford in an unpublished Plate of the Fauna
Antiqua Sivalensis. Figs. 1 to 5 are rather less than one-
half (about 4) of the natural size; figs. 6-8 are two-thirds of
the natural size. All of the specimens are in the British

Museum.

Figs. land 2. Enhydriodon Sivalensis. Anterior portion of cranium of
a young and probably female individual. B. M. Cat. No. 87,154.
(See pages 334 & 553.)

Figs. 3 and 4. Enhydriodon Sivalensis. ‘Two views of cranium, pro-
bably of a female. B. M. Cat. No. 37,153. (See pages 331
& 552.)

Fig. 5. Enhydriodon Sivalensis. Anterior portion of cranium of an
old individual, with very perfect alveolar ridges. B. M. Cat. 4
No. 87,155. (See pages 331 & 553.)

Figs. 6 and 7. Lutra Paleindica. Two views of cranium with very —
perfect alveolar ridges. B. M. Cat. No. 37,151. (See page —
552.)
Fig. 8. Lutra Paleindica. Lower jaw with a large and very perfect —
carnassier. B. M. Cat. No. 37,152. (See page 552.)

VoL, I.

Vol. 1. Plate 27.

6.8, Lutra Palzeindica.

Ss.

iodon Stval

tees

GH¥ord del. JDi

ENHYDRIODON. 3395
very different family, the Ruminantia, where the upper
lateral incisor puts on the form and development of a canine,
in the Camel. In one of the specimens (fig. 2), the base of
this tooth, with the other incisors, remains in the jaw, broken
off on a level with the alveolus, and enables us to determine
its relative size with precision. The section of the fang
shows a very broad oval; the dimensions, as contrasted with
those of a large Indian Otter, are! :—

Enhydrio-| Indian

on Otter

Inch Inch
Antero-posterior diameter outer incisor. ‘ a “45 All
Transverse ditto ditto : 37 “12
Antero-posterior diam. middle incisor : 33 14

Transverse ditto : < . : 14 ‘1

Antero-posterior diam. of inner incisor . : "25 09
Transverse of ditto . : : 5 09 08

Tn the younger animal, to which the head (fig. 2) belonged,
the incisors are all present; but in the older head (fig. 5),
the middle incisors are not only wanting, but the alveoli are
completely filled up and obliterated, there being nothing but
a blank space between the outer incisors. In this deciduous
character of the middle incisors Enhydriodon agrees with the
modern Ursus (Prochilus) labiatus, or so called Sloth Bear of
India, in which the incisors drop out as the animal increases
in age. None of our specimens show what the number and
character of the incisors were in the lower jaw; but such a
strongly-developed lamarian tooth in the upper must in all
probability have had something corresponding to oppose it
in the lower jaw; we therefore infer that there were two
large outer incisors, one on each side, below, and that the
four middle incisors were also present, for in Enhydra, where
there is a suppression of two incisors, it is the outer one on
each side that disappears, the central ones remaining. This
is proved by the form of the exhibited incisors.2

The canines of the upper jaw, like the lateral incisors,
were proportionately large, and of great strength and mas-
siveness. A section of the right one is got in the head

fig. 5, of a circular form. The dimensions are, antero-posterior
diam. 54 lines, transverse do. 43. In the head, fig. 2, both

‘ The measurements, which were | June 14, 1836, appears to consider that

omitted from the original MSS, have
been filled in by me from specimens in
the British Museum.—[Ep.]

2 wr. Martin, in the abstract of his
paper on Knhydra, contained in the
Proceedings of the Zoologl. Society for

the suppressed incisors in the lower jaw
of Enhydra belong to the middle ones,
But a close examination of the form of
the teeth leads to the conclusion stated
above.

336 FAUNA ANTIQUA SIVALENSIS.

the canines had dropped out, and the two alveolar cavities
are exposed, showing that the fang was comparatively short,
and much dilated, evincing a resemblance in this respect to
the canines of the Seals.

We have nothing to indicate what the lower canines were,
but it is legitimate to infer their existence and correspondence
with the upper teeth.

The molars in the upper jaw of Enhydriodon were, as in
Enhydra, four—two false molars, a carnassier, and a tuber-
cular. This is seen to have been the number in both the
fossil species with which we are acquainted. The first, or
precanine, was a very small and rudimentary tooth, deci-
duous in the adult animal. The alveolus for it is dis-
tinctly shown on both sides of the head, fig. 2, which
belonged to rather a young animal, and probably a female,’
as the tooth is retained longer by this sex than by the other.
The second had two fangs, and instead of having the flattened
form which it has in the Otter, terminating in a poited
cusp, it appears to have been short, thick, and blunt, as in
Enhydra. It was encircled by a rugged basal ridge, or burr.
None of the specimens show more than a section of it, the
crown being broken off in all of them ; but the form and burr
are best seen in the left side of the old head, fig. 5.

The upper carnassier is singularly remarkable in the form
of its coronal lobes generally, and in the extent and com-
plicated development of its interior portion. It constitutes
the most typical feature of the genus, and has nothing that
we know of analogous to it in the family of Carnivora, fossil
or recent. It was remarked both of Lutra and Enhydra, that
their carnassier was more or less triangular in its outline.
In Enhydriodon it is nearly square ; and instead of the cusps
and trenchant ridges of Lutra, or the flattened inequalities
of Enhydra, the coronal lobes are developed in conical
mamuille, somewhat like those of the mastoid-toothed Pachy-
dermata. Fig.? represents a perfect carnassier detached,
of the left side of the smaller species. It is divided by a
deep longitudinal flexuous hollow into two portions; the
outer or typical body of the carnassier is distinctly three-
lobed, the middle lobe being the largest and elevated into a
conical mammilla, the other two being blunt tubercles. The
inner portion, equalling the size of the outer, is composed of
two thick conical mammille, separated by a deep hollow; the
anterior mammilla has its posterior margin notched near the
base into two additional denticular lobules. The posterior

1 This conjecture and the grounds ? This specimen has been unfortu-
upon which it is founded were suggested | nately mislaid, and was never figured,
by Prof. Owen. It is not in the British Museum.—[Ep.]

ENHYDRIODON. 337

mamunilla is nearly of the same size, but simple. It is this
posterior lobe, of which there is no trace in the Otter or
Enhydra, that, so far as regards the composition of the
coronal, distinguishes the carnassier of Enhydriodon so
strongly from that of other Carnivora. Inits great features
this tooth looks as if made up of three nearly equal conical
mamumnille disposed in triangle, two occupying the inner side,
and one at the outer bearing a lobe at each end. In the
larger species! it has precisely the same form as in the small
one, showing it intact from wear, the only difference being
that the inner portion is, if anything, larger. The manner in
which the ‘wear’ of the tooth goes on is seen in all the
specimens, the apices of the mammille being truncated off
horizontally, so as to form depressed discs, surrounded by
a thick ring of enamel resembling what is shown upon the
teeth of old Hyzenas, and on the molars of the Phacochcerine
Hogs. We have noticed that the teeth of Enhydra and Lutra
do not show a ‘ wear’ of this sort.

APPENDIX TO MEMOIR ON ENHYDRIODON.

I.—MS. Memoranpvum By Dr. Fatconer, 1845.

Compared the head of Indian Otter (recent) with four heads in College
of Surgeons. It is about half inch longer than any of them; the naso-
frontal oblong in Indian; nearly square in European, or approaching
square. In one head, No. 295, of an old animal, it is very square. Post-
orbital part of frontals is more elongated and more contracted in Indian,
but great difference in this respect in the European. In No. 295 it is
very broad and short, wider than long ; while in another, No. 55, from
the side cases, the sides converge backwards into a very narrow con-
traction, half the width of No. 295; this is also an old head.

II.—Memoranpum 1n MS. mave at Dr. JAMESON’s, SUHARUNPOOR,
JAN. 21, 1841.

Dimensions of a Lower Jaw fragment. Enhydriodon.

Fragment consists of the posterior part of the lower jaw, left side,
with the ramus broken off above the base of temporal fossa; con-
dyles, &c., removed. ‘The carnassier tooth and a solitary tubercular
are in situ.

Dimensions.
Inches

From the anterior margin of carnassier to the rear of the tuber-

cular. : : 2 ° . 5 : : c . %r4
Length of the carnassier . c ° : . . wees
‘Width of ditto . . : : : : ; : ¢ . 4
Length of tubercular c : : 5 co . c » 46
Width of ditto . : : ; : : ¢ c : by <taes)
Height of jaw to base of carnassier . ; é A c SL
Greatest thickness of ditto 3)

1 Here and at page 331 allusion is made to more than one species of Hnhy-
driodon.—[ Ep. |

VOL. I. Z

338 FAUNA ANTIQUA SIVALENSIS.

The carnassier consists of three lobes: the anterior one is a low
three-sided pyramid; the middle one is an elevated acute lobe, as in
the Otter (it is partly broken off in the specimen), with a small tu-
bercle at the rear of the inner side, which is smaller than in the Otter ;
the rear lobe consisting of a flattened, nearly horizontal tubercle,
nearly plain, as in the Bear, but with a worn low maxillary process or
facet at the outside and inside. The tubercular or rear molar is an
oblong tooth with an oblique facet directed forwards; then comes an
anterior lobe, and then a posterior lobe with a re-entering angle
between; the posterior lobe consists of a single flattened point. The
corresponding tooth in the Otter is nearly circular in either central
point, from which four small facets are given off.

CARNIVORA. 339

XVIII. ON THE FOSSIL CARNIVORA OF THE
SEWALIK HILLS.

[AutHoveH the Felis cristata, Ursus (Hyenarctos) Sivalensis,
and Enhydriodon Sivalensis are the only Sewalik Carnivora
which have been carefully described, the Sewalik Fauna
included representatives of many other genera belonging to
this class. Two unpublished plates of the ‘ Fauna Antiqua
Sivalensis ’ (L. and M.) are devoted to the illustration of the
Hyena Sivalensis (Fale. and Caut.), the specimens of which
are in the British Museum, while a third unpublished plate
(N.) illustrates the remains of the Machairodus Sivalensis
(Fale. and Caut.), also in the British Museum ; and a fourth
(Q.) those of Ursitaxus, or a species of Ratel. (See descrip-
tions of these plates.) The British Museum collection also
contains specimens of fossil Canis from the Sewalik hills,
which, as well as Gulo, was briefly described by Messrs.
Baker and Durand in the Journal of the Asiatic Society for
Sept. 1836, vol. v. p. 581. These descriptions are annexed,
as well as an account by Dr. Falconer of the Sewalik Carni-
vora in the Museum of the Asiatic Society in Calcutta. A
reference to Machairodus Sivalensis, from one of Dr. Falconer’s
note-books, will also be found in vol. ii., under the head of
* Felis speleea.’—ED. |

I.— Description spy Messrs. BaAkeR AND DuRAND oF Fossit CARNIVORA
FROM THE SuB-HIMALAYAHS.!

GULO.

Of this genus we possess the fossil skulls of two individuals, one of
which, represented in fig. 4, is nearly perfect : the lower jaws have
been separated at their symphysis and otherwise somewhat mutilated,
but as they were not found attached to the cranium, we may consider
ourselves fortunate in having obtained them at all. The second
cranium, fig. 7, has suffered considerable mutilation, and is without the
lower jaws ; we have, however, inserted it in the plate, because though
otherwise less perfect, it has escaped a crush, which appears to have
flattened fig. 4. Some differences of proportion between the two fossils
would be accounted for under this supposition.

The recent skulls with which we have compared the above-mentjoned

fossils belong to an animal known by the Hindustani name Biju xv
identical, or nearly so, with the Cape Ratel (Gulo Capensis, Desm.)

1 The illustrations referred to are those in the ‘Journ. As. Soc.’—-[Eb.]
Z 2

340 FAUNA ANTIQUA SIVALENSIS.

In classing the Biju and Ratel under the genus Gulo, we follow the
common system of arrangement ; though, as remarked by Cuvier,!
both the number and character of ‘their teeth would rather place them
with the Mustela Putorius. They appear, in fact, to be indebted to
their plantigrade motion alone for a place among the Gluttons. The
fossils now under consideration correspond in dentition with the Ratel
and Biju, and the following table will show that their resemblance to
the latter in most other respects is very strong.

Bocent
Bim | pig. 4 | Fig. 7
Extreme length from posterior of occipital condyles to an-
terior of incisors, taken as the modulus and assumed at | 1-000} 1:000} 1:000
Breadth measured across mastoid processes. 0°581) 0592} 0°610
Greatest breadth of cranium opposite mastoid processes 0-467 | 0.443) 0-474
Least ditto ditto at temporal fossz : 0°226| 0°258 | 0°262
Height of occiput from inf. margin of foramen magnum
to sup. of occipital ridge . 0°318 | 0°307 | 0°343
Breadth of ditto from point to point of styloid processes 0°335 | 0°361| 0°394
Ditto across the occipital condyles . : 0°243 | 0°241} 0°248
Ditto of frontal from point to point of post. orbital
apophyses . : ’ - | 0°286| 0°327| 0.313
Ditto greatest across zygomatic arches . : - | 0°546) 0°643)| 0°561
Ditto exteriorly across the superior canine teeth c - | 0°220} 0°238 | 0°236
Perpendicular diameter of occipital foramen . 0088 | 0:091 | 0-108
Length from anterior of canine teeth to oe of tubereu-
lar teeth, measured externally . F 0°282 | 0:287| 0-287
Breadth of upper jaw measured across carnivorous meine 0°347 | 0°339 | 0°325
Greatest antero-posterior diameter of canine teeth . . | 0058) 0°071 | 0:057
Width of tuberculous molars . . ‘ : - | 0°081 | 0:073| 0:069

The two fossils, though differing considerably from each other, agree
in the following points of dissimilarity from the recent skull. Their
canine teeth are larger and stronger, and their tubercular molars
smaller; the two lines of molars converge towards the muzzle consi-
derably less in the fossil than in the recent animal, and the individual
false molars are set less obliquely to the line of maxillary. ‘The frontal
is wider between the orbits; the post-orbital apophyses more promi-
nent, and the depression of the cranium in rear of them less deep; the
exterior portion of the mastoid processes has a far greater development
(fig. 8); the transverse occipital ridge is thicker, more rugged and
more prominent, and projects considerably beyond the plane of the
occiput in the prolongation of that of the parietal bones (fig. 5). Mea-
surements of the recent and fossil lower jaws exhibit no differences
save in the canine teeth, which severally correspond with the same
teeth in the upper jaw. There is, however, in the fossil (fig, 6) a deep

1 Tl convient d’autant mieux de com- | et Vapophyse coronoide de sa machoire
parer le Ratel au Glouton, que ces deux | inférieure beaucoup moins haute, plus
quadrupédes sont a peu prés de méme | large, et plus obtuse. Les rapports du
taille; mais outre que le Glouton a six ratel avec le putois, d’aprés ses dents et
molaires de plus que Je Ratel, le crané | sa téte, sont certainement plus importans
de celui-ci est plus large en aEridre, son | que les différences de marche. — Ossemens
front moins élevé, son orbite moins cerné, | Fossiles, tom. iv. p. 0.
ses arcades zygomatiques moins hautes,

hl on wt |

CARNIVORA. 341

depression in the ramus, which in the recent species is nearly flat.
In our specimen this depression is as marked as in the tiger and other
feline animals. The differences above noted tend to prove that the
ancient species was even more powerful and savage than its present
representative, the Biju itself, which is by no means deficient in these
qualities. The three recent skulls examined on this occasion had all
suffered more or less from the violence to which the vigorous self-
defence of the animals had obliged their captors to resort.

Canis VULPES ?

The specimen represented in figs. 9 and 10, though fortunate in
possessing both lines of molars complete, has suffered much from a
erush by which the whole posterior portion of the head has been flat-
tened and disfigured. The dimensions selected for comparison in the
following table are those least likely to be affected by the accident.
Our recent specimen belongs to an adult male fox, of a species common
in the NW. provinces of India. Its size, the colour of its fur, and
other external characters, appear to correspond with the description of
C. Corsac (Pallas), which Cuvier is inclined to identify with C. Benga-
lensis of Pennant—figured also in ‘ Hardwicke’s Illustrations,’ Pl. II.,
Parts XV. and XVI.

Recent | Fossil

Length from occipital condyle to anterior of canine, taken as

modulus, and assumed at . 4 . P . Bs . | 1-000} 1-000
Breadth measured across mastoid processe : - - - | °3890| 433
Least breadth of cranium at the temporal fossze - - . | °205) +206

Breadth from point to point of styloid processes . - - | °253] :298

Ditto across occipital condyles . ga 5 4 + | 52021) 2or
Ditto of frontal from point to point of post-orbital apophyses . | +287] +295
Greatest breadth measured externally across both lines of molars | -338| °319

Horizontal diameter of occipital foramen . . : . - | °130) 133
Length occupied by line of molars and canine taken together . | 500} +470
Ditto ditto molars alone . . : : : : “ - | °410) 415

The chief differences here exhibited are those of greater breadth in
the posterior portion of the fossil’s head, and must, though they appear
natural, be liable to the suspicion of having been caused more or less
by the crush before alluded to; but there are some points of dis-
similarity which must be considered free from this objection. The
transverse occipital ridge is thicker and higher in the fossil; the post-
orbital apophyses are altogether broader and more prominent; the
hollow or depression in their upper surface, forming a valley between
the outer edge of the apophysis and the swell of the frontal (constituting
one of the distinctions between the fox and the other varieties of canis)
is in the fossil more marked. From the rear of these apophyses start
two ridges, which at first converge towards the occiput in a curvilinear
direction, until the distance between them is reduced to about half an
inch, after which they run nearly parallel for some distance, and then
converge again, till they unite near the occiput and become blended
with the parietal crest. We have been particular in describing this
formation, as a very similar one was observed by Cuvier in the skul]

342 FAUNA ANTIQUA SIVALENSIS.

of the ‘ Rénard tricolor,’ C. cinereo-argenteus, Linn. It will be
observed from the table of measurements that the length occupied by
the molars and canine teeth together is less in the fossil, while that
occupied by the molars alone is proportionally greater. This difference
is only apparent, and is caused by the advanced position of the first
false molar close to the root of the canine; the tooth is probably a
deciduous one, not yet replaced by the permanent molar; the unworn
state of the other teeth also attesting the youth of the fossil, at the
period of its demise. The lower jaws, figs. 11 and 12, are not suffi-
ciently perfect to afford any satisfactory measurements.

To the foregoing observaticns we have nothing further to add than
that, in our opinion, they point out sufficient proof of specific differ-
ences between the several objects compared ; but, for the reasons before
adduced, we must limit our conclusions to this, and cannot therefore
venture upon giving new names to our fossil species.

Measurements of Fossil Skulls.

lo le -
tgs 4 Pear pe
Details of Measurements 5&8 : 9& 10
In. In. In.
Extreme length from post. of occipital condyles to ante-
rior of incisor teeth . c . 5 : . | 5°51 | 5:08 | 4:09
Ditto ditto ditto canine ditto é : . . | 513 | 4:86 | 3°83
Breadth measured across mastoid processes : 3°26 | 3°10 | 1°66
Greatest breadth of cranium opposite mastoid processes 2°44 | 2:41 | 1°71
Least ditto ditto at temporal fossze : 1:42 | 1°33 | 0°79
Height of occiput from lower margin of occipital foramen
to top of transverse ridge os : . | 169 | 1:74 | 0-90
Breadth from point to point of styloid processes : . | 199 | 2:00 | 1:14
Ditto across the occipital condyles : . | 1°33 | 1:26 | 0:96
Ditto of frontal from point of post-orbital apophyses «| 2:80) |) aeose eats
Greatest breadth across zygomatic arches. 2°99 | 2°85 | —

Breadth measured externally across superior canine ‘teeth | 1:31 | 1:20] —
Ditto ditto most prominent points of line of molars . | 1:87 | 1°65 | 1°22
Perpendicular diameter of occipital foramen . : . | 0°50 | 0°55) —
Horizontal ditto ditto . : 0°62 | 0-60 | 0°51
Length from exterior of incisors to anterior of palatal sinus | 2°40 | 2:00 | —
Ditto from anterior of Bape! sinus to lower margin of

occipital foramen . 2°56 | 2°70 | 2-05
Ditto occupied by molars and canine e teeth, taken together 1:58 | 1:46 | 1:80
Ditto ditto molars alone : 1:25 | 1:17] 1:59
Diameter of orbit perpendicular, but measured in ‘plane

of orbit’s margin. 062 | — 0°58
Ditto ditto from point of post-orbital apophysis to anterior

margin of orbit c .| O79 | — 0°74
Greatest antero- -posterior diameter of canine tooth. - | 0°39 | 0°34 | 018
Width of tuberculous tooth (in Felis and Gulo) . . | 0°40 | 0°35 | —

Lower Jaw
Length from posterior of condyle to anterior of canine . | 3:20 | — —
Ditto from ditto ditto to posterior of molars . 14S5 pal
Ditto occupied by molars and canine teeth, taken together 174) — —
Ditto ditto molars alone : Lay Se ts
Depth of lower jaw taken in front of carnivorous tooth . 067); — 0°45
Greatest antero-posterior diameter of canine tooth. . | 088 | — 0713

gs eee One oe

CARNIVORA. 343

I].—Descrietion By Dr. Fatconer or Fossit Rematns oF CaRNIVORA,
FROM THE SewALtk Hitis, 1n THE Museum oF THE ASIATIC
Society oF BENGAL.

No. 847. Hyena.—Cranium nearly entire, deficient only in the
right zygomatic arch, and in the basal part of occiput, but covered
with matrix, and the teeth so mutilated that only the second premolar
is distinctly seen ; the incisors and canines in situ, but broken off.

No. 848. 7 of cranium a good deal crushed, and
mutilated behind the orbits, showing all the teeth on the right side ex-
cept the small premolar; the teeth are considerably larger than those of
the existing Indian Hyena. On the left side canine seen in section.

No. 849. Hyena.—Fragment of lower jaw, right side, containing the
four molars in situ; teeth larger than in the existing Hyena, but a
good deal concealed by matrix.

No. 851. Feline—Fragment of horizontal ramus, lower jaw, right
side, containing the two last molars in situ ; would be about the size
of Felis cristata (F. and C.).

No. 852. Canis.—Upper maxilla, left side, comprising a part of
palate and zygomatic portion of orbit.

No. 859. Hyena.—Fragment of hard matrix containing part of
the cranium of a carnivorous animal, and showing canine ? and por-
tion of premolar, but much concealed by matrix. An Hyena?

II].—Descrietion py Dr. Fatconer or Fossin Hyrana. From Note-
BOOK, JAN. 21, 1841.

Dimensions of a large Hyena, Dr. Jameson’s, Suharunpoor.
Inches

Interval between the outer edges of the third or last false molar
of the lower jaw j

Length of the three anterior false molars .

Ditto from anterior side pica incisor to the posterior ditto third

false molar . 5 :

Width of the line of i incisors

Length of posterior tubercular .

Thickness of ditto about .

Height of ditto. 5 - : ; : :

Length of second ditto . 5 : c : -

Thickness of ditto . :

Height of ditto about ; c : : - C

Interval between the bases of canines “ . . 5 5

Atos CON et em eg
OCORERUIAARSUR CHE

_

This specimen consists of the anterior half of the lower jaw both
sides, as far back as the rear of the third false molar. The incisors,
canines, &c., in situ. The teeth considerably worn. The incisors and
canines show nothing peculiar. The first false molar is very small, and
the second much smaller than the third. They are placed obliquely
in the jaw. Section of the canines oval. A very large species.

344 FAUNA ANTIQUA SIVALENSIS.

XIX. ON THE FOSSIL CROCODILES OF THE
SEWALIK HILLS.!

BY CAPTAIN P. T. CAUTLEY,

Or the skulls of the existing Crocodile, from which the mea-
surements accompanying this note have been taken, one
belongs to an animal 7 feet long, of which we have the per-
fect skeleton ; and the other was stated by the person from
whom it was procured to have belonged to an animal of 12
feet. We havea correction, however, in the smaller specimen,
which was carefully measured; and taking this as a type,
the animals being of the same species (C. biporcatus, Cuvier),
a mean of four measurements gives us a length of 132-09
inches, or 11 feet -09 inches, for the latter. In fixing this
specimen as belonging to an animal of 11 feet, we shall not
therefore be far from the truth.

There is so much difference in the few comparative mea-
surements that we have been able to obtain of the fossil, with
these two skulls, that it is hardly possible to take any pro-
portion of the existing animal as a guide to that of the fossil ;
the measurements taken separately would in some cases
reduce our fossil to that of an animal of 11 feet, with distinct
ocular proof to the contrary; in others the fossil animal
would be 17 feet long, which may probably be somewhere
near the actual size; while an assumption of 20 feet would
be extending the dimensions to their utmost limit, our esti-
mate being guided by the proportions of the species now
existing in our rivers.

The fossils from which the measurements were taken
consist of two very perfect fragments; First, the anterior
portion of the skull of a large and adult animal, the posterior

1 Reprinted from the Asiatic Re- | ‘Asiatic Researches’ are also in the

searches, vol. xix. p. 25, 1836. The ori-
ginal memoir was illustrated by rough,
though characteristic, drawings. The
drawings illustrating this reprint (Plates
xxvii. and xxix.) have been made by
Mr. Dinkel from more perfect speci-
mens in the British Museum. Most of
the specimens, however, drawn in the

British Museum. Thus, Plate ii. figs.
2 and 4, As. Res., is No. 39,801 in Brit.
Mus. C. biporcatus (sic) ; and Plate iii.
figs. 3 and 5, As. Res., is No. 40,206 Brit.
Mus. Lep. Gangeticus. By referring to
appendix it will be seen that the iden-
tification of C. biporcatus in memoir ought

to have been C. bombifrons.—[Ep.]

DESCRIPTION OF PLATE XXVIII.

CROCODILUS BOMBIFRONS AND LEPTORHYNCHUS GANGETICUS.

The figures in this Plate have been drawn by Mr. Dinkel
from the original specimens in the British Museum, which
have been identified with outline tracings named by Dr.
Falconer.

Fig. 1. Lateral view of fossil skull and lower jaw of Crocodilus bombi-
Jrons, a species still existing in India. The drawing is one-fifth
of the natural size. The number of the specimen in the British
Museum Catalogue is 39,798. (See page 355, and vol. i.
page 485.)

' Figs. 2 and 3. Another skull of Crocodilus bombifrons, showing upper

and palatal surfaces, and drawn one-fifth of the natural size.
The number of this specimen in the British Museum Catalogue
is 39,795.

These two specimens of C. bombifrons belong to the same species as
that identified in Capt. Cautley’s memoir as C. biporcatus, but are more
perfect than the specimens which he described, but which are also in
the British Museum. (See page 344, note.)

Figs. 4 and 5. Leptorhynchus Gangeticus. Upper and under surfaces
of cranium, drawn one-sixth of the natural size. The number
of the specimen in the British Museum Catalogue is 59,809.
The species is the same as that described in Capt. Cautley’s
paper as Croc. longirostris, and is the existing Gavitl of the
Ganges ; and the specimen closely resembles, but is more perfect
than, that figured by Capt. Cautley in the ‘ Asiatic Researches,’
vol. xix., Plate IL, figs. 3 and 5, the Catalogue number of which
in the British Museum is 40,206.

VoL. I.

Vol. 1 Plate 28 .

W.West imp

1,2,3.Crocodilus bombifrons. 4, 5. Leptorhynchus Gangeticus.

CROCODILES. 345

part from the palatal sinus being wanting ;! in this the front
of the lower jaw consisting of the left dental from the 1st to
the 8th tooth, and of the right from the 1st to the 4th, is
fixed—the fragments of the upper and lower jaw being
united: a point proving that some at least of these remains
were inhumed before the disintegration of the muscles and
integument connecting them; and, secondly, the posterior
part of the skull, from the occipital to the front of the orbits,
of an adult, but of a smaller animal than the preceding.

The 4th tooth in each side of the lower jaw being received
into a groove into the upper, the form and size of the cranial
foramina, together with those of the protuberances and in~
dentations of the muzzle, place our fossil amongst the true
Crocodiles, the species being allied to Cuvier’s Biporcatus, or
the Crocodile, ‘a deuw arrétes’ now existing in these rivers.

The following measurements will facilitate the reference to
Cuvier’s ‘Ossemens Fossiles,’ and be perhaps of still further
use in providing the means for general reference on points
relating to the existing Crocodile (See p. 346).

From the sutures being obliterated on the upper rugged
surface of the fossil, the junction of the lachrymal and
anterior frontal on the maxillary bone is not observable, and
as this is one of the points upon which the Cayman and Croco-
dile differ, it is perhaps to be regretted that this must remain
at present doubtful; for although the characters above given
distinguish the fossil from the Cayman, the bluntness of the
muzzle and the proportions of the bones of the head do, in
some respects, assimilate it with the latter sub-genus.

On the lower smooth surface the sutures are well defined,
and it is on this measurement that we observe the remark-
able distinction between the existing and the fossil animal ;
the shortness of the maxillary bones and the length of the
intermaxillaries, including the nasal aperture, are peculiarities
that will be observed in the table of measurements, the
former (maxillary) in the existing animal being to the latter
(intermaxillary) as 3:9 is to 3:2; in the fossil as 3°8 is to
4°3. The length of the maxillary bones on the lower suture,
or the space separating the palatine from the intermaxillaries,
is rather greater in the existing animal of 11 feet than in
the fossil. The comparative measurement from the point of
the muzzle to the maxillary extremity of the palatine bone,
together with those across the skull at the 10th and 4th
teeth, will point out in a still clearer way the bluntness and
breadth of the muzzle of the fossil animal. By the extension
of the intermaxillaries and the great length of the connecting
suture between the point of the muzzle and the nasal aper-

1 Brit. Mus , No. 39,801.—[Ep.]

FAUNA ANTIQUA SIVALENSIS.

346

Ie lala)
eee

ec}
=
=
S
=
nN

halehal
felled ed

9&T-0 Gé.g
061-0 9-8

06-0 6L
60-0 SLT
960-0 O-L
010-0 8-6
610-0 Lé
960-0 8-8
601-0 ben
6SL-0 0-9
PEs-0 6-6

11
IT

een | sayouy

0-F LOT-0 | ¢&-9
G.0 260-0 | OGL
LT 8F0-0 | S8-L
6-0 660-0 | G&L
8-1 FLO-0 166
£4 880-0 GE
LL 966-0 LIL
€.9 PS6-0 0-01
GFL 8F9-0 S16
G86 OL-0 | 96-8
LL €F0-0 LT
8-h O8T-0 | G0-L
GP-0 160-0 | $8-0
06-0 PLO-0 G0
SP-0 910-0 | 99-0
OL-0 620-0 60
06-0 Bae 0 €-1
6-0
€-1
G6-0 £P0-0 Lt
OF-T GG0-0 | G0%
0-8 SIT-0 | 99-P
LL 8F0-0 6-1
GG 112-0 ¥-3
B+, 066-0 GIL
VP Z81-0 GL
GL T 980.0 VS
L& GPL-O | G9.G
HG 660-0 6-8
GPG 080-0 GE
La 611-0 LY
GP OLT-0 LY
0.g 092-0 L8
Lg $86:0 GIL
IL SEP-0 1-LT

or}

soyouy | ageyy | soqour

TSS,

Suol "FL SUT “FFLL

OTLpooorn SurqsrxgL

auoq TelnoyTe Jo Aqrureryx9 Toyr1e3s0d pus amqrede TeAO. yo Aqqune79xo IOT10}S0d WoaMgoeq souweAsIcT

* 0}91p 0991p Jo yadoq.
5 : : ouog TBItOp JO Agrureayxo torz1aysod ay} 4% orng1ede [Bao Jo yASUET
ee ea * * T9804 496 DUB TIE TeeAgeq Avul Jo yydoq
: 5 . U u L . ° . stisAqdurds Jo y4ysuorT
. . . . . . . . . . . . al jo yadep qsoyvar4)
: * — sotoq cart amy Jo uworount ey Aq pourioy orngizode peao 949 98 YQpeerg
. . * 90g IB[NoZIB 014 9B YApBeIq oUMEAAXT
bs . 0 ir euog wwlnoy7e Jo AqpUMe14xe ToLIeqsod 07 o[zznu WMOIT YASUO

(og ‘qg007, JO oquIN\y)—MvP TOMO'T

° . 1990} TOMOT YAP OY} Sutataoar roy Mel roddn uo soaoors w99aA4eq WIPT A,
oyzznur Jo quod puv oinjtode [esru UeaMgeq ‘a'r ‘QIMJNS TO SdLLB[[FXBULIE, UT JO ygpeorg.
: _euod ouyred Jo Agrue14x9 Axeypexeun 09 e[zznw Jo guyod uory YgSueryT

§ 2 . 0391p o771D WIT 9}

: as) 0 U Z : 4 snTOoATS sar 98 aval aeddn 1900} WF JO LojouVIC,

. . . . . . . . . 0791p yo yydoq.

: . D : 0 * a B 0 s O * MOUTBIOF [B4Id1I90 Jo 0991,
s n : . 3 # 4 , : C * ayApmoo [eqtd1000 Jo yA pearg.

0391p 0791 0791p 0391p JO YIpeerg,
somoq [ejolivd puv ‘pyoyseur ‘yequoTy *roqsod ayy Aq pepunod esoy} Io vuTUTETOS aqrydvqjo.10 Jo yASuary

079Tp 079Tp JO UFpBVarEg,
, ti . ‘ . i s * 9angiode [vsvuU IO [L14SOU [BU1e}xXe Jo Y4SsuaT
” a seu0q pyoyseur OU} WIIM worqoUNl meyy 4B STVIUOIZ LOTIOySod 944 WO 0491q
‘ 2 * [eordryo@y ayy woount sy Wo otIOg [eIUOTy OYY Jo YYpeerg
. . . . * gmoq [BUIATYORT JO 0991p 07 0991p 09791
D . s * : g * qIqito JO 1Op10jUB 0} O[zZZNUT Jo Jutod woary YQSUIT
© L : x promeyds qo soshydods prosdreqdl 04 Sso10B YApPBeAq OUMAIy XG

‘ . . . * 0991p prloueyds Jo 0791q
. . . . . : x ui oye oangns 99 WO souo0g eulqeyed Jo WASue'T
. . . . . . . . < * 0497p SOLLB[[IxBUt Jo 0991

. . . . . : D * MOA OINJNS WO SOLV[[TXBULIAZUT JO YASUAT
£ ‘i . : : : . ‘ * OF9TP UIP 0991p 0791p WoaAgoq 0491.
‘ 0 a : ‘ , : - : 0FF1P UOT OF9TP 0991p W9aAZoq 0991
: * 4900} 94T O49 JO 1TO9ATB O49 JO Squrod yuoUTtMOAd 4sour oy UAdAIEq 0991
E , * mel TOMO] TILM TOQR[NdT418 844 4B Souog [vtodtme, oY4 UO YIpraIg
o[Aptoo jezId19000 Fo uLsavIM 10 AQIUI0I}xXO TO[A94sod 04 9[zznw Fo guIod woay YYSuET

(gg ‘qqeag, Jo Toqumyr)—Mvp reddgQ. ’

“ynyy fo suorsuaucy

CROCODILES. 347

ture, this aperture is thrown further back, so that a line
drawn transversely through the grooves for receiving the 4th
lower teeth (which in the existing animal would cut the
posterior extremity), in the fossil passes through the centre,
or rather in front of the centre of the nasal aperture.

The skulls of the true Crocodile and Cayman differ in the
following points. 1. That of the Cayman is less oblong,
shorter, and flatter at the muzzle. 2. The 4th tooth of the
lower jaw enters into a hole in the upper, instead of a groove
as in the true Crocodile. 3. They differ in the number of
the teeth. 4. The cranial foramina bounded by the posterior
frontal, mastoid, and parietal bones, are smaller, and some-
times altogether wanting inthe Cayman. 5. The lachrymal
and anterior frontal bones descend lower in the Crocodiles
than in the Caymans. 6. In the Cayman a part of the vomer
is visible in the palate between the maxillaries and the inter-
maxillaries. 7. The palatine bones advance more in the
palate and are wider in front in the same animal. 8. The
posterior nostrils are wider than they are long.

With regard to the cranial foramina of the fossil, and their
proportion relatively to the surrounding bones, we are enabled,
by having in our possession a very perfect fragment of the
occipital region and that portion of the skull bounded by the
orbits, to give the comparative measurements here also;
noting that this fossil is a portion of the skull of an animal
of much smaller dimensions than that from which the former
measurements were taken.

Existing Crocodile
Fossil Crocodile
11 feet long 7 feet long

In. Met. In. Met. In. Mét.

Length of crotaphite foramina . | 1°30 | 0:032)| 0-9 0:022} 1:40 | 0:036
Breadth of ditto . : < . | 0°90 | 0-022} 0°70 | 0-017] 1:0 0°026
Breadth of the frontal bone on its

junction with lachrymal 2 |) i 0048 | 1-1 0°027 | 2:1 0°053

Ditto on posterior frontals at their
junction with mastoid bones . | 4:55 | 0°115| 3-0 0°076| 4°85 | 0°122

Breadth of occipital condyle .| 13 0-032} 0-9 07022} 1:20 | 0:081
Ditto of occipital foramen . . | 09 0°022| 0-7 0°017| 0:90 | 0:022
Depth of ditto . . : - | 0°65 | 0°016} 0°45 | 0-012} 0-60 | 0-015

It will be observed from the above that the fossil and the
existing animal of 11 feet very closely correspond in dimen-
sions, although the crotaphite foramina are rather larger,
and the width of the bones in their neighbourhood greater in
the fossil than in the existing one. This proportional excess

348 FAUNA ANTIQUA SIVALENSIS.

of breadth, however, is not so striking as in the measure-
ments of the muzzle before given; although it still bears
us out in the general expanded dimensions of the fossil
animal.

Of the lower jaw the only comparative measurement that
our discoveries have enabled us to make is of a small portion
of the anterior extremity, showing an extreme contraction
and narrowness of the symphysis; that of the fossil being
actually less than that of the existing 11 feet Crocodile. The
form of the suture is similar in each, and the internal process
equally well defined.

Further than from an inspection of the plates and descrip-
tion of the varieties of the Cayman and the Crocodile, in the
5th volume of the ‘Ossemens Fossiles,’ I am unacquainted
with the form of any other head than that of the Crocodile
which inhabits the Ganges and Jumna rivers in this part of
India, and presume that I am correct in placing our existing
animal amongst the Crocodiles ‘ &@ deus arrétes.’? The pecu-
harities of the skulls in my possession resemble those of this
species, although there is a point relating to the ridges
which may as well be noted, more especially as the same
feature is most prominently marked in the fossil, thereby
assimilating our existing and fossil animal still more closely to
each other.

The ridges (in C. biporcatus) are described as ‘ proceeding
from the anterior angle of the orbit and descending in almost
. a parallel line along the muzzle, and gradually disappearing.’
In both the fossil and existing specimens now under descrip-
tion the above note applies distinctly, with this exception,
that the ridges partially disappear at a point half way on the
nasal bones, from whence they strike off in an oblique direction
right and left towards the alveoli of the 10th tooth, this
oblique ridge showing itself as prominently as that at the
anterior angle of the orbit.

There is a general resemblance between the fossil and the
head of the existing Crocodile, which is striking. The
rugosities and position of inequalities on the upper surface
closely correspond; the cranial foramina, the number of
teeth, the foramina in the upper intermaxillary bones for
receiving the two front teeth of the lower jaw, the grooves
for the 4th teeth of the lower jaw, and the general form of
the nasal aperture, are features similar in both. We may
therefore fairly conclude from analogy that the Crocodile
now found fossil in the upper strata of the Sewaliks is of a
species closely allied to the present one, with the simple
difference of possessing greater width in its proportions: in
which view we must be satisfied with establishing it as a

CROCODILES. 349

fossil variety of the C. biporcatus of Cuvier and the C. porosus
of Schneider.

In taking the numerical proportion of the fossils already
found as a guide to that of the animals existing on this tract
previous to the upheavement of the line of country, it would
appear that the Mastodon and Hlephant were the most
numerous. 2. Ruminants. 3. Hippopotamus of varieties,
the largest of which, with six incisors (H. Sivalensis), is in
the greatest proportion. 4. Crocodile, Gharial, and Tortoise.
5. Rhinoceros, Hog, and Horse. 6. Carnivora. 7. Fish.

The remains of Saurian animals, although standing high
on this list, consist chiefly of fragments of the osseous plates
of the neck, vertebre, detached teeth, articulating ex-
tremities of bones of the arms and legs, as well as other
portions of the skeleton; while portions of the head have
been rarely found. Those referred to in this note are very
perfect ; others are crushed and distorted ; but the leading
differences which have been adverted to are fairly marked in
all those that have come under my observation.

There appears to be a local disposition in the deposits of
these remains, as would be natural to expect on a varied
surface of plain, forest, and marsh. The Mastodons, Ele-
phants, &c., in great abundance at some points, give place
to the Hippopotami and Saurian reptiles in others. At many
places the latter, with the Tortoise, are totally wanting; at
others, as in a stratum of an impure marl attached to this upper
series where freshwater shells (chiefly like the Unio of the
present day) were found in great abundance, nearly the
whole of the remains accompanying them were those of the
Crocodile and Gharial. In considering the length of the
fossil species we see no signs of anything beyond that of the
animal now existing. ‘The largest remain in our possession
is a vertebra, which is one-third larger than that figured in
the London Geological Society’s ‘ Transactions’ amongst the
Ava fossils, and described as the remain of an enormous
animal; the vertebra above mentioned is either one of the
dorsal or lumbar, but the processes are broken and imperfect.
The dimensions of the barrel or cylinder in this specimen,
and the measurements of the third lumbar vertebra of the
existing Crocodile, are here given :—

Existing Crocodile

7 feet long Fossil
In. Met. In. Mét.
Extreme length . 1:4 0-036 4-9 07124

Breadth under transverse apophysis,
taking a mean measurement . - | 0:95 0°024 3°) 0:089

350 FAUNA ANTIQUA SIVALENSIS.

Large as the fossil may appear, the animal to which it
belonged did not in all probability exceed 25 feet in length,
whereas the Gangetic Crocodile of the present day is said to
arrive at the enormous length of 30 feet, and in the pages of
the ‘ Calcutta Journal’ an animal of 28 feet long is recorded
as having been killed by a gentleman of the Civil Service (I
believe) now residing in Calcutta.

Tur Fossin GHARIAL OF THE SEWALIK HILLS.

Amongst the numerous remains of the Crocodilean Saurians
which have been found in such abundance, from the oolite
up to the more recent strata, it would appear that the
greatest proportion has been allied to the Gharial,! and that
the existing Crocodile and Cayman have been almost without
their prototypes. It is only in the strata above the chalk? at
Montmartre, and the freshwater-formation at Argenton,
where remains have been found which were considered by
Cuvier as appertaining to the latter sub-genera; in these
strata, however, the remains of animals of this description
are scarce, and in those still more superficial abounding in
the remains of the larger mammalha, in Mastodons, Hippo-
potami, &c., where we might naturally expect to find the
Crocodile, the remains of this family have hardly I believe
been found at all.

Of the fossil Crocodile brought by Crawfurd and Wallich
from Ava, and figured in the London Geological Society’s
‘Transactions,’ the drawings show a much nearer approach
to the living congenera than had, up to the period of that
discovery, been found; and although we are unacquainted
with the geology of the country from which they were
brought, the new varieties of the Mastodons, which appear
to be common both to the Sewaliks and the Irrawaddi
deposits, may establish an identity between the two forma-
tions.

In the Sewaliks we have upheaved alluvium, or débris
from the great Himalayahs upheaved at a considerable
angle; at those points especially between the Jumna and

1 The French mode of writing this
word, Gavial, appears to have originated
in a misreading of the manuscript of
some naturalist; the 7 and wv being
nearly similar in form. As Gharial is
the correct native name, there seems no
reason for perpetuating the misnomer.—
J. Prinsep. Sec. As. Soc.

2 Jn the London clay the remains of
either the true Crocodile or Cayman
with the concayo-convex vertebra are
said to have been found, the species
allied to C. a museau aigu, vide Parkin-
son Int. Org. Rem. p. 387, and also the
head of an Alligator in the London clay
of the Isle of Sheppey, found in 1832.

math

- CROCODILES. 351

Ganges rivers, where the shingle and sand are the most
developed, their appearance is similar to what we might
imagine the beds of the present rivers to exhibit, were they
to undergo a similar convulsion. The presence of the fossils
has not been satisfactorily determined on the line between
the Jumna and Ganges; those that have been already col-
lected in such great abundance are from the prolongation of
the same line between the Sutlej and the Jumna rivers. Up
to the present time they have generally been collected from
the slopes of the mountains, slips, water-courses, &c. They
have been dug out near the village of Deoni, in the Nahun
Rajah’s territory, but at this spot the position of the stratum
from which they were excavated was not satisfactorily deter-
mined. In the Ambwalla Pass, however, we had the satis-
faction of finding a large fragment of bone in situ, in a
stratum of sandstone rock, in the face of a cliff, terminating
one of those tortuous little streams that drain the steep
slopes of the mountains into the main channel. The sand-
stone stratum in which this was found was inclined as usual
in an angle of from 20° to 30°; and the position of the fossil
was perhaps 600 feet from the bed of the main river. In the
present state of the inquiry this fact is interesting, for it
appears that in the many slips that have been visited and
most carefully examined, no fossils have been found actually
m the rock, with the exception of the instances above
mentioned. The fossils are evidently not confined to the
sandstone ; the clays and clayey conglomerates have their
proportion also.

Of the Crocodile of these strata I have attempted in the
preceding section to show, as far as measurements and my
limited means point out, that the main difference between
the fossil and the existing animal of the present rivers is in
the breadth ; a difference that might tend to an opinion of
its being allied to the Cayman, did not other more distinct
characters separate it at once from that sub-genus. In the
Gharial now under review I am unable to recognize any
difference from the living animal; and there are certain
peculiarities about the external surface of the skull of the
existing Gharial, in slight indentations and rugosities, which
are singularly coincident with those of the fossil. The fol-
lowing measurements are taken from two recent skulls, one
of an animal 10 feet 5 inches long, and the other 8 feet 8
inches long; and from a very perfect fossil skull with the beak
broken off, which is evidently the remain of a large animal.

352 FAUNA ANTIQUA SIVALENSIS.

Upper Jaw
(Number of Teeth, 56)
Extreme length from point of muzzle to
outer margin of occipital condyle .
Breadth on the temporal bones at the ar-
ticulation with lower jaw
Ditto on the 28th tooth .
Ditto on the 20th tooth .
Length of intermaxillary on suture (below)
Ditto maxillaries ditto . ¢
Ditto palatine bone ditto ‘
Ditto sphenoid ditto to anterior margin of
the posterior nasal foramen .
Extreme breadth on pterygoid apophysis
of sphenoid ;
Length from tip of muzzle to anterior of
orbit .
Ditto ditto ditto of lachrymal bone .
Breadth of frontal between orbits at the
junction with the anterior frontal
Ditto of parietal bones between crotaphite
foramina
Length of external nostril
Breadth of ditta :
Length of crotaphite foramina, or those in
rear of orbits
Breadth of ditto ditto
Ditto of occipital condyle
Length of occipital foramen
Depth of occipital foramen
Length of palatal sinus .
Breadth of ditto
Length from point of muzzle to anterior
extreme of palatine sinus . c
Length of upper table of cranium between
the anterior margin of the orbits and
the posterior mastoid apophysis .
Breadth of ditto ditto ditto
Lower Jaw
(Number of Teeth, 50)
Length of symphysis
Ditto on prolongation to posterior extreme
of articular bone.
Breadth on articular bones
Ditto on 23rd tooth : - -
Ditto 15th ditto . : : ‘ :
Ditto 3rd ditto A : 5 - ;
Ditto 2nd ditto .
Ditto 1st ditto
Length of oval hole at posterior extreme
of dental . ; : .
Depth of ditto :
Ditto of jaw on this oval hole .
Ditto on the 15th tooth .

Existing Gharial

Fossil

Gharial
10ft. din. long | Sft. Sin. long
In. Met. In. | Mét. | In. | M et
\22°7 |0°577 {19° |0-495} — | —
8:2 |0-21 65 |0°166| — =
4:3 |0-108 3°4 |0°086 | 9°3 |0°236
1°95 |0°049 15 |0°0388| — =F
ol (07129 44 |0-111) — Gaal
8:8 |0:223 79 |0:200)| — are
4:8 /0-°121 3°8 |0:096) — =
1°65 |0°042 1-4 |0°085] 38:2 |0°081
5:3 /0°133 4°15 |0°106 | — —¥
16-4 (0-416 |16°4 |0°366| — ——
14-4 |0°366 {12-2 |0-309| — =
2°4 |0°060 1:95 |0:049 | 4:9 |0°124
0°55 |0°014 0°65 0°016 | 1:0 |0°026
11 |0:027 0°85 0°021 | — ==
0-9 |0:022 0°60 0:015 | — =
2:0 |0:061 14 |0:035 | 3°3 |0°083
2°15 |0°055 16 (0:040] 3:8 |0°096
1:2 |0°030 0°95 0:024 | 2°5 \0°064
0-9 |0:022 0°80 /0°020 1:4 0°035
0°65 '0°016 045 0°012)| 0:9 0°022
2°65 |0-068 2°35 10°059 | 4°9 0124
14 |0:035 1:10 0°046 | 2:4 |0-:060
16°45 |0°418 |14°2 (0-361!) — =
58 |0°147 4°7 |0°120 |10°4 |0°264
60 |0°162 4:7 |0°120 |11°4 0-289
15'1 |0°384 |18°1 |0°332 | — _—
10°7 |0°271 9-3 |0°2386| — ——
86 |0°218 5:8 107147} — —_—
3:0 |0°076 2°5 |0°064 | — —_
16 |0°040 1:3 |0°032 | — —_
1°55 |0°0389 1:3 |0°0382 | — —_
1:85 |0°047 1-6 |0°040 | — =
14 |0°035 1:2 |0:080| — _
| 1°35 |0°0384 1:3 {0-032 | — —_—
0°6 |0:0155| 0°45 |0°012 | — —_—
2:0 |0°050 1:5 |0°038 | — =
0-9 |0°022 0:7 1007 = as

CROCODILES. 353

This fossil is water-worn, and some of the projecting bones,
especially of the sphenoid, are mutilated at the extremities ;
but the general character of the head and the form and
position of the foramina, &c., appear to correspond com-
pletely with those of the existing Gharial.'

There is no approach to any of the peculiarities pointed
out by Cuvier in the Caen and other fossils. On the upper
surface we have, in the frontal, the same concavity between
the orbits; the same form of the crotaphite foramina, with
the parietal surface between them of the same comparative
width; the posterior frontal separating the orbit from the
crotaphite foramina corresponds; the form of the mastoid
bones, both in themselves and at their articulation with the
apophysis of the os tympani, strictly agree with the existing
Gharial of the present rivers. The same may be said of the
lateral and lower faces, in the external widening out of the
pterygoid apophyses in the situation of the hinder nasal
fossa, in the elevation of the orbital edge of the frontal, with
the deep emargination, in the form and proportions of the
jugal, and the temporal fossa, and in the sharp elongated
internal process of the squamous bone: the form of the
palatine holes, and the relative situation of the teeth to these
holes, are also points which agree with the living animal !

The animal to which this fossil belonged was not quite
20 feet long; the complete head, from the tip of the muzzle
to the posterior margin of the occipital condyle, being about
47 inches. The measurements which I have made of the
existing Gharial show the proportion of the head to the length
of the animal as 1 to 51.

The following measurements of another fragment, con-
sisting of the anterior extremity of the beak or muzzle of the
upper jaw, will still further go to establish the resemblance.?

Existing Animal
Re Se OE Pa RE es eS Fossil
10ft. 5in. long 8ft. 8in. long
In. Met. In. Met. In. Met.

Length of intermaxillary on

suture below. ‘ 6 - | S11 | 07129} 4:4 0-111} 8:30 | 0:210
Breadth on 9th tooth . 5 «| 1455 | 0°038| 1:25 | 0:081)| 2-70 | 0-069
Ditto 4th ditto 5 5 . » | 2°00 | 0°050} 1°60 | 0-040} 3°40 | 0:086
Ditto 3rd ditto ‘ é ; - | 2°20 | 0:055} 1:80 | 0:046} 4:00 | 0°101
Ditto 1st ditto . 5 A - | 0-70 | 0°017| 0°60 | 0015.) 1:40 | 0:035
Depth on 9th tooth , " - | 0°90 | 0°022) 0°70 | 0°017} 1°40 | 0:035

The above is the remain of a smaller animal than the

* Brit. Mus. Cat., No, 40,206.—[Eb. | ? Brit. Mus. Cat., No. 39,811.—[Ep.]
VOL. I. AA

354 FAUNA ANTIQUA SIVALENSIS.

former one, showing the alveoli and some of the teeth, as far
back as the 10th, on each side of the maxillaries. A more
perfect resemblance to the living animal than this could not
well be conceived; and it moreover establishes, in the absence
of a connected beak and skull, that the fossil animal had
precisely the same number of teeth with the living species.
The suture connecting the intermaxillaries with the maxil-
laries is fortunately strongly marked in the fossil; the
posterior point of the suture occurring opposite the ninth
tooth, exactly as it does in the existing animal. The teeth,
the form of anterior extremity of muzzle, the outer nasal
aperture, with the lower indentations, correspond in every
way; and, to descend still further to minutiz, at the com-
mencement of the suture connecting the intermaxillaries and
maxillaries, at a point in the former bone immediately in
front of the sixth tooth, is a small hollow or indentation ;
this hollow exists in the same situation and bears the same
form in our fossil Gharial.

Of the lower jaw we have only an imperfect fragment of
the two branches connected at the commencement of the
symphysis; from the extreme hardness of the crystalline
rock in which it is embedded we are unable to see further
than that the angle of these branches corresponds with the
existing animal, a point however which is proved by the
fragments of skull which are in our possession, and which,
imperfect as they are at the muzzle extremity, show dis-
tinctly the commencement of that tapering form which is
peculiar to the Gharial of the present rivers.

In volume v. of the ‘Ossemens Fossiles,? Cuvier, in re-
capitulating the peculiarities and differences between the
Crocodiles and Gharials, says of the latter: ‘Les ptéry-
goidiens forment au-dessus des palatins des espéces de
grosses vessies renflées et ovales de la grosseur d’un ceuf de
poule, au lieu d’une simple voute cylindrique comme dans
les crocodiles et les caimans,’ &c., and then, ‘Je n’ai point
observé cette vessie dans le petit Gavial, mais je suppose
d’autant plus qu’elle est un produit de VPage, que dans les
vieux crocodiles des Indes cet endroit est beaucoup plus
renflé que dans les jeunes.’

These demi-cylindrical swellings are highly developed in
the 10 feet 5 inches specimen, of which the measurements
have been given; whereas in the smaller and younger
animal, measuring 8 feet 8 inches, there is no appearance
of them ; the sphenoid portion lying under the palatine and
extending up to the anterior frontal’s apophysis, in a flat
uninflated laminated bone. From the little difference that
exists between the bones of the Gharial and of the Crocodile,

BA

=
'
“
\
4
.
}
Mh Lie gs
ate |

The figures in this Plate have been drawn by Mr. Dinkel.
from the original specimens in the British Museum, which
have been identified with outline tracings named by Dr.
Falconer.

Fig. 1. Profile view of skull and lower jaw of Crocodilus crassidens,

Fig.

Fig.

Fig.

VOL. I.

CROCODILUS CRASSIDENS, CrocopiLus LEPTODUS, AND

. Is another specimen of a fragment of cranium of the same

. Shows a portion of the cranium, the palate, and teeth of the

. Another fragment of muzzle of Croc. Leptodus, drawn one-

. Very perfect specimen of anterior portion of muzzle of Leptor-

DESCRIPTION OF PLATE XXIX.

LEPTORHYNCHUS GANGETICUS.

an extinct species found in the Sewalik Hills. The specimen
bears the number 39,802 in the British Museum Catalogue, —
and is drawn one-sixth of the natural size. (See pages 297

& 355.)

species (C. crassidens), showing the palate surface and teeth.
It is numbered 59,803 in the British Museum Catalogue, and is”
drawn one-sixth of the natural size.

elongated muzzle of Crocodilus Leptodus, an extinct species
found in the Sewalik Hills. The figure is drawn one-sixth
of the natural size, and the specimen is numbered 39,805 in the
British Museum Catalogue. (See page 355.)

fourth of the natural size. (Brit. Mus. Cat. No. 39,806.)

hynchus Gangeticus, an existing species (the Gavial) found fossil
in the Sewalik Hills. The specimen has been drawn one-fourth
of the natural size, and is numbered 39,81la in the British
Museum. The artist has represented the under surface of the
muzzle instead of the upper. The latter shows well the cha-
racteristic external nasal aperture. The corresponding specimen
described in Capt. Cautley’s memoir is numbered 39,811. ( See
page 353.) Bs

CROCODILES. 355

we are unable to separate the remains of one from those
of the other. A great quantity have been found—teeth, osse-
ous plates, ribs, vertebrae, &c.; the latter having the concavo-
convex body, and the sacral vertebre, with their transverse
processes compressed and cylindrical, agree in every respect
with the existing animal.

NortHern Doas: October 1, 1835.

APPENDIX TO MEMOIR ON CROCODILES.
I.—NorveE sy EpiTor.

Subsequent investigations in the British Museum and elsewhere
caused doubt in Dr. Falconer’s mind as to the correctness of the
identifications by Capt. Cautley in the above memoir. This is indicated
by the following passage from a letter dated London, Jan. 5th, 1844,
addressed to Capt. Cautley in India :—

‘ Our Crocodile of the Upper Provinces is not the C. biporcatus of
Cuvier—of that I am certain. There appear to be three existing
Indian species of Crocodile; as distinct from Gavial, viz.: C. bipor-
catus, with a long sharp head; C. palustris, with a shorter and broader
head; and C. bombifrons, undescribed, very short and blunt. Of these,
the head figured by you and given with dimensions in the “ Asiatic
Researches,” appears to me to be C. bombifrons, undescribed, but so
named in the British Museum.’

The following extracts, however, from letters also addressed to Capt.
Cautley, show that Dr. F. had no doubt as to the identity of the fossil
with the existing Crocodile and Gharial :—

Aug. 5th, 1843. ‘He (Mr. L.) doubted my confident assertion that
our fossil Gavial was identical with the existing species.’

Dec. 6th, 1843. ‘I gave C. a flat contradiction on his saying that
existing species of Crocodile had been found before, with extinct genera
of animals. This in reference to our assertion of the fossil Crocodiles
being recent species.’

Four species of Crocodile among the Sewalik fossils were afterwards
determined by Dr. Falconer in the British Museum, viz.: Crocodilus
bombifrons ; Croc. (Leptorhynchus) Gangeticus ; Croc. (Leptorhynchus)
Leptodus; and Croc. (Leptorhynchus) crassidens. The two first are
identical with existing species. The last was described, in 1837, as
‘an immense species, far exceeding existing ones, and forming a passage
from the Gavials into the true Crocodiles. It has the cylindrical muzzle
and synostorized lower jaw of the former, with the blunt, thick teeth
of the latter.’ See antea, page 297.

Outline figures of these species, found among Dr. Falconer’s papers,
have been identified with the specimens in the British Museum; and
from these specimens the illustrations in Plates XXVIII. and XXIX.
have been drawn by Mr. Dinkel.—[Ep. ]

AA2

356 FAUNA ANTIQUA SIVALENSIS.

IJ.—MemoranDA ON THE Betrast Crocopites. From Dr. Fatconer’s
Nore-zBoox, May 22, 1845.

Made a comparison with Thompson of the Belfast specimen of the
long-muzzled Crocodile head, procured by Dr. McCormac from the river
Sierra Leone. This specimen certainly belongs to the Crocodilus
cataphractus of Cuvier. It corresponds exactly with the description
of the Crocodilus cataphractus given by Dumeril and Bibron, in the
number of teeth, form of the head, muzzle, &c. Compared it with the
figures of Crocodilus Schlegelii given in Miiller’s Atlas. It comes
pretty near this, but is certainly different, having fewer teeth in both
jaws, and the commissure of the lower jaw being much shorter in
proportion than in the C. Schlegelii. It is more of a Crocodile than the
C. Schlegelit which comes nearer the Gavial. The nasal bones in both
join with the intermaxillaries, but in the C. Schlegeli the slips of the
intermaxillaries run higher up between the maxillaries: and the
lateral outline in C. Schlegelii is more cylindrical, being strongly
undulated in C. cataphractus. The Crocodilus cataphractus is evidently
identical with the Mecistops Bennettii of Gray. The Belfast specimen
corresponds very closely in size with the stuffed specimen in the British
Museum. The head is about 17 inches long ; the extreme breadth at
the condyles, 7 inches; do. behind the orbits, 4? inches ; do. in line
with anterior border of do., 55 inches; do. at seventeenth or last tooth,
32 inches. This last tooth extends to within + inch of the orbit.
The muzzle contracts between the eleventh and twelfth teeth to 12
inch; it then expands to 2 inches opposite the ninth teeth, which are
the largest in the head. From the ninth it goes on conengiiae to the
space between the fourth and fifth (this interval being ? inch) where it
gets contracted to 1 inch. At the extremity of the muzzle, between
the second and third, it expands to 1? inch.

Its detailed dimensions are as follows :—

1. Length from the tip of the muzzle to the ieee ue » 155
2. Length from ditto to maxillary condyle : glide
3. Extreme width of cranium at condyles . : :
4. Interval between the orbits é 3 0°75
5. Length from the occipital ridge to the extremity of the
nasals 6
6. Length from tip of muzzle to ditto 9
7. Length of orbit : é 5 : : . . 1
8. Width of ditto . ’ . “ : : . 1
9. Length of crotaphite foramen . 5 3 - : 1
10. Width of ditto . : ’ 5 . : 0
11. Width of muzzle at the tip of nasals : ’ : 5 2°
12. Ditto at 9th tooth : 6 . . , . 2°
13. Ditto of contraction behind 4th tooth . f : 1
14. Width of dilatation of muzzle 1
15. Length of nasal opening 0
16. Width of ditto ‘ 0
17. Number of teeth in the “upper jaw, vey
18. Largest teeth in upper jaw, 3rd and 9th.
19. Length of intermaxillary bone at the apa : . 3
20. Length of maxillary ditto . 5 : : ¥ ’ 6°25
21. Number of teeth in lower jaw, 16.
22. Largest teeth in lower jaw, Ist, 4th, 10th, and 11th.

CROCODILES. 357

TI].—Memoranpum By Dr. FaLconer on CrocopILus SCHLEGELII.

The Crocodilus Schlegelii is not a true Gavial, but a Crocodile.
1. The cranial tablet, instead of being square to behind the orbits
converges on either side. 2. The crotaphite foramina are smaller than
the orbits considerably. 3. The nasals descend so as to join on with
the long wedge apophyses of the intermaxillaries, whereas in the Gavial
they are separated, and are a third of the entire length of the muzzle.
The cranial tablet in C. Sehlegelii has very much the form as in the
figure of C. dwcius in Cuvier’s ‘ Ossemens Foss,’ 4. The upper wedge-
shaped terminations of the intermaxillaries are larger and more slender
than in Gavial. 5. The outer border of the orbits do not overhang the
maxillaries as in the Gavial, but are within them as in the Crocodiles.
6. The greatest diameter of the orbit is not transverse as in Gavial, but
lengthwise, and the form is very much as in C. biporcatus. 7. There
is no considerable contraction behind the fifth tooth of the muzzle in
the upper jaw as in Gavial, and no appearance of the great enonarrhine
cartilaginous sac of Gavial. In C. Schlegeli the anterior ends of the
palatines terminate in a line with the great palatine holes, while in
the Crocodiles they descend in slips between the palatine portion of
the maxillaries, as also in the Gayial.

The form of Croc. Schlegelii agrees very closely with that of C.
Journet, described by Bibron and Dumeril. In C. Schlegelit there are
twenty teeth in the upper jaw and nineteen or twenty in the lower. In C.
Journei there are eighteen in the upper and only fifteen in the lower. The
length of head in an old C. Schlegelit is to its greatest width as 22 to 1,
which is exactly the same proportion asin C. Journe?. Width at anterior
border of orbits } nearly in C. Schlegelii, and 4 in C. Journet; opposite
ninth tooth in C. Schlegelii ¢ to 4, in C. Journei } at tenth tooth ; tip
of muzzle in C. Journei }, in C. Schlegelii } to }. (But the dimensions
of muzzle in C. Schlegelit vary with the age. The muzzle is longer in
proportion to its diameter in the young than in the old animal.) The
longest teeth in C. Journei are the first, fifth, and tenth, in upper jaw,
and the first and fourth in lower. In C. Schlegelii they are the first,
fifth, and ninth, in upper jaw, and the first and fourth in lower,

ITV.—Descrirtions By Dr. FALCONER oF Foss REMAINS OF CROCODILES,
From Ava, Perim IsLAND, AND THE NERBUDDA, IN THE CaTA-
LOGUE OF THE MusEuM OF THE ASIATIC SOCIETY OF BENGAL.

The following from the Sewalik hills are most noteworthy [ Ep. | :—

No. 642. Leptorhynchus Gangeticus.—Fine fragment of cranium with-
out the lower maxilla, deficient only from the muzzle portion in front
of the orbits; the whole of the occipital region with the occipital
condyle nearly entire. The transverse articular surface of the tym-
panic bone and a part of the squamous portion adjoining removed; the
cranial tablet perfect, as also the two crotaphite foramina, together with
both orbits; palatines shown underneath, but pterygoids broken off.
The alveoli of only three small teeth shown on the broken part of the
right maxilla. The contraction of the muzzle, the form of orbits, their
elevation above the frontal, the cranial tablet, and the form and pro-
portion of the crotaphite foramina to orbits and the other adjoining parts,

358 FAUNA ANTIQUA SIVALENSIS.

correspond so exactly in every respect that no determinable difference
is perceivable. The skull is larger than that of a Gavial’s head, which
measures 19 in. The specimen infiltrated with iron and very heavy.

No. 652. Leptorhynchus.—Fragment of muzzle-portion of lower jaw
of an enormous-sized animal ; the left side showing the empty alveoli of
the five anterior teeth of very large size, broken off obliquely back-
wards; the right side broken off nearly opposite the termination of the
last alveolus on the left side and continued backwards, showing the
alveoli of five teeth, four of which contain more or less of the teeth, its
anterior portion wanting.

Dimensions. Paphos
Length of fragment . é ; : : . Sead . 13°5
Width about middle. ¢ : : : : 3 : : 9°5
Thickness : : : - : : . . . 46

The alveoli of the two first teeth are very large; the third and fourth
are still larger, the third showing a diameter of 1°8 in. by 1:4 in.; the
sixth tooth is smaller, being 1 in. in diameter ; the seventh is 1:4 by
1-2 in. in diameter; the eighth and ninth are smaller.

Marked ‘Fossil Alligator’ in J. Prinsep’s hand-writing; no other
mark or No.; but vol. v. p. 180 of Jour. As. Soc. contains a notice of
a Crocodile of gigantic size with a cylindrical muzzle, and with teeth
projecting three inches above the jaw and one inch two lines in diameter.

No. 653. Leptorhynchus.—Portion of lower jaw left side, comprising
a part of the horizontal ramus behind the symphysis, and containing
the alveoli of two teeth. The suture of the opercular bone is visible ;
alveoli of enormous size, the anterior being about 1 in. 5 lines in
diameter.

Inches
Height of jaw taken saat last tooth . . : . ° oy (6r%
Thickness of ditto . : . . . ¢ . 56
Length of fragment ° : . . : 5 . : . 10°65

Probably belongs to an animal of corresponding size with 652, but
it differs in weathering and colour of surface.

COLOSSOCHELYS ATLAS.

XX. ON THE COLOSSOCHELYS ATLAS, A GIGANTIC
TORTOISE FROM THE SEWALIK HILLS.!

I.—Own some Remains OF THE MEGALOCHELYS SIVALENSIS,
A GIGANTIC TORTOISE FROM THE TERTIARY STRATA
OF THE SEWALIK HILLSs.?

BY H. FALCONER, M.D.

TuE class of Reptilia has been prolific in colossal and surpris-
ing forms, beyond all others; nor are its wonders at an end.
Year after year has continuously added something new, till
air, land, and sea have alike vindicated by their representa-
tives the pre-eminence of Saurians above all other monsters.
The Iguanodon of Europe had not ceased to create astonish-
ment by the vastness of its dimensions, when the Basilo-
saurus of America comes forth to claim a higher rank, and ex-
hibit a reptile still more enormous; and, when the fossiliferous
deposits of the tropics shall have unfolded their contents,
something still more may be produced.

The analogy of nature would a priori lead to the presump-
tion that gigantic types of the Chelonian reptiles must have
formerly existed, as well as Saurian ones. We see the
existing species extending through the same range of habit
and organization, some carnivorous and some herbivorous,
some marine or fluviatile and others terrestrial. The same
analogy would lead us to infer that the largest types would be
found among the terrestrial and herbivorous groups, as this
holds among the Saurians. But, so far as we know, no very
immense J'estudinata have hitherto been discovered, although
such might have been expected to be associated with the
Iguanodon and Megalosaurus. So much is it the reverse,
that the Hmydes and Trionyches of the strata which have

1 Unfortunately no complete memoir
on the Colosscochelys was ever published,
and there are no drawings of this re-
markable fossil in the ‘Fauna Antiqua
Sivalensis.’ The documents, however,
now brought together remedy in a great
measure this defect. The remains of
this gigantic tortoise are preserved in
the British Museum, where there is also
a restoration of the shell. The illus-
trations in Plates xxx. and xxxi. have
been drawn by Mr. Dinkel, from the

specimens in the British Museum.—
[ Ep. ]

2 This memoir was commenced about
1837, but was never finished. The fossil,
however, was referred to under the
designation Megalochelys Sivalensis, in
Journ. Asiat. Soc. vi. 358, 1837. The
name was afterwards changed to Ce-
lossochelys Atlas, ‘as the term ‘“‘ Mega-
lochelys” was thought not to convey a
sufficiently expressive idea of the size,’
[Ep.]

360 FAUNA ANTIQUA SIVALENSIS.

_ yielded these are not larger than existing species of the same
genera. The largest remain on record, within our means of
knowing, is mentioned by Cuvier,! as a radius of a sea-turtle
which indicated an animal with a buckler eight feet long,
procured from the quarries near Luneville, of the age of
the jurasic limestone.

In a previous communication? we intimated our haying
become possessed of bones of the extremities, with corre-
sponding fragments of bucklers, of a Chelonian from the
tertiary deposits of the Sewalik hills, as large as the corre-
sponding bones of the Indian Rhinoceros. These form the
subject of the present notice.

The most perfect specimen comprises somewhat less than
the upper half of the left humerus, and is, so far as it goes,
very completely preserved. The articulating head, except
where splintered by the chisel, is entire, and so also are the

two great crests. The dimensions are:—
Inches

1. Extreme length of the fragment . . . 14°4

2, Greatest. diameter from the convexity of ‘the articulating head
to margin inner tuberosity . 9°7

3. Length of the inner tuberosity, from where it rises from the shaft
of the bone . 10°4

4. Greatest width of ditto from its outer margin to the border of
the neck. Pie ig?
5. Projection of ditto above the level of the articulating head + 25
6. Greatest diameter of articulating head _ . : . «OL
7. Transverse ditto between the roots of the crests . » 45
8. Antero-posterior diameter of the shaft where broken across Ae ell
9. Transverse ditto ditto . : 5 . : . - 89d
10. Width of smaller crest . = : : . . . «gO:

The humerus and femur of the Testudinata are so much
alike that it is difficult to distinguish them, except by the
lower end. Our specimen we make out to be a humerus,
and it is of some importance to ascertain this point in esti-
mating the size of the animal:—Ist. From the form of the
inter-tubercular hollow, which has a ridge across dividing it
into two fosse: this is peculiar to the humerus. 2nd. By
the spherical form of the articulating head, which in the
femur is oblong. 3rd. By the smaller crest, which runs
partly across the direction of the articulating head, exactly
as in the humerus of the Hmydes. It differs from the humerus
of any of the known sub-genera of the family. In the land-
tortoises the deltoid crest hardly rises above the level of the
articulating head ; and the smaller crest reaches only as high
as the inferior level; the fossil is strongly contrasted in both
respects. In Chelys and Trionyx the crests are wide apart,
so as in the former to point in opposite directions with the
same plane. In the fossil they are approximated, and partly

? Ossemens Fossiles, tom. y. part ii., ? On the remain of a fossil Monkey.
p. 526. [See antea, p. 297.—Ep.]

a

COLOSSOCHELYS ATLAS. 361

parallel. The humerus of the sea-turtles is so peculiar as to
have no analogy with the fossil except in the olecranon-like
production of the deltoid crest.

It has more resemblance to the Emys, or the box-tortoises,
than to any of the others. The two crests are approximated.
in both—as also in Testudo, and the smaller crest in both
reaches half way up, and well across the articulating head.
With this resemblance, however, it differs sufficiently. The
deltoid crest is greatly more developed than in Emys; it is
more pointed upwards, and projects more; and runs down with
a convex outline where it falls off in Emys. This convexity
and pointed form of crest is not seen in any of the existing
Testudinata. The articulating head is very globular, the
greatest diameter being 5:1 inches, and across between the
roots of the crests, 4°5. Calculated on the proportions
of a small Emys, the dimensions would give a buckler of
12 feet in length. We possess similar fragments of the right
humerus and a femur; but the crest and articulating head
are broken off, and have no character for description.

Of the bucklers we have a great variety of fragments ; some
of them indicating animals of prodigious size, but it is difficult
to make anything intelligible out of them, as the ribs and
plates of the plastron have become completely synostosed,
and afford no trace of suture. But some of them are suffi-
ciently distinct to show that the animal to which they
belonged differed remarkably from all known Testudinata.
One of these, which we consider to be the caudal extremity of
the plastron, has a wide and deep re-entering angle between
two projecting horns. The only thing approaching this is
the slight emargination of the plastron in Emys.

Some other pieces are very remarkable. They rise out of
the inner surface of the shield and project beyond its margin
with a rugged and gradually thinning convex edge, like the
ribs of Trionyx. The dimensions are :—

Inches

Width of the rib-like extremity , : . . » 8-2 and 8°6
Greatest thickness . fj ; i Mice, = . 6°5 and 4:2
Least thickness where hollowed out . 2 3 : : 08

Supposing them to be ribs, which, however, is by no means
certain, what prodigious animals must they not have come
from? But when we are certain of such large tortoises as
those indicated by the humerus described, it were easy to
imagine others with humeri as large as those of the largest
mastodons. ‘The surface of the shield-plates in all the great
fragments is rough and loosely cancellated, like the interior
structure of the articulating heads of bones. There is none of
the shagreening of Trionyx, nor of the smooth surface of Emys.

To what description of Testudinata do the humerus and

362 FAUNA ANTIQUA SIVALENSIS.

these bones belong? This is a question not easily to be solved
at present. The characters of the humerus would determine
the animal to have been allied to Emys and Testudo; but
on the ragged edge of the buckler fragments, as described
‘above, we see every appearance of the attachment of a soft
substance, and the resemblance is very strong to Trionyx.
It would appear that part of the case was cartilaginous, as in
that genus, and not completely ossified. Some large frag-
ments have the character of a line of marginal plates, which
would separate them at once from Trionyx, and possibly the
cartilaginous parts were confined to the plastron.

IL—On THE OSTEOLOGICAL CHARACTERS AND PALMHONTO-
LoGicAL History oF THE COLOSSOCHELYS ATLAS.
ABSTRACT OF AN ExTEMPORE COMMUNICATION TO
ZoouocicaL Society oF Lonpon, Marcu 26TH AND
May 147H, 1844.}

BY H. FALCONER, M.D.

Part I.

A communication was made by Dr. Falconer, conveying the
substance of a paper by Captain Cautley and himself on the
osteological characters and paleontological history of the
Colossochelys Atlas, a fossil tortoise of enormous size, from
the tertiary strata of the Sewalik hills in the north of India
—a tertiary chain apparently formed by the detritus of the
Himalayah mountains.

A great number of huge fragments derived from all parts
of the skeleton except the neck and tail were exhibited on
the table, illustrative of a diagram by Mr. Scharf of the
animal restored to the natural size. (See Fig. 12.)

RESTORATION OF COLOSSOCHELYS ATLAS, REDUCED.

1 Reprinted from Proc. Zool, Soc. Lond. 1844, Part xii. pp. 54 and 84,

‘COLOSSOCHELYS ATLAS. 363

The communication opened with a reference to the reptilian

forms discovered in the fossil state, among which colossal
-representatives have been found of all the known tribes, such
as the Iguanodon, Megalosawrus, Labyrinthodon, &c., besides
numerous forms of which no living analogues exist, such as
the Enaliosawrian reptiles and Pterodactyles. No fossil Tes-
tudinata remarkable either for size or deviation from existing
forms have hitherto been found in the fossil state. The
Colossochelys supplies the blank in the first respect, while it
differs so little from the land-tortoises in the general con-
struction of its osseous frame, as hardly to constitute more
than a sub-genus of Testudo.

The plastron or sternal portion of the shell affords the
chief distinctive character. The episternal portion in the
adult is six and a half inches thick, and contracted into a
diameter of eight inches, bifid at the apex, and supplied with
a thick cuneiform keel on its inferior surface ; this keel con-
stitutes one of the principal features in the fossil. The
entosternal portion exhibits exactly the form of Testudo, the
same being the case with the xiphosternal or posterior por-
tion. The plastron in the adult animal was estimated to be
nine feet four inches long. (See Plates XXX. and XXXII.)

The carapace or buckler of the shell coincides exactly with
the general form of the large land-tortoises, of which it
exhibits only a magnified representation, flattened at the top
and vertical at the sides, with the same outline and recurved
margin. The shell was estimated to have been twelve feet
three inches long, eight feet in diameter, and six feet high.
The head was estimated to have been two feet long, and the
whole animal upwards of 20 feet.

The extremities were described as constructed exactly as
in the land-tortoises, in which the form of the femur and
humerus is marked by peculiar characters. These bones in
the fossil were of a huge size, corresponding to the dimen-
sions of the shell. The ungual bones indicated a foot as

1 On a slip of paper Dr. Falconer | the Testudo elephantopus, in the College
draws the following comparison between | of Surgeons, and the Colossochelys Atlas.
Test. eleph. Coloss. Atlas.

From the episternal to end of entosternal 10 inches 2 feet 11 inches
Length of plastron . ; : Fig see Le eed
Length of carapace. ; a ee Leen anes
Ditto along the curve . A F 2-04 * 35 Lars OMe

The following calculation is then made as to the total length of the Colossochelys :
Feet In.

Length of carapace . : - Z 3 : - : pe}

5 neck c : é : : : ; a eaGu.O

5 head 2 0

A tail : é . 4 ome, O

Total : ; . 22 38

364 FAUNA ANTIQUA SIVALENSIS.

large as that of the largest Rhinoceros. The humerus was
more curved, and the articulating head more globular and
deeper in the fossil, from which it was inferred that it had a
stronger articulation and greater rotation, and that the Co-
lossochelys was enabled to bring its anterior extremities
more under its weight than is the case with existing tortoises.

The affinities with Testudo shown in the shell and extremi-
ties were found to hold equally good in the construction of
the head, of which a comparatively small-sized specimen,
inferred to have belonged to a young or half-grown Colosso-
chelys, was exhibited. The head of the adult to correspond
with the dimensions of the shell, and according to the pro-
portions furnished by a large Testudo Indica, was deduced to
have been two feet long.

There were no ascertained cervical vertebrae to afford
direct evidence as to the length of the neck, which was con-
structed in the diagram relatively to the proportions of
Testudo Indica. The entire length of the Colossochelys Atlas
was inferred to have been about eighteen feet,’ and it was
believed to have stood upwards of seven feet high.

The generic name given by the discoverers has reference to
the colossal size of the fossil (codoccds et xédvs), and the
specific one to its fitting representation of the mythological
tortoise that sustained the world, according to the systems of
Indian cosmogony.

The anatomical details occupied so much of the evening,
that space was not left for Dr. Falconer to enter on general
points connected with the fossil, such as its possible con-
nection with the mythological fables of the Hindoos and the
zera of its extinction, which will form the subject of another
communication.

The results of a chemical analysis of the bones by Mr.
Middleton were communicated, showing that they contained
a very large quantity of fluorine. Some rough sketches of
the Colossochelys were exhibited, etched on glass by means of
the fluorine yielded by its own bones. The analysis indicated
the presence of 11 per cent. of fluoride of calcium.

Parr If.

On a former meeting we went through the anatomical
characters presented by the remains of the Colossochelys _
Atlas. Commencing with the plastron, we traced the modi- —
fications of form through the costal elements of the carapace —
and the dorsal vertebre, all of which bear the closest resem- —
blance to the ordinary type.of the Chersite Chelonians, or —

1 See note, p. 363.—[ Ep. ]

~

as

DESCRIPTION OF PLATE XXX.
CoLOssocHELYS ATLAS.

The figures in this Plate have been drawn by Mr. Dinkel
from the original specimens in the British Museum.

Fig. 1. The bifid episternum of Colossochelys Atlas, seen from the
under surface, showing the thick cuneiform keel. The specimen
has been drawn one-sixth of the natural size, and bears the
number 40,630 in the British Museum Catalogue. (See pages
363, 374, & 378.)

Fig. 2. Shows the xyphosternal or posterior portion of the plastron
of Colossochelys Atlas, one-sixth of the natural size. The pos- |
terior cleft is well seen. The specimen is numbered 40,629 in
the British Museum Catalogue. (See pages 363 & 378.)

Figs. 3, 4, and 5. Represent three different views of one of the ungual
bones of Colossochelys Atlas, drawn one-half of the natural size.
Brit. Mus. Cat. No. 36,731. (See pages 364, 374 & 381.)

VoL. I.

Vol. I. Plate 30.

W-West imp .

de].et lith

J.Dinkel

Golossochelys Atlas.

COLOSSOCHELYS ATLAS. 365

true land-tortoises. A like result followed the examination
of the extremities, which, as exhibited in the remains of the
humerus, femur, and ungual phalanges, were seen to be con-
structed exactly on the plan of Testudo, with columnar legs
and truncated club-shaped feet, as in the proboscidean
Pachydermata. The same direction of affinity was observed
throughout the confirmation of the head. The only portions
of the skeleton from which more or less direct evidence was
not derived were the neck and tail vertebrae, of which there
were no specimens in the collection. The general result of
the examination showed that the Colossochelys Atlas was
_ strictly a land-tortoise in every part of its bony frame; and
the impressions of the horny scutes proved the like in regard
to the arrangement of its dermal integument.

_ The principal distinctive characters were found in the
sternum, which is enormously thickened at its anterior ex-
tremity, along the united portion of the episternal bones, and
contracted into a narrow neck, so that the width of the com-
bined episternals does not much exceed their thickness ; this
thickened portion bears on its under side a deep massive
cuneiform keel, which terminates upon the commencement of
the entosternal piece. There is more or less thickening of
this part in all the species of Testudo, and the amount of it
is very variable in different individuals of the same species ;
but there is nothing approaching the same degree of con-
_ traction in reference to the thickness, nor aught like a
developed keel, in any of the existing land-tortoises which
we have either had an opportunity of examining, or seen
described in systematic works on the tribe. The keel in the
fossil is feebly shown in the young animal, but strongly
marked in the adult. Conceiving that generic distinctions
are only legitimate in the case of well-defined modifications
affecting some of the leading characters in the organization of
an animal, we do not consider ourselves warranted in attaching
a higher systematic importance to the Colossochelys than as
a subgenus of Testudo, which may technically be defined

_ thus (the distinction resting mainly on the form of the

sternum) :—

Subgen. CoLossocHELYs.

Testa solida, immobilis, sterno antice in collum valde incrassa-
tum, subtus carina crassd cuneiformi instructum, angustato.
—Testudo terrestris, statura et mole ingente (inde
nomen xodooods et yédvs) sui tribus prodigium! Olim in
Indiz Orientalis provinciis septentrionalibus degebat.

Colossochelys Atlas.
The first fossil remains of this colossal tortoise were dis-

866 FAUNA ANTIQUA SIVALENSIS.

covered by us in 1835 in the tertiary strata of the Sewalik
hills, or Sub-Himalayahs skirting the southern foot of
the great Himalayah chain. They were found associated
with the remains of four extinct species of Mastodon and
Elephant, species of Rhinoceros, Hippopotamus, Horse,
Anoplotherium, Camel, Giraffe, Sivathertum, and a vast
number of other Mammalia, including four or five species of
Quadrumana. The Sewalik fauna included also a great
number of reptilian forms, such as crocodiles and land and
freshwater tortoises. Some of the crocodiles belong to extinct
species, but others appear to be absolutely identical with
species now living in the rivers of India: we allude in par-
ticular to the Crocodilus longirostris or Gharial (Gavial), from
the existing forms of which we have been unable to detect any
difference in heads dug out of the Sewalik hills. The same
result applies to the existing Hmys tecta, now a common
species found in all parts of India. A very perfect fossil
specimen, presenting the greater part of the evidence of the
dermal scutes, is undistinguishable from the living forms,
not varying more from these than they do among each other.
Professor T. Bell, the highest living authority on the family,
after a rigid examination, confirms the result at which we
had arrived, that there are no characters shown by the fossil
to justify its separation from the living Emys tecta, There
are other cases which appear to yield similar results, but the
evidence has not yet been sufficiently examined to justify a
confident affirmation of the identity at present.

The remains of the Colossochelys were collected, during a
period of eight or nine years, along a range of eighty miles
of hilly country; they belong in consequence to a great
number of different animals, varying in size andage. From
the circumstances under which they are met with, in crushed
fragments, contained in elevated strata which have under-
gone great disturbance, there is little room for hope that a
perfect shell, or anything approaching a complete skeleton,
will ever be found in the Sewalik hills. It is to be men-
tioned, however, that remains of many of the animals asso-
ciated with the Colossochelys in the Sewalik hills have been dis-
covered along the banks of the Ivrawaddi in Ava, and in Perim
Island in the Gulf of Cambay, showing that the same extinct
fauna was formerly spread over the whole continent of India.

This is not the place to enter upon the geological ques-
tion of the age of the Sewalik strata; suffice it to say, that
the general bearing of the evidence is that they belong to
the newer tertiary period. But another question arises:
‘Are there any indications as to when this gigantic tortoise
became extinct? or are there grounds for entertaining the

COLOSSOCHELYS ATLAS. 367

opinion that it may have descended to the human period ?’
Any @ priori improbability, that an animal so hugely dis-
proportionate to existing species should have lived down to
be a contemporary with man, is destroyed by the fact that
other species of Chelonians, which were coeval with the
Colossochelys in the same fauna, have reached to the present
time; and what is true in this respect of one species in a
tribe may be equally true of every other placed under the
Same circumstances. We have as yet no direct evidence to
the point from remains dug out of recent alluvial deposits,
nor is there any historical testimony confirming it; but
there are traditions connected with the cosmogonic specula-
tions of almost all Eastern nations having reference to a
tortoise of such gigantic size, as to be associated in their
fabulous accounts with the elephant. Was this tortoise a
mere creature of the imagination, or was the idea of it
drawn from a reality, like the Colossochelys?

Without attempting to follow the tortoise feeding
through all its ramifications, we may allude to the interesting
fact of its existence even among the natives of America. The
Iroquois Indians believe that there were originally, before
the creation of the globe, six male beings in the air, but
subject to mortality. There was no female among them to
perpetuate their race; but learning that there was a being of
this sort in heaven, one of them undertook the dangerous
task of carrying her away. A bird (like the Garada of
Vishnoo, or the Eagle of Jupiter) became the vehicle. He
seduced the female by flattery and presents; she was turned
out of heaven by the Supreme Deity, but was fortunately
received upon the back of a tortoise, when the otter (an im-
portant agent in all the traditions of the American Indians)
and the fishes disturbed the mud at the bottom of the ocean,
and drawing it up round the tortoise formed a small island,
which gradually became the earth. We may trace this tra-
dition to an eastern source, from the circumstance that the
female is said to have had two sons, one of whom slew the
_ other; after which she had several children, from whom
sprang the human race.

In this fable we have no comparative data as to the size
of the tortoise, but in the Pythagorean cosmogony the infant
world is represented as having been placed on the back of an

elephant, which was sustained on a huge tortoise. It is in the
Y Hindoo accounts, however, that we find the fable most cir-

cumstantially told, and especially in what relates to the
second Avatar of Vishnoo, when the ocean was churned by
means of the mountain Mundar placed on the back of the
king of the tortoises, and the serpent Asokee used for the

868 FAUNA ANTIQUA SIVALENSIS.

churning rope. Vishnoo was made to assume the form of
the tortoise and sustain the created world on his back to
make it stable. So completely has this fable been impressed
on the faith of the country, that the Hindoos to this day
even believe that the world rests on the back of a tortoise.
Sir William Jones gives the following as a translation from
the great lyric poet Jyadeva: ‘The earth stands firm on thy
immensely broad back, which grows larger from the callus
occasioned by bearing that vast burden. O Cesava! assum-
ing the body of a tortoise, be victorious! Oh! Hurry, Lord of
the Universe !’

The next occasion in Indian mythology where the tortoise
figures prominently is in the narratives of the feats of the
bird-demigod ‘ Garida,’ the carrier of Vishnoo, After stating
the circumstances of his birth, and the disputes between his
mother Vinita and ‘ Kudroo,’ the mother of the serpents, it
is mentioned that he was sent on an expedition to bring
‘Chundra’ the moon, from whom the serpents were to derive
the water of immortality. While pursuing his journey
amidst strange adventures, Garida met his father, Kishyipa,
who directed him to ‘appease his hunger at a certain lake,
where an elephant and tortoise were fighting. The body of the
tortoise was 80 miles long—the elephant’s 160. Garada
with one claw seized the elephant, with the other the tor-
- toise, and perched with them on a tree 800 miles high.” He
is then, after sundry adventures, stated to have fled to a
mountain in an uninhabited country, and finished his repast
on the tortoise and elephant.

In these three instances, taken from Pythagoras and the
Hindoo mythology, we have reference to a gigantic form of
tortoise, comparable in size with the elephant. Hence the
question arises, are we to consider the idea as a mere fiction
of the imagination, like the minotaur and the chimera, the
griffin, the dragon, and the cartazonon, &c., or as founded on
some justifying reality? The Greek and Persian monsters
are composed of fanciful and wild combinations of different
portions of known animals into impossible forms, and, as
Cuvier fitly remarks, they are merely the progeny of uncurbed
imagination; but in the Indian cosmogonic forms we may
trace an image of congruity through the maze of exaggera-
tion with which they are invested. We have the elephant,
then as at present, the largest of land animals, a fit supporter
of the infant world; in the serpent Asokee, used at the
churning of the ocean, we may trace a representative of the
gigantic Indian python; and in the bird-god Garida, with
all his attributes, we may detect the gigantic crane of India
(Ciconia gigantea) as supplying the origin. In like manner,

COLOSSOCHELYS ATLAS. 369

the Colossochelys would supply a consistent representative of
the tortoise that sustained the elephant and the world
together. But if we are to suppose that the mythological
notion of the tortoise was derived, as a symbol of strength,
from some one’of those small species which are now known
to exist in India, this congruity of ideas, this harmony of
representation would be at once violated; it would be as
legitimate to talk of a rat or a mouse contending with an
elephant, as of any known Indian tortoise to do the same in
the case of the fable of Garaida. The fancy would scout the
image as incongruous, and the weight even of mythology
would not be strong enough to enforce it on the faith of the
most superstitious epoch of the human race.

But the indications of mythological tradition are in every
case vague and uncertain, and in the present instance we
would not lay undue weight on the tendencies of such as
concern the tortoise. We have entered so much at length on
them on this occasion, from the important bearing which the
point has on a very remarkable matter of early belief enter-
tained by a large portion of the human race. The result at
which we have arrived is, that there are fair grounds for
entertaining the belief as probable that the Colossochelys
‘Atlas may have lived down to an early period of the human
epoch and become extinct since:—Ist. From the fact that
other Chelonian species and crocodiles, contemporaries of the
Colossochelys in the Sewalik fauna, have survived. 2nd. From
the indications of mythology in regard to a gigantic species
of tortoise in India.

Some of the bones were analyzed with great care by Mr.
Middleton, and yielded a large proportion of fluorine, the
constituents being :—

Phosphate of lime . : : c : : . ; . 64:95
Carbonate of lime . : : § i i 5 ‘ 22:36)
Fluoride of calcium . : ; é F F ; ; e lk6s
Oxide of iron . 5 : A : , 3 3 1:00

A trace of chloride of sodium:

99:99

Other Sewalik fossil bones were at the same time sub-
jected to analysis, such as the Mastodon Hlephantoides, Camelus
Siwalensis, Horse, Ruminants, &c., and the whole of them
yielded similar results, with a proportion of fluoride of
calcium varying from 9 to 11 per cent. This is much above
the usual quantity found in fossil bones ; the utmost that has
been met with having been in bones of Anoplotherium from
the Paris basin, viz. 14 per cent.

Manuscript Note by Dr. Falconer.|\—Fivorine.—There is

1 Found among Dr. Falconer’s papers.—[Ep.]
VOL. I. BB

370 FAUNA ANTIQUA SIVALENSIS.

another point of much interest connected with the mineral
condition of these remains to which I am desirous of calling
your attention. It may be in your recollection, sir, as well
as that of other members of the Society now present, that on
the first evening of the session, when our meeting opened so
auspiciously, with Professor Owen’s great paper onthe Dinornis
birds of New Zealand, it was suggested by myself, in regard
to the doubtful antiquity of those remains, that some aid
might be derived from chemical analysis, and ascertaining
the presence to any extent, or the absence, of fluorine in
the bones; as it had been determined by some of the most
distinguished authorities in chemistry that fluorine frequently
exists in very large quantity in fossil bones, as compared
with recent ones ; that its presence had even been ascertained
in the bones of Pompeii and Herculaneum; and that it had
been found in the greatest abundance in the most ancient
remains. This suggestion was met with a very confident
assertion by another speaker—a grave and worshipful signor,
Mr. President—a pillar of the State, and the present head of
a sister Society, that the opinions and the professed facts as
to the existence of fluorine in fossils were mere delusions ; the
baseless fabrics of an air vision, and deserving to be con-
signed to the bottomless gulf of oblivion, along with trans-
mutation and such like heresies. The general negation was
founded on a single affirmed experiment by Sir H. Davy, who
had failed to detect fluorine in some Kirkdale cave bones. It
was satisfactory, under such circumstances, to find that one
was connected in the so-called delusions with some of the great-
est of modern chemists, such as Berzelius, Thénard, Vauque-
lin, &c.; and it has been entered on the record of science as
an established fact. Fluorine has even been found to as
great an extent as 14 per cent. in the fossil bones of Anoplo-
therium from the Paris basin. I was naturally desirous to
ascertain how the case stood with the fossil bones from the
N. of India, and in numerous analyses which have been made
by my friend Mr. Middleton he has found it in every
instance in abundance. Here is the evidence, on a piece of
glass, in regard to its abundance in the fossil Tortoise. You
know, sir, to what strange uses the gravest things may be
turned. The philosophic Hamlet has traced the dust of
Alexander the Great till he landed it as a plug to the bunghole
of a beer barrel. In like manner, these tortoise remains, after
many thousand years of repose, have been pounded in a mortar,
stewed in a crucible, and so curiously dealt with as to be
made to distil, as it were, their most hidden humours for the
express purpose of engraving their own image on a plate of

glass. The prevailing idea in the artist’s mind, judging ~

ee or

COLOSSOCHELYS ATLAS. 371

from the design, appears to have been that the tortoises
derived their fluorine from the vicious habit of smoking
tobacco; but I beg to repudiate any connection with such an
opinion.

The following is Mr. Middleton’s Analysis of the Plastron of
the Colossochelys Atlas! :—‘ Viewed through a magnifying lens,
minute crystals, and here and there nests of minute crystals
of carbonate of lime are seen to occupy the pores of the fossil
associated with oxide of iron, which latter gives to it a
slightly brownish tinge. It is readily soluble in nitric and
hydrochloric acids. Five grains, finely powdered, were slowly
raised to a red heat, and continued in that state for some
time without having undergone from first to last any change
of colour. It was found, on re-weighing, to have lost =, of a
grain, a loss so small that it might easily have arisen from
other causes than diminution in the weight of the substance.
A piece, if exposed to the fumes evolved in the usual way, gave
with remarkable readiness and decision indications of the
presence of fluorine.

Judging from the quantity of fluorine present, I am disposed
to assign to the existence of this animal a date much anterior
to that of the Mastodon Hlephantoides, as the bones of the
former contain nearly twice as much fluorine as those
of the latter. How far dissimilarity of structure in the
tooth of the Mastodon (which I analyzed) and the plastron of
the Tortoise may go to account for the disparity in the quantity
of fluorine in them ; how far, for instance, the latter be better
suited for the absorption or development of fluorine than the
former, I am not prepared to say ; but as no reason suggests
itself to me, other than that of the lapse of time, which abundant
experiments have shown to be attended with the presence of
that substance in bones, I do not think it necessary to with-
hold the conclusion at which I have arrived.

The analysis was performed by me in the laboratory of
University College, with every possible care, and for greater
certainty repeated. The following are the results :—

Per cent.

Phosphate of lime 5 é . ‘ 3 ; . 64:95
Carbonate of lime ; ‘ 5 ; 2 5 . 22°36
Fluoride of calcium. 6 : 2 : ‘ . 11°68
Oxide of iron : F : 5 i > 00

A trace of chloride of podium.

99°99

' Found among Dr, Falconer’s papers.—[ Ep. ]
§ paz

BB 2

372 FAUNA ANTIQUA SIVALENSIS.

IIT.—Apstract OF AN ExtEMPORE DISCOURSE ON THE Co-
LOSSOCHELYS ATLAS, DELIVERED BEFORE THE BRITISH
ASSOCIATION AT YORK, ON SATURDAY, SEPTEMBER
287TH, 1844, av 8 p.mM., THe DEAN or Evy, PRESIDENT,
IN THE CHAIR.!

This evening the Association, as on Friday, assembled in
the Great Concert Hall of York to hear an extemporary
discourse by Dr. Falconer, on the Colossochelys Atlas or
Gigantic Fossil Tortoise, discovered along with numerous
other extinct animals, by Captain Cautley and Dr. Falconer
in the tertiary strata of the Sewalik hills, in the north of
India. The temporary screen in front of the orchestra
recess was sheeted over with a display of diagrams, maps,
and sections, in illustration of the discourse, among which
the most prominent was a very spirited and cleverly executed
restoration drawing of the Sewalik colossus, done by Mr.
Scharf, under the superintendence of Dr. Falconer, to the
natural size, measuring, exclusive of curves, about 184 feet.
On stands beside the lecturer there was also a fine array of

1 This abstract was written by Dr.
Falconer himself, but was never pub-
lished. The following notice, however,
appeared in ‘Chambers’ Edinburgh
Journal’ for Noy. 28, 1844:—‘ Another
evening was rendered still more agree-
able by an account of certain recent dis-
coveries in India. The demonstrator on
this last oceasion was Dr. Falconer, a
young medical man recently returned
from India on leave. The members, on
entering this evening, were surprised by
the picture of a tortoise, displayed on
the green screen above the speaker's
head, exhibiting an animal the same in
form as ordinary land-tortoises, but
about twelve feet long. Strange as it
may seem, remains of this huge animal,
to which the name of Colossochelys Atlas
has been given, are found in the super-
ficial gravel upon the Sewalik hills;
some of these were shown, particularly
one of the leg bones, the similarity of
which to the corresponding bone of the
modern diminutive species was easily
recognized. It appears that this and a
vast number of other animals, elsewhere
found in the tertiary strata, are, in that
part of the world discovered in the more
recent gravels, showing that the tertiary
species may have lived in certain dis-
tricts down to a time nearer to our own
era. And this idea Dr. Falconer con-
nected in a very interesting manner with
mythic traditions of India, descriptive of

enormous tortoises, one of which was
fabled to support the elephant by which
the world was supported. It seemed
not unlikely that these legends referred
to animals which had been living in the
early ages of mankind, but which have
for many centuries been extinct. The
plain and perspicuous, yet arresting
address of Dr. Falconer, was universally
allowed a high place among the scientific
affairs of the week. He has made a
most important contribution to geology ;
and the ample specimens which he has
brought home enrich the museums to
which they have been presented. His
services are the more creditable to him-
self, that, placed in charge of the Botanic
Garden upon the Sewalik hills, he had
little means of cultivating the science in
any of the more ordinary methods.
When a canal excavation near the garden
exposed to him a rich treasury of fossil
bones, he had no means of studying in
order to ascertain what these were; but
he took an original method. He went
off to the woods and wilderness and shot
animals, from which he might study
comparative anatomy, and by a reference
to these he was able to refer the fossils
to their proper species. What a crown-
ing to years of toil, thus to be able at
length to come before one of the most
intelligent audiences in Europe, and
enchain them with descriptions of such
novelties in human knowledge !’

COLOSSOCHELYS ATLAS. 373

specimens of huge fragments derived from all parts of the
skeleton of the Colossochelys.

Dr. Falconer commenced with a reference to the history
of the reptilian forms discovered in the fossil state (chiefly
in the older strata, such as the lias and oolite), among which
colossal representatives have been found of almost all the
known tribes, such as the Megalosaurus, Iguanodon, &c.,
besides numerous remarkable forms of which no analogues
are now in existence, suchas the Ichthyosaurus, Plesiosaurus,
and Pterodactylus. Even the Batrachian order, or Frogs, was
represented in the fossil state by the gigantic Labyrinthodon
of Professor Owen. But no fossil Testudinata, remarkable
either for size or deviation from existing forms, had hitherto
been met with. The existing species of sea-turtle were
larger than any which had been found of the order in the
fossil state, and the most remarkably modified known form
Was an existing species, the Chelys Matamata. . Nature,
however, consistent in her course, had not left the Chelonian
order without a grant. The Oolossochelys fills up the appa-
rent blank, and yet differs so little from the ordinary land-
tortoise, in the general construction of its osseous frame; as
hardly to constitute more than a sub-genus of Testudo, while
at the same time it furnishes one of the most gigantic forms
found in any section of the reptilian order.

The anatomical structure was then briefly described. The
plastron or sternal portion of the shell affords the chief dis-
tinctive character, and the elements for determining the size,
form, and affinities of the fossil. Dr. Falconer explained
that the diagram was not drawn from any one entire speci-
men of the shell and skeleton, of which there was probably
none in existence either above or below ground, but that it
was what is called a Cuvierian restoration founded on the laws
of relation between special and general structure; in fact a
reconstruction of the form determined upon the collation
of a vast number of specimens of the individual parts.
Allusion was here made to the laws of reciprocal connection
between special and general structure, which involves the
necessary condition that every individual part has a definite
relation to the aggregate form, and to every other part of it;
or, to express the law in mathematical language, that each
proportion of an animal is a function of every other propor-
tion and of the sum of those proportions. The application
of this law to the reconstruction of the Colossochelys was
next considered, and it was shown how far the sternum,
carapace, extremities, and head had common tendencies,
and how far they warranted the size, form, and generic
representation in the figure. The anatomy of the Colosso-

374 FAUNA ANTIQUA SIVALENSIS.

chelys was then described. The episternal, or- anterior por-
tion of the plastron, is 64 inches thick, contracted into a
diameter of 8 inches, bifid at the apex, and supplied on its
inferior surface with a thick cuneiform keel. This keel
constitutes one of the principal features in the fossil. The
entosternal portion presents exactly the same character as in
the ordinary forms of Testudo, the like being the case in
regard to the ziphosternal or posterior pieces, of which there
was a magnificent specimen exhibited on the table. The
‘ plastron’ or breast-plate in the adult animal was estimated
to have been 9 feet 4 inches long. The ‘ carapace’ or buckler
portion of the shell was described as coinciding in every
leading character with the general form of the land-tortoises,
of which in fact it exhibits but a hugely magnified repre-
sentation. The same resemblance was found to hold good
in regard to the dermal covering or scutes. The shell was
estimated to have been about 12 feet 3 inches long measured
along the chord, and 15 feet 9 inches measured along the
curve of the back, 8 feet in transverse diameter, and about
6 feet high.

The extremities were then described, and shown to follow
the same line of affinities as was indicated by the shell: viz.,
to be constructed precisely as in the land-tortoises, in which
the form of the femur and humerus is marked by peculiar
characters. These bones in the fossil are of a prodigious
size, corresponding to the dimensions of the shell. A
specimen comprising the upper part of the humerus was
exhibited, exceeding in size the same portion of the corres-
ponding bone of the Indian Rhinoceros, yet so like the
humerus of an existing Tortoise that the outline of the fossil
could pass for a magnified representation of the small recent
bone, done by means of the pantograph. The same was
found to hold good in regard to the construction of the feet,
which are formed exactly on the plan of Testudo: viz., with
columnar legs and truncated club-shaped hoofs, as in the
proboscidean Pachydermata. The ungual phalanges or bones
which support the nails were shown from specimens to equal
the size of those of the largest Elephant.

Dr. Falconer then described the form of the head, which,
like the rest of the skeleton, was closely allied to the ordinary
forms of Testudo. The collection did not contain an adult
specimen of the head, but from the proportions of a half-
grown perfect specimen, the head was deduced to have been
2 feet long (see Plate XX XI.). The entire length of the Colos-
sochelys Atlas, from the tip of the head to the extremity of
the tail, was inferred to have been close upon 20 feet, and
the animal was said to have stood upwards of 7 feet high.

DESCRIPTION OF PLATE XXXI.

CoLOssocHELYS ATLAS.

The figures in this Plate have been drawn from the original

specimens in the British Museum by Mr. Dinkel.

Fig. 1.

Fig. 2.

Cranium of Colossochelys Atlas, one-half of the natural size.
The Brit. Mus. number of the specimen is 39,819. (See pages
363 & 374.)

Fragment of upper end of humerus, with globular articulating
head of Colossochelys Atlas, one-fifth of the natural size. The
specimen bears the number 16,913 in the British Museum

Catalogue. (See pages 360, 364, 374, & 381.)

. Fragment of lower end of humerus of Colos. Atlas, numbered

39,825 in the Brit. Mus. Cat., and drawn about one-fifth of the

natural size.

Figs. 4 and 5. Represent two different views of upper end of femur,
- with articulating head of Colos. Atlas, one-fifth of the natural

size. The specimen is numbered 16,518 in the Brit. Mus. Cat.
(See pages 863 & 374.)

VoL. I.

Vol. 1. Plate dl.

SDinksl dal. et Lith, Colossochelys Atlas. W.Wast mp.

COLOSSOCHELYS ATLAS. 378

The whole of its organization proved that it was strictly a land
animal, with herbivorous habits, and probably of the most
inoffensive nature. The generic name given by Capt. Cautley
and Dr. Falconer to the fossil has reference to its colossal size
(xorossos et yédus), and the specific designation of Atlas, to
its fitting representation of the mythological Tortoise that
sustained the world in the systems of ancient cosmogony.

The principal part of the remains of the Colossochelys were
collected during a period of eight or nine years along a range of
about 100 miles of hilly country. They belong in consequence
to a great number of different animals, varying in size and age.
From the circumstances under which they are met with, in
crushed fragments, contained in upheaved strata which have
undergone considerable disturbance, there is no chance that
an entire shell, or anything approaching a complete skeleton
in a single specimen, will ever be disentombed from the
Sewalik hills. When the first fragments, in huge amorphous
masses, were found by the discoverers, they were utterly at a
loss what to make of them, and for many months could do
nothing more than look upon them in bewildered and nearly
hopeless admiration. But no sooner was the clue found to
the nature of a single specimen than every fragment moved
into its place, so as to form a consistent whole.

Dr. Falconer then entered upon the question ‘whether
are there grounds for determining the period when the
gigantic Tortoise became extinct, or are there any indications
that it may have lived down to be contemporary with any
portion of the human period ?’

The Moa bird or Dinornis of New Zealand—a parallel
instance of an equally gigantic form in another order—
appears from very strong evidence only to have become very
recently extinct, and any apparent a priori improbability
that an animal so hugely disproportionate to existing species
as the Colossochelys should have survived, so as to have been
coexistent with man in the Hast, is removed by the fact that
other species of Chelonians, which were coeval with the
Colossochelys in the Sewalik fauna, have reached to the present
time; and what is true in this respect, in regard to one
species of the tribe, may be equally true of any or every
other placed under the same circumstances. Some of the
Crocodiles—in particular the Crocodilus longirostris or
‘Gavial’—now existing in India, appear to be absolutely
identical with forms dug out of the Sewalik hills. The same
result was found to apply to the existing Hmys tecta,
now a common species in all parts of India. Dr. Falconer
mentioned that there were other cases which appeared to
yield similar results, but that the evidence had not been

370 FAUNA ANTIQUA SIVALENSIS.

sufficiently examined to justify a confident affirmation of
identity at present. There is no direct evidence from history
or from remains dug out of recent alluvial deposits that the
Colossochelys has recently ceased to be a tenant of the globe ;
but there are traditions connected with the cosmogonic
speculations of almost all Eastern nations having reference
to a Tortoise of such gigantic size as to be associated in their
fabulous accounts with the Elephant. Was this Tortoise
merely a creature of the imagination, or was the idea of it
drawn from a reality like the ‘ Colossochelys ?’

Besides a tradition current among the Iroquois Indians,
in regard to the important share which the Tortoise had in
the formation of the earth, there are several cases in ancient
history bearing on the same point. In the Pythagorean
cosmogony the infant world is represented as having been
placed on the back of an Elephant which was sustained on a
huge Tortoise. In the Hindoo accounts of the second Avator
of Vishnoo the ocean is stated to have been churned by
means of the mountain Mundar placed on the back of the
King of the Tortoises, and the Serpent Asokee used as the
churning-rope. Vishnoo was made to assume the form of
the Tortoise, and sustain the created world on his back to
make it stable. So completely has this fable been impressed
on the faith of the country, that the Hindoos to this day
even believe that the world rests on the back of a Tortoise.
Again, the Tortoise figures prominently in the narratives of
the feasts of the bird-demigod Garida, who is represented on
one occasion to have appeased his hunger at a certain lake
where an Elephant and a Tortoise were fighting. The dimen-
sions are given in the extravagant style of cosmogony.

In these three instances there is distinct reference to a
gigantic form of Tortoise comparable in size with the Elephant.
Hence arises the question, are we to regard the idea as a mere
fiction of the imagination, like the Minotaur, the Chimera, the
Griffin, Dragon, and Cartazonon, &c., or as founded on some
justifying reality. The Greek and Persian monsters are
composed of wild and fanciful combinations of different por-
tions of known animals into impossible forms, and are merely
the progeny of uncurbed imagination. But in the Indian
cosmogonic forms there is an image of congruity which may
be traced through the maze of exaggeration with which they
are invested. We have the Elephant then, as at present, the
largest of land animals, a fit supporter of the infant world;
and the Colossochelys would supply a consistent representative
of the Tortoise which sustained the Elephant and the world
together. But if we are to suppose that the mythological
idea of the Tortoise was derived as a symbol of strength,

nat Gow

COLOSSOCHELYS ATLAS. 377

from some one of those small species which are now known
to exist in India, this harmony of representation would be at
once violated ; it would be as legitimate to talk of a rat or a
mouse contending with an elephant as of any known tortoise
to do the like in the case of the fable of Garada. Imagina-
tion would scout the image as incongruous, and the weight
even of mythology would not be strong enough to enforce it
on the belief of the most superstitious period of the human
race. Dr. Falconer stated that he would not lay undue stress
on the tendencies of mythological tradition, which are gene-
rally so vague and uncertain ; but that he had entered so fully
on the point from the important connection which it had with
a very remarkable matter of belief entertained by a large
portion of mankind.

Dr. Falconer then described the animals associated with
the Colossochelys in the Sewalik fauna, including no less
than five extinct species of Mastodon and Elephant peculiar
to India, viz., the Mastodon latidens (Clift), M. Hlephantoides'
(Clift), M. Sivalensis (Fale. & Caut.), Hlephas planifrons
(F. & C.), H. Hysudricus (F. & C.).

1 Subsequently subdivided by Dr. Faleoner into Hlephas Cliftii, and Elephas
insignis —| Ep. |

THE ELEPHANT VICTORIOUS OVER THE TORTOISE, SUPPORTING THE WORLD, AND
UNFOLDING THE MYSTERIES OF THE FAUNA SIVALENSIS. FROM A SKETCH IN PENCIL
IN ONE OF DR. FALCONER'S NOTE-BOOKS BY THE LATE PROF. EDWARD FORBES.

378 FAUNA ANTIQUA SIVALENSIS.

ITV.— ANATOMICAL DESCRIPTION OF THE COLOSSOCHELYS
ATLAS, BEING MS. Norks For THE LECTURE DELIVERED
AT THE BritTisH ASSOCIATION IN SEPTEMBER 1844.

The Sternum.—In the Chelonians the sternum is a highly
developed apparatus, and consists in all the species invari-
ably of 9 pieces, 8 of which are pairs and 1 an odd piece;
2 episternals, 2 hyosternals, 2 hyposternals, 2 xiphosternals,
and 1 entosternal. In the Chelonia and Trionyx the pieces
are detached; in the Land Tortoises and Hmyde they are
solid. The fragments of the Colossochelys show that the
episternals were of an amorphous thick form, cleft at the
apex, with a deep cuneiform keel underneath. The ento-
sternal shows an exact correspondence in form and in every
particular with several Testudines. In the specimen in the
Zoological Society the length of sternum is 32 inches, and
the length of epi- and ento-sternal pieces 10 inches. In Colos-
sochelys, the epi- and ento-sternal pieces measure 35 inches,
and the inferred length, on the same proportions, of the entire
plastron is 9 feet 4 inches. The lateral curved expansions
are also to be noted. All these concurrent marks are found in
no other Chelonians than the Testudines. Hence the first in-
dication of its being a Tortoise is exhibited by the sternum.

The xiphosternals are cleft behind under the tail. In
the fossil, in thickening, marginal ridge, and general charac-
ters they resemble the living Tortoise. In the hyposternals
there is a similar correspondence and thickening. The whole
of the sternum indicates a testudinate animal of enormous
size, and a true Land-Tortoise.

Character of the Carapace.—The dorsal shield consists of 10
pairs of ribs in 8 sutured pieces (the Ist and 10th being con-
fluent with the 2nd and 8th), connected with as many
vertebral plates; there are 11 vertebral plates in all (the last
or supra-caudal having no attachment to the ribs), and 11 mar-
ginal or sterno-costal pieces corresponding to the cartilages of
the ribs. There are 10 dorsal vertebrae, the first free and
homomorphous with the cervical vertebre, of which there are
7 (apparently 8).

The carapace of the Land Tortoises is short, very elevated
and arched, and more or less vertical at the sides, as com-
pared with other Chelonians. The thickness of bone in
the convexity is almost in an inverse ratio to the size. The
physiological reason of this is that the smaller the animal the
more liable it is to injury, and it requires a greater arch to
sustain it. There are no complete pieces in the fossil
remains, and thereasons of this are the curvature and thinness.
The most marked pieces in the carapace are the sterno-costal,

oe Ne ee re

COLOSSOCHELYS ATLAS. 379

of which a large fragment is attached to the ribs, and gives
the form of the inguinal sinus for the emission of the leg,
exactly as in the Testudo Indica. 'The engroined apices
indicate the age to be young, and afford an excellent element
of measurement, as the sterno-costal pieces bear a direct
ratio to the development of the ribs to which they are
attached. Another crucial proof of the size is derived froma
comparison of corresponding measurements in Colossochelys
and Testudo Indica. (See antea, pp. 364 and 374.)

The correspondence shown in the symphysized portion of
the sterno-costal pieces is borne out by other pieces all round
the rim, viz., the great curvature anteriorly, and the abrupt
reflexion as in the Land Tortoises, and also the great thicken-
ing of the rim, and the furrows of the epidermal scutes.
The concave or spoon-shaped form of the supra-caudal plate
is also exactly alike.

The rib pieces are less common among the remains dis-
covered. They show a total disappearance of sutures. One
large piece, comprising the second vertebral plate and several
vertebree of the back, with the keel, indicate that the back
was flattened out and had little convexity. Other pieces
indicate the same character in the sacral vertebre. In
short, the whole gives the form of a gigantic Tortoise, with a
flattened back, compressed sides, vertical in front, with an
abrupt marginal reflexion.

There is a peculiarity in the bodies of the vertebree being
detached, requiring to be artificially held together in the
recent skeleton. None are available for description, but the
vertebral pieces to which they are attached, the sacral
ones, are among the specimens. This is also indicated in
the keel.

As to the relation in size of the plastron to the carapace in
the smaller species, for safety the plastron is elongated so as
to close up the openings before and behind; this goes on
increasing nearly in the ratio of the reflexion of the limb.
Taking the Testudo Indica as the standard, we find the
plastron to the carapace being nearly in the ratio of 3 to 4,
according to which the carapace of Colossochelys would be
12 feet 6 inches long, and the curvature 15:9 inches, the size
intended to be given in the figure.

We shall now see how the vertebree agree. The bones of

_ the skeleton of the Testudinata, viewed as a whole, undergo

more singular modifications than we observe in any other of
the higher reptilia, and more than is observed throughout
the whole order of birds or mammalia. The number of
vertebree, excepting of the tail, is pretty constant in all; and
the principal modification is observed in those that are

380 FAUNA ANTIQUA SIVALENSIS.

attached to the buckler. In Turtles, the soft Tortoises, and
the Terrapins, the bodies of the dorsal vertebre are connected
with the vertebral plates by a bony lamella on either side
forming a continuous osseous canal upon the upper surfaces
of the bodies in which the spinal marrow is lodged. The
bodies of the vertebre are consequently attached with con-
siderable strength to the buckler, and do not separate readily
when the soft parts decompose. But in the Land Tortoises the
only connection between the vertebral plates and the shafts
of the bodies is established by means of a membranous
lamella, there being no osseous canal for the spinal marrow,
which is lodged in a cavity formed by a duplicature of this
membranous lamella and a groove on the upper surface of
the bodies. The only osseous connection is established by
means of a thin slender splint-shaped process, given off on
either side from the vertebral plates to the common point of
union between the bodies of two vertebre. The bodies are
consequently but feebly attached, their principal strength
arising from their being arranged in an arch, and they
generally fall off when the soft parts decompose. The line
of attachment of the membrane is indicated by a ridge along
the axis of the plates. In all the vertebral fragments of the
fossil we observe precisely the same arrangement ; there is a
well-marked keeled ridge along the axis, indicating the line
of attachment of the membrane, and no indication of any
bony canal—nothing besides the thin splint-shaped processes,
or new apophyses, indicating, as in the rest of the skeleton,
the strictly Testudo affinities of the fossil.

The same leading resemblances are found in the sacral
vertebrae, which show the surfaces for the attachment of the
posterior and upper angles of the iltum exactly as in Testudo.

While on the vertebre of the trunk, I may mention that
there are no caudal vertebrz in the collection, and that, as
the number of these vary in the Chelonians from 18 to 26, the
restoration in the drawing has been assumed from the type
of the Land Tortoises, in which the tail is thick and the higher
number prevails. In certain species of this genus whose
habits have been well observed, the tail has been noticed to
be used as a pivot or fifth leg to sustain them, especially
during the act of defecation; and, from the huge size of
the fossil, and the weight which its limbs had to sustain,
it is at least highly probable that nature gifted it with a
similar aid in its tail, although we have no direct evidence on
the point.

We shall now pass to the special organs of progression,
and see how the extremities of the fossil were constructed.
In regard to the anterior member, no remains have yet been

EE ee eee

COLOSSOCHELYS ATLAS. 381

detected in the collection of the scapular arch, although they
will in all probability be found yet to give direct evidence as
to its build.

The humerus was of large size, its estimated length being
28 to 30 inches; but it corresponded exactly with that of the
existing Tortoise in its singular curve, globular head, large
size of inner tuberosity, and foramen for the brachial nerve
at lower end. The concentric arrangement of the corpuscles
of Retzius round the Haversian canals seen in the ungual
bones of the toes is a certain mark of a reptilian.

V.—Descrietion By Dr. FAaLconer oF Fossit REMAINS OF COLOSSOCHELYS
AND TesTUDO IN THE Muserum oF THE ASIATIC Society oF BENGAL.

[ The catalogue contains numerous references to remains of the Colos-
sochelys :—from Ava, Nos. 196, 241; from the Sewalik hills, Nos. 654
to 674, and Nos. 860 to 863; and from Perim Island, Nos. 98 and 112.

The chief of these is the following, supposed to. be from the Sewalik
hills.—Ep. | :—

No. 654. Colossochelys Atlas—Fine fragment, comprising the anterior
portion of the thickened episternum (of the plastron); it is bifid at the
apex, behind which it is contracted and supplied underneath with a
thick cuneiform keel. No portion of the thin part of the bone near the
entosternum is present; the specimen is tinged black, highly infiltrated
with iron, and crusted over with a thin coat of gritty sandstone like
some of the Ava specimens; but it bears no mark. The tip of the left

bifurcation is broken off; the right is entire.
Inches

Length of fragment . ¢ . : 16-
Ditto of thick portion of episternum, ‘to tip of bifurcation ‘ : 13°
Width of thickened part behind bifurcation . 5 : - 6°5
Greatest thickness . : : : . 5 c ; 51

Col. Colvin is stated cae A.S., v. p. 184) to have presented to the
Society some huge fragment of this gigantic Tortoise, but the specimen
in mineral condition is unlike the generality of the Sewalik fossils.
It may be from Ava (Col. Burney), as Dr. F. found numerous frag-
ments of the gigantic Tortoise among the specimens presented to the
Geolog. Soc. by Mr. Crawfurd. This episternum is not of a full-sized
animal, Dr. F. having met with it 64 in. thick with a diam. of 8 in. at
the contracted part.

[ The description of a specimen of fossil Testudo in the same collection,
also from the Sewalik hills, is also worth quoting.—Ep. |

No. 675. Testudo.—Left side of plastron comprising the episternal,
entosternal, and mesosternal, separated along the median suture.
Episternum thick and overarching the entosternal, in form and markings
very much like TYestudo stellata. This was a small Land-Tortoise,
having no connection with the Colossochelys, and probably a young
animal, judging from the openness of the sutures.

382 FAUNA ANTIQUA SIVALENSIS.

XXI. ON A FOSSIL SPECIES OF EMYS FROM THE
SEWALIK HILLS, REFERABLE TO THE
EXISTING EMYS TECTA (BELL).!

Tue almost constant distinctness of species between forms
observed in the true fossil state and recent ones, in the
mammalia and reptilia, requires that every apparent excep-
tional case should be rigidly examined, and have the strongest
evidence to sustain it. The land and fresh-water Tortoises,
when well preserved in the fossil state, supply the required
character, for determining a case of this kind, under much
more favourable circumstances, and in greater abundance,
than any other tribe of the reptilia or the mammalia. For,
with the latter, the evidence is almost invariably limited to
the bony skeleton: while, in the Tortoises we have, in addi-
tion, the characters presented by the dermal integument or
horny scutes which leave well-marked impressions indicating
their form and extent, these scutes furnishing to naturalists
the best marks for the distinction of species in the living
state.

The small Indian species of Emys, well known under the
name of E. tecta,? is readily distinguished from every other in
the genus by the peculiar form of its carapace, which, in-
stead of being vaulted, as is usual in the tribe, slopes down
on either side from the ridge of the back, with a pitch nearly
equal to a right angle, resembling somewhat the roof of a
house. In addition, the three anterior vertebral scutes are
elevated each into an obtuse point of which the posterior one
is the most prominent, so as to form an uninterrupted tri-tuber-
cular keel. The plastron is slightly curved lengthwise with
a well-marked keel at either side, and throws off its ale in-
clined at a considerable angle upwards. The EH. Belli is
the only other species that has anything of a ‘tectiform’
carapace—but in a much slighter degree—and it is further

1 This essay was written in 1844, but ; Dinkel. Among the Sewalik specimens
is now for the first time published. The | in the British Museum there are likewise
specimen described in it is now in the | specimens of other species of Hmys as
British Museum, and from it the drawings | well as of Zrionyx and Testudo.—|Ep.]
in Plate xxxii, have been made by Mr. 2 Emys tectwm in MS.—[Ep.]

ee has en

4

EMYS TECTA. 383

readily distinguished by a more depressed form, and the ab-
sence of the three vertebral prominences.

Plate XXXII. represents the shell of a fossil Emys from
the tertiary strata of the Sewalik hills. The specimen has all
the mineral characters of an ancient fossil. It was embedded
ina hard sandstone matrix which fills the hollow of the shell,
and the bony part is densely infiltrated with hydrate of iron
and siliceous matter, so as to give it a dark colour and great
specific gravity. It comprises nearly the whole of the cara-
pace as far back as the commencement of the last vertebral
scute. The margin is broken off in front, and the gular and
anal portions of the plastron are also wanting. All the rest
of the shell is distinctly shown. It had been exposed to a
erush which has altered a little the form of the shell at the
junction of the right sterno-costal pieces with the ribs, and
caused a longitudinal ‘ fault’ on the left side of the plastron
between the median line and the keel.

The fossil agrees so closely with Hmys tecta in size and
general form that the resemblance is observed to be very
striking at the first glance. It has the same high-pitched,
roof-shaped carapace, the same tri-tubercular keel occupying
the first three vertebral scutes, and a similarly formed plas-
tron. A like agreement runs throughout the other details of
the characters.

The carapace of the H. tecta is comparatively high for its
size, the transverse diameter being about two-thirds of the
length, and the height in the same proportion to the dia-
meter. The profile from back to front along the vertebral
ridge is a very depressed ellipse segment, and the horizontal
outline is oval with greater width at the inguinal than at the
humeral region. In all these particulars the fossil, in so far
as it has the character to show, agrees closely with the recent
form. It is not so high in proportion to its width as the
smaller sized specimens of H. tecta, but in the latter the
shell is observed to become more depressed with the increase
of size and age, the fossil being in an intermediate condition
between the young and the old form. The nuchal scute in
Fimys tecta is always small, but subject to great difference
in size, in different individuals. In some specimens, and
often the largest, it is a minute rectilineal lobule, while in
others it expands behind into two angular wings, the differ-
ence being probably sexual. It isnot shown in the fossil.
The first vertebral scute is also subject to a considerable

range of difference in form. It is an irregular quadrangular

pentagon, wider in front than behind, the convergence varying
greatly in amount in different individuals. In the fossil the
exact form is not distinctly seen, though it seems to converge

384 FAUNA ANTIQUA SIVALENSIS.

less, and to be squarer, than generally holds in the recent
forms ; but probably it does not differ in a greater degree than
the extremes of these do from each other. The second and
third vertebral scutes are hexagonal, both in the recent Hmys
tecta and in the fossil. Both agree alike in the relative pro-
portion of the ridge borne by each of these scutes, which is
most protuberant in the third or central one of the series. The
outline of the fourth scute is not distinguishable in the fossil,
and the fifth one is wanting; but it is obvious that these
two scutes were more elongated than the anterior ones, a
point which constitutes one of the distinctive characters of
the recent species, in which the two last vertebral scutes
are as long as the three first. In this there is another close
agreement between the fossil and the existing form.

The outline of the costal scutes is obscured in the fossil,
and not satisfactorily distinguishable from the sutures, so that
a minute comparison with the corresponding parts of the
recent form is not in this instance practicable.

The margino-collar scutes vary a good deal between
triangular and pentagonal in the existing Hmys tecta, the
latter form prevailing in those which have the broadest
nuchal scute. The fossil has the margino-collar pentagonal,
and with the same arrangement of the sides as in a recent
form. A like correspondence runs‘through the margino-
humeral and margino-lateral scutes, which are oblong, rec-
tangular, or square both in the Hmys tecta and in the
fossil. The agreement is close, down to minute particulars
in the size of the angle of junction between the first margino-
lateral scute and the humeral scute of the plastron. The
same holds with the form and. extent of the axillary and
inguinal scutes, which differ greatly in nearly allied species.
The minute axillary foramen has exactly the same position,
relatively to the second margino-humeral and axillary scutes,
in the fossil as in the recent specimen.

The crushing force to which the fossil has been subjected
has caused a longitudinal fracture and ‘ fault’ along the left
side of the plastron, and a depression along its median line.
Its anterior and posterior extremities are wanting. There
is no direct evidence, therefore, in regard to the proportion
of length to breadth, but the impressions of the sternal scutes
are fortunately well preserved, so as to admit of minute
comparison with those of the Emys tecta. The plastron in
the latter has a strongly marked keel on either side, along
the line of margin, and the ale are given off at a considerable
angle upwards, to unite with the sterno-costal pieces, so that
the rim of the carapace is raised high above the level of the
sternum. In both particulars, and in the relatively less

= ita ee

ee

DESCRIPTION OF PLATE XXXII.
EMyYs TECTA AND VARANUS SIVALENSIS.

Figs. 1 and 2. Represent the upper and under surfaces of the shell of
Emys tecta, a species of Tortoise still existing in India, but also
found fossil in the Sewalik Hills. The specimen appears to be
the identical one described in Dr. Falconer’s memoir (page 382).
It has been drawn by Mr. Dinkel from the original in the Bri-
tish Museum (Cat. No. 39,837), one-third of the natural size.
Another fossil specimen, apparently of the same species, is
numbered 17,435 in the Brit. Mus. Cat.

Fig. 3. Is a diagrammatic view of the same specimen, endwise, to show
the roofed shape of the shell.

Figs. 4, 5, 6,and 7. Lower end of humerus, with articulating surface, of
Varanus Sivalensis, or fossil Monitor of the Sewalik Hills, two-
thirds of the natural size. This specimen appears to be unique.
The specimen itself, with drawings, was found thus designated in
Dr. Falconer’s handwriting in his private collection, and has
been deposited in the Paleontological Gallery of the British
Museum.

VOL. I.

VoL I. Plate 32.

“W.West imp.

ensis.

Hgs.1-3. Emys tecta. Figs 4—7. Varanus Sival

JDinkel delet ith.

EMYS TECTA. 385

amount of alar to marginal surface, there is the closest
agreement between the fossil and the recent forms. The
pectoral scutes in the recent form are rectangular and oblong
transversely, and the abdominal scutes nearly square ; the
anterior edge of each makes a sweep forwards as it rises
over the lateral keel, and then bends back to join on with the
corresponding marginal scute at an angle. In all these
there is a very close agreement between the plastron of the
fossil and of the recent form.

The last point to be noticed concerns the interior of the
shell. In the Land Tortoises and the genera allied to Emys
the hyposternal pieces of the plastron give off, on either side,
an apophysis which ascends on the inside of the third sterno-
costal piece as far as the first rib, and acts as a brace or
buttress to strengthen the arch of the dorsal buckler. In
like manner the hyposternal pieces send off a similar apo-
physis on either side, which ascends along the inside of the
seventh sterno-costal piece up to the sixth rib, strengthen-
ing the carapace behind. In the small Hmys tecta these
apophyses are greatly developed and expanded into wide
plates, like incomplete diaphragmata, which project into the
cavity of the shell, and contract the front and back apertures.
The anterior apophyses ascend nearly as high as the head of
the first rib, converging like the spandrels of a gothic arch.
These plate apophyses are constructed in the fossil on exactly
the same plan as in the recent species.

In short, after a rigid comparison conducted through all
the details of size, form, relative proportion, and the dermal
characters, there is the closest agreement between the fossil
and the recent Hmys tecta. 'The only discernible difference
is a slight one of form, in the anterior vertebral scutes,
which are known to be very variable in different individuals of
the existing species, and the fossil does not appear to differ
more from these than individuals among them do from one
another. Hence arises the question, are we entitled to
regard the fossil Emys as belonging to the same species with
the Hmys tecta? The point is one of much importance,
both as it affects the established principles in Natural History
regarding the ideas comprehended under the name of species
and, in the present instance specially, from the fossils with
which the specimen was found associated.

. The received opinion in regard to the meaning of species
is, that all forms are to be considered of one and the same
species, which derive their origin from one and the same
kind of ancestors, or which do not differ more from these
than they do among themselves. Tried by this test, the
fossil as above described is clearly referable to the Hmys
VOL. I. co

386 FAUNA ANTIQUA SIVALENSIS.

tecta. But a great authority, one of the most profound
naturalists of the age, M. Agassiz, after running over a
vast field of observations, has arrived at a very different
conception of the meaning of ‘Species.’ He states his
opinion to be, ‘that no so-termed character, that is, no
observable mark can be so striking as to indicate an absolute
specific distinction; but at the same time it should never be
regarded as so trifling as to point to absolute identity; that
characters do not mark off species, but that the combined
relations to the external world in all circumstances do.’ He
believes that he can show ‘that many organic beings are
specifically distinct, or at least that they stand in no genea-
logical relations to one another, although the individuals are
so like as to be confounded.’ He further advances the
opinion ‘ that no species occurs in two geological formations—
nay, not even in two different parts of one formation ;’” and he
believes ‘ that hereafter it.will become necessary to express
the specific difference of organic remains by the circumstances
of their occurrence without its being possible to assign distinc-
tions to them.’

If we are to apply this conception of ‘ Species’ to the fossil
it must at once be considered as distinct from Hmys tecta,
for it comes from a different geological formation, where it
is associated with numerous extinct forms. But it is obvious
that this part of the statement is nothing more than a
begging of the question. The distinctness of the species
must first be proved, in every case, on independent grounds
before the asserted law regarding the order of their occur-
rence can be admitted. And, in regard to the other state-
ment, that organic forms may be so like as to be undistin-
euishable and yet be specifically distinct, and that no ex-
ternal characters can be so marked as to determine specific
difference, they seem to imply little short of a sort of Natural
History negation of the universal axiom ‘ that things which
are equal to the same are equal to one another.’ M. Agassiz
appears to entertain the received opinion among naturalists
that common genealogy constitutes the essence of a species,
but the quoted expressions would imply that he abandons
external characters as the test by which such genealogy is to
be judged of. Ae

But naturalists have no right, in determining the specific
nature of an organic form, to take into any account the
geological position whence it came, any more than geogra-
phical place in regard to existing forms. The species ought
to be ascertained from the observable marks which it presents
apart from every extraneous consideration ; the proper order _
of succession being to apply the result thus arrived at to —

PMS. 2k ps

EMYS TECTA. 387

geology and geography, and not to determine the species
in any degree from either. Anything short of this is but
tampering with the right evidence.

Unless we are to continue to take modifications of form
and constant external characters as sufficient tests of common
origin or like descent, there is an end to the power of dis-
criminating species, and the foundations of the science are
undermined. But the majority of naturalists are agreed
that with existing forms the evidence is sufficient, and no
adequate reasons have yet been brought forward to show
why the same should not be believed to hold in the case of
extinct forms. In fact, a firm belief in it is the bases on
which all our investigations are conducted.

Under this view, conceiving that we are not justified in
constituting a difference where we do not find it, we infer
that the fossil Emys is specifically identical with the existing
Emys tecta, for it agrees as closely with the forms grouped
under that name as these do with one another.

The Emys tecta, according to our present state of know-
ledge, is confined to India, where it is found in great abund-
ance in the Ganges and in ponds in the North-western
Provinces, and in Bengal.

The fossil came from the tertiary strata of the Sewalik
hills, where it was associated with the remains of Colosso-
chelys, Trionyx, and other Chelonians; and of Mastodon,
Hippopotamus, Sivatherium, Camel, &c. The precise locality
is not known.

If the view taken of its identity with Hmys tecta prove
correct, the fossil will furnish a good instance of the per-
manence, and, within certain limits, invariableness, of species
during a long course of time extending through more than
one geologicalepoch. Such instances are so abundant among
the Mollusca, as to have constituted the basis of Mr. Lyell’s
classification of the tertiary strata. But they are so ex-
tremely rare among the Vertebrata, that there is perhaps
hardly a single well authenticated case on record. Agassiz
goes the length of stating that he has found his conclusion
regarding the absence of a species from two geological
formations, and even from two different paths of one forma-
tion, to have been. ‘invariably confirmed by fossil fishes and

Echinodermata.’

388 FAUNA ANTIQUA SIVALENSIS.

APPENDIX TO MEMOIR ON EMYS.

Manuscript Memoranpum By Dr. Fatconer, Aprit 1844.

Mr. T. Bell's Opinion of the fossil ‘ Emys tecta.’—‘ On comparing the
fossil with two specimens of recent Emys tecta, it looks a little flatter,
but in this respect it does not differ more from either than these do
between themselves, and certainly not more than would lead me to
think, were it a recent one, that it was the same species. In regard
to the first vertebral plaque, it is broader, flatter, and squarer; but
this is liable to variation in the recent species to an equal extent. If
the three specimens were all recent, I would merely consider them as
varieties of the same species.’

‘This,’ adds Dr. F., ‘isan admirable fact ; for if this species has lived
down to the human period, why might not the Colossochelys have done
so? Refer to Ward’s “ History of the Hindoos,” to the article ‘‘ Garuda
or Giroora,” the king of the birds, for the fable of his having found an
elephant and a tortoise contending together, and having made a repast
upon them. Joining the indication given by the Hindoo mythology
with the determined fact of the little Hmys tecta having survived from
the fossil period down to the present day, I have put forward the
opinion that the big Tortoise may have survived also, and only become
extinct within the human period. This is a most important matter in
reference to the history of man.’ (Letter to Captain Cautley, April 1844.)

FOSSIL FRESHWATER SHELLS.

XXII. NOTE ON THE

389

FRESHWATER SHELLS

FROM THE SEWALIK HILLS.

BY THE LATE PROFESSOR EDWARD FORBES.!

RECENT.

Pulmonifera :
Planorbis 1 (identical with speci-
men unnamed in Mr,
Lyell’s collection).
» 2 (in Mr. Lyell’s collec-
tion unnamed).

Pectenibranchiata :
’ Paludina unicolor
7 Bengalensis

” (in Mr. Lyell’s collec-
tion unnamed)

Melania thiarella
a corrugata

Ampullaria glauca ?

Lamellibranchiata :
Unio favidens, Benson (Mr. Lyell’s
specimen; U. corrugata,
Lam. ?)

» marginalis, Lam.

» (undescribed ?)
» (undescribed ?)

Cyrena (small species).

1 N.B. The materials for this compari-
son in recent forms of India are very li-
mited and incomplete.—[H. F.]

This note, by Professor Forbes, was

written about 1846. It has not been

Fosstn.

Planorbis identical with No. 1.

Pupa? Not resembling any specimens
I have seen in the Kast.

Paludina unicolor
allied to P. Bengalensis, but ap-
parently distinct,

Melania, quite distinct from any
recent one I have seen.

Ampullaria glauca
identical with specimens from
the Jumna. Differing some-
what from the recent Sewalik
specimens.

Unio favidens, a variety.

» (near a Jumna specimen in
Mr, Lyell’s collection).

published before. Several of the Sewalik
fossil freshwater shells were figured in
the Journal of the Asiatic Society, vol.
iy. Plate xlviiiim[Ep].

390 FAUNA ANTIQUA SIVALENSIS.

REMARKS ON THE List.

The total number of fossil species belonging to freshwater
genera is 15.

The total number of recent species brought from the same
country is 15.

Four of the fossils are identical with recent species.

The remainder are probably extinct forms. They consist
of :—

Ist. A Paludina, which, though extremely near P. Ben-
galensis, must be regarded rather as its representative than
its homologue.

2nd. A Melania which is very distinct from any Indian
form I can find a record of, and which appears to have taken
the place of the existing Melania thiarella. I have seen 10
Indian Melanie; it is none of them.

3rd. Of a series of Uniones which are different from any spe-
cies of which I can find arecord. In 1838 Mr. Lea enumerated
13 species as the total number of known Asiatic Uniones.
Of the 10 Sewalik fossil species, 1 only is identical with any
of these 13 species, so that 9 remain to be accounted for.
The common Unio of India appears to be U. margimalis of
Lamarck, which is not to.be found among these fossils.

Of the species found fossil which are identical with recent
forms, the only Unio appears to be identical with one of the
commonest existing species of India. The Paludina is the
P. unicolor which is found among the shells of the Paris basin,
and also in the mammilliferous crag of England. At present
this species ranges from Egypt to India. The Ampullaria is
remarkable for bearing a more close resemblance to the
individuals of the species from the Jumna than to those
existing at present in the Sewalik district.

Besides these fossil freshwater shells there is a species of
Pupa or Bulimus with which I am unacquainted.

On the whole I am inclined to regard the evidence of the
Sewalik Molluscan fauna as indicative of the older Pliocene
period—at latest.

" Hi. Forses.

ete

FOSSILS FROM PERIM ISLAND. 391

XXIII. DESCRIPTION OF SOME FOSSIL REMAINS
OF DINOTHERIUM, GIRAFFE, AND OTHER
MAMMALIA, FROM PERIM ISLAND, GULF
OF CAMBAY, WESTERN COAST OF INDIA.!

(Chiefly from the Collection presented by Captain Fulljames, of the
Bombay Engineers, to the Museum of the Geological Society.)

BY H. FALCONER, M.D.

Dorine the late meeting of the British Association, at Cam-
bridge, I made a communication? to the Geological section
on some new additions to the Fossil Fauna of India, from
Perim Island, in the Gulf of Cambay. Among these were
mentioned a species of Dinotherium and of Giraffe, and a new
Ruminant genus of a size nearly equalling the Sivatheriwm,
found associated with remains of Mastodon, Elephant, Rhino-
ceros, Hippopotamus, and several species of Ruminants.
The occurrence of Dinotherium in the extinct fauna of India
is a point of such interest that no delay ought to take place
in laying before paleontologists the evidence upon which

1 Note by the President.—The follow-
ing letter with reference to the sub-
ject of this Memoir has been received
from Dr. Falconer :—

To the President of the Geological
Society.

Sir,— Having had occasion to examine
the Indian fossils in the Museum of the
Geological Society with reference to the
work on which I am at present engaged
on the extinct Fauna of Northern India,
and having found in the collection of
fossil bones from Perim Island, presented
to the Society in 1840 by Captain Full-
james, several remains of the highest in-
terest which have not yet been described,
I have in compliance with your desire
made a brief description of them, which I
have now the satisfaction of forwarding.

I have the honour to be, Siz,
Your most obedient Servant,

H. Fatconer.
London : July 1845.

The collection of fossil bones presented
to the Geological Society by Captain
Fulljames not having been accompanied
by any memoir, and no description of it
having yet been given in the publications
of the Society, the President and Coun-
cil have not hesitated to deviate from
their usual course with regard to the
publication of memoirs, and have directed
the insertion of Dr. Falconer’s communi-
cation in the present number of the Jour-
nal of the Society.

[This memoir was communicated to the
Geological Society of London in July
1845, and is reprinted from the Journal
of the Society. In addition to the fossils
herein described the reader is referred
to the figures and descriptions in the
‘Fauna Antiqua Sivalensis’ of Mastodon
Perimensis (Plates xxxviii., xxxix., and
xl.), Rhinoceros Perimensis (Plates lxxv.
and lxxyi.).—Ep. |

2 This communication was read on Tues-
day, 24th June, 1845.

392 FOSSILS FROM PERIM ISLAND.

the statement is founded; and as the Geological Society
possesses the largest collection of Perim Island fossils to
which I have had access, including remains of most of the
species to be noticed in the sequel, the pages of its journal
are the fittest place for this communication, the main object,
indeed, of which is to do justice to the meritorious labours
of Captain Fulljames, of the Bombay Engineers, one of the
earliest and most successful explorers of the ossiferous beds
of Perim Island. This is the more called for, as considerable
delay has occurred in the description and determination of
the remains which that officer collected and transmitted to
England several years ago.

Fic. 14.

Map or THE Districr NEAR Prerim JsLAND.
Long. 72° 30 E,

ae

Lat. 22° Ne Lat. 22° N,
CUZERAT.
SMa ay
MT :
f
\
U
al, Z oy RT
C. Ae ie ZnSO
Whoa) Ave i
PONIES Sh
cot N IN é k
i ov
= Fe A
iN
Gurr Pr | CamBay {
References.
a. Perim Island. e. Bhownuggur, a. Surat.
6. Gopenaut. Jf. Cambay. k. Tapti River.
e. Politana. g. Baroda. 2. Nerbudda R.
d. Gogah. h. Baroche. m. Mhye R.

n. Gulf of Gutch.

Note.—Between the Nerbudda and Tapti rivers (% and 2) is the river Keem, the
mouth of which is seen, but whose course is not indicated on the map.

Perim' is a small island, situated in lat. 21° 31’, in the
Gulf of Cambay, nearly opposite the estuary of the Nerbudda

* Captain Fulljames, Journ. Asiat. Soc. of Beng., vol. v. p. 289.

oe eee ee

FOSSILS FROM PERIM ISLAND. 303

River, and separated about 500 yards from the coast of Kat-
tiwar in Guzerat, by a channel which Captain Fulljames
states to be 75 fathoms deep. The island is about three
miles in circumference, being from one and a half to two
miles in length, and in breadth one half to three-quarters of
amile. The only particulars regarding its structure, with
which I am acquainted, have been given by Captain Full-
james and Dr. Lush.’ The highest point of the land is said
to be not more than 60 feet above high-water mark. The
western side presents cliffs of conglomerate, of about 30 feet
above the sea, ‘the upper strata being of compact sandstone,
all perfectly horizontal.’? Captain Fulljames describes
the order of succession, commencing from the surface, as
thus :—

. Loose sand and gravel.

. Conglomerate, composed of sandstone, clay, and silex.
Yellow and whitish clay, with nodules of sandstone.
. Conglomerate, as above (No. 2).

. Caleareo-siliceous sandstone, with a few fossils.
Conglomerate.

Indurated clay, more or less compact.

. Conglomerate, being the principal ossiferous bed.

ONAAPR wore

No precise measurement is given of these beds, but the
deepest strata of conglomerate are described to be about 3
feet thick, although in general they do not run more than
18 inches to 2 feet, and for the most part are horizontal.
“On the western side of the island, however, the strata are
much disturbed, being fractured and dipping at an acute
angle to the east. On the southern end of the island, sand-
stone appears below the fossil stratum of conglomerate,
dipping to the north at an angle of 25°.’ ‘Capital fresh
water is procurable on the island, rising from 20 feet below
the surface; it is found below the stratum of sandstone.’
Dr. Lush states that ‘proceeding from the south point
towards the eastward, layers of kunkur are met with below
the sandstone.’ He also adds that shells and other fossils
are found in the conglomerate, besides the osseous remains.
But none of those shells are to be seen in the specimens to
which I have been able to refer in the Geological Society’s
collection or at the British Museum.

Our information regarding the geological structure of both
sides of the Gulf of Cambay is at present exceedingly imper-
fect; but much may be expected when the unpublished
researches of the lamented Malcolmson are brought out,
as he is known to have carefully determined the succession

1 Toc, citat. Lush, idem, vol. v. 2 Lush, loc. citat.
p. 767, 3 Fulljames, loc. citat.

394 FOSSILS FROM PERIM ISLAND.

and age of the tertiary beds along the coast of the northern
Concan. In regard to what is known, Dr. Lush describes
the sandstone of Bombay as appearing as Mahim, Seergaum,
and Danu in horizontal strata, and ‘evidently above the
trap. At Gundavie the shell sandstone disappears, and
beds of clay and kunkur present themselves in the line of
section from Gundavie to Surat. From this point to the
Keem River nothing is seen but the ‘ black cotton soil ;’ on
the right bank of the Keem, sandstone and conglomerate
are exposed, according to Dr. Lush, in the following order :—

Section on the Right Bank of the Keem.

1. Alluvial soil with masses of conglomerate ; : ; . 6 feet
2. Horizontal beds of sandstone in thin layers : : 2 . 8 feet
3. Sandstone. é c 5 5 . 5 feet

4. Coarse conglomerate (bed of the river).

Respecting the Kattiwar coast, nearest which Perim Island
is placed, Dr. Lush mentions the conglomerate as reappear-
ing at Gogah, close to the island, where masses of the rock
containing shells are dug out of the beach. This conglome-
rate appeared to him to contain no fragments of trap,
although the central ridge of Kattiwar, including the hill of
Politana, is composed of trap, which is also seen at Bhow-
nuggur.! Captain Fulljames states that he has found ‘a simi-
lar formation to that of Perim all along the coast from Gogah
to Gossnath Point, where a firm sandstone is quarried, and
of which the splendid Srawak temples of Politana? are built.’
Captain Fulljames, in a separate paper,*® gives an account of
the strata passed through, in an experimental boring at the
townof Gogah. Of the 320 feet mentioned in the section, the
uppermost 74 include of sand and gravel, 11 feet; stiff black
clay, 6 feet; sand and clay, 10 feet; soft sandstone, alternating
with thin seams of different coloured clays, sand, and gravel,
13 feet; and lowermost, a very hard siliceous sandstone, 9
feet thick. The inferior portion of the section is composed of
a great bed of dark clay, which has been penetrated down to
246 feet, containing pyrites and broken shells. The whole
of this mass appears to be above the conglomerate, but it is
not shown whether the absence of the clay deposit at Perim
is owing to denudation or to its upheaval before the clay
on the coast was deposited. Captain Fulljames states that
he had discovered fossil remains, like those of Perim Island,
down the coasts towards Gossnath, and in a similar formation.

The first announcement of the Perim fossils is given in a

1 Dr. Malcolmson, however (vide postea, 2 Pattitona (of Fulljames).—[Ep.]
p- 404), mentions the occurrence of trap 3 Jour. Asiat. Soc. of Beng., vol. vi.
pebbles in these same tertiary beds. p. 787.

FOSSILS FROM PERIM ISLAND. 395

communication, dated 17th April, 1836, by Baron Carl von
Hiigel,' in which he mentions their having been discovered
by Dr. Lush. Among the remains which he enumerates are
bones of the Mastodon latidens, the core of the horn of a
species of Bos, the head of a boar, and a rodent. Captain
Fulljames concedes the priority of discovery to Dr. Lush ;
but immediately after followed up the inquiry by more ex-
tended researches, commencing in April of the same year ;
and it is to him that we are indebted for the greater part of
the Perim fossils, which are to be found in the Museums of
the Asiatic Societies of Calcutta and Bombay, and of the
Geological Society of London. Among those which he first
met with he mentions ‘ teeth of Mammoth, Mastodon, Palwo-
therium, Hippopotamus, Rhinoceros, and a number of other
smaller animals ; elephant’s tusks; the head of some large
saurian animal; tortoise; one half of a deer’s foot; and a
shell in siliceous sandstone.’ In the collection which Captain
Fulljames sent to the Asiatic Society of Calcutta, Mr. James
Prinsep? enumerates ‘many jaws of the Mastodon in fine
preservation; also teeth or jaws of the Hippopotamus,
Elephant, Rhinoceros, a large animal assimilating thereto
(Lophiodon ?), Sus, Anthracotheriwm (?), Deer, Ox, many ver-
tebree and unidentified bones and horns; Tortoise fragments,
and a peculiarly perfect saurian head.’? These identifica-
tions are not to be considered, in several of the instances, as
more than approximative ; for neither of these gentlemen
profess to be familiar with the subject of fossil bones.

No further account of these remains has appeared in any
of the Indian journals since that time. In 1840 Captain
Fulljames sent his donation to the Geological Society, and
about the same time some specimens from the same locality
were presented by Miss Pepper to the British Museum.

Judging from the matrix which adheres to them, the
Perim fossils seem to be embedded, in most cases, in a caleareo-
ferruginous conglomerate, composed of nodules of indurated
yellow clay, cemented together by a paste of sand and clay.
Some of them are attached to patches of a hard argillaceous
sandstone. Many of them have had the matrix washed off
by the action of the sea, and are in this case generally covered
over with the remains of small species of Serpula and other
recentmarine shells. The mineral character of the bones shows
that they are penetrated with siliceous infiltration, like a
great portion of the Sewalik fossils ; and in consequence they
present a great degree of hardness. The same character

' Jour. Asiat. Soc. of Beng., vol. v. $ Jour. Asiat. Soc. of Beng., v. vi. p. 78.

p. 288. (May 1836.) (Jan, 1837.)
* Loe. citat., vol. v. p. 290.

396 FOSSILS FROM PERIM ISLAND.

holds in many of the osseous remains from the crag; like the
latter, the Perim bones, under the action of the sea, wear
down into a polished vitreous surface.

DINOTHERIUM.

The first of these remains to be noticed is a fragment
(Plate VI. fig 3) consisting of the posterior half of one of the
inferior molars of a species of Dinotheriwm. The correspond-
ence of the specimen with the teeth of the large European
Species is so complete in the form of the gable-shaped grind-
ing ridge, its transverse direction, and the reflected marginal
bulges into which it swells out on either side, together with
the characteristic crenulation of the edge, that there can be no
doubt of its belonging to the genus Dinotheriwm. The pecu-
liar ‘talon’ or heel ridge is developed in the same degree and
with a like amount of crenulation along its edge. The frag-
ment is represented in section in Plate VI. fig. 3, the internal
structure exhibiting the same agreement with that of the
European Dinotherium (fig. 4) indicated by the external form.
The centre is occupied by a rhomboidal core of arenaceous
matrix marking the form of the unossified pulp nucleus. I
have compared it minutely with a corresponding section of
the same tooth (the penultimate of the lower jaw) of Dino-
thertum giganteum (fig. 4) from Eppelsheim; and the only
perceptible difference is, that the angle formed by the ridge
of the ivory is more acute, and the enamel thicker in the
Indian than in the European form. Perhaps no conclusion
can be safely drawn from this observed difference of angle in
the ivory ridge; as it may be a peculiarity of the individual.
The greater thickness of enamel is probably of more import-
ance, and may represent a mark of specific distinction. The
specimen, however, is much too defective to warrant any
opinion in regard to the relations of the Perim fossil to the
European species, except that it was quite as large as the D.
giganteum. Weare fortunately able to determine the position
of the tooth in the jaw with some confidence. The upper
erinders in Dinotheriuwm have a long low basal ridge in front
and behind; while the same teeth in the lower jaw have
hardly any ridge in front, and the hind one is consider-
ably more developed than in the upper grinders, so as to form
a strongly marked ‘talon’ or heel. The Perim fossil ex-
hibits this heel of large size, while the presence of an impres-
sion on the posterior surface proves that there was a tooth
behind it. It, therefore, belonged to the penultimate molar
of the lower jaw, and apparently to the left side.

In short, there can hardly be a doubt about the specimen
belonging to a species of Dinotheriwm. The only question

DESCRIPTION OF PLATE XXXITI-

BRAMATHERIUM PERIMENSE AND DiInoTHERIUM INDICUM.

Fig. 1.

Fig. 2.

Fragment of left side of upper jaw of Bramatherium Perimense,
presented to the Geological Society by Capt. Fulljames, showing
the three premolars, and below them the broken remains of the
first true molar. The figure is two-thirds of the natural size,
and has been reproduced from a drawing by Mr. Scharf in the
Quart. Journ. Geol. Soc., vol. i., Plate XIV., fig. 3. (See page
399.)

Shows the rugous surface of the enamel on the second premolar
of the same specimen, of the natural size.

Figs. 8 and 4. Another fragment of upper jaw, left side, of Brama-

Fig. 5.

therium Perimense, containing the last premolar (at upper end of
figures) and the three true molars. The specimen was referred
to Dr. Falconer by Major Jervis, and is copied two-thirds of the
natural size from the original drawing by Mr. Scharf in the
Quart. Journ. Geol. Soc., vol. i., Plate XIV., figs. 4 and 4 a.
(See page 400.)

Dinotherium Indicum. Fragment of left half of lower jaw con-
taining nearly the whole of the adult series of five molars in
situ. ‘The specimen is among the Perim Island fossils presented
to the British Museum by Miss Pepper. The figure is one-
fourth of the natural size, and has been copied from a drawing
by Mr. Ford in Plate XXXV., fig. 6, of the Fauna Antiqua
Sivalensis. (See page 404.)

VOL. I.

Vol. I. Plate 33 .

Fig. 4.

W.West imp.

1-4. Bramatherium Perimense.

G. Scharf del. J Dinkel lith.

5. Dmothertum Indicum.

FOSSILS FROM PERIM ISLAND 397

which can arise is in regard to the correctness of the locality
whence the specimen is said to have come. It was presented
to the British Museum by the lady whose name is mentioned
above, asa Perim Island fossil, along with teeth specimens of
a species of Mastodon known to be found in the Perim de-
posit. Mr. Konig, the eminent conservator of the Palzon-
tological department, who had early recognized the generic
relations of the fossil, is confident about the donor and the
mentioned locality. An additional confirmation is met with
in the mineral condition of the specimen. It exhibits the
silicified appearance, which is so prevalent in the Ava, the
Sewalik, and Indian fossils generally. The ivory core is
fissured into a vast number of radiating minute segments
which have been re-cemented by a siliceous paste (as has
happened to certain agates), and the whole of the structure
—enamel and ivory—has become so thoroughly penetrated
with siliceous infiltration that it resists the knife and takes
on the highest degree of vitreous polish in the section, while
the external surface of the enamel, from the same cause,
presents an opaline appearance. All the Eppelsheim speci-
mens of Dinotheriwm which I have had an opportunity of
examining are, on the other hand, unsilicified, softer, and
of less specific gravity. In section their ivory cuts under
the knife, and yields a dull earthy surface; while the harder
enamel takes on but a very imperfect polish. This circum-
stance strongly confirms the Indian origin of the fossil. It
is very possible that the large animal—‘ assimilating to the
Rhinoceros (Lophiodon ?)’—mentioned by Mr. James Prinsep
in the quotation above given, may also belong to Dinotheriwm.
This conjecture is thrown out for the guidance of those con-
nected with the Museum at Bombay and that at Calcutta,
who have access to the original specimens. What we know
at present must serve in a great measure as an index merely
to further inquiries. I would suggest in the meantime de-
signating the Perim fossil provisionally by the specific name
of Dinotherwwm Indicum.

The following are the dimensions of the fragment compared
with those of the same tooth of the Dinotheriwm gigantewm
from Eppelsheim.

Eppels-
heim
Specimen

_ Perim
Fossil

Inches Inches

Length of penultimate molar, lowerjaw . c ; : -- 3°75
Width of ditto, at the posterior ridge . c col) eral 2°75
Length in section of the per ee of 1 Ory, ‘atbase .| 15 1-75
Height of ditto . : : : - 63 63

Thickness of the enamel . ; : 2 a ; A 25 19

398 FOSSILS FROM PERIM ISLAND.

The fossil is represented in the figure about two-thirds of
the natural size (Plate VI. fig. 3).

GIRAFFE (Camelopardalis).!

The specimen is a fragment comprising the posterior half
of the second cervical vertebra of a Giraffe, a good deal mu-
tilated. It shows the characteristic form of the body of
the vertebra in this genus, and the cup-shaped articulating
surface for the head of the third cervical vertebra. The
upper half is wanting, and the posterior oblique processes are
broken off. Along the middle of the body there is a well-
marked longitudinal ridge, corresponding exactly in form
and development to that mentioned as characterizing the
third cervical vertebra of the Camelopardalis Sivalensis de-
scribed in the ‘ Proceedings’ of the Geological Soc. (See antea,
p- 199). The same remark applies to the lateral ridges of the
body, which are decurrent from the inferior’ transverse pro-
cesses, terminating at the posterior end of the bone in thick
expansions. This part of the vertebra is differently formed
in both respects in the existing species. The same resem-
blance is further shown by the spinous process, the projecting
part of which, as in the Sewalik specimen, is placed lower
down on the arch than in the living species. The mutilated
condition of the fragment prevents the form of this process
from being well ascertained; but the very low position and
shape of the most salient part determines the vertebra to
have belonged to the second of the neck series. There is
enough remaining to indicate that there was a like corre-
spondence with the Sewalik fossil in the curve of the body
on its under surface, which is more arched than in existing
Giraffes. The specimen is so weathered and abraded as to
present only few points for measurement; but such as may
be taken indicate the closest agreement between the fossils :

Perim .
Island Sewalik

Fossil Fossil

Greatest width at the posterior end of the pee between the | Inches | Inches

transverse processes : : : ol 31
Vertical diameter of articulating cup - : : 2-1 2°
Transverse diameter of ditto - - - : - : 21 2°

The Perim fossil, like the Sewalik one, is proved to have
belonged to an adult and even aged animal, by the marked
1 The specimen is now in the British | vol. i. Plate xlv. fig. 5, and in an un-

Museum (Cat. No. 39,748). It is also | published plate of the‘Fauna Antiqua
figured in the Quart. Jour. Geol. Soc., | Sivalensis’ (E. 2).—[ Ep. ]

FOSSILS FROM PERIM ISLAND. 399
relief of the ridges, the depth of the muscular depressions,
and especially by the circumstance that the posterior articu-
lating surface is completely synostosed with the body of the
bone, which is not the case in young animals. With this
united correspondence in form, size, and other particulars, I
have little hesitation in referring the Perim Island fossil to
the second cervical vertebra of the Camelopardalis Sivalensis.
This specimen is from the collection sent by Captain Full-
james to the Geological Society.

BRAMATHERIUM. Pl. XXXIII. figs. 1, 2, 3, 4.

The next of these remains to be noticed are of great in-
terest, as they appear to indicate a large and peculiar form
of Ruminant, nearly equalling the Sivatherium in size, but
at the same time essentially different. The remains consist
of two fragments of the left side of the upper jaw, including
the entire series of the superior grinders.!| Although pro-
bably of the same species, they are certainly not derived from
the same individual. The first fragment (fig. 1) is from
the collection sent by Captain Fulljames to the Geological
Society. It contains the three false or premolars nearly
perfect, together with the broken remains of the first true
molar. The surface of the enamel (fig. 2) shows the
rugosely furrowed character, which is found in the Siva-
therium ; but the whole of the teeth in the fossil are at once
distinguished from those of that genus by the absence from
all of them of the sinuous plaited flexures, which the inner
crescent of enamel presents in it; they also want the basal
collar or ‘burr’ on the inside, which is seen in those of
Sivathervum. With these discrepancies, which are of consider-
able importance in the Ruminantia, from the constancy of
such modifications in the different groups of this order, the
premolars of the fossil correspond in general form, and in

1 The following remains of Brama-
therium are described by Dr. Falconer in
the Catalogue of the Museum of the
Asiatic Society of Bengal :—

No. 38. Perim Collection. Lower
jaw, left side, in two pieces, containing
2nd and 38rd true molars, truncated in
front and behind.

No. 39. Lower jaw, left side, con-
taining four molars, worn, but more or
less broken.

No. 40. Proximal ungual phalanx
of left leg. It corresponds almost
exactly in size and in proportion with
the same phalanx of the fore leg of
existing Giraffe in the Asiatic Society’s
Museum.

No. 41. Horn of the Bramatherium (?)
resembling the horn of Sivatherium,
figured in Journal of the Asiatic Society,
but different. This is a specimen of
very great interest.

No. 42. Portion of the shaft of a long
bone, apparently the tibia, left side, of a
ruminant resembling the Giraffe and
hence inferred to be of Bramatherium
from its larger size.

For an account of the bones of the
anterior and posterior extremities of
Bramatherium the reader is also referred
to the description of an unpublished plate
of the Fauna Antiq. Siv. (F.)—[Ep.]

400 FOSSILS FROM PERIM ISLAND.

the relative proportion of width to length with those of Siva-
therium. The only other genus of Ruminants which shows
the peculiar rugose enamel furrowing, in a marked degree,
is the Giraffe, which agrees with the Perim fossil in the simple
direction without fold, of the inner crescent of enamel. But,
in this genus, the upper premolars are distinguished from
those of all other Ruminants by their great excess of width
compared with their length. In this respect, and further in
being considerably more oblique, both in form and in their
relative position in the jaw, these teeth in the Perim fossil
differ from those of the Giraffe. The dimensions of the fossil
contrasted with those of the Sivatheriun gigantewm, and of
the skull of an adult male Giraffe in the collection of the
College of Surgeons, are as follow :—

peri | rium | Mate
- irate

No.5) eee

Inches Inches Inches
Length of the three premolars. . : - “4 4:5 2°9
Length of the Ist premolar . 15 1:75 9
Width of the 1st * : ; § 1:3 1:63 11
Length of the 2nd ___,, : = 1:4 15 ifs
Width of the 2nd i, 15 1°75 11
Length of the 3rd __s,, 1:25 1:75 it
Width of the 3rd 49 5 . : ; 3 _— 19 1:2
Length of the Ist or antepenultimate true molar .| 1°6 16 1:33

The second specimen (figs. 3 and 4), (for an examination
of which I am indebted to the kindness of Major Jervis, of
the Bombay Engineers), is also from Perim Island, and, shows
the hindmost premolar, together with the three back or true
molars nearly perfect. Like the premolar of the other spe-
cimen, these teeth, besides being smaller, differ from their
equivalents in Sivatheriwm giganteum, by the absence of the
flexuous direction of the enamel, and of the basal ridge at
the inside. In these particulars, and also in the presence of
a minute or rudimentary cone of enamel, on the inner side
at the base, between the barrel divisions of the teeth, but
attached only to the posterior lobe, they correspond with the
other molars of the Giraffe. But the anterior pillar of
enamel, on the outer surface of the front half of these teeth,
is considerably thicker in proportion in the fossil than in the ~
Giraffe ; while the outer surface of the posterior half is more —
expanded in length, and is more hollow than in the latter —
genus. A still more important difference is, that in the —
fossil there is no tendency to a basal mammilla or enamel lobe ~
at the outside between the barrel divisions of the two back- —

FOSSILS FROM PERIM ISLAND. 401

most molars, as in the larger fossil Giraffe of India (See
antea, p. 201, and Pl. XVI. figs. 5, 5 a, and 6); while the
middle of each of these divisions at the inner side is so
compressed vertically as almost to present an obsolete or
indistinct form of keel. The following are the comparative
dimensions, as in the case of the previous specimen :—

Perim i 5 ale

Fossil | “Zigant. | Giraife

Inches Inches Inches
Length of the series of three back molars. .| 463 a 39
Length of the 1st molar 5 c : < : 1-6 1:63 1:33
Width of the 1st, 5 - 1:75 2° 1:97
Length of the 2nd _ ,, : : : ‘ : 1°75 We 1:37
Width ofthe 2nd __s,, ; FE 5 . é 19 2° 1:37
Length of the 3rd _,, P : . 3 ‘ 1:6 Bp 1:37
Width of the 3rd __,, ; . - : sy — 1:75 1:37

i

It is not necessary to follow up the comparison of the fossil
teeth with those of the Bovide, Cervide, and other families
of the order, from all of which they appear to be more re-
moved than from the Sivatheriwm and Giraffe. The molars
of the Ruminantia generally are formed so much upon the
same plan, that it is not easy to draw sufficient generic dis-
tinctions from them alone. The characters presented by these
Perim fossils, so far as they go, certainly distinguish them
from the Sivatherium, and also from the Giraffes, fossil and
recent; but their nearest affinity appears to be with the
latter genus, and they probably belong to the same family.
The materials presented here, as in the case of the Dinothe-
rium, are much too scanty at present for any conclusive
opinion on the subject. Meanwhile, under the conviction of
the generic distinctness of the Perim Ruminant, I propose
considering it as a genus under the name of Bramatheriwm,'
with the specific title of B. Perimense, to mark the rich and
interesting fossil locality where it was found.

The Dinotherium, Giraffe, and Bramatherium, are the only
Perim fossils which it is intended to particularize by descrip-

~ tion in this communication. But Captain Fulljames’s collec-
_ tion includes specimens of a great many other forms, which

prove that the clay conglomerates of the Gulf of Cambay
contain entombed in them the remains of a very extensive

_ and varied fauna. Among them there occurs one species of

1 The name Sivatherium, derived from | niently be applied to another; the

the Hindoo God Siva, having been ad- | ordinal relationship of the two will thus

mitted for one great fossil Ruminant | be easily remembered, together with

from India, Bramatherium, derived | their common Indian origin.

' from the God Bramah, may conye-

VOL. I. DD

402 FOSSILS FROM PERIM ISLAND.

Mastodon, one of Elephant, a large species of Rhinoceros,
Hippopotamus, Sus, Equus, several species of Antelope, Bos,
two species of Crocodile, one of which is of the Gavial type;
several forms of freshwater Tortoises, with fish vertebrzee two
and a half to three inches in diameter. These will be noticed
in detail, in the work upon which Captain Cautley and myself
are engaged, on the fossil fauna of the Sewalik hills. The
principal point of interest about them requiring mention on
this occasion is, that the mass of the Perim fossils belong to
the same genera and species which are found in the Sewalik
hills, and in the ossiferous beds of the Irrawaddi in Ava.
We have thus conclusive proof that, in the later tertiary
period, as at present, one and the same vertebrate fauna
ranged from the banks of the Irrawaddi, on the eastern side
of the Bay of Bengal, 1,700 miles up along the foot of the
Himalayahs to the Indus, where it escapes from these moun-
tains, and thence across the continent to the western side of
India. We are now getting the first glimpse of the evidence,
regarding the range and distribution of the species. Some,
as at present, were common over the whole extent of country,
while others appear to have been limited to, or had their
force of development in, a particular tract. The prevailing
species of Mastodon from Perim is identical with one of
the forms described by Mr. Clift, under the name of IM.
latidens, in his excellent memoir in the Geol. Transact.,
2nd. ser. vol. ii. p. 371, this nominal species appearing to in-
clude two very distinct forms. One of these (Mr. Clift’s, Pl.
XXXVI. figs. 1-4) seems to have been common on the western
side of India and in Ava, while it is but rarely found in the
Sewalik hills. The Perim Sus is identical with a Sewalik
species (Sus Hysudricus, Fale. and Caut.) ; and a like agree-
ment has been noticed as holding with one species of Giraffe.
The Dinotherium and Bramatherium have not yet been
observed amidst the fossils of the Sewalik hills, while the
gigantic Tortoise (Colossochelys Atlas) ranged from the Se-
walik hills to the Irrawaddi.t The Hexaprotodon form of
Hippopotamus occurs in Perim Island, Ava, the valley of the
Nerbudda River, and the Sewalik hills.

I have had occasion, in more than one instance, in joint
communications with Captain Cautley to the Geological
Society, to refer to the singular richness of the ancient fauna —
of India, in mammiferous forms. As a general expression of —
the leading features, it may be stated that it appears to have |
been composed of representative forms of all ages, from the —
oldest of the tertiary period down to the modern, and of all

1 There are fragments of this great Chelonian among the fossils brought by —
Mr. Crawfurd from Ava.

FOSSILS FROM PERIM ISLAND. 403

the geographical divisions of the old continent grouped to-
gether into one comprehensive fauna in the countries along
the valley of the Ganges. The Dinotherium of the miocene
period of Europe was, till now, a notable exception; but the
fossil described in the preceding pages shows that ancient
India was not without a representative of this most remark-
able genus. In addition to most of the known types of
Ruminants we have now evidence that the same country had,
in the Sivatherium and Bramatherium, at least two colossal
forms of this order.

In regard to the precise determination of the age of the
ossiferous deposits of India, the problem still remains to be
solved. The western coast of the peninsula will, in all pro-
bability, furnish the most certain and numerous data for its
solution ; as we may expect there to find deposits and organic
contents corresponding to the numerous alternations of up-
heavement and submergence which the land on that side of
the continent has undergone. Fossil remains of Elephant,
Hippopotamus, Equus, &c., were discovered by Dr. Spils-
bury,' in the valley of the Nerbudda, near Jubbulpoor, in a
bed of limestone capped by a thick mass of basalt, and traces
of mammiferous remains have been found in other parts of
the basaltic district of central India. Extensive lacustrine
deposits disrupted and altered by the same igneous rock have
been met with over a wide extent of the Deccan, containing
the same species of Paludina, Physa, Limnea, Unio, and
Cypris.”

Reasoning from these facts, Dr. Malcolmson was led to the
inference ‘that the part of the Vindhya range near Mandoo
was elevated during the same comparatively recent epoch as
the Sichel hills, between the Godavery and Tapti, the Ga-
wuleurh range, and the Satpoora mountains, south of the Ner-
budda.’ He adds also the following startling generalization:
‘ Over all these tracts, then, I am justified in believing that
at one time extensive lakes and marshy plains existed, full
of the ordinary forms of lacustrine life. The precipitous
and thirsty mountain ranges which intersect India, and which
now rise bare and burnt up in inaccessible cliffs, which for
- months of every year hardly afford water for the birds of the
air, must then have exhibited vast plains, full of freshwater
lakes and marshes, on the muddy shores of which multitudes
of gavials, crocodiles, and tortoises must have preyed; and
amidst the rank luxuriance of the bordering vegetation the
Mastodons, Hippopotami, Bisons, and Sivatheria, must have

1 Jour. Asiat. Soc. of Bengal, vol. i1. | 2 Malcolmson, Geol. Trans., series 2,
p- 583, vol. v. p. 570.
DD2

404 FOSSILS FROM PERIM ISLAND.

ranged, whose bones are now found so abundantly scattered
over India.’?! Unfortunately, this excellent observer’s re-
searches on the Gulf of Cambay have never been published ;
but in a note appended to the paper quoted above he men-
tions the occurrence of trap pebbles in the tertiary sandstones
of Perim Island and Kattiwar (see antea, p. 394), and in the
cornelian conglomerates of Rajpeepla and Broach, which are
said to be remarkably altered by the intrusion of igneous rocks
of a late date.

Supplementary Observations.—Since the preceding remarks
were in type I have had occasion to examine some other
Perim Island fossils presented to the British Museum by Miss
Pepper, one of which has furnished additional and most un-
equivocal evidence of a huge Indian species of Dmotherium.
The specimen is a superb fragment of the left half of the lower
jaw, containing nearly the whole of the adult series of five
molars in situ.2 The contour of the body of the jaw is shown
in the most perfect state of preservation, the fossil having
fortunately been mineralized by means of a very hard siliceo-
ferruginous infiltration. But it has evidently been long
rolled about on the sea-beach as a boulder, so that the crowns
of the whole series of molars have been hammered off nearly
level with the alveolar margin of the jaw; the surface of the
fossil is jet black, and almost all of the matrix has been
cleared away, probably by the long-continued action of the
sea, which has given it a semi-vitreous polish. That it had
latterly been in the sea is distinctly proved by adherent
patches of recent marine shells identical with those found on
others of the Perim fossils; and the testaceous remains being
white, pearly, and fresh-looking, are seen in marked relief
upon the black surface of the fossil. The symphysis of the
jaw is broken off about 2} inches in front of the anterior
premolar, and the bone is truncated behind exactly opposite
the point where the coronoid margin of the ramus begins to
rise up, the fracture passing through the middle of the last
molar, the anterior ridge of which is visible in situ in the jaw.

The dimensions given below will indicate most distinctly
the characters by which this fossil differs from the jaw of the
D. gigantewm of Kaup. In relative length, the two agree
very closely, the four anterior molars measuring but half an
inch more in the Indian than in the European species. But
the other proportions are very different. The depth of the
jaw measured to the alveolar margin of the second premolar, _
where the deflexion of the symphysis begins alike in both, is —

1 Malcolmson, Journal Bombay Geo- 2 See Plate xxxili. fig. 5, copied
graphical Society, vol. for 1841-1844, | from Fauna Antiq. Siv., Plate xxxy.
p- 371. figs. 6 and 6 a.—[Ep.]

FOSSILS FROM PERIM ISLAND. 405

9-2 inches in the former, while it is but 6°9 in the latter, and
at the back of the third tooth or first true molar, 8°7 inches
to 6:2 inches. The Perim fossil exhibits a like excess of
dimensions in relative thickness, the jaw measuring 5:1
inches in diameter under the second premolar, and 6-4 inches
at the middle of the penultimate molar, while in the European
species the corresponding dimensions are respectively 4 and
5 inches. In consequence of this great depth and thickness,
the jaw of the Indian fossil approaches very closely the
massive and turgid form seen in the typical Mastodons, such
as the M. giganteus; while that of the European Dinotherium
is comparatively much thinner and more compressed. The
inner side of the jaw in the latter is very flat, differing in
this respect widely from the Mastodons generally ; in the
Perim fossil this flatness is much less, not exceeding that of
the Mastodon giganteus, and, behind, the body of the jaw
bulges out on either side, so as to yield nearly a circular
outline in section, and exactly to represent the form in the
American Mastodon. This resemblance is so great, that in
the absence of the crowns of the teeth, and from its huge
proportions, the fossil, when presented to the Museum and
up to this time, has always been regarded as the jaw of a
Mastodon. The relationship indicated by the shape of the
jaw is further borne out by the form and structure of the
penultimate lower tooth, as described in the preceding part
of this paper. The enamel, which is thinner in the D. gigan-
tewm, is as thick in the Indian species as in the lower
penultimate of the American Mastodon; the outline of the
ivory ridge beneath the enamel is the same in both; the ©
crown ridges have the same transverse, continuous, crenu-
lated, and trenchant form; and what is most important and
significant of all, the hind talon, in respect of form, amount
of development, and the characteristic crenulation of its
edge, is so precisely similar, that this part in the one exactly
represents the corresponding part of the same tooth in the
other. The same direction of affinity is further indicated
by the nearly horizontal line of protrusion and horizontal
plane of wear in the teeth, by the form of the ramus, coronoid
process, and angle of the jaw, and by the absence of antero-
posterior curvature in the outline of its lower surface, in all
of which particulars the American Mastodon deviates widely
from its congeners, and from the Elephantine type generally,
and approximates towards the Dinotherium. This tendency
is also shown in the very reduced formula of the teeth-ridges,
in the deflexion of the symphysis, its thick bluff termination,
and in the inferior tusks. JI shall soon have occasion in
another place to follow this subject at greater length, and in

406 FOSSILS FROM PERIM ISLAND.

the meantime must content myself with the simple state-
ment, that the North American Mastodon and the Indian
Dinotherium are the nearest connecting forms of the two
genera yet known, and that their relationship is far from
being remote, perhaps even nearer than that of the American
Mastodon to the Indian Elephant or the Mammoth.!

The deflection of the symphysis commences immediately
behind the second molar, as in the Din. gigantewm, and it is
evidently produced into a great bluff mass, bent downwards
as in that species. The section at this point does not include
any part of an inferior tusk, or of an alveolus for it; but
Dr. Kaup? tells me that the large tusks of the Eppelsheim
species, with their alveoli, always terminate considerably in
front of the anterior premolar. There is no reason, there-
fore, to conclude that the Indian had not tusks resembling
those of the European species; and although there is no
direct evidence to the point, everything in the construction
of the symphysis goes to support the presumption that there
were tusks. The posterior mentary foramen is of large size,
and situated at the outside under the anterior premolar,
exactly as in the Eppelsheim fossil, but at a greater distance
from the alveolar border of the jaw. It is much larger than
the foramen seen in the cast of the Eppelsheim lower jaw ;
but no faith can be. put in the dimensions of a foramen
measured on a cast.

In regard to the teeth, nothing is seen of their crowns,
which have been broken off close to the alveolar margin ;
but the bony partitions between five teeth are distinctly
visible, showing the usual complement in Dinotherium, and
proving that the fossil was derived from an adult animal.
These five teeth consist of two premolars and three true
molars. They diminish in width from the backmost forwards,
as in the European species. The anterior premolar has two
lobes, the front one being compressed and sharpened off
forwards into a cuneiform edge, the rear lobe being shorter
and broader. This tooth is upwards of half an inch longer
than that belonging to the jaw of the great specimen figured
by Kaup. The second premolar is nearly square in outline,
but wider behind. It appears to have had two ridges, and
four fangs. The third tooth or first true molar presents a
length of 4 inches by 2°8 of extreme breadth; while that of

1 See antea, p. 85.—[Ep.] guished paleontologist, the founder of

“2 T have had the advantage, while | the genus. Dr. Kaup was at once
engaged on the examination of this | convinced of the generic relations of
fossil, to benefit during his present resi- | both the fossils, but he is in nowise
dence in London, by the intimate | responsible for any of the opinions here

lmowledge of the structure of the | advanced regarding the distinctness of
Dinotherium, possessed by this distin- | the species, or its affinities,

in eee inn ss

Sr ee

FOSSILS FROM PERIM ISLAND. 407

the Eppelsheim cast measures 3°6 by 2°6. We have in this
excess of length conclusive proof that the Indian, like the
Huropean species, possessed the remarkable character of
having the first true molar three-ridged, and more complex
in its form than the two backmost grinders. The crown is
so utterly mutilated as to afford no evidence regarding the
form of these ridges. The second or penultimate true molar
is nearly square in its plan outline, but more than half an
inch longer and wider than in the European species. The
tooth specimen described in the body of this paper was in-
ferred to be the penultimate inferior, and it was probably
derived from a female or small-sized individual. The remains
in the jaw appear to indicate that this tooth was two-ridged,
with a talon as in the European species. Of the third and
last molar only the anterior half remains, and we have no
direct proof how many ridges it bore; but the number was
most probably two, with a talon, as in the Kuropean species.
The portion which remains presents two distinct and slightly
divaricating fangs, indicating, among many others which
could be added, another character of resemblance to the North
American Mastodon.

The following are the dimensions of the fossil compared
with those of a cast of the jaw of the great head specimen, sup-
posed to have been a male, figured and described by Kaup :—

Dinothe- | Dinothe-
Mastodon rium rium
gigan- | Indicum, | 8!822-
teus." [from Perim eae ue
Island. ee
Inches Inches Inches
Length of the fragment : : : . ¢ — lee a
Ditto of the four anterior teeth . 5 c : — 13°5 13°
Ditto of the first premolar . : é . < _- 3°5 2-9
Width of ditto behind . 3 ; : 3 0 — 2°2 2°2
Length of 2nd ditto. . 5 : : ¢ — 2:9 33
Width of ditto behind . : : : — 2°6 2°7
Length of 8rd or first true molar tooth . . . — 4: 3°6
Width of ditto behind . : 0 : — 2°8 3°
Length of 4th tooth (2nd true molar) : : — 39 33
Width of ditto. — 3°5 2:9
Depth of jaw to alveolar. margin at the 2nd
premolar : 6 : : 75 9-2 6-9
Ditto at 3rd tooth (1st true molar) : 2 .| 66 8-7 6-2
Width of jaw at 2nd premolar. 57 51 4:
Ditto at the middle of 4th tooth (or penultimate
true molar) . 6:5 6-4 5
Distance between the upper margin mentary a
foramen and alveolus of lst premolar . : — 3°6 2-2
Ditto from inferior margin to ditto 3 5 —_ 4°75 34

1.N.B. The four measurements of | son, are taken at points of the jaw corre-
Mastodon giganteus, given for compari- | sponding to those of the Dinotheria,

408 FOSSILS FROM PERIM ISLAND.

The Dinotherium of Eppelsheim is known to range through
a very wide difference of size, dependent on sexual or indi-
vidual peculiarities, and several nominal species, chiefly
founded upon this character, have been described by authors.
But Dr. Kaup informs me that he now admits but two species,
D. giganteuwm and D. Kenigii, as he regards all the rest, such
as D. Cwwieri, D. Bavaricum, D. proavum, &c., to be merely
dwarfed varieties, or females of D. gigantewm. M. de Blain-
ville has arrived at nearly the same conclusion in his ‘ Ostéo-
graphie.’ It would be unsafe, therefore, to found any opinion
regarding the Indian fossil merely on a difference of size.
But, in addition to the larger dimensions, the very remark-
able peculiarities in the form of the jaw, indicated by its
great depth in front, the excessive width, massive form, and
circular outline in section behind, togetherwith the absence of
the flattening of the inner side, which is so marked in every
specimen of D. gigantewm, taken in conjunction with the very
significant difference in the thickness of the enamel, appear
to furnish the strongest evidence that the Indian fossil
belongs to a distinct species. It is to be kept in mind, also,
that all these differential characters tend, in a remarkable
manner, in the direction of greater affinity with the Mastodon
giganteus. In corroboration of this view, it deserves to be
stated that of the numerous fossil Proboscidea discovered in
India we! have found that all the forms are specifically
distinct from those which occur in Europe. I have now no
hesitation in regarding both the Perim fossils to belong to
a distinct species of Dinotherium, larger than the D. gigan-
teum, and more closely allied to the Mastodons, which, as
proposed in the preceding pages, may be called D. Indicwm.?

Notr.—In the ‘ Athenzeum,’ No. 923, p. 662, there is an
abstract of a paper by Mr. A. Bettington, read to the Royal
Asiatic Society, on June 21 of this year, giving an account
of a finely-preserved cranium of a huge Ruminant, found by
that gentleman in Perim Island; I have repeatedly seen the
specimen, which was exhibited at the anniversary meeting of
the Geological Society on February 17 last; but as unpub-
lished material, which I had no authority to quote, I have not
felt at liberty to refer to it in the descriptions given in this
paper. Mr. Bettington institutes a comparison of his fossil with

1 In stating this, l use the plural pro- | were discovered in the valley of the
noun we, intending to intimate that the | Indus, below Attock, by Lieut. Garnett,
opinion is one in which my colleague | and identified by Dr. Falconer. See

Captain Cautley also concurs. postea, p. 414.—| Ep.]
2 In 1857 remains of Dinotherium

.

1
%
f

|

FOSSILS FROM PERIM ISLAND. 409

the Sivatherium and Giraffe, and considers it, so far as the ab-
stract above quoted indicates, to be distinct from both. The
circumstance that this cranium and the fossils here described
are from the same locality creates a strong presumption that
they may belong to the same genus or even to the same
species; but I am unable to say in how far the teeth agree,
as I have not had an opportunity for making the necessary
comparison. Mr. Bettington, as quoted in the abstract,
appears to consider that, in addition to horn buttresses behind
the orbits, there was a pair of recurved rear horns in his
fossil, at the side of the occiput, placed as in the buffalo.
This inference, if well founded, would be against the affinities
here attributed to Captain Fulljames’s fossil, should it prove
to belong to the same species. Among the remains men-
tioned as having been found associated with this cranium by
Mr. Bettington are species of Mastodon, Rhinoceros, besides
several forms of Ruminants, Crocodiles, &c.

APPENDIX TO MEMOIR ON PERIM ISLAND FOSSILS.

I.—DEscrIPTION OF A FRAGMENT OF Fosstu Bone From Pert IsLanp,
GouLF oF CAMBAY, IN THE COLLECTION OF Mr. Henry Ducx-
wortH, Liverpoot (Lasers 12 anp 47). From Dr. F.’s Nore-
Book.

This specimen consists of a mutilated portion of the right ramus of the
lower jaw, comprising the fang portions of one or more molar teeth ;
together with about two inches in length of the entire and sharp edge
of the diasteme. The internal alveolar wall is broken off, and polished
by rolling; the crowns of the teeth are also entirely wanting. The
specimen is truncated in front through the diasteme; and behind, in a
line with what is inferred to be the second molar, the surface of the

bone has acquired a vitreous polish, and is covered here and there with
white undetermined incrustations. The fragment is chiefly remarkable
- for the compressed character of the ramus, and for the great concave
curve of the lower margin in front, which is deflected downwards in a
bold curve intermediate between that of Dinotheritum Indicum and
M. (Trilophodon) angustidens. The compression of the ramus is too
great for an adult jaw of Dinotherium, and the downward deflection
greater than is met with in any species of Mastodon. The anterior end
of the fragment is curved slightly inwards, to meet the corresponding
opposite side; but no portion of the symphysis remains. Regarded from
the anterior end, the section shows a deep compressed elliptical hole,
considerably in advance of the mentary foramen, and which therefore
cannot be regarded as the dentary canal, but rather the alveolus of a
very large and solitary recurved incisor, as in Dinotherium giganteum,
or deflected, as in Mastodon angustidens. The mentary foramen is
situated posteriorly to the anterior fangs, and at about two-thirds of the

410 FOSSILS FROM PERIM ISLAND.

vertical height of the outer surface of the jaw. Part of the outer layers
of the bone has recently been denuded from the external surface; the
mentary foramen is of inconsiderable size. The specimen consists of
two fragments, the larger of which bears the label No. 12, to which a
smaller fragment (No. 47) was fortunately added, yielding the first
indication of the downward deflection of the symphysial portion. The
mentary foramen is of no great size; the diastemal edge is raised, and
very sharp, and follows an outward direction, where the anterior part of
the ramus commences to be bent inwards.

Unfortunately all evidence derivable from the crowns of the teeth
as regards the degree of composition and complexity is entirely wanting.
The principal dimensions are as follow :—

Inches

775

1. Extreme length of the fragment taken at the inferior margin (as
a chord to the are) . 4 : ; : 2 - ;

2. Versed sine of the concave are of the lower jaw, when rested upon
the ends of the fragment. : 5 : : : ;

. Vertical height of ditto, at anterior termination of the diasteme .

. Vertical height of the jaw, in a line with the anterior molar

. Greatest thickness of the jaw behind . < - ;

. Ditto in a line with the mentary foramen . 5

. Length of the remains of the diasteme - : A

Vertical height of the supposed alveolus, of the incisor = .

. Transverse ditto : : . c . “ 7 :

op to BN ENS

CO COMI D VP
Anwnoncoa

The supposed alveolus of the incisor is filled up with conglomerate
matrix, and its definition imperfectly shown.

This is a very important specimen, and thin sections, both of the
ivory of the teeth, and of the compact bone, ought to be taken for
microscopical investigation. As an approximative opinion, I am inclined
to consider it as of a Dinotherian form, and it may be, of a young Dino-
therium Indicum; but the contour of the lower jaw in that species is so
imperfectly known that nothing more positive can be asserted in regard
to it.!

Lonvon, 31 SAckviILLE STREET :
March 1858.

TI.—MS. Note sy Dr. FAatconer.—Microscopic EXAMINATION OF Mr.
Tart’s SUPPOSED VERTEBRA OF DINOTHERIUM FROM PERIM ISLAND.

Vertical Section.—Viewed with } inch object glass, the length of one
of the cells of Purkinje is 5+; inchand the width 5,,,. Their form
is long and ragged, and attenuated at either end, like an insect with in-
numerable legs. The measurements of another cell were 74, in. and x34

: : i F500"
The calcigerous tubes are +;455 of an inch apart. Seen in transverse

1 Among other specimens from Perim | lection from Perim Island, forwarded by
Island in Mr. Duckworth’s collection, | Dr. Buist, of Bombay, in 1860, Dr. Fal-
Dr. Falconer identified remains of | coner identified remains of Mastodon
Mastodon, Hippopotamus, Deer, Antelope, | Perimensis, Rhinoceros, Bramatherium,
Crocodile, and Tortoise. In another col- | &c.—[Ep.]

FOSSILS FROM PERIM ISLAND. 411

section their outline is oval with a diameter of =1,, to z545. In some
places the radiating calcigerous tubes are cut across, and take the form
of numerous dots in the field, like the ends of ivory tubes, and under a
low power have amoniliform or beaded appearance. The diameter of
the concentric circles is about #5 inch, there being from five to eight
rings of cells in each circle.

412 FOSSIL REMAINS FROM AVA.

XXIV. NOTE ON CERTAIN SPECIMENS OF
ANIMAL REMAINS FROM AVA.!

PRESENTED BY JAMES CaLDER, Esq., V.P., To tHE Museum
OF THE AsrIATIC SOCIETY.

BY HUGH FALCONER, M.D.

THESE specimens of Ava fossils were procured, with some
difficulty, from the neighbourhood of Prome, where the col-
lection by Dr. Crawfurd had been previously made. We may
hope ere long to have a richer assortment, through the
exertions of Major Burney; but in the meantime it may be
interesting to the Society to know what individual fossils of
the number have been identified with those taken hence by
Dr. Crawfurd.

It must be premised that the following attempt at dis-
criminating the animals to which the fossil bones belonged
is submitted to the Society with great diffidence, and is con-
fessedly imperfect. Had the means of comparison been more
extensive, an accurate list might have been made out; but
no recourse could be had to the skeletons of any analogues,
and the only available sources of information were Cuvier’s
‘Ossemens Fossiles,’ and Memoirs in the Geological ‘ Trans-
actions,’ by Buckland and Clift.

Further, this note does not profess to contain anything
original, or to make any addition to the list of fossils from
Ava already discovered.

The remains are confined to quadrupeds. There are no
marine nor freshwater shells, nor any specimen of the
deposit in which the remains are found; but it is believed to
be the same diluvial formation in which the remains of
mammalia are found in Europe and America.

The specimens consist of fragmental portions of bone;
many of them coloured withiron. The fragments are mostly
angular, and few of them bear marks of attrition.

1 This note was communicated to the | ings in Science,’ vol. iii. p. 167, Calcutta.
Asiatic Society, on April 20th, 1831, | Although fragmentary, it is here repro-

soon after Dr. Falconer’s arrival in | duced, as it has been quoted by Kaup
India, and was published in the ‘Glean- | and other authorities. Ep. ]

Re RES ea cn

FOSSIL REMAINS FROM AVA. 413

Of the Pachydermata there are bones belonging to two
genera, the Rhinoceros and Mastodon.

No. la. Isa fragment consisting of a longitudinal and
vertical section, of a molar tooth of a Mastodon, with a por-
tion of the attached jaw, and a nearly entire fang. The
characters are not sufficiently marked to determine the
species, but it seems to approach most nearly the Mastodon
latidens of Mr. Clift (Geol. Trans. vol. ii. p. 11).

No. 16. Isa portion of the middle of the femur of a
Mastodon.

No. 2a. Is a portion of the lower jaw of the left side of
a Rhinoceros, containing a perfect tooth, which has been
accidentally divided. It has belonged to an animal now
extinct, and of larger size than the Rhinoceros wnicornis of
this country. The form of the tooth is different from that
found by Mr. Crawfurd, but as the characters vary, from
wearing, with the age of the animal, it is no easy matter to
determine whether or not it belongs to a different species.
Tt resembles very closely in form the Rhinoceros eriques of the
“Ossemens Fossiles’ of Cuvier.

Nos. 3 a,b, and ¢ are vertebre from different parts of the
spinal column of Crocodilee. Some of the bones carried to
Kurope were found to belong to the Leptorhynchus of the
Ganges, or a species very nearly allied.

Nos. 4 a, b, c, and e are osseous fragments of two large
genera of Turtle, the Hmys and Trionyx. The remains of
these animals bear a large proportion to the other bones.

There are several specimens, comprising the greater part
of the collection, to which no names have been attempted to
be given. Some of them are well marked, and the in-
dividuals to which they belonged might be hazarded with a
tolerable degree of confidence; but when the evidence fell
short of certainty, it appeared better to present the bones
without attempting to name them than run the risk of
giving misnomers and misleading others.

A correspondence has been discovered between a specimen
from the Himalayahs and those of the Trionyx or Hmys from
Ava; there can be little doubt that when we have an op-
portunity of making a comparison of the fossil bones from
both places further coincidences will be brought to light.

414 FOSSIL REMAINS FROM ATTOCK.

XXV. NOTES ON FOSSIL REMAINS FOUND IN THE
VALLEY OF THE INDUS BELOW ATTOCK,
AND AT JUBBULPOOR.!

I nave examined the fossils from Attock presented to Dr.
Oldham by Lieutenants Garnett and Trotter. They are very
interesting. Among them I have found the following :—

1. A species of Dinotheriwm, probably new. Of this there
are two undoubted molars. The first is apparently the pe-
nultimate or first premolar, upper jaw, right side. At least,
this is inferred from its form and an obscure disc of pressure
on the posterior side, and from there being no disc of pres-
sure in front. The tooth consists of a longitudinal ridge
on the outer side, the enamel edge of which is rounded off,
but not much worn. It is convex in its antero-posterior
direction, and separated by a broad valley from two inner
mastoid points. These points are sub-pyramidal, the apex
of the anterior one being slightly pitted by wear. There are
no transverse ridges. In the first penultimate milk molar of
Kaup (Plate I.) the anterior ridge is bifid; in this it is quite
entire. The tooth also shows very large fangs.

European Attock

form form

Antero-posterior diameter, outer edge. A ; . 24 2°95
Do. do. inner ,, : : : anleil 2:1
Transverse do. 5 : A “ - a et 2°8

The Indian fossil is smaller than the European. It differs
also in the following respects. The surface of the crown is
free from the vertical groove; the anterior talon is much
more salient; the crenulated bourrelet is limited to the ends
and there is a bridge between the inner points; whereas in
the European fossil the crenulated bourrelet runs all round.

The second specimen, judging from its square proportions,
is the antepenultimate true molar, upper jaw, right side.
The two anterior ridges are well worn; the last ridge is
barely touched ; only the crenatures are gone. In form it
is very like the tooth of the European species. The last |
ridge is somewhat concave and crescentic across, as compared

! These fragmentary notes are extracted from Dr. Falconer’s note-books.—[ Ep. ]

FOSSIL REMAINS FROM ATTOCK. 415

with one of Kaup’s casts (marked d. 9 and 14 f). The

measurements are :—
European Indian
fossil fossil

Extreme length . 5 : 5 A) a8 2-7
Extreme width of crown, first ridge 3 3 6 a Boe 2-4

John Hunter’s original specimen (Cat. Coll. Surg. Eng. No.
800), called ‘the third deciduous upper molar, oe side,’
has these dimensions: extreme length, 2:7 ; width, 2

The most prominent distinctive “character is the uniform
convexity of the outer surface of the first true molar.

The species would probably be too small for the Dino-
therium Perimense.

2. Tapirus Pentapotamie (Falc.). This species is also new
to the Indian fossil fauna. One specimen is a tooth which
is probably the right penultimate true molar upper jaw.
Both ridges are slightly worn, and there is a disc of pressure
in front and behind.

Inches

Length of crown . : : 5 C “ : . : . 0°85
Width at base in front 3 5 : 5 : . P . 08
Do. behind . 3 . 5 ; Ms ‘ : - F . 0°84

The tooth corresponds with the penultimate true molar
upper jaw, right side, of the American Tapir (Tapirus
Americanus), but it is a trifle smaller.

Attock Malay | American

Tength . ‘ 0°82 1-1 “9
Width in front . 08 1-2 95
Ditto behind 0:8 11 8

The two smaller teeth would agree with the form of the
first premolar. They are distinctly bi-cuspid, as in the
Tapirs. The definition of the cusp is much more pronounced
in the fossil, and the form of the tooth more triangular.

3. Sus pusillus (Fale.). The specimen is a fragment of
the right ramus of a lower jaw of a small swine animal. It
consists of the upper part of the ramus and the base of the
coronoid, and it contains the posterior two-thirds of the last
molar in situ; the tooth is three-barrelled and the crown has
the rough warty swine character. The enamel is very thick;

the anterior barrel is broken off.
Inches

Length of crown from alveolus’. : : . « : . 0°65
Width of middle at base. . 2 c c : . 03
Height of enamel crown : : > : : - 02

The talon is nearly as large as the middle barrel. The
jaw is clearly of a full-sized animal, and therefore it belongs

416 FOSSIL REMAINS FROM ATTOCK.

to a small species of a genus undetermined. It is not
Hippohyus. I have named it provisionally Sus pusillus.

4. Merycopotamus nanus. The difference in the dimensions
of the last molar from Attock is so great from those of
Merycopotamus dissimilis that the former must have been
a distinct species (See antea, p. 138).

Merycop. Merycop.
dissimilis of Attock

Length of last molar : : 5 : : seeds 0°75
Width of do. in front > : : . 5 » te 08

5. Amphicyon. The specimen is the tubercular tooth of
a large carnivorous animal, as large as the Polar Bear and
allied to the Ampicyon, which occurs in the Huropean
miocene strata along with Dinotherium.

6. Antoletherium.! Among the fossils discovered by
Lieut. Garnett, near Attock, is a portion of the lower jaw of
a tapiroid animal containing what appear to be the last
premolar and the first and second true molars. Of this
specimen I have received an excellent drawing executed by
Col. Baker, who regarded the species as allied to Tapir.
(Plate XXXIV. figs. 1 and 2.) The teeth certainly differ
generically from those of Dinotheriwm in the massive con-
necting bridge between the two ridges, speedily running them
into one confluent disc. The bend of the first ridge is very
tapiroid. The specimen appears to me to be of an undes-
cribed genus for which the provisional name Antolethervwm
(avToAy the east, and @nypiov) would be appropriate.

[Col. Baker’s drawing was transmitted by Dr. Falconer to
Professor Owen, who replied on Nov. 19, 1856, as follows.—
Ep.]:

‘In the sketch, which I now return, B is more worn than
A, and A than C. B may therefore be m. 1, and G, p. 4, or
the last premolar, A being the second true molar, m. 2.
From the wearing it would seem as if A and B had had
more connecting matter between the two ridges than C has,
and that C was a more simple or decidedly two-ridged tooth.

The bend of the first ridge appears to be tapiroid, or lophio- —

dontoid.’

[Among Dr. Falconer’s notes there is a further memo-
randum respecting the same specimen, dated 9th Novem-
ber, 1857, which is appended. Dr. Henry Walker had
formerly been a ‘colleague of Dr. Falconer’s as Professor in

1 Tt does not appear from Dr. Fal- | men, which is here described from a
coner’s notes that he had seen the speci- | drawing made by Col. Baker,—[ Ep. ]

ae

DESCRIPTION OF PLATE XXXIV.

Figs. 1 and 2. Show, in plan and profile, the teeth from Attock of the
Tapiroid animal described at page 416 under the designation of
Antoletherium. The figures have been copied from the pencil
drawings by Col. Baker referred to in the text, and are three-
fourths of the natural size.

Figs. 8, 4, and 5. Three different views of the fossil vertebra from
Jubbulpoor, described at page 418, one-third of the natural size.

Figs. 6 and 7. Two views, in plan and profile, of the antepenultimate
upper left molar of Mastodon (Trilophodon) Pandionis. The
specimen thus identified in Dr. Falconer’s handwriting was found
in his private collection, and has been deposited in the Pale-
ontological Gallery of the British Museum. The drawing is
about two-fifths of the natural size. (See page 124.)

VOL, I.

Vol. I. Plate 34.

J. Dirks lth -
¥ 6,7. ‘Mastodon Pandionis.

1,2. Antoletherrum. 3,45, Fossil Vertebra from Jubbulpoar, WWestamp.

¥ Pr” 4 . ne @ :
é % ‘ rs

el

FOSSIL REMAINS FROM ATTOCK. Al?

the Medical College of Calcutta, and had assisted him in
editing the Catalogue of Fossils in the Museum of the
Asiatic Society. He died in 1857.—Eb.]

‘Found among Walker’s papers two rough sketches—side
and plan. The one is evidently of a form closely allied to
Dinotherium. Oldham’s specimens comprise two teeth,
both apparently three-ridged, considered by Walker between
Dinotheriwm and Diprotedon, the other regarded by him as
of Mastodon latidens, but evidently the same specimen as
Baker’s drawing of the Attock tapiroid form from Lieut.
Garnett. Walker’s sketch throws a new light on it. The
tooth (A), as in Baker’s drawing, shows two ridges, both worn
and connected by means of a bridge. The middle molar (B)
shows three ridges, all much worn and confluent into a com-
mon disc; the third molar (C) shows only a single ridge,
slightly worn, the rest of it being apparently broken off.
From Baker’s drawing I had formed the impression that the
last ridge of the middle molar (B) formed part of the tooth
(C). But Walker’s side-sketch draws the line of demarcation
very marked. It is very puzzling to say which is the anterior
molar and which the posterior. B is the most worn and
therefore the oldest tooth, and C is much less worn than A,
and therefore younger than it in appearance, i.e. later. The
distal (free) ridge of A appears to be more worn and to be
of wider transverse diameter than the prominent ridge (i.e.
that nearest the middle molar). Hence A is inferred to be
p-m. 4, or the last premolar; B to be the first true molar,
m. 1, three-ridged ; and C the anterior ridge of the m. 2, or
the penultimate true molar. This is the reverse of the view
first taken by me from Baker’s drawing, and from that which
was suggested to Owen by it.’

[Dr. Falconer subsequently wrote to Col. Baker in Calcutta
_ for further information respecting the specimen, who on
_ Jan. 13, 1858, replied to the following effect.—Eb. |
_ ‘Your first impressions regarding the subdivision of the
- fossil teeth, represented in my sketch, coincides with my own.

Tooth B, in my opinion, consists of two ridges and termi-
nates at #.

Tooth C extends from x to b”. The mark at y, in Walker’s
sketch, indicates his opinion that there existed on that line
_ a division between two teeth. I am quite certain that there
does not exist any external visible mark more distinct than
what is shown in my sketch.’

VOL. I. EE

418 FOSSIL REMAINS FROM JUBBULPOOR.

MEMORANDUM OF TWO REMARKABLE VERTEBRA, SENT BY
Dr. OLDHAM FROM JUBBULPOOR—SPILSBURY’S Brp.

29th August, 1862.

The larger vertebra consists of a compressed body, very
considerably compressed sidewise, and contracted in diameter
between the articular surfaces, both in the vertical and
transverse direction. The anterior articulation is elliptical
vertically in its outline, and the cup as deep as in the Cro-
codilia; the posterior articulating surface is of a correspond-
ing reversed form, i.e. very convex, flattened laterally, the
ereatest convexity being towards the middle or axis. The
inferior surface of the body at either end bears, immediately
behind the rim of the cup in front, and in front of the ball.
behind, a pair of surfaces for the articulation of a chevron bone,
ie. each chevron has been articulated to two adjoining
vertebree.

The spinous process, which is broken off, is flattened and
of considerable size towards the base; it is given off from
the body backwards at an angle of about 45°. Between it
and the body there is a semicircular niche about 1-2 inch
deep. From the anterior part or base of the spinous pro-
cess two articular apophyses are given off nearly horizontally,
or inclined upwards at a small angle, and diverge, but the
divergence is small. The articular surfaces are on the axial
side (inside).

It would appear that the next anterior vertebra passed
its spinous process between these articular surfaces; but no
marks of such articulation are seen in the spinous process of

the vertebra.
Dimensions.
Inches
Extreme length of body 54
Height in middle to Wolly between spinous and articulating
“apophy sis : . . e . . -
Height of anterior coneave end 0 5 :
Width of do. . -
Length of body from rim to base of ball : : : : 2
Height at base of ball behind c . A ‘i “ :
Transverse diameter of do. . . - . -
Height of body where constricted behind .
Greatest constriction of do. . . . . : . .
Length of articular process. . . . .
Do. from base of spinous process to Bey of do. : 5
Length of lamina, right side . 4 . . . : .
Vertical diameter of spinous process F : . : 5
Transverse diameter of do. . - : . .

Vertebral canal small and constricted; not a trace of a
suture remaining (See Plate XXXIV. figs. 3, 4, and 5).

The other vertebra, No. 2, is shorter and less perfect. The
spinous process is broken off at its base, and the articular

NwWDOWReE RES

SE RIDE tt ot

ooo

FOSSIL REMAINS FROM JUBBULPOOR. 419

processes, if any, are gone. The body is shorter and less
constricted; there are the same ball and socket ends, but
they are not so deep ; there are also the two pairs of articular
surfaces for chevron bones.

Dimensions.

Inches
Length of body . ‘ : ; ; : : : : 43
Do. between articular ends . “i ; : ; : erase
Vertical height of body in middle - ; c AE
Constriction of do. ; i . PF ; ; \ . 1:55

This vertebra is more square in form than the other, and
is much shorter.

EE2

420 FAUNA ANTIQUA SIVALENSIS.

XXVI. NOTE ON AN EXTRAORDINARY VARIETY
OF ELASTIC SANDSTONE.

BY H. FALCONER, M.D.

T nave lately had sent to me to look at, by Captain
McNaghten, of Kurnal, a specimen of rock which has sur-
prised me beyond measure. It is a slab of sandstone 14
inches long by 54 wide, and two inches thick, and looks like
along brick. It exactly in appearance resembles the building
sandstone used at Agra. It is flexible and elastic m every
direction! If you place it flat on a table, and press the
hand on one end and raise the other, you can bend it to a
certain extent, and see the undulations moving along to
the fixed end. If you seize it by both ends, one in each
hand, and make an action as if you intended breaking it,
you can see and feel it hard like a piece of whalebone, but
of course in an infinitely smaller degree, and the undulations
are observed propagated from end to end: If you tap it on
the side with the finger as you would a mussak of water, it
yields pretty much in the same fashion, propagates an un-
dulation, and instantly recovers its form. If you press it at
the sides it gets narrower, and if you pull at the ends it
elongates! but it always recovers its original form. Is there
any account on record of so extraordinary a sandstone ?
Should there not, I may send you some notes about it. It
is not known where the specimen came from.
BB.

1 This note was communicated to the | in the ‘Journal’ of the Society, vol. vi.-

Secretary of the Asiatic Society of | p. 240.—[Ip.]
Bengal in April 1837, and was published

eS a

DESCRIPTION OF PLATES. 42]

XXVIII. A DESCRIPTION OF THE PLATES IN
THE FAUNA ANTIQUA SIVALENSIS.

(Compiled by the Eprror, from Notes and Memoranda by H. Faxconer, M.D.)

[Tuts description has been mainly compiled from the following
sources :—1. Memoranda in Dr. Falconer’s note-books and papers ;
2. References to certain of the figures in his published memoirs on
Hlephant, Mastodon, &c.; 3. References to other figures in his
correspondence with scientific friends; and 4. Labels in his handwriting
on the specimens figured which are now in the British Museum.
Although the figures are drawn to scale, the actual measurements have,
as far as practicable, been introduced into the description of each
figure. It has been thought that by their means, the value of the
descriptions would be increased to those who have not an opportu-
nity of consulting the Plates, and that even to those who possess
the Plates the comparison of specimens would be facilitated. The
measurements are given in English inches, and in tenths of an inch.
The letters B.M. indicate that the specimen referred to is in the
British Museum.—Ep. |

Plates I., II., and III. are intended to represent, by careful copies of
nature, the modifications in structure and form exhibited by the molar
teeth of the Proboscidea. 'They show in vertical sections a series of
gradations, commencing with Dinotherium and Mastodon Ohioticus at one
extremity, and running through the other species to Hlephas primigenius,
in which the greatest deviation from the ordinary form of a grinding
tooth is met with.

Puate I.

Fig. 1.—Hlephas primigenius, or the true Mammoth: longitudinal
and vertical section of last upper molar, left side, from an English
specimen found near Kingsland, and formerly in the Museum of the
Geological Society. Shows the ‘ridge formula’ and the form and
relative proportions of the alternate layers of ivory. The section
closely resembles that of the corresponding tooth of the Indian
Elephant, but the ivory segments are even thinner, more vertical, and

more approximated. The disposition of the plates presents the ex-

treme degree of ‘pectination’ seen in the molars of any known

‘species of elephant.—B.M. (Reproduced in Plate V. fig. 3.)

Length, 11 in. No. of plates, 21. Depth of enamel at tenth plate, 6:2 in.
Length of space to 10 plates, 43 in.

Fig. 2 a.—Elephas Indicus. Vertical section of an upper penulti-

mate molar of the existing Indian Elephant. It is composed of

seventeen ridges, with a reduced talon splent behind, the anterior

422 FAUNA ANTIQUA SIVALENSIS.

talon being confluent with the first ridge. The anterior eight plates are
inclined forwards, and by the process of wear they are ground down, so
that the front part of the tooth is truncated obliquely before the
posterior lamella have come into use. The plates are very thin and
vertical, and the enamel is thin. The gradual attenuation of the plates,
successively exhibited from E. insignis to E. Hysudricus, is here carried
to excess, eighteen being comprised within the space occupied by about
nine in the equivalent tooth of the African species. The pectinated
arrangement contrasts strangely with the chevron-formed ridges of
E. insignis and the cuneiform plates of EH. planifrons. The mass of
ivory at the base of the tooth is much thinner than in the corres-
ponding molar of £. Hysudricus. (Reproduced in Plate V. fig. 2.)

Length of crown, 8°2 in. Space occupied by 10 plates, 44 in. Height at tenth
plate, 6 in,

Fig. 2 b.—Elephas Indicus. Vertical section of unusually large spe-
cimen of last lower molar of an Indian Elephant from Assam, in India
House collection. The entire length of the crown is about fifteen
inches, and it includes as many as twenty-seven ridges, of which the
anterior thirteen are more or less abraded. The first five or six ridges
incline a little forwards, while the posterior ridges incline so much in an
opposite direction, that the hindermost are nearly horizontal, producing
the flabelliform character that so readily distinguishes in most instances
the last from the penultimate lower molar. The same disposition and
proportions of the dental substances are observed as in the upper grinder.

Fig. 8 a.—Elephas Hysudricus, from the Sewalik hills. Vertical
section of penultimate upper molar, left side. The tooth is in the
middle stage of wear, eleven of the thirteen plates of which it is
composed having been in use, and the two anterior ridges being worn
out. The same vertical disposition of ivory, enamel, and cement is
presented as in the African Elephant, but the plates are thinner and
more vertical; the layer of enamel is proportionally thicker; and the
interspaces occupied by the cement are wider in general than the ivory
plates—B.M. (Reproduced in Plate V. fig. 1.)

Length, 7°7 in. Length of 10 plates, 5-75 in.

Fig. 3 b6.—Elephas Hysudricus. Vertical section of portion of last
molar of lower jaw, comprising about fifteen plates. 'The same general
character, in the disposition and relative proportion of the ivory,
enamel, and cement are exhibited as in the upper molar, bearing in
mind that the latter is a younger and consequently smaller tooth. The
layer of enamel, however, is thinner than in the upper molar. The
ivory segments curve back near their base, and the apices of the pos-
terior plates lean towards the front of the tooth, a disposition still more
marked in the existing Indian Elephant. The dark shade below the
ivory indicates a core of sandstone, occupying the place of the unossified
part of the pulp nucleus, and of the undeveloped fangs.—B.M.

Puate II.

Fig. 4 a.—Elephas Africanus. Vertical section of a penultimate
grinder, upper jaw, of the existing African Elephant, in the possession of
Mr. C. Stokes. It is composed of nine principal divisions and a subordi-
nate talon ridge, the four anterior of which are partly worn, the rest being
entire. The ivory segments consist of long narrow wedge-shaped plates,

PRPs rece mde 'y Ia Asie

DESCRIPTION OF PLATES. 423

the height of which is many times greater than the width of their base.

The interspaces are deep and filled up with copious cement. The enamel

and common basal mass of ivory are much less than in either L. insignis

or EL. planifrons, the latter being only suflicient to establish a common

connection between the bases of the segments, and a foundation for the

offset of the fangs, which are numerous. (Reproduced in PI. IV. fig. 3.)
Length, 8°7 in.

Fig. 4 b.—Elephas Africanus. Vertical section of penultimate molar
of lower jaw, belonging to Mr. C. Stokes. It is composed of nine
cuneiform plates. This tooth had been a long time in use, all the
plates, except the last being affected by wear. The anterior part of the
crown has been ground down to nearly one-third of its original height,
so that the enamel divisions between the two anterior ivory plates have
disappeared, and the latter are confluent into a common mass. The
section exhibits the same kind of wedge-shaped ivory plates, a similar
amount of cement in the interspaces, and an analogous thickness of
enamel as in LH. planifrons, fig. 5 b.

Length, 7:2 in.

Fig. 5 a.—Elephas planifrons, from the Sewalik hills. Vertical section
of penultimate upper molar, with nine ridges, the three anterior of
which alone have been in use, the two first being worn down to a
single disc of ivory. The ridges are seen to be much more elongated
vertically than in Z. insignis (fig. 6a), but to be considerably less so
than in the African Elephant. From the latter it also differs in the
enormous quantity of cement, fillmg up the valleys and enveloping the
ridges, and in the much greater thickness of the folded plates of enamel.
The enamel is reflected over the ridges of ivory, and down into the
hollows zig-zag wise, exactly as in E. insignis—B.M. (Reproduced in
Plate IV. fig. 2.)

Length, 8°7 in.

Fig. 5 b.—Elephas planifrons. Vertical section of portion of last
molar of lower jaw, with nine ridges, and presenting the same general
characters as fig. 5 a. The lower tooth, however, had been longer in use,
and all the ridges are more or less worn, except the two last.—B.M.

Fig. 6 a.—Elephas insignis, from the Sewalik hills. Vertical section
of last upper molar. The four anterior ridges are affected by wear ;
the six posterior ridges are entire, the fangs are fully developed, and
their mode of implantation in the jaw is distinctly shown. The white
mass in the centre represents the body of ivory, which is projected
upwards in ten angular lobes, terminating in a sharp edge. The height
of these lobes does not much exceed the width of their base, and closely
applied over them is a thick layer of enamel reflected up and down in
a continuous zig-zag plate. The interspaces of the five posterior ridges
of enamel are completely filled up by a mass of cement much exceeding
the enamel in thickness (vide Plate VI. fig. 7). This is the best illustra-
tion of the intermediate type of a proboscidean molar tooth, from which
those of the other species diverge in opposite directions. It belongs to
the Mastodon Elephantoides of Clift. The dark granulated shade,
below the portion of the ivory nucleus sustaining the five posterior
ridges indicates the hollow of their common fang, which in the fossil is
occupied by a core of sandstone—B.M. (Reproduced in Pl. IV. fig. 1.)

Length of tooth, 10°3 in,

424 FAUNA ANTIQUA SIVALENSIS.

Vig. 6 6.—Flephas insignis. Vertical section of anterior portion of
adult tooth of lower jaw. The two front ridges only have been touched
by wear. The ivory, enamel, and cement present the same characters
as in the upper molar, but the common curve of the crown is slightly
concave instead of convex. The posterior part of the basal portion of the
pulp nucleus has not completed the stage of calcification, its place being
occupied by a nest of calcareous crystals. The figure also shows two
ridges of the preceding molar, with their common fang implanted in
the lower jaw.—B.M.

Puate III.”

Fig. 7 a.—Elephas Ganesa, a fossil Indian species. Vertical section
of last upper molar. The crown consists of ten principal ridges, with a
subordinate talon ridge in front and behind. The anterior seven ridges
have their summits worn. A small portion is broken off at the anterior
end. ‘The disposition and relative proportions of the ivory, enamel,
and cement bear the closest resemblance to those of the corresponding
tooth of #. insignis, and the number of ridges agrees. In fact, there are
no good characters by which the teeth of these two species can be
satisfactorily distinguished, although the crania are so remarkably
different.—B.M. (Reproduced in Plate VI. fig. 1.)

Length of tooth, 9°25 in.

Fig. 7 b.—Elephas Ganesa. Vertical section of posterior molar of
lower jaw. A small portion of the anterior end of the crown has been
broken off, but the presence of-the anterior fang proves that the section
includes the whole length of the tooth, except the first ridge, the
posterior end being entire. It appears to have consisted of eight prin-
cipal ridges, with a talon ridge behind, and a subordinate ridge in front.
Five of the ridges have been in use, the anterior two being worn down
close to the common base of ivory ; the three last ridges are entire. It
bears a close resemblance to the corresponding inferior tooth of E.
insignis in the form of the ridges, thickness of enamel, and proportion
of cement.—B,M. ;

Fig. 8.— Mastodon latidens (Clift). Vertical section of two last
molars of upper jaw. The specimen of which the section was made
was formerly in the collection of the Geological Society, and is figured in
Clift’s memoir (Plate XX XVII. fig. 1). The last tooth shows five prin-
cipal ridges with a posterior talon ridge and a subordinate ridge in front.
The ridges are transverse and divided by a longitudinal cleft into two
pairs of principal points without intermediate mammille in the hollows,
The enamel is very thick and the cement is reduced to a thin layer,
only observable in the bottom of the hollows. The ivory lobes resemble
those of E. Ganesa, but are less elevated, with a broader base. The
anterior tooth had been a long time in use, and the ridges are nearly all
worn out; they were four in number. Mastodon latidens is the form
most nearly alhed to KH. Cliftii, and, through that species, to the true
Elephants.—B.M. (Reproduced in Plate VI. fig. 2.)

Length of last tooth, 5:5 in.

Fig. 9.—WMastodon Ohioticus. Vertical section of last upper molar. |
It consists of four principal ridges and a small talon lobe. The ridges
are transverse, terminating in a trenchant edge; the ivory segments are
in regular angular lobes; the layer of enamel is of uniform thickness, .

a

DESCRIPTION OF PLATES. 425

and the hollows between the ridges are very wide and open, being
almost rounded at the bottom. There is only an exceedingly thin crust
of cement, continued over the fangs in greater thickness. The common
plane of the grinding ridges of the crown is nearly horizontal. M.
Ohioticus constitutes the terminal link in the chain, and through
Dinotherium establishes a passage into the ordinary Pachydermata.—
B.M. (Reproduced in Plate VII. fig. 2.)

Fig. 10 a.—Mastodon Sivalensis, from the Sewalik hills. Vertical
section of last upper molar. The ridges are more complex in their
composition than in WM. latidens. The crown is bisected by a longitudinal
furrow, each division of the ridge being composed of a pair of contiguous
conical mammillez placed more or less alternately. The hollows are in
consequence interrupted. There are five principal ridges, with a subor-
dinate ridge in front, and a talon ridge behind. Eight divisions of the
ivory may be counted in the figure, the smaller segments arising from
the direction in which the section has been made through the alternate
mammille. The ridges are approximated, and the enamel bears a
large proportion to the conical lobe of ivory which it invests. The
cement is entirely wanting, except in the bottom of the clefts—B.M.
(Reproduced in Plate VII. fig. 1.)

Length of tooth, 7 in.

Fig. 10 b.—Mastodon Sivalensis. Vertical section of greater part
of last lower molar. The tooth differs from the corresponding upper
molar only in being complicated with an additional ridge.—B.M.

Fig. 11.—Dinotherium Indicum (Fale.), from Perim Island. Ver-
tical section of posterior ridge and talon of the penultimate lower
molar, left side. The internal structure exhibits the same agreement
with that of the European Dinotherium, as is indicated by the external
form. The only perceptible difference is, that the angle formed by the
ridge of the ivory is more acute, and the enamel thicker in the Indian
species. The centre is occupied by a rhomboidal core of arenaceous
matrix marking the form of the unossified pulp nucleus. This tooth
is described and figured in Dr. Falconer’s Memoir on Perim Island
Fossils (Vide page 396, and Plate VI. fig. 3.)

Fig. 12.—Dinotherium giganteum (Kaup), from Eppelsheim. Ver-
tical section of entire penultimate lower molar, consisting of two
transverse crenulated ridges, and a talon ridge, while in the equi-
valent molar of Mastodon Ohioticus there are three principal ridges.
Corresponding to the smaller number of divisions the ridges are more
widely separated, less elevated, and broader at their base, while the
interspaces are also wider and more open than in the North American
Mastodon. The layer of enamel is of similar thickness, and there is
no appreciable crust of cement. The correspondence is followed out
in the form of the subordinate heel ridge. D. Indicum, however, is
the species most nearly allied to MW. Ohioticus:—B.M. (Reproduced in
Plate VII. fig. 3.)

Puate IV.

_ Llephas Hysudricus. (Fale. and Caut.), from the Sewalik hills. Front
view of skull, one-fifth nat. size. This fine specimen was purchased
from Conductor Dawe.—B.M.

Length of the cranium from the protuberances of the occipital to the broken tip
of ieft incisive, 45° in. Length from broken occipital condyles to anterior border of

426 FAUNA ANTIQUA SIVALENSIS.

alveolus, 28: in. Vertical height of head, from broken condyles to the pyramidal
bulge of sinciput, 26-in. Vertical height from surface of occipital to the tip of the
nasals, 27°75 in. Extreme width of the head restored on left side, 38°5 in. Width
at narrowest part of forehead between zygomatic fosse, 10°5in, Width of naso-maxil-
lary fissure, 18-5in. Depth from tip of nasals to anterior margin of naso-maxillary
fissure, 3°5 in, Depth of rami of naso-maxillary fissure, 4" in. Width between middle
of the orbits, mesial, 26: in. Greatest width of zygomatic fossa, 12° in. Depth from
hollow of frontal to condyles, 20'in. Depth from posterior border alveolus to margin
of naso-maxillary fissure, 21°5 in. Length of alveolus of last grinder, 10°5in. Depth
of hollow of frontal below mesial plane, 4°5 in. Extreme width of alveolus, 4°75 in.
Width of incisive sheath in front of the alveolus, 18°5 in. Transverse diameter of the
left tusk, 7-5 in. Antero-post. of the left tusk, 7°75 in. Depth below mesial plane of the
occipital hollow, 8°5 in. Width of bottom of occipital hollow, 5°75 in. Depth of pos-
terior bulge of the cranium from the occipital bone to surface of zygomatic fossa,
15°5 in. Least width at back part of cranium behind the alveoli, 8:5 in. Depth from
posterior broken surface of condyle to the posterior border of the alveolus, 19° in.
Depth of infra-orbital foramen, 2°5 in. Transverse diameter of foramen, 1°75 in.
Length of infra-orbital canal, 6: in. Depth of the left orbit, 6°75 in. From
anterior margin auditory foramen to anterior border of the orbit, 20 in. Vertical
diameter auditory foramen, 1°5 in. Depth of the fossa between incisive sheaths at
the top of it, 65 in. Width across fossa, 3-in. Depth of the naso-maxillary vault,
12:in. Depth of skull from posterior end of socket to the orbit, 22°5 in.

PuaTE V.

Figs. 1, 2, 8, and 4.—Elephas Hysudricus. Four different views of
same skull as in Plate IV., } nat. size —B.M.

Puate VI.

Figs. 1 and 2.—Elephas Hysudricus. Perfect small head from the
Geol. Soc. Museum, with the second and third milk molars, and first
true molar in germ. The second milk molar on left side is much
worn. The infra-orbital hole is very iarge. The tusks are oval on
section, as shown in fig. 3 of another animal of same age. The palatal
bones are divergent in front. The tusks are also a little divergent,
and are very near in size those of the young Indian Elephant, but are
narrower in front and more convex. ‘The palate is not figured.—B.M.

Extreme length from occipital crest to broken incisor, 23°3in. Length from
occiput to tip of nasals,13-8in. Width of nasal opening, 7°1 in. Depth of nas. op. at
sides, 2°8in. From lower end of nasal opening to tip of incisives, 9°5in, Semi-
diameter from tip of nasals to left orbit, 6-°3in. Width of brow, 12°6in. Width
at contraction of incisive sheaths, 5‘6in. Extreme length of orbit, 3°5in. Width
at tips of incisive sheaths, 5'6in. Vertical diameter of left tusk, 1:8 in. Transverse
diameter of left tusk, 1-5 in. From outer margin of orbit to occiput, 16°8.in. Width
of brow at temporal contraction, 6°in. Length of second milk molar, 2-1in. Width
of second milk molar behind, 1‘6in. Number of plates, 5. Length of third milk
molar, 4°3 in. Width in front, 2*lin. No. of ridges 7, with a front and heel ridge.
Interval between second milk molars in front, 1:9in. Interval between third, 2°2 in.
From niche of palate to commencement of diasteme, 5-3in. Length of diastemal ridges
to tip, 63in. Interval between ridges at base, 1-4in. Expansion at tip, 2:9 in. .

Fig. 3.—Elephas Hysudricus, under surface of young skull. This
specimen agrees in age and characters with that shown in figs. 1 and 2,
except that the third milk molar has 8 principal ridges, with a front
and back heel, instead of 7 as in the other.—B.M.

Length of second milk molar, 2:2 in. Width of second milk molar, 1:6in. No.

of plates about 5. Length of third milk molar, 4-3in. Width, 2°in. Interval
between second teeth, 1:2in. Between third, 2-1 in.!

1 Notes by Dr. Faleoner of other speci-] 1. A small head with second and third
mens of £, Hysudricus, not figured, milk molars, of exactly the same age as.

TORN

DESCRIPTION OF PLATES.

427

Figs 4, 5, and 6.—Hlephas planifrons (Fale. and Caut.), from the

Sewalik hills.

third milk, and first or antepenultimate true molar.

Portion of cranium with palate containing premolar,

The left pre-

molar consists of three principal ridges and an indistinct front and

back ridge.
and aft.
untouched by wear.

Their direction is so oblique that they point nearly fore
This little tooth is nearly globular in form and is quite
The crown is composed of a number of tubercles

irregularly huddled together, somewhat in a botryoidal manner, and

presenting no distinct indication of transverse ridges.

A. hollow filled

with matrix is seen on the right side, where the corresponding tooth

had dropped out.

The third milk molar is very broad, all the six

ridges worn, enamel thick, The first true molar is entirely in germ.

—B.M.

Length of premolar, 1-2 in. Width, 1:1 in. Height of crown, ‘8 in. Length

of third milk molar, 4: in.
front ridge.
at fourth ridge, 3° in.

between third milk molars in front, 2°3 in.

2°8 in}

Fig. 7.—Elephas insignis.

the small head, Plate vi. fig. 3, and
Plate vii. fig. 1, if anything younger,
as only the first ridge is touched by wear.
Shows the plates of the second milk molar
better than any other.

Length of second left milk molar,
25in.; width, 1:6in.; has 5 distinct
ridges anda heel. Length of third milk
molar, 4:1 in.; width, 2:in.; has 7 main
ridges with a large front ridge and large
heel, or 9 good plates. Interval between

teeth in front, 1‘6in.; between last teeth

behind, 2°1 in.

2. Another imperfect head of young
E. Hysudricus, of same age as fig. 1 of
Plate vi. Has second and third milk
molars in use and first true molar in
germ. The second milk molar is well
worn, the third has the five first plates
worn.

Length of second milk molar, 2°3 in. ;’

width, 1:7 in. ; number of plates 5, and a
heel. Length of third milk molar, 4°6 in. ;
width, 2°3in.; number of plates 8, with
front and heel plates in addition. Height
of sixth plate, 2°6 in.

3. Fragment of a very large cranium
comparatively as regards the age of the
teeth. Contains the third milk molar
and first true molar. The third milk
molar is well worn, with 7 ridges and a
back and front ridge; the two front ridges
worn. The tooth is very broad, broader
even than the third milk molar of £.
planifrons (Plate vi. figs. 4 and 5);
the enamel is thin, and finely crimped.
The first true molar is equally remarkable

Width, 2:4 in,
Length of first true molar, 5°5 in.
Number of plates 7, with front and back ridges.

Number of ridges 6, with a heel and
Width, 2°7 in. Greatest height
Interval
Ditto at niche of palate behind,

Section of molar showing laminated
character of cement filling up the valleys.

In some sections as many

in being broad and short and in having
few ridges. The first four ridges are
touched by wear. It is proved to be L.
Hysudricus by the great size of the nasal
opening, and the downward direction of
the rami.

Length of third milk molar, left side,
3°8in.; width, 2°6in. Length of first
true molar, 6*in. ; width, 2°6 in.; number
of plates 8, with a large front ridge and
a yery large heel. Interval between the
front teeth on either side, 1:6 in.; be-
hind at the niche, 2°7in. Depth of
cranium from posterior surface of molar
to brow between the orbits, 13:8 in.
Width of brow between middle of orbits,
13°6in. Width of naso-maxillary open-
ing, 9'4in, Width of muzzle at sub-
orbital foramen, 10: in.

1 Another valuable specimen of palate
of . planifrons is of same age as that
in Pl. yi. figs. 4, 5, and 6, but is a little
further advanced and-~- belonged to a
larger animal. The first true molar is
an inch longer and is much broader and
higher. The points of the plates are
few, being about 6 to the fifth and sixth
plates. The apices of the plates are
somewhat incurved or bent forward.
The specimen consists of the back
portion of left side of palate. The last
ridge of the third milk molar is in front.

Length of first true molar, 6:5 in. ;
greatest width, 3°4in.; height at fourth
ridge, 3°5 in.; has 7 principal ridges with
front and back ridge,

428 FAUNA ANTIQUA SIVALENSIS.

as eleven distinct strata of this substance may be counted. The
section is a portion of the tooth represented in Plate II. fig. 6 a, com-
prising the sixth and seventh ridges, and drawn to natural size.

Puate VII.

Fig. 1.—lephas Hysudricus. Fragment of upper jaw containing
second and third milk molars. Age of individual about same as in
Pl. VL. fig. 3.—B.M.

Figs. 2 and 2 a.—EHlephas Hysudricus. Fourth tooth, or first true
molar, upper jaw, right side; 12 plates. Vertical Section in B.M.

Figs. 8 and 3 a.—Hlephas Hysudricus. Fragment of upper jaw con-

taining fifth tooth or penultimate true molar, 13 plates. Vertical

section in B.M.

Figs. 4 and 4 a.—Elephas Indicus (erroneously designated £.
Hysudricus in Plate). Last grinder of upper jaw. The anterior
plates are inclined forwards, and by the process of wear they are ground
down, so that the front part of the tooth is truncated obliquely, before
the posterior lamelle have come into use. The plane of detrition
makes a large angle with the unworn plane of the crown, and slopes
from the inside outwards. On the worn surface the digitated summits
of the anterior ridges are found ground down into circular rings of
enamel enclosing a pit of ivory.—B.M.

Fig 5.—Elephas Hysudricus. Fragment of lower jaw, left side, with
second milk molar, vertically divided, and showing 7 or 8 plates.
Specimen shows also remains of alveolus of first tooth_— B.M.

Fig. 6.—Elephas Hysudricus. Inferred to be lower jaw ? left side ?
with second milk molar. Large variety, with 9 plates—B.M.

Length, 3-4in. Width in front, 1-lin. Width behind, 1:8 in.

Figs. 7 and 7 a.—Elephas Hysudricus. Fragment of lower jaw, left

side, with second milk molar, showing 7 or 8 plates.—B.M.
Length, 3‘lin. Width, 1°5 in.

Fig. 8.—Elephas Hysudricus. Lower jaw, left side, with third milk
molar.—B.M.

Length, 55in. Width, 2:2in. No. of plates, 9.

Fig. 9.—Elephas Hysudricus. Fragment of lower jaw, with third
milk and first true molars. The former has 9 plates; the latter is im

erm.
Figs. 10 and 10 a—Elephas Hysudricus. Fragment of lower jaw,
with the first true molar, presenting 12 plates.

Figs. 11 and 11 a@.—Elephas Hysudricus. Fragment of lower jaw,
with penultimate ? true molar, presenting 12 plates.

Figs. 12 and 12 a.—Elephas Hysudricus. Fragment of lower jaw,
with last molar, entire, im situ. The tooth is more elongated, and
includes a greater number of divisions (17 or 18) than is usual in the

last inferior grinder of H. Hysudricus. The specimen is now cut

into sections.—B.M.

Puate VIII.

Fig. 1.—Elephas Hysudricus. Specimen of cranium in Mr. W. Ewer’s
collection. Shows the palate with the first and second true molars and
tusks on both sides; the first well worn, and partly ground away in

¥

}

DESCRIPTION OF PLATES. 429

front; the second has the first four ridges well worn. The molars
consist of ten ridges and a large heel ridge, eleven in all. The tusks of
the opposite sides do not correspond, the left being nearly circular and
the right oval. One large sub-orbital foramen.

Depth from back molar to the front at top of incisives, 17:5 in. Contraction of
muzzle at sub-orbital foramen, 13°8in. Breadth, outer surface maxillaries, 9-1 in.
Vertical diameter, left tusk, 5-7 in. Transverse diameter, 5-1in. Greatest diameter,
right tusk, 6-2in. Least, 4°3in. Interval between teeth in front, 22 in. Behind,
at niche of palate, 3-in. Length of anterior (first true) molar, 4-lin. Width, 2°7 in.
Number of ridges remaining, 5 and a heel. Length of left back molar, 8: in.
Width in front, 3-in. Number of ridges, 10 and a large heel.

Figs. 2 and 2 a.—Elephas planifrons (misnamed E. Hysudricus in
plate). Very perfect specimen of lower jaw. Has three mentary fora-
mina on the right side, only two on the left ; none on either side at the
symphysis. The beak is very deep and thick, and appears to have
terminated bluntly. The enamel is very thick. The teeth are certainly
the last of the lower jaw, with few points to the back ridges. The slope
of wear inclines very much from the outside inwards, the difference
being nearly ? inch, at the third ridge of the left side. The front fang
portion has dropped out. Nine ridges remain in the left tooth; on the
right side are the remains of ten or eleven. The teeth are very broad,
and there is considerable mesial expansion.—B.M.

Extreme length of right side, 24: in. Divergence of rami behind, 21-6 in. Height
to front of alveolus, right side, 8-3 in. Greatest thickness behind, 6:7 in. Length
of right molar, 8°8 in. Greatest width, 3-8in. Distance between the teeth in front,
2Sin. Divergence of teeth behind, 5:6 in.

Fig. 8.—Elephas Hysudricus. Lower jaw of small-sized adult. The
inside only of this specimen has been figured, and only the portion from
the last ridge backwards as a fragment. It is a very old jaw with the
last molar. The anterior part of the tooth had dropped out. The last
ten ridges remain, all of them worn. The enamel is thick, but very
much crimped, and the plates are close together. The condyle is
broad and very convex, and the long axis, instead of being transverse,
runs obliquely fore and aft. The edge connecting with coronoid
commences immediately below the condyle, instead of sloping down
with a narrow neck as in E. planifrons (See Plate XI. fig. 3).—B.M.

Extreme length of jaw, 19: in. Height of ramus to top of condyle, 17°7in. Trans-
verse diameter of condyle, 4-5in. Antero-posterior diameter, 2-Sin. Greatest
thickness of ramus behind, 5-7 in. Antero-posterior extent of ascending ramus,
96in, Height to alveolus, 6-2in. Length of remaining portion of molar, 9°5 in.
Width, 3°1 in.

Fig. 4 (and Plate XIII. A. fig. 7).—Elephas Hysudricus. Beautiful
specimen of entire lower jaw, with two molars. The number of plates
in the anterior molar is nine, with a front ridge and a smull heel, inner
side. Nine plates of the next following tooth are seen in germ. From
H. F.’s collection —B.M.

Extreme length, including beak, 16-6in. Extreme divergence of rami behind,
141in. Height to alveolus, 4-4in. Height of condyle, 11-2in. Antero-posterior
extent of ascending ramus, 7‘4in. Greatest thickness behind, 3‘8in. Length of
anterior right molar, 54in, Width, 2°3in. From inner side of symphysis to tip
of beak, 3:3 in.

Fig. 5.—Hlephas Hysudricus. Extremely old lower jaw, right side,
with the last tooth nearly worn out, and showing about five remaining

430 FAUNA ANTIQUA SIVALENSIS.

plates extremely distorted. Belonged to an individual of small size.
The figure is chiefly intended to show the distortion.—B.M.

PriaTe IX.

Elephas planifrons (Fale. and Caut.), from the Sewalik hills. Front
view of skull, one-third of natural size. The forehead of this species
is very flat; the naso-maxillary opening very small, and the occipital
fissure very low.—B.M.

PLaTe X.

Elephas planifrons. Four different views of same cranium as figured
in Plate IX. The last true molar is seen in germ and intact on the
right side, and well worn on the other, so that the corresponding tooth
on the right side of the lower jaw had probably been wanting. It has
eleven ridges and a heel. The pterygoids are very low.—B.M.

Extreme length of cranium from occiput to broken incisives, 265°in. Extreme
width of occiput, 21-7in. Height of occiput (condyles broken), 13°7in, From
middle of occipital notch to tip of nasals, 11:in. Transverse diameter of nasal
opening, 8°7 in. Vertical, 2°8in. Interval from posterior orbital process to margin
of nasal opening (partly broken), 8°3in. Estimated width at posterior orbital
processes, 27°8in. Greatest contraction between the temporals, 14-7in. From
occiput to anterior margin of orbits, 20°7 in. Width of muzzle at orbital
foramina, 12°5in. Depth from surface of molar to brow at contraction between
the temporals, 16°in. Antero-posterior diameter of orbit, 46in. Transverse
ditto, 4:5in. Length of right molar, 9°7in. Greatest height of crown plates,
unworn, 4*in, Width of crown in.worn tooth, 3°65 in.

Prave XT

Fig. 1.—Elephas planifrons. Fine specimen of old palate, with last
molar of either side. A section was made of the right molar, which
consisted of ten ridges, back heel inclusive. The section showed the
anterior fang complete; the enamel very thick; general expansion of
the plates ; points in the plates very few, fewer even than in LE. insignis.
Specimen in Mr. W. Ewer’s collection (See note, page 433, No. 1).

Length of last molar, left, 11‘in. Width in front, 45in. Width at seventh
ridge, 3:5in. Interval between the teeth in front, 2°6 in. Interval behind at niche
of palate, 5:4 in.

Fig. 2.—E. planifrons.—Superb specimen of lower jaw. It has two
mental foramina placed, as in fig. 8, much worn in front. The last true
molar is beautifully preserved on either side. They have thirteen
principal ridges, and a back heel and front ridge; enamel very thick ;
points few ; an intermediate mammilla, the detrition of which causes the
mesial expansion ; the tooth curves a good deal out. No long spout as
in EH. Africanus. From Sir Proby Cautley’s collection (See note, page
434, No. 2).—B.M. (Reproduced in Plate VI. of vol. ii.)

Extreme length of jaw, 19:2in.- Divergence of rami behind (outer surfaces),

19:in. Height to anterior, margin of alveolus, 8*in. Greatest thickness behind,
6:5 in. Interval between teeth in front, 3°6in. ‘Interval behind, 5-6in. Length
of right molar, 11°8in. Width of right molar in front, 3-in, Width behind, 2°8 in.

Fig. 3.—E. planifrons. Superb specimen of left half lower jaw,
entire. The coronoid crescent slopes downwards from the condyle.
Crown of the tooth very low. Seven last plates of tooth only remain; —
great thickness of enamel and abundant cement, and mesial expansion ;

DESCRIPTION OF PLATES, 431

enamel plates projecting. Proved to be EL. planifrons by the distance
between the plates, the very low crown, thick enamel, and two mental
foramina.—B.M.

Extreme length of jaw, 24:2in. Height of alveolar margin, 7-5in. Height of
ascending ramus to top of the condyle, 20-2in. Width of ascending ramus from
coronoid margin to posterior edge, 10°5in. Greatest thickness, 6*lin. Transverse
measurement of condyle, 4:°2in. Length of molar, 10°in. Width of molar at
middle, 3°6 in.

Fig. 4.—£. planifrons. Is a most remarkable fragment of the Iast
molar, upper jaw, right side, taken out of a palate in H. F.’s collection.
It is figured to show how diversified the species may be, and also the
dedalian line of flexure.

Length of fragment, 5:4 in. Width, 2:5 in.

Fig. 5.—Elephas planifrons. Enormous tooth-fragment, with very
thick enamel, low plates, and mesial expansion. This specimen is
twice figured (See Plate XVIII. A. fig. 2). It is the last molar, lower
jaw, right side.—B.M.

Length, 10°5 in. Greatest width, 4:2 in. Height of ninth plate, 3°5 in.
Number of ridges, 9.

Fig. 6.—Elephas planifrons. Lower jaw, left side, with first and
second (antepenultimate and penultimate) true molars: The first
tooth is much worn; shows about six plates; enamel transverse with
little crimping. The penultimate has nine ridges and a small heel, or
eight and a double heel; the two front ridges barely touched. The
ridges have few points, the fourth having only five. (Vide E. insignis,
Plate XVIII. fig. 7).—B.M.

Length of front molar, 6-in. Greatest width, 2°8 in. Length of penultimate,
8-in. Width at fourth ridge, 2°8 in.

Fig. 7.—E. planifrons. Left. side of lower jaw, with last molar very
old. All the first half of the grinding ridges worn out. Very great
expansion of the plates. Three mental foramina outside.—B.M.

Length of molar, 10°2 in. Greatest width, 4: in.

Fig. 8.—E. planifrons. A magnificent typical specimen,. consisting
of a fragment of the lower jaw with whole length of penultimate true
molar; the anterior fang exposed; the three first ridges on this fang
gone by wear; eight other ridges, making eleven ridges and a heel;
enamel very thick; plates wide apart; much cement; few points;
three mental foramina.—B.M.

Height of jaw to alveolar margin, outer side, 86 in. Greatest thickness,
64 in. Length of molar, 121 in. Width at fourth ridge, 3-6 in. Greatest
width, 3-8 in.

Fig. 9.—Elephas planifrons. A fragment. of last. lower molar, left
side, intended to show the large digitations and few points of the
species. It has the three last ridges and a heel; points very distinct,
and enamel very thick; ridges very low; resembles Plate XVIII. A.
fig, 1.—B.M. ,

Fig. 10.—£. planifrons. Lower (upper in MS.) jaw, left side, with
penultimate and last molar. The whole of the penultimate much worn ;
the two front ridges worn out; has eight ridges and a very small heel.
The tooth is very broad for its length; has a great abundance of
cement; the enamel is very thick; figured for the remarkable fact of

432 FAUNA ANTIQUA SIVALENSIS.

there being no crimping whatever, only a little flexuosity, and no
mesial expansion. Last tooth is quite untouched by wear.—B.M.
Length of penultimate, 7: in. Width of ditto behind, 3:7 in. Length of first
five ridges of last tooth, 56 in. Width at third plate, 3:7 in. Height of fifth
plate unworn, 3°8 in.

Puate XII.

Figs. 1 and 1 a.—Elephas planifrons. Antepenultimate milk molar
of upper jaw, with four ridges; drawn of natural size. Fig. 1 6 shows
a vertical section of same tooth.

Fig. 2.—E. planifrons. Section of second milk molar, upper jaw, right
side. It has six main ridges, and back talon and front heel. The four
front ridges are touched by wear; great quantity of cement and thick
enamel, Resembles the third milk molar in Plate VI. figs. 4 and 5.—B.M.

Length, 3:8 in. Greatest width, 2°3 in.

Fig. 3.—(None in the plate).

Figs. 4 and 4 a.—E. planifrons. First true molar, upper jaw, much
worn, showing five ridges and a heel remaining ; two ridges probably
gone.—B.M.

Length, 5-2 in. Width, 2°8 in.

Fig. 5.—E. planifrons. Fragment of upper jaw with first and penul-
timate true molar. The first molar is that numbered as fig. 4. The
penultimate has eight ridges and a front heel. The specimen is broken
behind, but the artist in the drawing has repaired the eighth ridge
and added a little. Another first true molar in Plate VI. fig. 5 (See
also note, page 427).

Length of penultimate, 8° in. Width, 3: in.

Fig. 5 a.—E. planifrons. Is a distinct specimen from fig. 5. It is a
most valuable palate specimen, showing entire the penultimate or second
true molar on either side. On the left side the cavity for the last molar
is seen. The penultimate has eight distinct ridges and a front and
back heel; all the ridges are more or less worn; the points are few and
large, and the enamel thick. Has all the characteristic marks of E.
planifrons. Specimen in Mr. W. Ewer’s collection.

Length of left molar, 7°5 in. Width in front at third plate, 3-2 in. Width at
last or eighth ridge, 2°6 in. Interval between teeth in front, 3-2 in. Inter-
val behind, 5:4 in.

Figs. 6 and 6 a.—E. planifrons. Penultimate true molar in stu in
upper jaw ofa large animal. Tooth has eight main ridges.

Figs. 7 and 7 a.—E. planifrons. This is a beautiful little spe-
cimen of right side of lower jaw, containing the second milk molar.
It shows a very small front splent, with six main ridges and a small
heel limited to the inner two-thirds of the width of the last plate. Has
exactly the same number of ridges as the corresponding tooth in young
African Elephant (six main plates), but is a larger and broader tooth.
The first three plates are worn. The specimen also shows at 0 the
fang-holes of the first milk molar.—B.M.

Length, 6:2. Height to alveolar margin in front, 3°2 in. Ditto behind, 2°7 ins

Greatest thickness, 2°6 in. Length of second milk molar, 2-4 in. Greatest width — 4

of crown at fourth ridge, 1-4 in.

Figs. 8 and 8 a. Llephas planifrons.—Left side of lower jaw.

This is a superb specimen. It displays three teeth in situ, viz. in the

i te! Ae ope,

DESCRIPTION OF PLATES. 433:

posterior extremity the last milk molar; in front of it the penultimate

milk molar (4), nearly worn out, and emerging from below the latter a

small vertically succeeding premolar (c). The third or last milk molar

has seven main ridges, with a double front heel and a small splent
behind ; the four front ridges are worn; it is broader behind than in
front; the reverse in the upper.—B.M.

Length of last molar, 4-4 in. Width at second ridge, 1:8 in. Width behind
(greatest), 2°4 in. Length of small premolar, 1: in. Width behind, -8 in.; does
‘not show the ridges.

Fig. 9.—Elephas planifrons. Shows at c the penultimate premolar.
It is considerably smaller in all its dimensions than the antepenulti-
mate milk molar (fig. 1 a), drawn to the same scale. It is of a roundish
form, and shows no distinct indication of ridge-divisions. It was,
therefore, of small importance, functionally, in the economy of the
species.

Figs. 10 and 10 a.—Elephas planifrons. This is an invaluable
specimen. Proved by its size and development to be the first true
molar, lower jaw, left. Shows seven main ridges and a small ridge in
front; no heel behind, or only a very small one; is broader behind
than in front; the five first ridges are worn; enamel very thick with

_ mesial expansion ; few points to the plates; much cement. The most
interesting point is the third premolar (0) in front zn situ. The back

part of it only seen; it had not protruded through the jaw. Shows a
last plate of three points and a small heel.—B.M.

Length of fragment of jaw, 8° in. Height to alveolar margin, 5:9 in.; greatest
width, 4:4 in.; greatest height to crown behind, 68 in. Length of the first true

_ molar, 67 in.; width in front, 2°3 in.; greatest width behind at fourth-ridge,
_ 26 in. Length of premolar fragment, 1-in.; height of crown, 1°5 in.; width, 1-in.
‘Fig. 11.—#lephas planifrons. Last premolar (0) vertically divided
_ through the middle, the anterior portion being wanting. Although
partly emerged, it is still embedded in the alveolus and intact, while
_ the tooth behind it is well worn. It is of comparatively small size,
but presents distinct indications of two transverse ridges terminating in
the thick digitations characteristic of the species. ‘This figure refers to
the same specimen as fig. 10, but is drawn on a larger scale.

Figs. 12 and 12 a.—Elephas planifrons. This appears to be the last
true molar, lower jaw, right side; has ten main plates, with a front
_ plate and heel; is apparently of a small sized individual; has the
a) enamel straighter in the bend than usual; ridges low.—B.M.
Length, 10-in. Width in front, 3:5 in.; ditto behind, 2°9 in. Height of crown
# at seventh plate, 4: in.

Fig. 13.—Elephas planifrons. Lower jaw, left side, with last true
molar entire; crown not figured; very thick enamel plates reclined ;
_ considerable mesial expansion; points in the back plates few; in the
if front plates a good deal of crimping; shows about thirteen ridges and a
a heel, or possibly fourteen. Resembles very much an untigured speci-~
~ men in H. F.’s collection (See note, page 434, No. 3).1—B.M.
Length of molar, 12-7 in.; width, 3°6 in. ; height at tenth ridge. 45 in.

1 The following notes refer to unfigured | molar, upper jaw, has not been figured.
specimens of Elephas planifrons :— Its measurements are: extreme length,
1.—The most characteristic specimen | 1l:in.; width in front at second ridge,
of this species consisting of the last | 3°7 in.; ditto at eighth ridge, base, 3°8

VOL. I. FF

434

Fig. 13 a.—Elephas Hysudricus.
Is misnamed Z. planifrons on plate).

FAUNA ANTIQUA SIVALENSIS.

(Has no connection with fig. 13.
It is the last molar, right side,

lower jaw. Has a peculiar slew or twist in the wear, in front from the
inside out, and ‘behind from the outside inwards. Is an enormous tooth.
Shows the anterior fang in section, only one or two plates gone;

enamel very thick and plaited; mesial expansion.

a heel.—B.M.

Has ten plates and

Length of tooth, 11-3 in.; width in front, 3-9 in. ; in middle, 4-4 in. Height to

alveolar margin, 9° in.
thickness of Jaw, 6°3 in.

Height of jaw to crown of molar behind, 9° in. Greatest

in.; ditto at eighth ridge, near apex,
2-4 in.; height at eighth ridge, measured
from reflection of euamel plates below,
4°8 in. Number of principal ridges
twelve, with a front ridge and heel.
This tooth resembles very much in wear
Mr. Ewer’s specimen, Pl. xi. fig. 1,
and H. F.’s specimen, Pl. xii. fig. 5.
Of the 12 plates composing it, the first
eight are touched by wear. The front
subordinate ridge is joined on by a neck
or reduplication of enamel to the first
principal ridge. The enamel is very
thick. There is a good deal of crimp-
ing in the first three ridges, but no great
amount of mesial expansion. The points
are few in number, there being only six
which are worn into round rings to the
seventh ridge. A comparison of this tooth
with the last upper molar of Hlephas

Hysudricus is as follows :
E. planifrons EB. Hysudricus

Inches Inches
Extreme length of

last upper molar 11:0 . 11-0
Width in front at

2nd ridge 4 Sr . 3:4
Width at 8th ridge 2-4 3°3
Heightat8thridge 48 5-4

2.—A superb specimen of the lower
jaw in H. Fs collection. The specimen
has the left molar entire; of the right,
only the first eight ridges remain; all the
ridges on to the heel are worn. Enamel
very thick with beautifully marked me-
sial expansion, forming a sharp loop.
The back loop of one plate nearly in
contact with the front lamina of the next
ridge! Is a most beautiful and charac-
teristic specimen. The beak, although
broken, projects sufficiently to prevent
the ramus from resting on its lower sur-
face. The diastemal ridges are not
raised as in EL. antiquus (EL. meridionalis
in note of date about 1846, see note,
p. 448), but form a broad flat beak (not

sharp and narrow, as in E. insignis)
which projects dcwnwards as in the
African Elephant, although it is more
abruptly bent down, shorter and flatter,
something as in Z. primigenius. In this
respect the specimen is more perfect
than Pl. xi. fig. 2. The molars are
nearly parallel in front, and diverge
afterwards. There are three outer men-
tary foramina, and one on the inside.
The backmost foramen begins below the
front fang of the molar; the two others
are on the same sloping line in front.

Extreme length, 21°6 in. Height to
alveolus, 7°5 in. Greatest width, 7°7 in.
Interval between teeth in front, 3°8 in.
Width of diasteme, just below the be-
ginning of the symphysis, 3°d in.
Length of the left molar, 11:0 in. ; width
at third ridge, 2:9 in.; greatest width in
the middle, 3°3 in.; height of 8th plate,
3:0 in.; number of plates, 13 and a heel.

3.—Specimen of the entire last lower
molar, right side, contained in a mutilated
lower jaw. This magnificent specimen
shows the entire length of the tooth, and
a small portion of the penultimate in
front of it. The seven anterior ridges
are touched by wear. The enamel is
very thick with a mesial expansion and
somewhat crimped. The sixth ridge
shows six annular dises; the seventh
only five points. The tooth is very
broad ; much cement; the last plate or
heel is an oblique splent of only three
or four irregularly placed points. The
fang projects behind it. The tooth in
its direction curves much ‘outwards, and
is very nearly of the same width from
back to front.

Extreme length of last molar, 11°8 in.
Extreme width at sixth ridge, 3°9 in.
Height at eleventh plate, 4°6 in. Number
of principal ridges 14, and a small heel
of three points,

DESCRIPTION OF PLATES. — 435

DATE NCeee At

Elephas Namadicus (Fale. and Caut.) From the valley of the
Nerbudda. Probably a female, from small size of tusks. This
specimen was presented to the Museum of the United Service In-
stitution by Major Orlando Felix, and was received by him with other
Nerbudda specimens, from Lieut.-Colonel Ouseley. It was chiselled
out by Dr. Falconer, and determined by him to be a new species. In
a letter to Lieut.-Colonel Ousely, Dr. F. writes thus: ‘It is probably
the most perfect specimen of a fossil elephant’s cranium in Europe.
The species is especially interesting from the form of the cranium,
which is so grotesquely constructed that it looks the caricature of
an elephant’s head in a periwig. I have named the species H. Nama-
dicus, after the Nerbudda river, the Namadus of Ptolemy.’ There isa
very similar specimen in the Museum of the Asiatic Society of Bengal
(See antea, p. 115).—Specimen is now in B.M.

Extreme length from occipital bosses to molar surface, 29°8 in. Extreme width
of occiput, 30°0 in. From plane of occipital bosses to tip of nasals, 17-4 in. From
bottom of fossa of bosses to tip of nasals, 12-9 in. From bottom of bosses to
anterior margin of frontal bulge, 10°5 in. Depth of occipital fossa below the plane
of the bosses, 7°5 in. Length of crista galli-like plate, 11:2 in. Greatest con-
traction of brow between the temporals, 20-0 in. Projection of frontal bulge
above plane of forehead, 2°7 in. From anterior margin of orbit to occipital bosses,
24-8 in. Width of deep occipital fossa, 5-6 in. Transverse extent nasal opening,
15°0 in. Vertical height of nasal at sides, 5:3 in. Width of brow between
middle of orbits, 20°0 in. Width between tips of posterior orbital processes, 25:0
in. Antero-posterior diameter right orbit, 6°2 in. Width of base of muzzle at
contraction of the sub-orbitaries, 10°8 in. Interval between outside of maxillaries,
10:1 in. Vertical diameter sub-orbitary foramen, 3:1 in. From auditory foramen
to anterior margin of orbit, 14:0 in. Transverse diameter, right tusk, 2°9 in.
Vertical diameter of right tusk, 2°5 in. Depth of temp. fossa from ear (foramen
auditor.) to frontal margin of fossa at contraction, 13°3 in. Interval across
the occipital condyles, 9°7 in. Antero-posterior diam. left condyle, 4°6 in.
Transverse, 3°2 in. Antero-posterior diam. occipital foramen, 3:0 in. Transverse
diameter occipital foramen, 3°6 in. From anterior margin occipital hole to
posterior of surface palate, 13:9 in. Length of palate from niche to diasteme,
about 84 in. Width of base of skull at posterior end of zygoma, 26-2 in.
Width between ridge of pterygoids, 9:1 in. Height of pterygoid ala of sphenoid
above Vidian hole, 10°0 in. Length of articular surface for lower jaw, 5:1 in.
Across articular surface for lower jaw, 3°3 in. Length of remaining portion left
molar, 7°5 in. Width of remaining portion left molar, 3°7 in. N.B.—Twelve
plates in this extent. Width of palate in front (between molars), 2:8 in. Width
behind, 4:1 in.

1 Memorandum upon the Nerbudda Fossil Elephant, India House specimen.

Eleph Corse’s Elepha
Measurements of the sixth or last ee es Assam, large Hoan | Elephas
i a specimen y rimi e
Gece | Grmder dicus mecent poe Mae: dricus | Prmgen
Inches Inches Inches Inches Inches

Length of the eleven anterior
plates measured near the
base : : 3 5 81 57 5.6 aul 45

Width or thickness of the
ivory core, third plate, one
inch above the buse F 0°45 02 Fes 4-5 ose

Width or thickness of the
ivory core, fifth plate, one
inch above the base . 05 0-2 we 0°35 0:3

FF 2

436

FAUNA ANTIQUA SIVALENSIS.

Puate XII. B.

Figs. 1, 2, and 3.—Elephas Namadicus.
same skull as figured in Pl. XII. A.

Three different views of
The molars are less perfect than.

Memorandum upon the Nerbudda Fossil Elephant—continued.

Measurements of the sixth or last evn
True Grinder dicua
Inches
Width or thickness of the
ivory core, eighth plate,
one inch above the base 0.55
Width or thickness of the
ivory core, eleventh plate,
one inch aboye the base . 0:5
Average thickness of enamel ©
plates : : - 0:2
Height of enamel plate,
tenth ridge ; : 75
Greatest width of tooth, at
fourth plate, -

Corse’s
Assam, large a Elephas
recent specimen primigen.
Brit. Mus. dnicns
Inches Inches Inches Inches
0:2 Sic 0°35 oe
0:2 0:25
O01 0°15 AEE
74 ise 4:8 6:2
sn 35 35 Bor

The specimen is upper jaw, right side,
with the last grinder of which the eleyen
anterior plates remain ; there must have
been several more behind, from the great
height of the last plate. On comparing
the section with that of the EH. Hysu-
dricus and existing Indian species, it is
at once seen to differ from the former in
the extreme height of the plates, from
their slight amount of thinning upwards
and their nearly vertical direction. They
are as straight and vertical as in the
Mammoth. There is besides no loop
about the middle of the tooth plates, in
the enamel and comparatively thin
crusta. It is assuredly different from
the £. Hysudricus. As compared with
the existing species, the ivory is very
much thicker, with 2o curve towards the
apex; the enamel plates are very much
thicker also. The crown of the plates
resembles very much the last tooth of
Corse’s big head in the transverse di-
rection of the plate ribands, and in the
excessive amount of crimping or fine
plaiting of the enamel. From the
measurement given it will be seen,
however, that the enamel and ivory in
thickness indicate a wide difference,
which is further borne out by the yer-
ticality of the plates. Having seen
nothing among the existing teeth of a
range of difference at all approaching
this, I am compelled to consider the
species, as far as my present information
goes, as distinct. There is no possibility
of considering it a variety of E. Hysw-
dricus. I call it therefore provisionally
Elephas Namadicus (from the Greek
name of the Nerbudda ‘Namadus’). It

was found along with Hippopotamus,
Buffalo, &c., in the Nerbudda. There
must have been at least nine or ten
plates more, and it would rank in place
between the existing Indian elephant
and the 2. Hysudricus--
Thus, . primigenius,

E. Indicus,

E. Namadicus,

E. Hysudricus,

E. planifrons, §c.

The inferred distinctness of species
is further borne out by the excessive
width of palate in the other Nerbudda
specimen, seven inches behind. The
Perim species is probably the same.

N.B.--Prinsep, in the Journal of the
Asiat. Society of Bengal, vol. iii. p. 585,
describes and figures the lower jaw, one
side nearly entire, ofa fossil Elephant from
the Nerbudda, which he states to be so
like the existing Asiatic Elephant, judg-
ing from a comparison with a jaw in the
Caleutta Museum, that it was impos-
sible to distinguish them, although it
may be confidently distinguished from
the EL. primigenius. The figure shows
about fifteen or sixteen plates in wear,
and at least seven more behind, or
twenty-three to twenty-four in all. The
rami, however, as sketched by Prinsep,
are much more apart than in the Asiatic
species generally. Dimensions: length,
11} in., width in the middle, 33 in.;
transverse diameter of jaw at coronoid

”

dise, 6 in., and girth of jaw in front of

coronoid, 24 in.

This in all probability belongs to the
Elephas Namadicus, as also the prodigi-
ously large humerus at the India House.

DESCRIPTION OF PLATES. 437

in the corresponding specimen in the Museum of the Asiatic Society
of Bengal (See antea, p. 115).—B.M.

Fig. 4.—EHlephas Hysudricus. This fragment of skull, which is
probably female, and is but very slightly concave on the forehead,
yields very few good measurements. The figure is chiefly given for
the form. It has two small tusk sheaths; the tusks are broken off
near the base of the nucleus, and show only a thin plate. Only one
orbital foramen, very large. This specimen is very remarkable in
the molars having so few plates, only eight to the first true molar and
no heel.—B.M.

Length of penultimate molar, 5-2 in.; width, 2-3 in. The penultimate entirely
in germ shows eleven plates. Extreme length of the fragment, 28°5 in. Length
from occiput to tip of nasals, 18-0 in.; width of nasal opening, 12:5 in.; width of
brow across orbits, 216 in. Interval between the teeth in front, 2:1 in.; interval
at niche of palate, 3°8 in.; diameter of the right tusk, 2°5 in.

Puate XII. C.

Figs. 1 and 1 a.—Elephas Namadicus. A small fragment of lower
jaw, with three plates of what is probably the first true molar.—B.M.

Figs. 2 and 2 a.—Elephas Namadicus. Lower jaw, left side. This
specimen contains the third milk molar well worn and the first true
molar in germ.—B.M.

Extreme length, 12:5 in. Height at alveolus, 4:1. Thickness of jaw behind,
40 in. Length of anterior molar, 5:2 in.; width, 1:8 in. Number of plates
emaining, 7.

Figs. 3 and 3 a.— EH. Namadicus. Young lower jaw, right side, with
third milk molar, which has ten ridges and a heel. The crimped
character of Elephas antiquus' is well shown. A small vertebra is
attached to the ramus.— B.M.

Length of fragment, 11:0 in. Height at alveolus, 4:8 in. Thickness behind,
40 in. Length of third milk molar, 5:5 in.; width, 1-9 in.

Figs. 4 and 4 a.—K. Namadicus. Right lower jaw of adult, con-
taining last molar with twenty plates and a heel. The specimen shows
two mentary foramina. The broken coronoid portion of the ramus
shelves more out than in L. antiquus,? and the mentary foramina are
placed higher. Presented by C. Frazer, Esq.—B.M.

Extreme length of fragment, 23°6 in. Height at alveolus, 9:2 -in. Length of
the molar partly concealed and chiselled, 14:7; width, 3:11 in. Width of jaw
behind, 8-0 in.

Figs. 5 and 5 a.—E. Namadicus. Adult lower jaw of large size.
The specimen does not show the beak distinctly, and is more obtuse
there than in ZH. antiquus.? The number of outer mentary holes is
uncertain, as in H. antiquus* there is no inner hole. Presented by
C. Frazer, Esq., and described in Journ. Asiat. Soc.—B.M.

Extreme length, right side, 20:5 in. Expansion of rami, 24° in. Height of jaw at

beginning of alveolus, 10: in. Length of right molar, 14:in. Width, 3-7 in. Number
of plates remaining about 15. Greatest width of jaw, 81 in.

1* Hlephas meridionalis’ in original | land, is dwelt on in the memoir on
notes, written about 1846. The close | Elephants, invol.ii. See also note, p. 443.
resemblance of the £. Namadicus from 2 See last note.
the Nerbudda to the EZ. antiguus of the 3 See note 1.
oyster beds of Norfolk coast, in Eng- + See note 1.

438 FAUNA ANTIQUA SIVALENSIS.

Figs. 6 and 6 a.—Hlephas Hysudricus. Lower jaw, left side, with
first true molar which has ten plates, with a small heel and front ridge.
It is excessively like Plate VII. fig. 11, which is the penultimate of EF.
Hysudricus. The specimen is very remarkable as it is believed to have
come not from the Sewalik hills, but from the valley of the Nerbudda;
the mineral condition, however, is very hard, unlike the Nerbudda
specimens. Presented by C. Frazer, Esq.—B.M.

Length of molar, 8° in. Width, 2°6 in. Height at 8th plate, 4-2 in.

PuateE XII. D.

Figs. 1 and 1 a.—Elephas Namadicus. Beautiful specimen of lower
jaw, left side, containing the first true molar with thirteen ridges, and
a heel and front ridge, fifteen ridges in all._—B.M.

Extreme length, 16:2 in. Height at alveolus, 6°3 in. Length of molar, 7-4 in.
Width, 2-2 in.

Figs. 2 and 2 a.—E. Namadicus. Lower jaw, right side. This is
a little larger than the last specimen, and contains the first true molar
with about thirteen ridges. All these specimens show two highly-
placed mentary foramina. Part of the molar is concealed behind. A
portion of the third milk molar is seen in front.—B.M.

Extreme length, 19°5 in. Height at alveolus, 6-4 in. Length of molar, 7:3 in.
Width, 2°5 in. Width of jaw behind, 6:4 in.

Figs. 3 and 3 a.—E. Namadicus. Lower jaw, left side. This is a
most beautiful specimen, containing the second true molar. The
alveolus of the last tooth is shown behind. It contains about fifteen
plates, twelve to thirteen of which remain. ‘The whole length of the
tooth is present. It narrows very much in front, N.B.—Another
specimen of same jaw, opposite side, not figured, is exactly similar.—
B.M.

Extreme length, 13°6 in. Length of molar, 10°2 in. Width at middle, 3°3 in.

Figs. 4 and 4 a.—Elephas antiquus.! Lower jaw, left side, with first
true molar. This tooth is a beautiful specimen; shows twelve to
thirteen ridges, with front ridge and heel. It narrows excessively in
front and behind, like fig. 3of H. Namadicus! The crimping, &c., are
also exactly alike.—B.M.

Length of molar, 8: in. Width at middle, 2-6 in. Width in front, 13 in.

Figs. 5 and 5 a.—Elephas antiquus.? Last? molar of upper jaw,
right side, showing sixteen ridges and a small heel, much worn. Speci-
men belonging to the Canterbury Museum and labelled ‘2 Tooth of
Mammoth, Kent.’

Length, 10°8 in. Width, 3-3 in. Extreme height, 6 in.

Puate XIII.

Figs. 1, 1a, and 1 6.—Elephas Namadicus. Fragment of upper jaw,
right side, containing eleven plates of the 6th molar or last true
grinder. Fig. 1 a shows well the crimping of the enamel, and fig. 1

*Misnamed ‘ £. meridionalis’ on plate, | Museum. See also note, page 443.
but corrected by Dr. F. in copy of ‘ Fauna 2 See last note.
Antiqua Siyalensis,’ belonging to British

Cag

See

wee.

ie © oe.

DESCRIPTION OF PLATES. 439

shows a longitudinal vertical section of the tooth. Presented by C.
Frazer, Esq., to India House.—B.M.

Length, 7-9 in. Width behind, 2°5 in. Width in front, 4:2 in. Height anteriorly,
2:in. Height posteriorly, 8:4 in.

Figs. 2 and 2 a.—Elephas Namadicus. Palate with sixth or last
molar on both sides. Presented by C. Frazer, Esq.—B.M.

Length of fragment of grinding surface of molar of right side, 7°3 in. Greatest
breadth posteriorly, 3-in. Length of fragment of left side, 6-9 in. Width posteriorly
atfirst plate, 2°6in. Width at fifth plate, 3-8 in. Width of palate posteriorly, 5°83 in.
Width of palate anteriorly, 4°9 in. .

Figs. 3 and 3 a.—Elephas Namadicus. Fragment of upper true molar
with six plates; enamel crimped.

Length of fragment, 4:2 in. Width at second plate, 2°9 in. Width of posterior
plate, 2°in. Greatest height, 6°1 in.

PuaTE XIII. A.
Lower Jaws of Elephants viewed from above.

Fig. 1.—Elephas primigenius. Old. One mentary foramen inside
and three outside. Right true molar has thirteen plates, and measures
9-4 in. in length, and 3°6 in. in width.

Fig. 2.—E. primigenius. Young. Contains the antepenultimate or
first true molar on either side with twelve ridges, and a small heel and
front ridge, all of which, except the posterior talon, are affected by wear.
The plates are very fine. The tooth is not so broad relatively to the
length as in other specimens. The discs of wear form closely com-
pressed transverse bands, with attenuated plates of enamel. Some of
these plates differ from the ordinary type of the Mammoth in exhibiting
a certain amount of irregular crimping, but in no degree approaching
that seen in the Indian Elephant, this character concurring with a less
than ordinary width of crown. The penultimate true molar is seen in
germ behind. ‘There is one inner mentary foramen on either side.
A Rhine specimen from Dr. Kaup.

Extreme length of jaw, 16°8 in. Extreme expansion behind, 16: in. Height to
broken condyle, 125 in. Height to alveolus, 4°7 in. Thickness of jaw at middle,
48in. Length of molar, 5-4 in. Width, 2:2 in.

Fig. 3.—E. primigenius. English fossil specimen, with two last
true molars on either side. In the last left molar there are eighteen
plates in 7-7 inches. The jaw has a short beak, and one inner mentary
foramen on either side. In this, as in figs. 1 and 2, representing the
jaw at different ages, it is to be noted that the opposite lines of molars
are more or less convergent instead of being parallel, or nearly so, as
laid down by Cuvier.—B.M.

Extreme length of jaw, 23°6 in. Divergence of rami behind, 21-3 in. Height at
alveolus, 7:2 in. Greatest width of jaw, 63 in. Breadth of condyle, 10°3 in. Width
of last molar, 2°8 in. !

Fig. 4.—EHlephas antiquus.? Lower jaw with penultimate and last true
molars on either side. The last molar is very perfect and has seventeen
plates, of which the nine anterior ones only are worn. Only six plates of

1 Another lower jaw of E. primigenius, |in front and one inside. It is very
not figured, from Siberia, contains the | circular in outline in front.
last molar only, very much worn. It| Extreme length, 22°5 in. Height to
has thirteen plates in 8°9 in., is very | condyle, 18-7 in.
narrow, and has hardly any crimping.| 2” Misnamed ‘ E. meridionalis’ in Plate.
It-has one large outer mentary foramen | See notes pages 438 and 448.

440 FAUNA ANTIQUA SIVALENSIS.

.the penultimate are seen. No mentary foramen. Specimen in Geological
Society’s Museum. (Reproduced in Plate IX. of vol. ii.)

Extreme length of jaw, 26° in. Divergence of rami, 24:5 in. Height at alveolus,
9:2 in. Height to broken condyle, 16°3 in. Breadth of ascending ramus, 11°5 in.
Thickness of jaw, 7: in. Length of anterior molar, 3-9 in. Width, 3° in. Length of
last molar, 12° in. Width in front, 3-1 in. Number of plates 18.

Fig. 5.—Elephas antiquus.! Fragment of lower jaw with first true
molar on either side. The number of plates is twelve, with a heel.
There is no mentary foramen inside. This specimen formerly belonged
to the Earl of Aylesbury, but is now in B.M. (Vide Pl. XIV. A. fig. 7).

Length of right ramus, 14°5 in. Height, 5-1 in. Length of right molar, 6°7 in.
Width, 2°3 in.

Fig. 6.—Elephas Indicus. Existing Indian Elephant. Specimen from
Malacca in Museum of Asiatic Society. The jaw contains the last
molar on either side. The number of plates is twenty-two or twenty-
three, of which the eleven anterior are worn.

Extreme length of jaw, 19°3 in. Height of condyle, 18°4 in. Breadth of ascend-
ing ramus, 9°in. Thickness of jaw, 5:2 in. Length of molar, 10-4 in. Width, 2-9 in.

Fig. 7.—Elephas Hysudricus. Same specimen as figured in Pl. VIII.
fig 4. The description and measurements have been already given.—
B.M.

Fig. 8.—Elephas Africanus. Young lower jaw with two molars
(third milk and first true) on left side, and with first true molar and
alveolus of third milk molar, right side. The antepenultimate true
molar has seven ridges and a back and front talon. From Museum of
Asiatic Society.

PiaTe XIII. B.

Lower jaws of elephants in profile. The numbers correspond to the
eight specimens figured in P]. XIII. A. In these figures, which repre-
sent the groups Loxodon and Euelephas, the back of the symphysis is
seen to be a prolongation of the inferior margin into which the diastemal
ridges descend with great obliquity and also to be attenuated towards
the apex, to terminate in an obtuse point.

Fig. 1.—Elephas primigenius. Showing three outer mentary
foramina.

Fig. 2.—E. primigenius. With two outer mentary foramina on left
side. There were four on right side.

Fig. 3.—E. primigenius. Two outer mentary foramina.

Fig. 4.—E. antiquus. Two outer mentary foramina on left side.

Fig. 5.—£. antiquus.3 Three outer mentary foramina on left side.

Fig. 6.—#. Indicus. Five outer mentary foramina on left side.

Fig. 7.—#. Hysudricus. One small outer mentary on left side.
(See also Pl. VII. fig. 4.)

Fig. 8—E. Africanus. Three outer mentary foramina on left side.

Puiate XIV.
Figs. 1, 1a, and 1.—Elephas antiquus.4 Second milk molar, lower

jaw, left side, with six ridges and a front and back heel, from Grays, in

Essex.—B.M.
Length, 2°4 in. Width, 1-3 in.

1 Misnamed ‘ Elephas meridionalis’ in 3 See note 1.
Plate. See notes pages 438 and 443. 4 See note 1.
2 See last note.

ete:

DESCRIPTION OF PLATES. 441

Figs. 2, 2 a, and 2 b.—Elephas antiquus.1 Second upper right milk

molar, with six ridges and a heel, from Grays, in Hssex.—B.M.
Length, 2°4 in. Width, 1°3 in.

Figs. 3 and 3 a.—E. primigenius, probably (sic). Small milk molar:
9 ridges.

Fig. 4.—Elephas Africanus. Lower jaw of young animal with first,
or antepenultimate, and penultimate milk molars on both sides. The
right penultimate molar has six plates and a heel. This is the in-
teresting specimen figured and described by De Blainville.

Length of jaw, 9°7 in. Length of penultimate right milk molar, 2°38 in. Width,
9 in.

Figs. 5 and 5 a.—Elephas Africanus. Penultimate right lower
molar, with eight ridges and a heel. The ridges are broad, and the
fangs supporting the five posterior ridges are confiuent.

Length, 7:4 in. Width of crown, 2°4 in. Height of crown at eighth plate,
3-4 in.

Fig. 6.—Elephas priscus. Mutilated fragment of penultimate
lower molar, left side, from the brick-earth deposits of the valley of the
Thames. Shows only the entire discs of five partially worn ridges.
The outline of the discs corresponds very closely in form with those of
the posterior ridges of the larger specimen from Gray’s Thurrock,
shown in fig. 7. There is the same mesial angular expansion, and a
still greater tendency to the discs assuming a crescentic form. The
mutilated state of the specimen renders its identification somewhat
doubtful, but it is inferred to belong to EL. priscus.—No. 18,966 B.M.

Length, 5: in. Width, 3-in. Height behind, 2:8 in.

Figs. 7 and 7 a.—Elephas priscus. ‘This is a most interesting spe-
cimen from Gray’s Thurrock, purchased from Mr. Ball, of last lower
molar, left side, containing eight ridges, heel inclusive. It is slightly
concave on the outside, and convex on the inner side. A small portion
of the anterior end—two plates, perhaps—is broken off. The fangs of
the whole of the anterior part remain attached. The seven anterior
plates are worn down to wide discs; the hind talon has also begun to
wear. Is excessively like African Elephant in lozenge-shaped expan-
sion, and in the thickness of the enamel lamelle. The rhomb form is
not so sharply defined in the discs of wear, but the greatest expansion
is at the centre, where there is a tendency to an outlying loop (Vide
fig. 5). Besides the great expansion, this tooth differs from all true
i. antiquus specimens in the lowness of the crown ridges (Vide Pl.
XIV. A., passim, and especially fig. 8). If not a separate species, is a
very marked variety.—No. 39,370 B.M.

Fig. 7 6.—Longitudinal section of same molar of F. priscus. Shows
the closest relation to the existing African Elephant (Pl. II. fig. 4 6,) in
all that relates to the relative proportions of the alternate layers of
ivory, enamel, and cement, and in the cuneiform character of the
ridges.2_ (Reproduced in Plate VII. of vol. ii.)

Length of molar, 7°8 in. Width of crown at first plate, 2°35 in.; at second,
2'6 in.; at third, 2°75 in.; at fourth, 2°8 in. ; at fifth, 2°7 in.; at sixth, 2-45 in.;

at seventh, 1-8 in.; at talon, 13in. Height at seventh plate, 2°5 in. Expansion

?Misnamed Elephas meridionalis in | and FE. Africanus are considered in de-
Plate. See notes pages 438 and 443. tail in Dr. F.’s Memoir on Fossil Ele-
*The differences between E. priscus | phants, vol. ii.

442 FAUNA ANTIQUA SIVALENSIS.

of first plate at the middle, ‘8 in.; of second, “95 in.; of third plate, 85 in.; of
fourth, 75 in.; of fifth, -7in.; of sixth, -6 in.; of seventh, °d in.

Fig. 8.—Elephas planifrons. A very fine and characteristic speci-
men of the last upper molar, right side. Was found in contact with
the lower jaw specimen, fig. 9. Shows about eight or nine ridges and
a heel; the three first ridges worn out. The enamel is very thick,
with irregular lozenge-shaped expansion, and a loop in the middle. The
plates stand high in relief from the cement. The denticles are very
few and thick, the last ridge showing only three crown plates very low,
the Jast being but 24 inches high.—B.M.

Length of molar, 10: in. Width of crown, 3°5 in. Height behind, 2°5 in,

Fig. 9.—Elephas planifrons. Last lower molar, right side. Is very
valuable from having been found in contact with the upper molar
(fig. 8). Shows only eight ridges and a heel. The plates are very
thick and low. The sixth ridge shows only four denticles. In this
and the last figure the discs of wear form transverse bands, which are
broader, fewer in number, and wider apart than in the Indian Elephant ;
the bounding edges of enamel are sometimes nearly parallel; in other
cases they show a slight angular expansion, or throw out a salient loop
(or outlying) tubercle near the middle.—B.M.

Length of molar, 9°5 in. Width of crown, 3°5. Height of crown at sixth
ridge, 3° in.

Fig. 10.—Elephas planifrons. Fragment of lower jaw, with ante-
penultimate and penultimate milk teeth in sctu.'—B.M. ;

Puate XIV. A.
Molars of Elephas antiquus.?

Figs. 1 and 1 a.—Second upper milk molar, right side, from Kent,
with five plates. Specimen in Canterbury Museum.

Figs. 2 and 2 a.—Third upper milk molar, right side. Has about
ten plates. Specimen from Southwold, in Museum of Geological
Society.

Length, 5°5 in. Width, 2°3in. Height, 2:8 in.

Figs. 3 and 8 a.—Third milk molar, upper jaw, left side. Has ten
plates and a heel.—B.M.

Length, 6: in. Width, 2°2in. Height, 3-5 in.

Figs. 4 and 4 a.—First true molar, upper jaw, right side. Light
plates. From Southwold. Geol. Soc. Mus.

Length, 55 in. Width, 2°6 in. Extreme height, 4°8 in.

Figs. 5 and 5 a.—Last true mclar, upper jaw, right side. Has
fourteen plates and a heel, well crimped. From forest bed, Ostend,
Norfolk. Green collection.—No. 16,229 B.M.

Length, 10° in. Width, 3-4 in. Height, 6-6 in.
Figs. 6 and 6 a.—Third milk molar, lower jaw, right side, imperfect.

1 In the plate this specimen is desig- 2 Erroneously designated Elephas me-
nated E. Hysudricus, but this is doubt- | ridionalis on plate. The error is corrected
less an error, as the figure is referred to | in Dr. F.’s handwriting in the copy of
as that of molar of E. planifrons in | the ‘Fauna Antigq. Sival.’ belonging to
Dr. F’s Memoir on Fossil Elephants, | the British Museum. See also note,
vol. ii. page 443.

DESCRIPTION OF PLATES. 443
Has seven well crimped plates. Specimen is from Suffolk, and was
presented by Dr. Cooke to Geol. Soc. Mus. (No. 8,411).

Length, 4°2 in. Width, 2°1. in. Height, 3° in.

Figs. 7 and 7 a.—Third milk molar, lower jaw, right side, from
Kent; not quite perfect behind. Is narrow in front, broader behind,
and well crimped. Proves Lord Aylesbury’s specimen to be first true
molar (Vide Pl. XIII. A. fig. 5).

Length, 5-4 in. Width, 2°in. Height behind, 2°65 in.

Figs. 8 and 8 a.—First true molar, lower jaw, right side ; embedded
in fragment of jaw. Has twelve plates.—No. 18,967 B.M.
Length of molar, 8:3 in. Width, 2°5in. Height, 4-2 in,

Figs. 9 and 9 a.—Fragment of lower jaw, right side, with portions of
two last molars. The plates are enormously higher than in the EKlephas
priscus from Grays. The specimen is believed to be from Rome,
from Cardinal Gualteri’s collection B.M.

Length of anterior molar (imperfect), 4° in. Width, 3°in. Length of last molar,
7 in. Height of section, 7°3 in. Number of plates remaining, 9.

Figs. 10 and 10 a.—This is probably a second true molar, lower jaw,
right ride. Has twelve plates and a heel, five of the plates worn.—
No. 19,844 B.M.

Length, 10° in. Width, 25 in. Height, 6° in.

Figs. 11 and 11 a.—Last molar, lower jaw, left side, with fifteen to
sixteen plates and a heel. Portion in front gone. Specimen belonging
to Mr. Bowerbank. It was brought from Saffron Walden by Mr.
Sampson Hancock, and presented to the Mathematical Society at Dover,
in whose collection it remained until its dissolution. It then passed
into the hands of Mr. J. 8. Bowerbank, who has known the specimen
for about thirty years. (MS. Note on Plate, March 22nd, 1858.) Re-

produced in Plate IX. of vol. i.
Length, 12°3 in.

Width, 3: in.

Height, 5: in.

Figs. 12 and 12 a.—Last lower molar, right side, from Happisburgh.

Only the eleven posterior plates are present.

and bent.—B.M.
Length, 10°5 in.

Width, 3°4 in.

Plates very crimped

Height, 5:7 in.

Figs. 13 and 13 a.—Last lower molar, left side; fourteen plates

remaining, but some in front missing.

Via Appia, Rome.—B.M.

collection.
Length, 11: in.

From Cardinal Gualteri’s

Width, 3:4 in.

Puate XIV. B.!

“ Figs. 1 and 1 a.—Elephas meridionalis.
penultimate or second upper milk molar.

1 Great confusion has existed with
regard to this plate, which I hope to
have succeeded in now clearing up. In
the published plate (xiv. B.), figs. 1 to
9 and 11 to 16 are said to belong to HL.
meridionalis, and figs 10, 17, and 18, to
E. antiquus. While the plates of the
‘ Fauna’ were passing through the press
Dr, Falconer became satisfied that he

Plan- and side-view of the
It is a germ-specimen,

had committed a mistake in making the
fossil remains of the fiuviatile beds of
the Thames valley identical with the ex-
tinct Elephant of the Vald’Arno, instead
of the fossil remains of the ‘Crag.’ In
his memoir on British Fossil Elephants,
written ten years later (1857), but not
published until after his death, he says

| that although convinced that the ‘ Crag’

444 FAUNA ANTIQUA SIVALENSIS.

without fangs, and a good deal rolled. The crown is composed of six
principal ridges, besides front and back talons. It was compared with the
corresponding tooth of EH. (Loxodon) planifrons, which it resembles
very closely, but it has a broader crown. The dimensions are :—

Length, 2°6 in. Width of crown at first plate, 1°15 in. Width of crown behind,
1:-4in. Height of crown at fifth ridge, 1-56 in.

The corresponding tooth of LE. (Hueleph.) antiquus and of E. primi-
genius yields normally eight transverse plates. The precise origin of
the specimen is not recorded; but it is supposed to have belonged to
Mr. Samuel Woodward, and to have been derived from the Norfolk
coast. Norwich Museum, No. 11.

Figs. 2 and 2 a.—Elephas meridionalis. Another example of the
same tooth, a penultimate upper milk molar, right side, discovered in
the Norwich Crag at Easton, Suffolk, by Captain Alexander. It pre-
sents six ridges, well advanced in wear. Norwich Museum. The di-
mensions are :—

Length, 24 in. Width in front, 1:0 in. Width behind, 1°6 in.

Figs. 3 and 3 a.—Elephas meridionalis. Another well-worn penulti-
mate milk molar, probably of the lower (?) jaw, right side. It is of a
larger size than the others, but shows the same number of plates,
namely six, with talons. It is very broad in the crown relatively to
the length. The discs of the ridges are very wide, like the Italian
specimens. This molar belonged to the collection of Mr. Samuel
Woodward ; it is now in the Norwich Museum. It is heavy and
dark-coloured, and bears fresh patches of marine incrustation, and
may have come from the ‘ oyster-bed’ of Mundesley and Happisburgh.

Figs. 4 and 4 a.—Elephas meridionalis.
The crown is worn, and comprises eight
The ends and sides of the crown are partly injured. In

the lower jaw, left side.
ridges.

The last milk molar of

mineral condition it is black and heavy, but free from patches of

marine incrustation.

been procured from the coast (Norwich Museum, No. 10).

dimensions are :—

Length of crown, 3-9 in.
sixth ridge, 2°0 in.

Width of crown in front, 1:4 in.
Height of crown at seventh ridge, 21in.

It is supposed by Mr. Samuel Woodward to have

The

Width of crown at

molars were identical with those of £.
meridionalis, he had, in order to prevent
confusion, continued in the subsequent
plates the nomenclature adopted in the
earlier ones, intending to give a full ex-
planation of the whole in the letter-
press, and he concludes as follows :—‘I
beg leave to explain now that all the
plates bearing the name of EH. meridion-
alis in the “ Fauna Antiqua Sivalensis,”
including the outline figures of crania in
Plate xli., belong to E. antiquus, while
those that bear the latter name belong to
E.(Loxodon)meridionalis. In the descrip-
tions which follow they will be cited as
such.’—Quart. Journ. Geol. Soc., August,
1865, p. 281. According to this correc-
tion, all the figuresin Plate xiv. B., except
10, 17, and 18, should belong to E. anti-
quus, although mostly from the Crag

and some even from the Val d’Arno!
The correction, moreover, is incom-
patible with the description and identifi-
cation of every figure in Plate xiv. B.,
given in a subsequent part of the same
memoir, and extracted above, according
to which every figure in the plate, with
the single exception of fig. 16, belongs
to L. meridionalis. The fact is that the
descriptions in Dr. F.’s memoir on Ele-
phant were taken, in 1857, from a proof
copy of the plate, in which all the figures
were designated E.antiquus, but that in the
plate as published in 1847, Dr. F. had
actually corrected the designations of
most of the figures.

The proof copy has been deposited in
the Library of the Geological Depart-
ment of the British Museum.—[Eb. ]

cee

gery

DESCRIPTION OF PLATES. 445

The ‘ridge-formula’ in these specimens yields the same ciphers as
are found to hold in the Italian specimens; and they agree in the
other characters of a broad crown, with low ridges and thick plates of
enamel.

Figs..5 and 5 a.—Elephas meridionalis. A finely preserved entire
specimen of the antepenultimate or first true molar, lower jaw, left side,
composed of eight principal ridges, with front and back talons. The
six anterior ridges are worn. The discs of the first three ridges are
wide and open, but irregularly indented, with a tendency to mesial
expansion, and surrounded by margins of thick enamel, which is
vertically channelled externally, and slightly crimped; the posterior
ridges show the apices of six or seven digitations; the interspaces filled
with cement between the ridges are open, and the ridges well apart.
The dimensions are :—

Length of crown, 5°3in. Width in front, 1-6 in. Width behind, 2:3 in. Height
of the seventh plate, 2°5 in.

One of the distinctive characters of the species, namely, the low
height of the crown in reference to the breadth, is well exhibited.
The specimen is dark-coloured and heavy, from ferruginous infiltration.
It was discovered at Mundesley, and belonged to Mr. 8. Woodward
(Norwich Museum, No. 8).

Figs. 6 and 6 a.—Elephas meridionalis. Another left lower ante-
penultimate true molar of a larger individual, and more advanced in
wear. ‘The crown presents a front talon and eight ridges, all of them
worn; the discs are wide and open, and the vallecular interspaces
are also wide; the enamel edges thick, and in some of the plates
disposed to slight crimping, with irregular angular expansion. The
annular discs of the seventh ridge are of large size. This tooth bears
the large anterior fang. It is a very characteristic specimen of Z.
meridionalis. The dimensions are :—

Length of crown, 5:5 in. Width of crown at second ridge, 2°2 in. Width of
crown behind, 2:65 in. Height of crown at seventh ridge, barely worn, 2°0 in.

The specimen is hard, heavy, and dark-coloured, and is marked as
having come from Mundesley (Norwich Museum, No. 7).

Figs. 7 and 7 a.—Elephas meridionalis. A fragment comprising
the anterior two-thirds of the penultimate or second true molar of the
lower jaw, right side. It includes seven worn ridges. The discs of
wear are wide, and separated by broad bands of cement; the rings of
the digitations are large; the plates of enamel are thick, with angular
flexures and deep channelling on the outer surface, but free from
crimping. The specimen is black and heavy, and bears patches of
marine incrustation. The dimensions are :—

Extreme length, 5-2in. Width of crown at second ridge, 2:3in. Width of
crown at seventh ridge, 2-9 in.

No note was taken of the height of the last ridge. The specimen is
without fangs, and, although distinctly of H. meridionalis, the number
of ridges to the entire crown is not shown. This also belonged to Mr.
§. Woodward, and is now in the Norwich Museum (No. 13), It has
all the mineral appearance of the Mundesley and Happisburgh beds.

Figs. 8 and 8 a.—Elephas meridionalis. The anterior portion of a
lower right molar, comprising the remains of six well-worn ridges. It
is figured to show the angular flexures that are sometimes seen when

446 FAUNA ANTIQUA SIVALENSIS.

the plates are ground down low. The side view, fig. 8 a, exhibits the
thickness of the enamel. This specimen is too mutilated to fix its serial
position with confidence. It is heavy and dark from iron impregnation,
and corresponds with the fragments from Mundesley and Happisburgh
Norwich Museum, No. 18. :

Figs. 9 and 9 a.—Elephas meridionalis. The posterior two-thirds of
the crown of a lower molar of the right side. It is inferred to bea
penultimate, but without certainty, and may be the last true molar.
The crown shows six well-worn discs and a posterior talon; there are
no fangs ; the enamel is very thick, with large rings to the digitations ;
the discs are somewhat angularly expanded, and separated by wide
interspaces of cement. This is best shown by the side view, fig. 9 a.
From being worn low down, the plates exhibit a greater tendency to
crimping than is usual. The specimen is dark and heavy, and bears
fresh patches of marine incrustetion. It is one of Woodward’s specimens,
probably from the ‘ Oyster-ked’ (Norwich Museum, No. 14). The
dimensions are :—

Length, 5°3in. Width of crown at second ridge, 3-°2in. Width of crown at
fourth ridge, 3:1 in.

This is a characteristic fragment of EL. meridionalis.

Figs. 10 and 10 a.—Elephas meridionalis. A specimen in Dr.
Buckland’s collection from the Val d’Arno. Itis figured to demonstrate
how exactly the English specimens agree with the Italian form, as may
be seen by comparing figs. 8 and 9 with fig. 10. The fragment of
lower jaw, although mutilated, shows well the long symphysis, and the
gradual inclination of the diasteme into the beak.—B.M.

Figs. 11 and 11 a.—Elephas meridionalis. 'The posterior portion of
a last lower molar of the right side, including six discs of wear and the
back talon. The discs are broad, the interspaces of cement the same,
and the enamel plates are very thick, with deep external vertical
channelling, but without crimping. The specimen is black, heavy, and
bears patches of marine incrustation, indicative of its having been
procured from the ‘Oyster-bed.’ From Woodward’s collection (Norwich
Museum). The dimensions are :—

, Length, 56in. Width of crown in front, 2°8in. Width of crown behind,
‘Lin.

This is also a characteristic specimen of E. meridionalis.

Figs. 12 and 12 a.—Elephas meridionalis. A very notable fragment
of the posterior end of a last lower molar, comprising two discs of wear
and a talon. The crown is ground down low, the interspaces of cement
are very wide, and the annular discs of the digitations are so thick as
to approach the character of the worn ridges of some of the Stegodons.
The dimensions are :—

Length of the fragment, 2°7in. Width of crown, 4:2 in.

A solitary digitation is situated at the outer side of one of the valleys.
It bears the appearance of a Mundesley specimen.

Figs. 13 and 13 a.—Elephas meridionalis. A mutilated fragment of
a very old upper molar, formerly in the collection of the late Dr.
Mantell, and now in the collection of the British Museum (Old
Paleontol. Cat. No. 7,456), comprising the remains of ten discs of wear,
ground down nearly to their common base. The central discs exhibit
a certain amount of open crimping. The specimen is also remarkable

¥ PRPC OSN ‘i

DESCRIPTION OF PLATES. 447

for the breadth of the crown; it is understood to have been derived
from the ‘ Oyster-bed’ of Mundesley or Happisburgh. The dimensions
are :—

Length of crown, 8°2in. Width, 4°3 in.

I regard it as being of EL. meridionalis.—H.F. 1857.

Figs. 14 and 14 a.—-Elephas meridionalis. The crown of a fine last
upper molar, left side, of a very old animal, and in an advanced stage
of wear. There are nine ridges remaining, the first five of which are
ground down into transverse discs; the posterior four exhibit rings that
are not confluent. There isa talon behind enveloped by cement. In
front of the first remaining disc there is a broad depressed surface of
ivory, indicating the position of two or three worn-out discs in front.
The discs are expanded, with a slight tendency to a crescentic bend,
the cornua being bent forwards. The plates of enamel are very thick,
and deeply channelled exteriorly, so that there is a spurious appearance
of crimping on that surface ; but the edges in contact with the cores of
ivory are unplaited. The specimen in its mineral condition is black
and heavy. It is understood to have belonged to Woodward (Norwich
Museum, No. 10). The dimensions are :—

Length of crown, 9°2in. Width of crown at second remaining ridge, 3°6 in.

The antero-posterior convexity of the grinding surface determines
the tooth to be an upper molar. (Reproduced in Plate VIII. of vol. ii.)

Figs. 15 and 15 a.—Elephas meridionalis. A very remarkable
fragment of upper molar, of enormous width. It is worn down close to
the base, the grinding surface being somewhat convex from front to
rear. The remains of seven discs of wear are visible. They are
irregularly expanded, and the surrounding plates of enamel are thick
and deeply channelled on the outer surface, but with only a very slight
amount of crimping. The specimen is dark and heavy, and patched
over with fresh marine incrustations. From Happisburgh (Norwich
Museum, No. 13). ‘The dimensions are :—

Length of the fragment, 5-4in. Width of crown, 4°9 in. !

Figs. 16 and 16 a.—Elephas antiquus. ‘ The same plate, XIV. B., con-
tains a representation, fig. 16, of an entire upper molar, comprising from
sixteen to seventeen ridges within an extent of eleven inches. Only
three of the anterior ridges are worn, the rest being intact. I now
regard it as a molar of EH. (Huelephas) antiquus, and not of L.
meridionalis.’—H.F. 1857. (Norwich Museum.) Believed to be the
last true molar, upper jaw, right side, from its triangular form and the
way in which the ridges fall off in height very rapidly behind.

Height at fourth plate, 6°8in. Height at posterior ridge, 2°8in. Width of
crown in front, 3°5 in.

Figs. 17 and 17 a.—Elephas meridionalis. A Val d’Arno lower
molar of the same age, from Dr. Buckland’s collection in the Oxford
Museum, crown side. (Reproduced in Plate VII. of vol. ii.)

Length of crown, 10°in. Wadth of crown, 3°4in. Height of crown, 5: in.

Figs. 18 and 18 a.—Elephas meridionalis. ‘The finest detached
molar of this species that has come under my observation is a specimen
which was discovered in the ‘“‘ Mammalliferous Crag” on the Thorpe
Road, near Norwich, by Mr. Prestwich. The authority of so eminent
and accurate a geologist is a sufficient guarantee for the locality and

448 FAUNA ANTIQUA SIVALENSIS.

the formation. It is now lodged in the Museum at Norwich, and is the
specimen which first convinced me many years ago that the “Crag ”
yielded a species of Elephant entirely distinct from the Mammoth and
from E. antiquus. It is represented, one-third of the natural size, by
figs. 18 and 18 a of Pl. XIV. B., under the misnomer already explained,
of Elephas antiquus, in the “ Fauna Antiqua Sivalensis.” It is the last
true molar, lower jaw, right side, showing eleven principal ridges, an
anterior talon, and a back talon limited to a single thick digitation.
The first five ridges are slightly worn, the rest being intact. The fangs
are broken off, but the definition of the anterior large fang is distinctly
traceable. The cement over the surface generally has been decomposed
or denuded, and is replaced by a crust of Crag matrix, of a very rusty
appearance, filling the interspaces. The anterior talon thins off from
the outside inwards, and is considerably narrower than the first ridge,
of which the inner edge is broken. ‘The apices of the ridges, from the
second to the fifth inclusive, are all more or less fractured, and the
digitations present very thick enamel. The sixth, seventh, and eighth
ridges show each about four thick digitations ; the ninth and tenth from
four to five, converging; and the eleventh four digitations, the inner-
most of which is fractured. The definition of the base of the crown
behind is a little damaged, but nothing is wanting. The dimensions
are :—

Extreme length of crown, 11:25in. Width of crown in front, 3°3 in. Width at
fifth ridge, where the crown is broadest, 3°8in. Extreme height of ridge, 4:8 in.
Width of ninth ridge, 3°5in. Height of ninth ridge, 4°6 in.

‘From these dimensions it is apparent that, in a length of 11} inches,
there are eleven ridges, with talons, and the seven ridges from the
fourth to the tenth inclusive, measured along the inner wall of the
crown, yield a length of fully 7 inches, being an average of one plate
to an inch, and fully equal to the expansion of the ridges in the African
Elephant, or in EL. (Loxodon) planifrons. The terminal divisions of
the ridges form stout irregular cylinders, as thick as the little finger,
while in the Mammoth they are more slender and quill-shaped. The
digital lobes of the ridges in EL. meridionalis are so massive and distinct
that they have occasionally been figured and described as being of
Mastodon. —H.F. 1857. (Reproduced in Plate VIII. of vol. ii.)

PuaTeE XV.

Elephas insignis! (Fale. and Caut.). From the Sewalik hills.
This is the most remarkable of all the Indian fossil Elephants. The
cranium is as singular and grotesque in construction as that of the
Dinotheriwm giganteum.

The cranium is seen to differ remarkably from that of H. Ganesa
(Plates XXI. and XXII.) notwithstanding that the molars of the two
species agree so closely. That of H. insignis is flattened at the top,
elongated from side to side and singularly modified, so as to bear an
analogy to the cranium of Dinotherium giganteum, while that of EL.
Ganesa does not differ much from the ordinary type of the Elephants.
(See also Plates XLII., XLUL, XLIV., and XLV.)—Specimen is not in
B.M.

1 This is one of the forms included under Mastodon Elephantotdes by Clift. See
notel, p. 461.

DESCRIPTION OF PLATES. 449

Piate XVI.

Fig 1.—lephas insignis. Broken cranium, oblique antero-lateral
view. Left orbit, &c., gone. This head is very cubical in form, is old,
very concave in front and vertically ; teeth broken. Interval between
incisive sheaths deep. Notusks. A black specimen in Cautley’s collec-
tion.—B.M.

Extreme length from occiput to surface of molars, 26-in. Depth of brow from
occiput to upper margin of nasal opening, 5°5in. Antero-posterior diameter of orbit,
55in. Width of incisive sheaths at orbitary foramina, 11:4 in. From occipital
to brow between middle of orbits, 17: in.

Fig. 2.—Lateral view of same skull, as shown in fig. 4, showing
zygomatic arch.—B.M.

Fig. 3.—Posterior view of same skull, asin Plate XVII. figs. 1 and 2,
showing occiput, occipital foramen, and condyles.—B.M.

Fig. 4.—Palate view of skull with last upper molars, from a speci-
men in H. F.’s collection, the same as shown in fig. 2.—B.M.

Length of palate to commencement of diastemal ridges, 10°in. Length of
molar, 9°4in. Number of plates 10, and a heel; probably two plates dropped
out in front. Interval between molars in front, 1:9 in.; interval behind, 3-in,
Height of pterygoids to palate, 10°in, Length of articulating surfaces for condyles
of lower jaw, 54in.; width, 2°6in. Transverse diameter of right tusk, 3°65 in.
Length of zygomatic arch, 14°3in. Length of temporal fossa, 9*in.; width of
temporal fossa from pterygoids to maxillary surface, 7°8 in. Antero-posterior
diameter of orbits, 6°3 in.

Puate XVEI.

Figs. 1 and 2.—Elephas insignis. Anterior and lateral view of
cranium, same as represented in Plate XVI. fig. 3.—B.M.

Extreme length from occipital bulge to plane of molars, 23:in. From occipital
bulge to broken tips of incisives, 24:in. Extreme width of occiput, 25°5in. Width
of brow at post-orbitaries, 24-4in. Greatest contraction of brow between tem-
porals, 18-in. From occipital plane to tip of nasals, 7-9in. Width of naso-
maxillary opening, 11:°3in. Depth of opening at wings, 3:2in. Contraction of
incisives at orbitary foramen, 12°2in. Vertical height of orbit, 53in, From
occipital condyles to anterior end of palate, 22-in. From anterior margin of
occipital to posterior surface of palate, 12-in. Length of palate to commence-
ment of diasteme, 11°8in. Height from sphenoid to tip of pterygoids, 9:8 in.

_ Width of palate posteriorly between molars, 3-2in. Width of palate in front,

2°8in. From posterior surface of pterygoid to extremity of molar, 51in. Height
from occipital condyles to middle of brow between the orbits, 20°5in, Distance
between the outer margins of the occipital condyles, 7°2in. Vertical diameter
of occipital condyles, 2°8in. ‘Transverse diameter of occipital condyles (length
of one), 3°8in. Transverse diameter of occipital foramen,- 2°9in. | Vertical

diameter of occipital foramen, 2°in. From occiput to anterior margin of orbit,
156 in.

Figs. 3 and 4.—£. insignis. Anterior and lateral view of another

- eranium. Both zygomatic arches are missing, and the left side of the
cranium is deficient. Shows the great length of the incisive sheaths —
5. MM.

Pruate XVIII.

Fig. 1.—Elephas insignis. Very young skull.—B.M.

Fig. 2.—Elephas insignis. Young skull with milk dentition—B.M,

Fig. 3.—Elephas insignis. Skull of a middle- aged individual.—B.M,

Fig. 4.—Elephas insignis. Lower jaw with two (second and third)
true molars. The specimen comprises only the right side, with symphysis

VOL. I. GG

450 FAUNA ANTIQUA SIVALENSIS.

and beak. The left side has been restored in outline. The ascending
ramus is broken off.—B.M.

Figs. 5 and 5 a.—E. insignis. Lower jaw, left side, with ascending
ramus, but condyle broken off. Contains portion of last true molar.
Fig. 5 ais a view of inner surface, with large opening for nutritious
artery.—B.M.

Figs. 6 and 6 a.—E. insignis. Fragment of lower jaw, left side,
of smaller individual, including ascending ramus and condyle.—B.M.

Fig. 7.—Elephas planifrons. Fragment of lower jaw with portions
of two true molars.—B.M.

Puate XVIII. A.

Figs. 1 and 1 a.—Elephas planifrons. Last molar, lower (upper in
MS.) jaw, left side. An enormous specimen in H. F.’s collection.
Enamel very thick and denticles few. Hight plates. Is very like fig. 9
of Plate XI., and fig. 8 of Plate XIV.—B.M.

Length, 10-4 in.; width, 4°1in.; height, 3-2 in.

Figs. 2 and 2 a.—Z#. planifrons. Last lower molar, right side. A
very large specimen; enamel very thick; plates low; has a great fang
in front; true type of large E. planifr ons. This specimen is figured
in a former plate (XI. 5).——B.M.

Length, 10°5 in. Width, 4:2 in. Height, 3:2in. Height of plate where broken
behind, 3°8 in.

Figs. 3 and 3 a.—Flephas insignis. Lower jaw containing on either
side a molar with twelve ridges and a heel, the first six ridges worn ;
nine denticles to the seventh ridge; one large outer mentary foramen,
none inside.—B.M.

Length of jaw, 20-in. Height at alveolus, 9‘ in. Length of molar, 12-2 in,
Width of molar, 3°5 in.

Figs. 4 and 4 a.—E. insignis. Lower jaw, right side, with symphysis,
but ascending ramus broken off. Contains two molars im situ; four
ridges to first and seven to second; plates very deep. Hdge of diasteme
sharp; no mentary foramen shown.—B.M.

Extreme length of fragment of jaw, 18:2 in. Height at alveolus, 6°6 in. Greatest
width, 6.6 in, Width of anterior molar, 3-4 in. Of last molar, 3°65 in.

Figs. 5 and 5 a.—E. insignis. Lower jaw, left side, containing last
molar with eleven ridges and a heel of two points; the seven ridges in
front are worn; plenty of cement. One large mentary foramen.—B.M.

Extreme length of fragment, 19:in. Height at alveolus, 7-4 in. Greatest
width, 7-4 in. Length of molar, 11°3 in. Width in front, 4- in.

Figs. 6 and 6 a.—Mastodon Sivalensis. Lower jaw. Very indistinct
specimen ; teeth utterly worn out; enamel gone. One mentary fora-
men outside. Shows well the non-divergence of the rami behind.—B.M.

Height of jaw to alveolus, 8- in. Length of molar, 8:5 in. Interval between
molars in front, 2°8 in. Interval between molars behind, 2°8 in,

PLATE XIX.

Figs. 1 and 1 a.—Elephas insignis. Fragment of upper jaw containing
first (6) and second milk molars, in situ. The second milk molar shows
six ridges.—B.M.

Its length is 2°7 in., and greatest width, 1:6 in. It closely corresponds with a
specimen in the Asiatic Society of Bengal. See antea, p. 109.

DESCRIPTION OF PLATES, 451

Figs. 2 and 2 a.—Z. insignis. Third milk molar, upper jaw, n situ,
in same young skull as fig. 1. Length, 4:8 in—B.M.

Fig. 3.—E. insignis. Vertical section of third milk molar, in situ,
in a very young cranium, which also contains first and second milk
molars (fig. 1) and penultimate tusk on left side.—B.M.

Figs. 4 and 4 a.—E. insignis. Fragment of skull, showing palate,
with two molars on either side; the first (third milk) molars well worn.
The first true molar has seven ridges and a heel.—B.M.

Figs. 5 and 5 a.—E. insignis. Fragment of upper jaw with two
molars. Very similar to specimen of L. insignis (Pl. XXIV. fig. 6).
The molars are first and second true.-—B.M.

Figs. 6 and 6 a.—E. insignis. Portion of fine head with two back
molars. A small piece only of the front tooth remaining. The last
tooth has eleven ridges and a heel and a great abundance of cement;
the five front ridges are worn. The fossa between the tusk-sheaths is
very deep and narrow, as in the other specimens of this species, an
outward twist in the sheaths marking the curvature of the tusks.—B.M.

Length of the two teeth, 12°5in. Length of last tooth only, 11: in.; width, 3:8 in.
Interval between teeth in front, 1:2 in, Interval at niche behind, 4°7 in. Height
of pterygoids to palate, 10-4 in.

Fig. 7.—E. insignis. Transverse section of young tusk in situ, in
same cranium as figs. 1, 2, and 3.

PuLatTe XIX. A.

Figs. 1 and 1 a.—Elephas insignis. Fragment of upper jaw with
two molars; the front one much worn; the last has seven ridges.
Very doubtful whether they are the second and third milk molars, or
the third milk and first true molar; in all probability the latter.—B.M.

Length of front tooth, 3-6 in. Width, 18 in. Length of back tooth, 5°5 in,
Width, 2°7 in.

Figs. 2 and 2 a.—E. insignis. Portion of small head, showing palate
with two last molars on both sides. The front tooth has seven ridges,
all worn but the last; the last has eleven ridges and a front and back
heel, and its plates are very compressed, showing nine to ten denticles,

Length of penultimate, 6-6 in. Width, 3:lin. Length of last tooth, 11°3 in,
Width, 3:4 in.

Figs. 3 and 3 a.—E. insignis. Portion of very large skull. In this
splendid specimen some of the characteristic marks of the species are
shown, and especially the enormous height of the pterygoids, which are
10-2 inches from the Vidian hole to the summit, wrap over the maxil-
laries, and run up forming a very strong crested ridge into the base of
the orbit. Posteriorly, they form a flat disc-like surface, 3:8 inches
broad. There are also the indications of very deep trunk fossa. The
last left molar has nine ridges and a heel; a portion in front has dropped
out. The corresponding tooth on right side is very imperfect.—B.M.

Length of last molar, 10°7 in. Width at fifth ridge, 4: in. Interval between
teeth in front, 2-5 in. Interval between teeth behind, 4:5 in. Length of palate,
from niche to commencement of diasteme, 12°3 in. Height of pterygoids, 10-2 in.
Width of pterygoids outside, 9°5 in. Width of flattened surface of pterygoids,
3'8 in.

Figs. 4 and 4 a.—E. insignis. This, though mutilated, is a superb
and characteristic specimen of the skull. It shows the concavity of the

GG 2

452 FAUNA ANTIQUA SIVALENSIS.

brow, and the great depth of the trunk fossa. In this respect it
resembles Mastodon Sivalensis, but there is no great divergency
of the tusks, as in that species. The tusks are small and nearly
cylindrical. The front tooth, very fine, has eight main ridges and a
front and back heel; the back tooth entirely in germ, shows ten plates,
the hindmost reversed, and the ridges like compressed plates—B.M.

Length of anterior molar, left side, 9°7 in. Width at second ridge, 3:5 im.
Interval in front between the molars, 2:6 in. Interval between the molars behind,
3-6 in. Length of palate from niche to diasteme, 9°7 in. Depth of trunk fossa,
8-2 in. ! :

Figs. 5 and 5 a.—Elephas bombifrons. Portion of skull showing
palate with two teeth on either side. The front tooth has six ridges;
the back one nine ridges and a front and back heel, only one ridge
worn. ‘The teeth have a great quantity of cement and the enamel is
roughly fluted. These are characters of /. bombifrons rather than of
E. insignis, as the figure is designated in the plate. The back tooth is
very narrow behind, and so it is in H. bombifrons. The specimen
resembles Pl. XXIX. fig. 2.—B.M.

Length of front tooth, left side, 7-2 in. Width, 3-6 in. Length of back molar,
11l-lin. Width, 4:2 in.

PLATE XX.

Figs. 1 and 1 a.—Elephas insignis. Fragment of lower jaw with
two milk molars (second and third).—B.M.

Figs. 2 and 2a.—Z£. insignis. Fragment of lower jaw with milk
molar (third). The tooth has seven ridges and a front talon.—B.M.

Figs. 3 and 8 a.—E. insignis. Fragment of lower jaw with second
and third milk and first true molars. The first and last teeth are im-
perfect.—B.M.

Figs. 4 and 4 a.—E. insignis. Lower molar (second true?) with ten
ridges, five front ridges worn ; ten denticles in fourth ridge from back.
—B.M.

Figs. 5 and 5 a.—E. insignis. Fragment of lower jaw including
ascending ramus, with portion of back molar.—B.M.

Figs. 6 and 6 a.—Z. insignis. Fragment of lower jaw containing
second true molar with nine ridges and front and back heel; not at all
worn.—B.M.

Figs. 7 and 7 a.—E. insignis. Fragment of lower jaw containing
last true molar, with twelve or thirteen ridges, the five front ridges
worn.

Figs. 8 and 8 a.—ZE. insignis. Fragment of lower jaw with last true
molar containing about twelve ridges, of which only the three front
ridges are worn; the greater part of the tooth still in germ. (Re-
produced in Plate V. of vol. ii.)

Figs. 9 and 9 a.—Z. insignis. Fragment of lower jaw with (second?
true) molar; ten ridges, or nine ridges and a back talon. Fragment of
another tooth in front. Specimen in Geol. Soc. Museum. For further
description see Pl. XX. A. fig. 6.

Puate XX. A.

Figs. 1 and 1 a.—Elephas Ganesa (Fale. and Caut.). From the
Sewalik hills. Lower jaw with first and second true molars. Proved to

be so by fig. 2. A most remarkable jaw, very high in front and with

DESCRIPTION ‘OF PLATES. 453°

very divergent rami behind; diastemal edges very sharp; two outer
mentary foramina on right side. The front tooth much worn; has five
ridges and a heel; the last tooth has eight ridges. Nothing else like
this in the collection B.M.

Extreme length of jaw, 18°5 in. Height at alveolus, 8°5 in. Width of jaw in
front, 4-in. Width of jaw behind, 5-8 in.!' Length of front tooth 5:lin. Width,
27 in. Length of last tooth, 9°3 in. Width, 3-2 in.

Figs. 2 and 2 a.—Hlephas Ganesa. Portion of lower jaw with left
penultimate lower molar. The jaw in this specimen is also very high
and narrow in front, and low behind. The tooth has seven ridges
and a heel. Behind it is seen a portion of the last molar.—B.M.

Length of fragment, 14°5in. Height at alveolus, 8:°3in. Height behind, 5:8 in.
Width in front, 3-3in. Width behind, 6-6in. Length of molar,9:2in. Width at
back, 3° in.

Fig. 3.—Elephas insignis. Fine specimen of lower jaw, including
ascending ramus. Two outer mentary foramina. Second and third ?
true molars.

Figs. 4 and 4.a.—E. insignis. Fragment of anterior portion of lower
jaw. Is the only specimen that shows a beak entire to the tip. Is
very like the large specimen containing a molar with great number of
plates (Vide Plate XVIII. A. fig. 3).—B.M.

Figs. 5 and 5 a.—E. insignis. Fragment of lower jaw with last
molar, imperfect. The latter has eight plates remaining ; seven denticles
to the second (distinct) ridge.

Length of fragment, 16°in. Length of eight plates of last molar, 8-1 in. Width
in frent, 8-lin. Width behind, 3°64 in.

Fig. 6.—E. insignis? Fragment of lower jaw with second (third?)
true molar. The plates show seven to eight points. This is the
specimen described by Clift in the Geological Transactions. It is very
like H. insignis. It is also represented in Plate XX. figs. 6 and 6 a.—
Geol. Soc. Museum.

Length of fragment, 19:in. Height at alveolus, 6-6in. Length of molar, 12: in.
Width in front, 3:°2in. Width in the middle, 3-6in. Length of fragment of
anterior tooth, 4°2in. Width, 2°5 in.

Figs. 7 and 7 a.—E. insignis. Fragment of lower jaw, right side,
with first true molar nearly worn out, and six ridges of second molar.
The plates are very high and there is much cement. The teeth are
very broad in relation to the jaw, as compared with LH. Ganesa in fig. 1.
This is a beautiful specimen from Baker’s collection.—B.M.

Length, 14:7in. Height, 7°-7in. Width in front, 3-5in. Width behind, 6:7 in.

Length of front molar, 4°6in. Width, 2°7in. Length of back fragment, 5:3 in,
‘Width, 3 in.

Puate XXI.
Elephas Ganesa (Fale. and Caut.). From the Sewalik hills, in Colonel

Baker’s collection. Large skull, with fragment of left incisive in situ,
and corresponding fragment of right incisive detached. The incisive
alveoli are remarkably elongated, as in EH. primigenius. The plane of
the incisives is continuous with that of the frontal, but with a tendency
to obliquity forwards. The skull is very imperfect on right side. Pl.
XXI. gives a front view one-fifth of the natural size, and Pl. XXII. figs. 1
and 2, give a lateral and palate view of the same skull. Compare with
skull of H. insignis, Pl. XV.—B.M. The dimensions areas follows :—

454 FAUNA ANTIQUA SIVALENSIS.

Length of cranium from occipital protuberance to the end of incisive, left side
(four feet exactly), 48°in. From occipital condyles to left side, 39°in. From
occipital condyles to anterior border of molar alveolus, 25°5in. Vertical height
from condyles to sinciput, 24°5in. Diameter across the occipital condyles, 9°2 in.
Antero-posterior diameter of left condyle, 5-1in. Transverse diameter of right
condyle, 3°85 in. Transverse diameter of occipital foramen,3-0in. Antero-posterior
diameter of occipital foramen, 3°1in. From the surface occipital bulge plane to
anterior entire margin of naso-maxillary sinus, 19°lin. Semi-diameter of widest
part of occipital (making total of occiput, 29°2in.), 146in. Semi-diameter
(transverse) of naso-maxillary sinus (entire diameter restored, estimated 16: in.),
7°9in. Interval between naso-maxillary sinus and post-orbital margin of frontal,
4:55 in. Mesial width across forehead from post-orbital process to inner mar-
gin of incisive, left side, 13:1 in. (Width of forehead at this part restored,
26°25 in.) From tip of post-orbitary process to surface of occipital, 17°75 in.
From tip of incisives outside to post-orbitary process, 30°75in. Length of
incisive to margin of naso-maxillary sinus, 31:in. Depth of zygomatic fossa, 4°25
in. Estimated width of cranium between middle of zygomatic fosse, 19° in.
Height from lower margin of auditory foramen to the summit of sinciput, 18° in.
Depth or height of cranium from the posterior margin of molar alveolus (back
part of palate) to the summit of sinciput, 32° in. Depth from posterior and upper
margin occipital foramen to posterior margin molar alveolus, 9°5in. Height of
the orbit, 8°78 in. Length of anterior portion of the palate, from anterior end of
molar alveolus to tip of incisive, 16-in. Transverse diameter, left incisive at tip,
11:5in. (Estimated width of both incisives at tip, 24in.) Vertical diameter of
left incisive at tip, 10-6 in. Vertical height of sub-orbitary foramen, left, 3°85 in.
Width of incisive at contraction below sub-orbitary foramen, 10°75 in. Width of inci-
sive sheath at sub-orbitary foramen, 7-5in. (Estimated width of both incisive sheaths
at sub-orbitary foramen, 21:5in.) Interval between the posterior molars, anterior
end (width of palate in front), 2°7in. Interval behind, 3°25 in. Length of riglit
molar (backmost), 11°9in. Width of right molar in front, 4:05 in. Width of right
molar behind, 5-in. Vertical height from posterior outer margin of molar alveolus
to post-orbitary process, 21-in. Interval between outer surfaces of the molars at
fifth ridge from front, 12°6 in. Antero-posterior diameter auditory foramen,
1:15in. Transverse diameter auditory foramen, lin, Estimated height of occiput
at restoration from lower surface of condyles to sinciput, 24*5in. Length of left
tusk outside the incisive sheath, measured along lower surface, 10 ft. 6in. Length
of left tusk, inside sheath, 2ft. 3in. to 2ft.4in. Estimated total length, 12 ft.
9in. Total length of head from occipital protuberance to tip of tusk, 4 ft.+
10 ft. 6in.=144 ft. Length of right tusk, 10 ft. 8$in. Interval between the tusks
at emergence from incisive sheath, 6°in. Interval between the tusks at nearest
approximation, at 34 ft. from incisive, 3°5in. Divergence at tips, 5 ft. 3}in. as
restored. Height of versed sine of curve from tip to incisive border, right side,
23-in. Height of versed sine of curve, from tip to incisive border, left side, 23° in.
Vertical diameter left tusk (141m. from base) where greatest, 9°in. Transverse
diameter, left tusk, at ditto, 79in. Girth of left tusk, at ditto, 26°in.

Puiate XXII.

Fig. 1.—Elephas Ganesa. Lateral view of large skull figured in
Pl. XXI.—B.M.

Fig. 2.—E. Ganesa. Palate view of same skull. The right in-
cisive is seen in section. The posterior true molar is seen on either
side of palate. It has ten plates and a heel behind, and a small talon
in front; the hind heel has few denticles; the four front ridges are
worn. The alveoli are parallel as in the Mammoth.—B.M.

Fig. 3.—. Ganesa. Sketch showing restoration of skull, with
tusks, of £. Ganesay, profile view, one-thirteenth of natural size.

Puate XXIII.
Fig. 1—Elephas Ganesa. Sketch showing restoration of skull, with

DESCRIPTION OF PLATES. 455

tusks, of H. Ganesa, oblique antero-lateral view, one-thirteenth of
natural size.

Fig. 2.—Ditto, ditto, front view.

PLATE XXIV.

Figs. 1 and 1 a.—Elephas Ganesa. Fragment of right upper jaw
with first true molar. The tooth has six ridges and a heel; five ridges
worn. From Baker’s collection.

: Length of molar, 5-1 in. Width in front at second ridge, 2°5 in. Width behind,
“in,

Figs. 2 and 2 a.—E. Ganesa. Fragment of upper jaw with second ?
true molar presenting seven ridges.

Length of molar, 7:5 in. Width in front, 3:1 in. Width behind, 3:4 in.

Figs. 3 and 3 a.—E. Ganesa. Upper jaw, right side, with three
ridges of first true molar, and entire penultimate. The latter has seven
ridges and a heel; no cement; plates unworn. Closely resembles Z.
insignis.—B.M.

Length of penultimate molar, 9 in. Width, 4 in.

Figs. 4 and 4 a.—Z. Ganesa. Fragment of last molar, upper jaw,
right side. Fangs confluent; six ridges and a heel; discs of wear
very large; the last ridge has seven denticles.—B.M.

Length of molar, 9°2 in. Width, 4-2 in.

Figs. 5 and 5 a.—. Ganesa. Fragment of upper last? molar, right
side, very large; shows six ridges, or five anda heel, very compressed ;
nine denticles on fourth ridge.—B.M.

Length of molar fragment, 7:4 in. Width, 4:3 in.

Figs. 6 and 6 a.—Elephas insignis. This is a very instructive speci-
men, consisting of the palate, with penultimate and antepenultimate
true molars on both sides, from an animal of medium size. The ante-
penultimate has seven ridges, all worn but the last, and a small heel.
The penultimate or second true molar is entirely in germ; it consists
of eight main ridges and a front ridge, but has no heel. This is a cha-
racteristic type of true H. insignis, with the ridges very high and com-
pressed. There are about ten points or denticles to the fifth ridge of
the penultimate tooth, this being about the average; this tooth has no
cement between the plates, only matrix. The front, or antepenulti-
mate, tooth has plenty of cement. There are small tusks on both sides.
Compare with Plate XIX. 5.

Length of antepenultimate, 6°8 in. Width, 3-2 in, Length of penultimate,
85 in, Width, 3°6 in.

PLATE XXIV. A.

Figs. 1 and 1 a.— Elephas Ganesa. Fragment of skull with palate
and back molars on both sides. This is a most remarkable specimen.
T have called it H. Ganesa (H. F.), and it much resembles the molar
of the big Ganesa specimen (Plate XXII. fig. 2) in form and in the
compression of the ridges, but the ridges are few.'—B.M.

1 Another specimen of Mr. Cunliffe’s | with two last molars on both sides im
in Geological Society, not figured, is | situ. The front tooth has four plates,
also very remarkable. Shows the palate | much worn and very crimped. The

426 FAUNA ANTIQUA SIVALENSIS.

Length of back molar, 9°7 in. Width, 3°8 in. Width of palate in front, 2° in.
Width behind, 2°8 in.

Figs. 2, 2 a, and 2 b.—Elephas insignis. Small broken head. This
is a very remarkable specimen. The teeth, although small, are assuredly
the first and second true molars. The front tooth has six ridges and a
heel; the penultimate has only seven ridges and a large heel, with a
distinct talon in front; all the ridges of the front tooth worn; very
little cement; plates deep. The tusks, if any, have dropped out; two
large sub-orbitary foramina.—B.M.

Extreme length of skull, 18:in. Length of incisives, 15°8 in. Width between
orbits measured to lachrymal tubercle, 7-8 in. Contraction of muzzle at suborbi-
taries, 11:9 in. Width of muzzle at tip, 9°2 in. Width between outer surfaces of
maxillaries at the back molar (beginning), 8°-4in. Length from diasteme to tip of
tusk-sheaths, 9°3in. Depth from pterygoid to front, 14:7 in. Length of anterior
molar, 4°9 in. Width, 2°7 in. Width of palate between front molars anteriorly,
1:5 in. Ditto behind, 2-2 in. Length of penultimate molar, 7-4 in. Width,
3° in.

Figs. 3 and 3 a.—E. insignis. Lower jaw, with first and second
true molars. Five ridges remaining of front molar; other molar has
twelve main ridges anda heel. Besides the characters shown in the
figure, the diastemal portion is very much flattened in front of the an-
terior molar. This remarkable specimen is in the Museum of the Col-
lege of Surgeons; the drawing is taken from a cast.

Extreme length of fragment of jaw, 24: in. Height to alveolus, 9° in. Width
of jaw in front, 48 in. Width behind, 7-5 in. Length of ascending ramus, 12°8
in. Length of front molar, 5-in. Width, 3°5 in. Length of last molar, 11:5 in.
Width, 4° in.

Figs. 4 and 4 a.—Elephas Namadicus (Fale. and Caut.). From the
Nerbudda. Small head with two molars in situ; one worn, but the back
one in germ; probably the first and second true molars. The right
tusk is present; is very large for size of head; is nearly circular in
outline, and diverges greatly in front. The tusk-sheaths are long, as in
E. Indicus and E. primigenius. The brow-ridge, &c., are exactly as
in the large head of H. Namadicus (Plate XII. A.). The space between
the tusk-sheaths is very shallow, as in E. planifrons. Is probably a
young male head.—B.M.

Extreme length from broken occiput to broken incisive, 29° in. From groove of
brow to tip of nasals, 6-1 in. Across nasal opening, 125 in. Height of nasal

opening at sides, 3-6 in. Width of muzzle (incisive sheaths) at orbital foramen, _

13:in. From top of incisive sheath at fissure to diasteme, 9°8 in. Antero-poste-
rior diameter of left orbit, 5-9 in. Transverse diameter of right tusk, 3°6 in.; ver-
tical, nearly the same. Width of palate in front, 1°6 in. Width of palate behind,
2-7 in. Width of front tooth, 2°6 in.

PLATE XXV.

Figs. 1 and 1 a.—Elephas Ganesa. Lower jaw, with last lower
molar. Shows two mentary foramina on either side; is very sharp in

last tooth has ten ridges, or nine and a Length of anterior molar, 4°5 in.
large heel; only the two front ridges | Width, 4-in. Length of last molar (very
are worn; the tooth is very convex from | much curved), 10-4 in. Greatest width,
back to front, and the ridges are very | 3°9 in. Interval between front teeth,
high and convex across, with a large| 1:9 in. Interval behind at niche of
quantity of cement. Looks very like (!) | palate, 4°5 in.

the EL. bombifrons fragment, Pl. xxix. ;

‘fig. 4. ;

es ee

py By ws

DESCRIPTION OF PLATES. 457

front at diasteme, and high with an edge; ascending ramus is inclined
forward and does not shelve out. Molar has seven or eight ridges re-
maining, but is imperfect in front; enamel very much crimped (Vide
Plate XXV. A. fig. 2).—B.M.

Extreme length of jaw, 19°3 in. Height to alveolus, 8: in. Thickness in front,
4-1 in. Thickness behind, 6°3 in. Length of right molar, 8-9 in. Width at
middle, 4°7 in. '

Figs. 2 and 2 a.—Elephas bombifrons (Fale. and Caut.). From the
Sewalik hills.. Fragment of lower jaw, showing the united symphysis,
parallel rami, and three large mentary foramina on outside. Only a
fragment of last molar seen on both sides.—B.M. :

Extreme length, 17:3 in. Height to alveolus, 7-8 in. Width of jaw in front,
3°6 in. Width of jaw behind, 6:1 in.

Figs. 3 and 3 a.—H. bombifrons. Magnificent specimen of lower
jaw; fault in jaw; two large mentary foramina; beak very thick ;
right molar has nine ridges and a heel; enamel very thick; hardly any
cement.—B.M.

Extreme length of jaw, 23°6 in. Height to alveolar margin, 9:in. Thickness
in front, 4°2 in. Thickness behind, 8°3 in. Length of right molar, 12-9 in.

Greatest width at fifth ridge, 4:4 in.

Figs. 4 and 4 a.—Elephas insignis. Lower jaw, partly distorted by
pressure, and containing two molars on either side. The anterior
(first true) molar is entire, though mutilated, and has seven ridges and
a large heel. The back (second) molar is fragmentary.—B.M.

Extreme length of jaw, 15°in. Height to alveolus, 5-8 in. Thickness in front,
3-5 in. Thickness behind, 5°5 in. Length of front molar, left side, 6°5 in. Width
at middle, 2°5 in.

PuaTe XXV. A.

Figs. 1 and 1 a.—Elephas Ganesa (Fale. and Caut.). Fragment of
lower jaw, thick behind, with last lower molar, showing seven ridges
and a heel, and great crimping, but no characteristic feature. —B.M.

Length of right molar, 8°8 in. Greatest width, 3°8 in.

Figs. 2 and 2 a.—H. Ganesa. Fragment of lower jaw, very old,
with last molar much worn. Resembles Plate XXV. fig. 1—B.M.

Height of jaw at alveolus, 7°8 in. Length of right molar, 10-2 in. Width,
4: in.

Figs. 3 and 3 a—EH. Ganesa. A dumpy, small-sized lower jaw,
with imperfect molar; seven ridges remaining ; the back ridges curved
and much crimped.

Length of molar, 8:4 in. Width behind, 3°8 in. Width in front, 34 in.

Figs. 4 and 4 a.—E. Ganesa. This little lower jaw is very remark-
able in being high and narrow; the ascending ramus is much bent
forward. It contains three teeth, one in front worn out; a second with
seven ridges and a heel, and a third in germ.—B.M.

Extreme length of jaw, 18-4in. Height to alveolus, 6:in. Height at front tooth,
2:5 in. Height behind, 4:in. Length of ramus (antero-post.), 8:7 in. Length
of front tooth, 2°5in. Length of second tooth, 5-2 in.; width in front, 2: in.;
width behind, 2°3 in.

Figs. 5 and 5 a.—£. Ganesa. Angle of lower jaw, left side, with
portions of ascending and horizontal rami, and posterior five and a half

r adbes of last irue molar.—B.M.

Length of fragment of molar, 7*in. ‘Width, 3:9. in.

458 FAUNA ANTIQUA SIVALENSIS.

Figs. 6 and 6a.—E. Ganesa. Portion of lower jaw, left side, with
fragment of last true molar, showing anterior 9 ridges.
Length of tooth fragment, 9°3 in.; greatest width, 3°8 in.
Figs. 7 and 7 a.—E. Ganesa. Lower jaw, right side, with last molar.
Shows the back part of the tooth, on to the anterior large fang.—B.M.

Length of fragment of jaw, 18in. Length of fragment of tooth, 8°5in.; width,
3°3 in.

PuateE XXVI.

Elephas bombifrons (Falc. and Caut.). From the Sewalik hills.
Anterior view of large head.—B.M.

Extreme length from occipital to broken incisive, 34°2 in. Length from oeci-
pital to commencement of diasteme, 32°2 in. Occiput to tips of nasals, 16:2 in.
Middle of naso-maxillary fissure, 14°7in. Semi-diameter of brow at the post-
orbitary, 12°2 in. Width of brow at post-orbitary, 24-4 in. Width of muzzle at
contraction near orbitary foramen, 16°8 in. Width of inter-incisive fossa, 5:3 in.;
depth of fossa below incisive sheaths, 6°in. Greatest contraction of brow, 13°3 in.
Antero-posterior diameter of orbit, 6°in. Width of palate in front, 2°in. Width
of palate at middle, 4-5in. Width of palate behind, 4°6in. Height of pterygoid,
11:2in. Transverse diameter of left tusk, 3°3 in.

Puate XXVII.

Elephas bombifrons (Fale. and Caut.). Very fine and perfect skull,
anterior view. Four other views of same skull are given in Plate
XXVIII. This head is very marked; it is convex from occiput to
front and also across, and is very narrow at the temporal contraction.
The bounding ridges sweep round by a bold curve into the post-
orbitary processes, as in EH. meridionalis. ‘There is a deep furrow
between the tusks. The nasal opening for the trunk is above the
line (or nearly so) of the post-orbitary processes of the frontal bone.
Above the infra-orbitary foramen on the right side there is another
smaller opening. (On a proof copy of the plate this species is designated
Elephas intermedius, or Mastodon Elephantoides of Clift.) —B.M.

Extreme length from occiput to broken incisives, 27°in. From occipital condyles
to anterior border of alveolus of molar, 23°2in. Vertical length of head from broken
condyle to tip of occiput, 17°7in. Greatest width of occiput, 25°5in. Interval
between auditory foramina, 21-5in. From anterior margin occipital foramen to
the posterior surface palate, 14-in. Length of palate from niche to the downward
berd of tusk-sheaths, 12°5in. Interval between outer surface of teeth, behind, 9-8 in.
Int. between outer surf. teeth, infront, 9°in. Length of right molar, 10°2in. Width
in front of molar, 3°7 in. Width behind, 3°4in. Interval between molars, in front, 1: in.
Interval behind, extremely divergent, 4:2 in.; number of ridges 9 and a heel, 8 front
worn. From occiput (middle) to tip of nasals, 13°8 in. Width of brow across post-
orbitary processes, 22'4in. Greatest contraction temporal fossa, 10°8in. Trans-
verse diameter nasal opening, 11°3in. Depth of nasal opening at sides, 2°5in. In-
terval between middle of orbits, 19°5in. Vertical height of orbit, 4-6in. Height
from posterior surface palate to the middle of bulge of frontal, 23°5in. From the
anterior margin orbit to surface of occiput, 20°in. Width across incisive sheaths
at base, 12°5in.; interval between, about middle, 2°8in. Depth of fossa between .
ine. sh., 4" in, Vertical diameter left tusk, 3°6in. Transverse diameter of left
tusk, 3-3in.

Piate XXVIII.
Fig. 1.—#lephas bombifrons. Lateral view of same skull, as figured
in Plate XXVIJ.—B.M.
Fig. 2.— 4%. bombifrons. Palate view of same skull, showing sections of
tusks, and last? true molar on either side, with 9 ridges and a heel;

DESCRIPTION OF PLATES. 459

the 8 front ridges worn. The interval between the molars in front is
very narrow ; behind they are extremely divergent.—B.M.

Fig. 3.—E.bombifrons. Antero-lateral view of same skull, with large
infra-orbitary foramen.—B.M.

Fig. 4.—E. bombifrons. Posterior view of same skull, showing occi-
put, occipital foramen and condyles, and pterygoids—B.M.

Fig. 5.—E. bombifrons. Detached specimen of occiput.—B.M.

Greatest width, 25:in. Vertical height, 17:2in. Diameter across occipital
condyles, 7°5in. Transverse diameter of occipital foramen, 2°5in. Vertical
diameter of occipital foramen, 2°4 in.

PuaTE XXIX.

Fig. 1.—Elephas bombifrons. Broken cranium, palatal surface,
with last true molar on either side, that on the right side presenting
8 ridges and a heel, and very fine.—B.M.

From anterior margin of occipital foramen to niche of palate, 14*in. Width of
skull across occipital foramen, 8'in. Length of palate from niche to commencement
of diasteme, 10°7in. Width of palate between molars in front, 2°in. Width
behind, 3°in. Length of right molar, 10°in, Width of other molar, 4: in,
Width of incisive sheaths at muzzle, 13° in,

Figs. 2 and 2 a.—£E. bombifrons. Broken cranium with last? true
molar on either side of palate. The right molar has 8 ridges, of which
the five anterior ones are much worn. The molars are approximated
in front, and very divergent behind (Vide Plate XIX. A., 5).—B.M.

Figs. 3 and 3 a.—Second milk molar, upper jaw, left side of Elephas
insignis. (Misnamed E. bombifrons on Plate.)

Figs. 4 and 4a.—E. bombifrons. Upper last true molar, right side.
This is a beautiful specimen, with a continuous transverse heel in
front; 8 plates remaining; the points are very numerous; no mesial
division. The tooth is convex across (See note to Plate XXIV. A., 1.)
—B.M. ;

Length of fragment, 9'in. Width in front, 4:in.; width behind, 3-7 in.

Figs. 5 and 5 a.—£. bombifrons. Portion of upper jaw with a mag-
nificent fragment of the last upper molar, right side, very large, discs
very wide; shows 7 ridges and a heel. Very like the Ganesa speci-
men formerly in India House collection.—B.M. ?

Length of fragment of molar, 10°4in. Width in front, 3°8in. Width behind,
3°8 in.

Figs. 6 and 6 a.—£. bombifrons. A magnificent palate specimen,
with portion 0” cranium and last molar, right side, showing 9 ridges
and a heel.—B.M.

Length of last molar, 10°‘9in. Width in front, 3°8in, Width behind, 3:8 in.
Greatest width in middle, 4:3 in.

PuaTe XXIX. A.

Figs. 1 and 1a—£. bombifrons. A mutilated lower jaw, short
and thick in its build, with what is probably the third milk molar,
and having the first true molar appearing in germ behind. The
milk molar shows 6 ridges and a heel. Baker’s collection.—B.M.

Thickness of jaw in front, 2°8in. Thickness of jaw behind, 5-in. Height at
alveolus, 5:in. Length of anterior tooth, 4:in. Width, 2° in.

Figs. 2and 2a.—E.bombifrons. Fragment of lowerjaw, right side, with
what is certainly the first true molar, showing 7 ridges and a heel; all the

460 FAUNA ANTIQUA SIVALENSIS.

ridges worn but the last; the two first worn out; a leng sloping
diasteme ; two outer mentary foramina.—B.M.

Height of jaw to alveolus, 7-in. Thickness in front, 3-4in. Thickness behind,
6:in. Length of molar, 6'4in. Width in front, 2°3in. Width behind, 2°9in. ~

Figs. 83 and 3a.—E£. bombifrons. Portion of lower jaw, left side,
with one large outer mentary foramen, and penultimate true molar,
showing 7 ridges and a large heel.—B.M.

Extreme length of fragment of jaw, 15-in. Height to alveolus, 73in. Length
of molar, 8°5in. Width in front, 2°8 in. Width behind, 3:2 in.

Figs. 4 and 4 a.—E. bombifrons. Fine specimen of anterior portion
of lower jaw, both rami with symphysis. Contains the penultimate
true molar of either side, with 7 ridges and a heel. Shows-also the
commencement of the last tooth behind, and proves the tooth (fig. 2)
to be the first true molar. Jaw is deficient on both sides behind the
penultimate.—B.M.

Length of fragment, 14:3in. Height to alveolus, 7: in. Thickness behind, 7-in.
Interval between rami in front, 2°7 in. Interval behind, 3°3in. Length of right
molar, 8°in. Width in front, 2-9in. Greatest width, 3-4 in.

Fig. 5.—E. bombifrons ? Fragment of molar, from lower jaw, right
side, with four ridges.—B.M.

Length, 5‘8in. Width, 46 in.

Fig. 6.—£. bombifrons ? Fragment of molar with three ridges and

a heel. ‘ Doubtful what figs. 5 and 6 are.’ —H.F.
Length, 4*4in. Width, 4:5 in.

Figs. 7 and 7 a.—E. bombifrons. Portion of lower jaw, with molar
showing 7 ridges, counting the last; no heel; another tooth in germ
behind ; is apparently the first true molar, with the second, or penul-
timate, coming behind it—B.M.

Length of anterior molar, 6°5in. Width, 2-7 in.

Figs. 8 and 8 a.—E. bombifrons. Lower jaw, right side, of a small
variety, but old. It is figured chiefly for its small size. The tooth is
certainly the last molar ; it is wide behind and very thick; the discs of
wear are peculiar.—Specimen in Geol. Soc. Museum.

Height of jaw to alveolus, 7-5in. Width behind, 6-5 in. Width in front, 4-6 in.
Length of tooth, 7°8in. Width behind, 3-in.

PLATE XXIX. B.

Figs. 1 and 1 a.—Elephas insignis. Fragment. of anterior half of
second milk molar, of left lower jaw, with two ridges and a front talon:
Length of fragment, 1-7 in. Extreme width, 1-4in. Height, 1-2 in.

Figs. 2 and 2a.—Elephas Ganesa. Fragment of lower jaw, right
side, with the third milk molar entire, presenting 7 ridges and
heel. :

Length, 4:6 in. Width in front, 1-4in. Width behind, 2-2 in.

Fig. 3—E. Ganesa. Fragment of lower jaw, with three molars
(second and third milk, and posterior 5 ridges of first true).

Figs. 4 and 4 a.—E. Ganesa. Fragment of lower jaw, right side,
with fifth or penultimate molar, presenting 8 ridges. ;

Length, 7:°25in. Width in front, 3-1lin. Width behind, 2-9 in.

Figs. 5 and 5 a.—Elephas bombifrons. Fragment of lower jaw, right

DESCRIPTION OF PLATES. 461

side, showing third milk molar with 5 ridges and an anterior and
posterior talon. The three anterior ridges are worn into a common
disc.

_ Length, 36in. Width in front, 1:-4in. Width behind, 1-8 in.

Figs. 6 and 6 a.—Elephas bombifrons. Fragment of lower jaw, right
side, with the antepenultimate or first true molar, presenting 7 ridges
and a heel.—B.M.

Length of molar, 5:9 in.
ridge, 2°7 in.

Figs. 7 and 7 a.—E. bombifrons. Last molar, lower jaw, left side,
presenting 9 ridges and a heel; 3 front ridges abraded.—B.M.

Extreme length, 13°4in. Extreme breadth of crown, 4°2in. Breadth at
posterior ridge, 3:'2in. Breadth of grinding surface, 3°2in, Extreme height,
6-4 in.

Figs. 8 and 8 a.—Elephas insignis. Fragment of lower jaw, right
side, with portions of penultimate and last true molars. Shows 4 ridges
and a large posterior talon of penultimate, and 3 ridges and anterior
talon of last molar. Two large outer mentary foramina.—B.M.

Length of grinding surface of penultimate, 6-3in, Extreme breadth posteriorly,
37 in. Breadth across anterior ridge, 3-2 in.

Width in front, 2-lin. Width at antepenultimate

PLATE XXX.

Figs. 1 and 1 a.—Elephas Cliftii! (Fale. and Caut.). Mastodon
Elephantoides of Clift. Palate, with third upper milk molar, and the
three anterior ridges of the antepenultimate or first true molar.2. The
third milk molar is entire on one side, but worn down to the common
base of ivory, so that the divisions of the crown have entirely dis-
appeared, leaving no certain data for determining the ridge formula.
Behind it the three anterior ridges of the antepenultimate true molar
are seen in situ, the posterior half being broken off. The plane of the
palate, on to the diasteme, is very flat. The mesial line of division of
the ridges in first true molar is not very distinct. This very interest-
ing specimen was brought from Ava by Colonel Burney, and presented
to the British Museum.

_ Extreme length of fragment, 10°6 in. Length of anterior tooth, 3-3 in. Width,
2-in. Length of second tooth, 3:in. Width, 3-4 in. Interval between teeth in
front, 3-2 in.

Figs. 2 and 2 a.—E. Cliftii. This is the first or antepenultimate

_ 1 Mr. Clift, in his excellent memoir, | #. (Stegodon) Cliftii; and the lower jaw

ineludes the Ava fossil Proboscideans
under two species, Mastodon latidens
and Mastodon Elephantoides. In the
‘Fauna Antiqua Sivalensis,’ the for-
mer name is retained for the speci-
mens of the Tetralophodon type, figured
by Mr. Clift in the Geol. Trans., vol. 11.
2nd ser., Plate xxxvii. figs. 1 and 4;
Plate xxxviii. fig. 1; and Plate xxxix.
figs. 1, 2, and 3. Of the others, the pa-
late specimen, Plate xxxvi. (Mastodon
latidens, Clift), together with the de-
tached molar, Plate xxxviil. fig. 6 (Masto-
_ don Elephantoides, Clift), are referred to

specimen, Plate xxxvili. fig. 2 (also M.
Elephantoides, Clift), is referred to £,
Stegodon) insignis. The specimens re-
garded by him as of his Mf. Hlephantoides
being here considered to belong more
properly to the genus Evephas, it became
necessary to resort to another specific
designation. Hence the origin of ZL.
(Stegodon) insignis.—H.F.

? Inthe Museum of the Royal College
of Surgeons (Cat. No. 664) is another
fine palate specimen of E. Cliftii with
the penultimate and last true molars on
both sides. :

462 FAUNA ANTIQUA SIVALENSIS.

true molar, upper jaw, left side, entire, detached, and beautifully pre-
served. It has six ridges and a small hind talon. The tooth is nearly
oblong. The enamel is somewhat fluted, and there is very little
cement. There is very little convexity of the tooth across, and no
distinct indication of the mesial dividing line. There are as many as
eleven to twelve denticles or points. The tooth is compressed and
angular in front, and the three front ridges are worn. This tooth is
also represented by fig. 6 of Plate XXXIX. of Mr. Clift’s Memoir
(Geol. Trans., vol. 1i., 2nd series). It is there described as Mastodon
Elephantoides, under which title Mr. Clift included specimens referred
by Dr. Falconer to two distinct forms.'! Its elephantine affinities are
indicated by the absence of a longitudinal line of division along the
crown, and by the great number of points that enter into the composi-
tion of the ridges.—Cast in B.M.
Length, 6:1 in. Width in front, 3-in. Width behind, 3-3 in.

Figs. 3 and 3 a.—Elephas Cliftii. Superb palate specimen contain-
ing the penultimate true molar in situ on both sides. The tooth is
proved to be the penultimate by its large dimensions, and by the cir-
cumstance that two or three ridges of another tooth (third true molar)
is seen behind it.2 The crown ridges are all more or less worn and
partly damaged by fracture, but enough remains to show that the tooth
was composed of six ridges and a hind talon. The palate is very flat,
and the teeth on either side (in the erect position of the skull) slope
upwards from without inwards. The teeth have very little cement.
The diastemal ridges are wide apart. The right ridge shows the tusk-
sheath; there is very little verticality of the tusk. The slope of grind-
ing surface from without inwards is a mastodontoid character, which is
very notably seen in Mr. W. Ewer’s specimen of M. Sivalensis (Plate
XXXIV. fig. 1). In the Elephas insignis the tendency of the grinding
surface is to shelve upwards from the inside outwards, being the reverse
of what holds in H. Cliftii.3 This very important specimen is from
Burmah, and is now in the Museum of the Geological Society. It is also
represented by Clift in Plate XXXVI. of his Memoir in the Geological
Transactions, vol. ii., 2nd series. The remaining portion of the upper
jaw containing the last true molar is believed to be in the Museum of
the Asiatic Society of Bengal (See antea, p. 114).

Extremelength of specimen, 19-in. Width of jaw at middle of molar, 8° in., doubled
=16:in, Length of anterior molar, 8-2 in. Width, 4-in. Width of palate in
front, 2°5in. Width of palate between diastemal ridges, 4°8 in.

Figs. 4, 4 a, and 4 6.—Elephas Clifiii. A fragment of the penulti-
mate molar, upper jaw, right side, with five ridges, shown also in
section (fig. 4b). Made out approximately to be #. Cliftzi. Cement —
in moderate quantity.—B.M.

Length, 6°6 in. Width, 3°8 in. ;

Figs. 5 and 5 a.—E. Cliftit. Last true molar, lower jaw, left side,
consisting of eight ridges and a talon. Five ridges are touched by
wear, and the two anterior ridges are nearly worn out. The anterior
large fang has been absorbed, but the portion of crown sustained by it

1 See note 1, page 461. | upper molars is as markedly from the |

2 See note 2, page 461. inside outwards and upwards, as it is the

3 Tn Col. Farquhar’s specimen of the | reverse from outside inwards in Masto-
Indian Elephant, the plane of wear in the | don Sivalensis,

DESCRIPTION OF PLATES. 463

remains. The six posterior ridges have their fung elements confluent
into a continuous plate or shell, thus maintaining the elephantine
affinity indicated by the crown characters. The crown is very flat;
the points are large and few in number, and there is no very distinct
mesial dividing line, but little cement. This magnificent specimen was
brought from Burmah by Col. Burney, and presented by him to the
British Museum (See antea, p. 114, and Plate V. of vol. ii.).
Length, 12:7 in. Width at middle, 4:5 in.

Figs. 6 and 6 a.—Mastodon latidens | (Clift). Fragment of lower
jaw, left side, with last or third true molar. This is one of the most
enormous specimens ever seen, the greatest girth over the molar being
no less than 27°5 inches. There is one mental foramen; no appear-
ance of tusk. The molar has five ridges and a double heel; the points
are very large, and the discs of wear very broad; the mesial line of
division is distinct; there is no cement. ‘The plane of wear in front
shelves from the inside out. There is a very similar specimen in the
Museum of the Asiatic Society of Bengal (See antea, p. 120).

Extreme length of fragment, 16:2 in. Height of jaw at anterior end of
alveolus, 10-2 in. Greatest thickness, 6:8 in. Length of molar, 11:3 in. Greatest
width, 4°5 in.

Pirate XXXI.
Figs. 1 and 1 a.— Mastodon latidens. Upper milk molar.—B.M.
Length, 2°7 in. Width, 2:4 in.
Figs. 2 and 2 a.—WM. latidens. Upper milk molar (third?).—Castin B.M.
Length, 3:8 in.- Width, 2°6 in.

Figs. 3 and 3 a.—WM. latidens. Fragment of upper jaw, right side,
with two molars (third milk and first true ?).—B.M.

Length of posterior molar, 5°6 in. Width, 2-9in. Length of anterior molar,
8°7 in. Width, 2°6 in.

Figs. 4 and 4 a.—M. latidens. Portion of palate, with two upper
molars left side.—B.M.

_ Length of anterior tooth, 3-9 in. Width, 2:8 in. Length of posterior tooth,
4-2 in. Width, 3-4 in.
Figs. 5 and 5 a.—WM. latidens. Fragment of upper molar.—Cast in
-M.

Length of fragment, 3:4 in. Width, 4:2 in.
Figs. 6and6a.—M. latidens. Upper true molar, very perfect.—B.M.
Length, 8°5 in. Width, 4:5 in.
Figs. 7 and 7 a.—M. latidens. Fragment of lower molar with fang
(See antea, p. 121).—B.M.
Length, 5:in. Width, 3:3 in. Length of crown fang, 5:3 in.
Figs. 8 and 8 a.— WM. latidens. Lower molar well worn (See antea,
p- 121).—B.M.
Length, 6-4in. Width, 3:3 in. Height of crown fang, 6:4 in.
Mastodon latidens, like the M. longirostris of Eppelsheim, presents a
Dinotherian type, in so far as the crowns of the molar teeth are con-
cerned; and in this respect it contrasts with M. Sivalensis and M.
Arvernensis, in which the molars have a hippopotamoid type. In M.
latidens the crown is broad, the mammille are thicker in proportion

1 See note 1, page 461.

464 FAUNA ANTIQUA SIVALENSIS.

to their height; the ridges are less elevated, and consist of a greater
number of coronal points, and the valleys are more open and trans-
verse (or interrupted only by an insignificant number of warty tuber-
cles) than in M. Sivalensis (See antea, p. 121).

Figs. 9 and 9 a.—Mastodon Perimensis (Fale. and Caut.). From
Perim Island. Portion of upper jaw, right side, with molar, present-
ing characters similar to those of M. Sivalensis and M. Arvernensis
(See antea, pp. 117, 122).—B.M.

Length, 6°3 in. Width, 3°3 in.

Figs. 10 and 10 a.—M. Perimensis. Portion of lower jaw, left side,
with penultimate and last true molars. Presented by Miss Pepper to
B.M.

Length of anterior molar, 4°8 in. Width, 2°6 in. Length of fragment of
posterior molar, 2°1 in. Width, 2°8 in.

Figs. 11 and 11 a.—Fragment of lower jaw, with molar.

Length of tooth, 8-4 in. Width, 3°65 in.

PLATE XXXII.

Mastodon Sivalensis (Fale. and Caut.). From the Sewalik hills;
anterior view. Three other views of same cranium in Plate XXXIII.
The brow is a little crushed between the temporals. The left tusk-
sheath is present, the right is withered; there is distortion of the
tusk-sheaths in consequence. There is an enormous and peculiar
projection of the lips of incisive anterior end of naso-maxillary fissure,
so as to project over the trunk, or inter-incisive fossa. This fossa is
very deep, enormously so. The depressions for the condyles of the
lower jaw are very vaulted. There are two infra-orbitary foramina on
the right side. The grinders are very much worn. On the left side
there would be five ridges, with a very complicated heel-series. Pos-
teriorly, the teeth show well the peculiar characters of Mastodon
Sivalensis—B.M. (Reproduced in Plate X.)

Extreme length from occipital bulge to tips of incisives, 28°3 in. Width of occiput,
22-4 in. Height of occiput from plane of condyles, 19:0 in. Interval between
outer edge of condyles, 66 in. Vertical diameter of occipital condyle, 4-0 in.
Transverse diameter occipital condyle, 2°7 in. Width of base of occiput, 21-0 in.
Transverse diameter occip. foram. 2°9 in, Vertical diameter occip. for. 2°5in. From
occiput to broken tips of nasals, 14°0in. From occiput to anterior margin of orbit,
21-0 in. From anterior margin nasal opening to the tips of the incisive, 14-0 in.
Width of brow between post-orbitary processes, 25:5 in. Width at contraction
between temporals, 11:6 in. Width of naso-maxillary (trunk) opening, 12-0 in.
Depth of naso-max. op., lateral, 3:0 in. Vertical diameter orbits (nearly circular),
43 in. Transverse diameter orbits, 4:3 in. Interval between orbitary processes
of one orbit, 2°7 in. Width of forehead at anterior margin of orbit, 20°6in. Con-
traction of muzzle at orbitary foramen, 13°0 in. Depth of trunk fossa below brow,
9°8 in. Greatest interval between zygomatic arches, 22-0 in. From base of
zygomatic to top of occiput, 20°5 in. Depth of anterior orbitary process of
maxillary, 45 in. Height of skull from palate to top of occiput, 25°0 in. From
posterior surface condyles to posterior surface palate, 16:0 in. From pterygoid
processes to anterior margin of palate, 12°2in. Height of pterygoid, 10-3in. Distance
from molar to posterior surface of pterygoid, 4:2 in. Width of pterygoid above, 6:0
in. Width of pterygoid below, 7-5 in. Width of fissure (between pterygoids), 2°3
in. Length of palate at middle, 9°6in. Width of palate between molars in front,
2°8 in. Width of palate between molars behind, 2°6 in. Length of left molar, 7°8
in, Width of left molar in front, 3-2 in. Width of left molar behind, 3°3 in.
Length of articulating condyles for lower jaw, 4:8 in. Greatest width of condyles
for lower jaw, 3°0 in. ;

:

DESCRIPTION OF PLATES. 465.

Pirate XXXII.

Fig. 1.—WWastodon Sivalensis. Lateral view of same skull as is
figured in Plate XXXII. Shows temporal fossa and left zygomatic
arch entire—B.M.

Fig. 2.—Palate view, showing palate, one molar on each side, both
zygomatic arches entire, occipital condyles, Gc. Same cranium as fig. 1.
—B.M.

Fig. 3.— Occipital view of same cranium, showing occipital condyles
and foramen, pterygoids, &c.—B.M.

Fig. 4.—Lateral view of another cranium of MW. Sivalensis, with
lower jaw in situ. The left zygomatic arch is broken away.—B.M.

Pirate XXXIV.

Fig. 1.—Mastodon Sivalensis. Portion of skull showing palate with
two molars on either side, the posterior molar on right side imperfect.
In this specimen there are two large orbitary foramina and there is
only a tusk on the right side; the left tusk-sheath is withered. The
tusk-sheaths are very divergent, with an enormous deep trough
between, and there is great concavity of the brow, as in HL. insignis.
The teeth diverge anteriorly, as in Mastodon Ohioticus, but in a
less degree. The molars are excessively like Smith’s specimen of the
last molar upper jaw, right side, of Mastodon Arvernensis (Plate
XXXVI. fig. 8). The slope of grinding surface from without inwards
is very notably seen (see Plate XXX. fig. 3). The anterior molar of
both sides is worn, but shows four ridges with a heel, the ridges so
advanced in wear as to exhibit the characteristic alternating discs of
wear, so remarkable in this species. The last molar is perfect, except
the heel ridge, on the left side. The front ridge is barely touched by
wear, while the four. back ridges are intact. The drawing does not
show the characters very perfectly. The front ridge has two confluent
ridges on the inside and two on the outside, to the inner of which the
intermediate pillar is attached, joining on with the innermost large
point of the third ridge. This third ridge shows but one thick point
on the inner division and two on the outer, with the intermediate pillar
connecting the outer division of the third with the inner division of

_the fourth ridge, and so with the last ridge. This diagonal connection

of the posterior surface of the outer division with the anterior surface
of the inner points of the next following ridge cause in wear the
alternate-like discs of detrition, which characterize the teeth of Mas-
todon Sivalensis. The same kind of arrangement holds in Smith’s spe-
eimen of Mastodon Arvernensis. The points in MZ. Sivalensis are very
high and obtuse, and the apex of the ridges is high as compared with
the specimen in Mr. W. Ewer’s collection. (See also antea, p. 117.)

Length of two molars, right side, 9:2 in. Length of front molar, left side, 4:7
in.; width, 2°8 in. Length of last left molar minus heel, 6-4 in.; width, 3:0 in.
Interval between teeth in front, 3°4 in.; interval between teeth behind, 2°7 in.
Height of enamel crown, 2°3 in. Height of inner mammilla of second ridge, 1:8
in. Length of palate from niche to diasteme, 9:0 in. Antero-posterior diameter right
tusk, 3:0 in.; transverse, 2°7 in. Width of base of muzzle at orbitary foramen,
14-0 in. Interval between the middle of the orbits, 19:2 in. Antero-posterior
diameter or height of left orbit, 4°8 in.

Figs. 2 and 2 a.—M. Sivalensis. Fragment of upper jaw with molar
presenting six ridges, with same characters asin fig. 1. The two anterior
ridges only are touched by wear. The tooth is now cut into sections,
—B.M.

VOL. I. H H

466 FAUNA ANTIQUA SIVALENSIS.

Figs. 8 and 3 a.—M. Sivalensis. Fragment of lower jaw with portion
of molar, four ridges and part of a fifth —B.M.
Extreme length, 5-4 in. Breadth in front, 2°5 in.
Figs. 4 and 4 a—M. Sivalensis. Fragment of lower jaw, with
three ridges of a molar.—B.M.
Figs. 5 and 5 a.—M. Sivalensis. Fragment of molar.—B.M.

PLATE XXXV.

Figs. 1 and 1 a—Mastodon Sivalensis. Fine specimen of lower
jaw, with one molar on either side. A portion of the right ramus
deficient, but restored in the drawing. The teeth show well the
alternating discs of wear characteristic of the species, and two outer
mentary foramina.—B.M.

Fig. 2.— Mastodon longirostris (Kaup), from Eppelsheim; lower
jaw in outline, profile view; from a cast.

Figs. 8 and 8 a.—Mastodon Andium (Cuv.). Perfect lower jaw of
an adult with two last molars in situ, from Buenos Ayres. The
anterior tooth confirms what is shown by the Canterbury specimen
(Plate XL. fig. 15) respecting the penultimate. It is in an advanced
stage of wear, but exhibits distinctly the discs of three ridges. The
crown is nearly rectangular in form; the dimensions being 5:1 in. in
length, 2°85 in. of width in front, and 3° in. behind. The posterior
tooth, which is the last or third true molar, has the crown composed of
four principal ridges, and a complex sub-triangular heel of several
points. The three anterior ridges are partly worn and exhibit well the
characteristic complex trefoil discs of wear. The two posterior ridges
are intact, and the sinuous hollows between them show the very
considerable layer of cement which is present in a greater quantity in
this than in any other species of true Mastodon. The dimensions of
this tooth are about 8 in. in length by 3°5 in. of width in front,
whence it narrows gradually towards the posterior end.—B.M. (Re-
produced in Pl. VIII. fig. 3.)

Fig. 4.—Mastodon Ohioticus (Blumb.), from North America. Fine
specimen of lower jaw with two last molars, viewed from above. The
anterior or penultimate tooth consists of three ridges separated by trans-

verse uninterrupted valleys; all the ridges are slightly affected by wear. .

The posterior tooth consists of four main ridges and a subordinate talon
ridge; all untouched by wear.—B.M.

Fig. 5.—M. Ohioticus. Profile view of same jaw. One large
and one small mentary foramen.—B.M.

Figs. 6 and 6 a.—Dinotherium Indicum (Falc.). Superb fragment
of lower jaw, left side, with molars, brought from Perim Island by Miss
Pepper. The specimen contains nearly the whole of the adult series of
five molars in situ. The contour of the body of the jaw is shown in
the most perfect state of preservation, the fossil having fortunately been
mineralized by means of a very hard siliceo-ferruginous infiltration.
But it has evidently been long rolled about on the sea-beach as a
boulder, so that the crowns of the whole series of molars have been
hammered off nearly level with the alveolar margin of the jaw; the
surface of the fossil is jet black, and almost all of the matrix has been
cleared away, probably by the long-continued action of the sea, which
has given it a semi-vitreous polish. Patches of recent marine shells
are also found on the surface. The symphysis of the jaw is broken off
about 24 in. in front of the anterior premolar, and the bone is truncated

q

:

DESCRIPTION OF PLATES. 467

behind exactly opposite the point where the coronoid margin of the
ramus begins to rise up, the fracture passing through the middle of the
last molar, the anterior ridge of which is visible in situ in the jaw.
A detailed description of the points of distinction between this fossil and
the Dinotherium giganteum of Kaup is given in the memoir on Perim
Island fossils (See page 404, and Pl. XXXII. fig. 5)—B.M.

Length of fragment, 17: in. Length of four front teeth, 13°5 in. Length of first
premolar, 3°5 in. Width of first premolar behind, 2-2in. Length of second premolar,
2°9in. Width of second premolar behind, 2°6 in. Length of third orfirst true molar,
4-in. Width behind, 2°8 in. Length of fourth tooth (second true molar), 3:9 in.
Width, 35 in. Depth of jaw to alveolar margin at the second premolar, 9-2 in.
Depth at third tooth or first true molar, 8-7 in. Width of jaw at second pre-
molar, 5-1 in. Width at middle of fourth tooth, 6:4in. Distance between the
upper margin of mentary foramen and alveolus of first premolar, 3-6 in. Distance
from inferior margin to first premolar, 4°75 in. :

Puate XXXVI.

Figs. 1 and 1 a.—Mastodon Sivalensis. Fragment of upper jaw with

(second) milk molar.
Length of tooth, 2°6 in.

Figs. 2 and 2 a.—WM. Sivalensis. Fragment of upper jaw with two
milk molars. (Second and third).—B.M.

Length of anterior tooth, 2°6 in. Width, 1:8 in. Length of posterior tooth, 4° in.
Width, 2-3 in.

Figs. 3 and 3 a.—M. Sivalensis. Portion of upper jaw, with frag-
ment of molar. (First or second true m.).—B.M.

Length of fragment of crown, 4: in. Width,2-9 in.

Figs. 4 and 4 a.—M. Sivalensis. Fragment of upper jaw, with
second? true molar, presenting five ridges, with alternately disposed
crown mammille.—B.M.

Length of molar, 5°6 in. Width, 2-9 in.

Figs. 5 and 5 a.—J!. Stvalensis. Fragment of upper jaw, with
last? true molar. Five ridges and a hind talon.—B.M,
Length of molar, 6:5 in. Width, 2°9 in.

Figs. 6 and 6 a.—Mastodon Sivalensis. Last molar, upper jaw,
left side ini plan and profile. It has six ridges and a hind talon, and in
this respect it differs from both I/. Arvernensis and M. longirostris, but
it most resembles the former in so far as the alternate disposition of the
crown mammille is concerned. The complexity of pattern is even
greater than in the English Crag Mastodon.—Cast in B.M. (Reproduced
in PI. IX. figs. 1 and 2.)

Length of tooth, 7°8 in. Width, 3°3 in.

Figs. 7 and 7 a.—Mastodon Arvernensis, or the English Crag Mas-
todon. Left upper jaw of a calf, with the last milk molar beautifully
preserved 7m situ, and the remains of the empty alveolus of the pe-
nultimate milk molar in front of it. The crown is composed of four
ridges with a front and hind talon, and a well-pronounced basal
‘pbourrelet.’ The three anterior divisions are more or less worn,
especially along the inner side; the last ridge is nearly intact. The
ridges are connected by one or two stout conical mammille, interrupt-
ing their transverse continuity, and alternating with the divisions of
the main ridges. The vertical furrowing of the enamel at } and ¢,

HH 2

468 FAUNA ANTIQUA SIVALENSIS.

presenting the appearance of a reeded column or of a number of cords
pressed close together, is remarkable. This character is not present in
the corresponding young molars of Mastodon longirostris, Pl. XL. fig. 6,
in which the enamel is irregularly wrinkled but never presents the
symmetrical fluting observed in the ‘Crag’ Mastodon. This dif-
ference indeed is sufficient to distinguish the young teeth of the two
species. Discovered in the ‘Crag’ at Portwick by Mr. Wigham, and
figured by Lyell, ‘Manual of Elementary Geology,’ 5th ed. 1855,
p- 166, fig. 133.

Length of tooth, 2°9 in. Width anteriorly, 1-7 in. Width posteriorly, 1:8 in.
Width of grinding surface, 1:2 in.

Figs. 8 and 8 a.—Mastodon Arvernensis. Last true molar, upper
jaw, right side, composed of five ridges with an anterior talon, and a
strong back talon. The crown is obscurely divided longitudinally by a
shallow cleft along its axis. Deep clefts or valleys intervene between
the ridges; but the valleys, instead of being transverse, are interrupted
in the middle by one or more large accessory conical mammille,
interposed between the ridges and alternating with the outer and inner
divisions. This is the famous Whittingham tooth forming the frontis-
piece of Mr. W. Smith’s ‘ Strata Identified,’ and of which a woodcut
(reversed) is given in Owen’s ‘ British Fossil Mammalia,’ p. 276.

Length of tooth, 7°in. Width, 2°9 in.

Figs. 9 and 9 a.—Mastodon Arvernensis. Another specimen of last
true molar, upper jaw, left side. This is Captain Alexander’s specimen
dredged up between Southwold and Easton, and of which there is a
cast in the Museum of the Geological Society. The specimen is very
black with a sandy matrix and no vertical pillaring. The crown con-
sists of five ridges and a heel ridge of four points. The anterior edge
is broken. The enamel is very thick. There are three sub-alternate
mammille in the first valley. The second and third ridges are very
closely approximated, with but one intermediate mammilla. The third
and fourth are wide apart with three mammille in the valley. The
fourth and fifth have but one intermediate mammilla.

Figs. 10 and 10 a.—Mastodon longirostris. Antepenultimate true
molar, upper jaw. From Eppelsheim. Cast in B.M.

Length, 4°5 in. Width, 2°5 in.

Figs. 11 and 11 a.—WMastodon longirostris. Penultimate true molar
from Eppelsheim. Cast in B.M.

Extreme length, 5-4 in, Width anteriorly, 3: in. Width posteriorly, 3°3 in.

Figs. 12 and 12 a.—Mastodon longirostris. Last true molar, upper
jaw. Shows five ridges and a talon. The crown is broader, and the
mammillz thicker in proportion to their height, than in M. Arvernensis.
The ridges also are less elevated, and consist of a greater number of
coronal points. The valleys are either entirely open and transverse, or
interrupted only by an insignificant number of warty tubercles. From
Eppelsheim. Cast in B.M.

Length, 9° in. Width, 3:8 in.

Figs. 13 and 13 a.—Mastodon longirostris. Last true molar, upper
jaw, presenting some characters as fig. 12. From Eppelsheim. Cast
in B.M.

Length, 6°8 in. Width, 2-9 in,

DESCRIPTION

OF PLATES. 469

PLaTE XXXVI.

Figs. 1 and 1 a.—Mastodon Sivalensis.

Portion of lower jaw, right

side, with first (~) and second milk molars.—B.M.
Length of fragment of jaw, 4°7 in. Greatest breadth, 2°6 in. Height opposite

posterior border of second milk molar, 2:1 in.
Length of second milk molar, 1:8 in.

Greatest breadth, -4 in.
1:1 in.

Figs. 2 and 2a.—M. Sivalensis.

Length of first molar, “6 in.
Greatest width,

Portion of lower jaw, left side,

with symphysis and two outer mentary foramina, and containing first

(x) and second milk molars.—B.M.

Length of symphysis (oblique), 2° in.
symphysis, 2°5 in.
Length of second molar, 1:9 in.

Length of anterior or first molar, -4 in.
Greatest width, 1:2 in.

From first molar to anterior margin of
Greatest width, “4 in.
Length of fragment of

jaw, 66 in. Greatest breadth, 2°1 in. Height at anterior margin of second molar,

2:3 in.

Figs. 3 and 3a.—M. Sivalensis.

Portion of lower jaw, right side,

with symphysis and third milk, and fragment of fourth, or first true,

molars.1.—B.M.

Length of fragment, 12° in. Length of symphysis (ant. post.), 4-4 in. Greatest

width of fragment, 3°8 in.

Height at posterior margin of third molar, 4:2 in.

Length of third molar, 3:5 in. Greatest width, 2° in.

Figs. 4 and 4 a.—M. Sivalensis.

Fragment of lower jaw, left side,

with first ? true molar, imperfect anteriorly —B.M.

Breadth of jaw, 5:7 in.
Width, 3-2 in.

Figs. 5 and 5a.—M. Sivalensis.

portion of true molar well worn.
wear.
Figs. 6 and 6 a.—M. Sivalensis.

Height, 5-4 in.

Length of tooth (imperfect), 8: in.

Fragment of lower jaw with
Shows well the alternate dises of

Fragment of lower jaw, left side,

with penultimate ? true molar, imperfect behind.2A—B.M.

Width of fragment of jaw, 6:1 in.
tooth, 6°in. Width, 32 in.

Figs. 7 and 7 a.—M. Sivalensis.

Height, 61 in.

Length of fragment of

Fragment of lower jaw, right

side, with last true molar, much worn, and imperfect in front.—B.M.

Length of molar (imperfect), 8°4 in.

1 In the Museum of the Royal College
of Surgeons (Cat. No. 669) is a beautiful
specimen of ramus of right lower jaw
of a young M. Stvalensis, with the third
milk and first true molar. The anterior
tooth is a little broken and worn out in
front ; it shows six dises of wear and a
large talon ridge. The talon and 4 last
ridges are quite distinct, the 5th and
6th are worn out, and probably an
anterior talon was included with front
ridge. This would make 6 ridges and
a back and front talon, or 8 in all. The
tooth has little cement, the ridges are
low, and there is great plaiting of the
enamel plates in wear. Length of tooth,
4-2in.; width at last big ridge, 2°4 in.;
width at fourth ridge, 2°05in.; width
at second ridge, 1:9in. The posterior
tooth has 5 emerged ridges with a talon
in frevt, extending only from the outside,

Width, 3: in.

half way across, but thick. The rest of
the tooth is concealed. The two front
ridges and talon are slightly touched by
wear. Width of tooth at first mdge,
2:2in. Width at second, 2°3in. Length
from middle of first ridge to middle of
fifth (excluding talon), 3-in. The third
ridge shows 11 little points. No. 670
is the left side of the same Jaw with
corresponding teeth. The mentary for-
amen is very much in advance of the
fourth tooth, and placed low.—H.F.

? Another specimen of left lower jaw
of M. Sivalensis, with what is probably
the penultimate true molar, is in the
Museum of the Royal College of Sur-
geons (Cat. No. 690). It has four trans-
verse ridges and a heel of two points,
not 5 ridges as stated in catalogue.
Length of tooth, 5:9 in.; width at first
ridge, 2°4in.; at fourth, 2°5 in—H.F.

470 FAUNA ANTIQUA SIVALENSIS.

Figs. 8 and 8 a.—W. Sivalensis. Fragment of lower jaw, left side,
containing a very perfect specimen of the last lower molar. The
alternate disposition of the mammillz of the crown is finely exhibited.
Cast in B.M.

Length of tooth, 8°8in. Width, 2:9 in.

Figs. 9 and 9 a.—Mastodon Arvernensis. Fragment showing pos-
terior half of the last inferior true molar. The mammille form two
alternate rows as in.J/. Sivalensis, each ridge being composed of a
pair of points. From a cast in Museum of Geological Society.

Length of fragment of tooth, 56 in. Width, 3.8 in.

Puate XXXVIII.

Mastodon Perimensis (Fale. and Caut.), from Perim Island.
Front view of skull. Other views of same skull are given in Plate
XXXIX. figs. 1, 2, and 3.

This cranium is in many respects singularly perfect, although it has
suffered from a crushing force, which has forced in the temples, so as
to have contracted to a few inches the inter-temporal portion of the
forehead. The ascending ramus of the lower jaw on either side is
én situ with the coronoid process and condyle, and, what is more
remarkable, the greater part of the hyoid bone lies upon the
sphenoid. The atlas also was found attached to the condyles. The
teeth are completely hammered down to the margin of the alveoli.
The most remarkable character of all about this head is the low height
of the pterygoid processes of the sphenoid, which are very little higher
than the condyles, and the comparatively little elevation of the condyles
above the palate. The interval between the plane of the lower surface
of the condyles and that of the palate is only 5 inches, the height of the
occiput being 22 inches. This is very much asin the North American
Mastodon, and even more so, so that the plane of the grinder does not
differ much from that of the condyles, thus showing a tendency in the
direction of Dinotherium and the Trilophodon Mastodon Ohioticus.
The pterygoids rise with a sharp posterior border, and do not spread
out into a flap over the posterior border of the maxillary. They are
not rugous as M. Ohioticus, nor are they so far (proportionally) ex-
tended behind. There are two large palatine foramina near the end
of the molar. The molars (allowing perhaps for some distortion from
pressure) run parallel, and do not at any rate diverge in the remark-
able way exhibited by I. Ohioticus; perhaps they are less divergent
even than in MW. Stvalensis. The palate looks long. On either side
are two molars, the penultimate and last true. The tusks exhibit an
oval outline on section. Both zygomatic arches are entire. Presented
by Captain Fulljames to B.M.

Extreme length from occiput to broken incisives, 27-in. From posterior surface
of occipital condyles to commencement of diasteme, 25°5in, Extreme width of
occiput, 20°in. Height of occiput from condyles, 22°2in. From occiput to
broken tips of nasals, 13:in. From tips of incisives (anterior end of nasal open-
ing) to commencement of diasteme, 14°5 in. Width of nasal opening (approximate)
98in.; antero-post. diameter of nasal opening, 4°3in. Estimated width of brow
at post. orbitaries, 19°in. Width of brow at middle of orbits, 15:in. Width of
inter-incisive fossa, 2°2in. Depth of inter-incisive fossa, 3°4in. Contraction of
muzzle at orbitary foramen, 11°6in. Vertical diameter of right orbit, 4-8 in.
From the auditory foramen to the anterior border of the orbit, 16°5. Transverse

_ penultimate, 4-4 in.; width, 3-1 in.

DESCRIPTION OF PLATES. 471

diameter of left tusk, 2-9in. Vertical diameter of left tusk (approximate), 4: in.
Width across the condyles, 7-6in. Antero-post. diameter of condyle, 375 in.;
transverse diameter of condyle, 2°5in. Vertical height of condyle, 2°6 in. Antero-
post. diameter of occipital foramen, 2°8 in. ; transverse diameter of occipital foramen,
2-9 in. From anterior border of occipital foramen to niche of palate, 10-9 in.
From niche of palate to beginning of diasteme, 12°7 in. Height of the pterygoids
above the body of the sphenoid, 5-5 or 6:in. Width of outer surface of pterygoids,
77in. Length of two broken molars (surface), 11:7 in. Length of anterior or
Length of back molar, 7°4 in.; width, 3-2 in.
Width of palate between anterior molars, 1:6in. Width of palate, behind, 3°5 in.

Puate XXXIX.

Fig. 1.—Mastodon Perimensis. Lateral view of same skull as figured
in Plate XXXVIII., described above.-—B.M.

Fig. 2.— Mastodon Perimensis. Palate view of same skull.—B.M.

Fig. 3.—Mastodon Perimensis. Occipital view of same skull.—B.M.

Figs. 4, 5, and 6.—Mastodon Sivalensis. Fragment of small black
head, three different views. The specimen is very perfect in forms
without crushing, so far asit goes. The plane of the occiput meets
that of the frontal in a slightly rounded manner. The ligamentary
depression is placed about the middle of the occiput, and is not deeply
marked, consisting of a dividing crest, separating two diverging pits,
having a heart-shaped outline. The occiput is slightly convex across
and from base to top. The condyles do not project backwards as in
M. Perimensis. The posterior boundary of the temple (edge of
occiput) is inclined to be sharp. There is no tendency to occipital
bosses as in Elephant. The occiput in some respects resembles that of
M. Ohioticus. 'There is a very obtuse convexity or boss on the middle
of the forehead between the temples.—B.M.

Greatest width of occiput (the half doubled) 22° in. Height of occiput from
surface of condyles, 17: in. Contraction of brow between the temporals, 11°8 in.
Interval across the condyles, 7:1in. Antero-post. diameter of left condyle, 4:1 in.
Transverse diameter of left condyle, 2°5 in.

Fig. 7.—Mastodon Sivalensis. Fragment of upper jaw with two
molars, broken end of incisives, and anterior portion of zygoma. From
a cast in B.M.

PuaTe XL.
Figs. 1 and 1 a.—Mastodon latidens. Second? upper milk molar
with two ridges.—B.M.
Length, 1:9 in. ; width, 1:4 in.!
Figs. 2 and 2a.—WM. latidens.

Third? upper milk molar with
four ridges.—B.M.

Length, 3:in.; width, 1:8 in.

1 Tn the Museum of the Royal College | in front. The second milk molar is 2: in.

of Surgeons is the left side of the upper
jaw of a young Mastodon latidens con-
taining the first and second milk molars.
The anterior tooth is about 1-in. long,
and ‘8 in. wide, and has two ridges with
aheel. The main ridge is transverse ;
the anterior one is an obtuse cusp.
The tooth is oval, the sharp end being

long by about 1-5in. wide. It has three
main transverse ridges and a small
bourrelet ridge in front, and a heel
ridge behind. It expands very widely
in the direction of the orbit. A vertical
section shows something like the enamel
of another small tooth, 3 inch long.—H.F.

472 FAUNA ANTIQUA SIVALENSIS.

Figs. 8 and 3 a.—WM. latidens. Upper molar (first true ?) with four
ridges and back and front heel.—B.M.
Length, 4° in.; width, 2°3 in.

Figs. 4 and 4 a.—M. Perimensis. Fragment of upper molar showing
two ridges and part of a third. The valleys are transverse, but are
interrupted in the middle by an accessory lobule in front of and behind
each ridge, and the outer termination of each ridge is bounded by a
large mammilla, exactly as in Mastodon latidens—B.M.

Figs. 5 and 5a. M. Perimensis. Fragment of lower jaw with
portion of true molar, presenting a similar arrangement of mammille
to that noted under fig. 4. (See also antea, pp. 117 and 122.)

Length of fragment of molar, 5-4 in.; width, 3-2in.

Figs. 6 and 6 a.—Mastodon longirostris. Fragment of right lower
jaw of young calf showing the series of three milk molars in situ. The
third milk molar is nearly intact; the four ridges of which it is com-
posed are seen to be transverse, compressed, and composed of a number
of little points ; the valleys are open, with the exception of a tubercle
in the first, and two or three minute tubercles in the last valley, which
in no way interrupt their transverse continuity. The back talon formsa
low transverse free ridgelet as in the Mastodon latidens of India. -The
enamel is irregularly wrinkled, but exhibits no vertical fluting, as in
M. Arvernensis (fee Plate XXXVI. fig. 7). The original specimen
from Eppelsheim was formerly in the Harl of Enniskillen’s collection,
but is now in B.M. It is also figured by Kaup (‘ Oss. Foss. de Darm-
stadt,’ Plate XX. fig. 2.)

Length of first tooth, 1:2in. Width, ‘9in. Length of second tooth, 1:8in.
Width, 15in. Length of third tooth, 2°6in. Width, 2-1 in.

Figs. 7 and 7 a.— Mastodon angustidens. Third? milk molar upper
jaw, the crown consisting of three transverse ridges and an accessory
talon of two tubercles. A single tubercle juts out into each of the
hollows between the ridges alternately with the principal points,
accounting for the trefoil-shaped discs, which the worn teeth pre-
sent in this species, so different from the lozenge-shaped discs of J.
Ohioticus. This specimen is from Mr. Edward Charlesworth’s collec-
tion, but there is no history as to its origin.—B.M.

Length of tooth, 2°8in. Width, 1-6 in.

Figs. 8 and 8 a.—M. angustidens. Antepenultimate or first true
molar, having the crown divided into three distinct ridges, with a small
back talon.

Length, 4-6in. Width, 2°6 in.

Figs. 9 and 9 a.—M. angustidens. Penultimate molar of upper
jaw, consisting of three ridges and a talon appendage of two tubercles
behind. The two anterior ridges are affected by wear; the last is
almost intact. The intervals, wide and deep, have only a single
mammilla connecting the ridges, about the middle. The crown is
very simple, each ridge consisting of two pairs of points. The tooth
has a strong impression in front, is narrow in front and widens behind.
The drawing is taken from a cast in B.M. The original specimen was
what Cuvier commenced his account of the species, and it is figured by
him in ‘ Divers Mastodontes,’ p. 255, and Plate I. fig. 4. The dimen-

DESCRIPTION OF PLATES. 473

sions correspond exactly with those of Lartet’s Gers specimen, viz. :
Length, 4°5in. Width in front, 2-1in. Width behind, 2°6 in.

Figs. 10 and 10 a.—Mastodon Andium.! Fragment of upper true
molar. Presented by Lord Shelburne. Shows four ridges and portion
of a fifth.— B.M.

Length, 6°in. Width, 3°5 in.

Fig. 11.—M. Andium. Fragment of upper molar, with discs of three
ridges much worn.—B.M.

Length, 5°2in. Width, 3-2in.

Figs. 12 and 12 a.—M. Andium. Last upper true molar with four
ridges and a complicated heel— B.M.

Length, 6°8in. Width, 3°5 in.

Figs. 138 and 18 a.—M. Andium. Fragment of left lower jaw, with
second and third milk molars in situ. The specimen is broken at the
symphysis and coronoid process. From the relative size of the jaw and
the development of the teeth, the animal corresponded to a sucking
Indian elephant of about two years of age. The second milk grinder
is fully protruded, but had barely come into use, the two front ridges
being but slightly abraded. ‘The third is in the state of an intact germ,
and although fully formed, it had not penetrated the gum when the
animal died. These teeth are both three-ridged, with a subordinate
crest in front, and a small bi-tubercular talon behind. They are exactly
alike in form, narrow in front, but broader backwards. The ridges, as
in M. angustidens, consist of two pairs of principal points, which
instead of being nearly simple, as in the latter species, are subdivided
into a vast number of superficial warty tubercles, which jut into the
valleys, forming a bridge or connection between the contiguous ridges,
and interrupting the transverse continuity of the valleys. In this
respect they more resemble the young teeth of M. longirostris. Specimen
from Buenos Ayres in B.M. (Reproduced in Plate VIII. fig. 1.)

Length of fragment of jaw, 7'6in. Breadth, 2°7in. Height, 2°9in. Length of
second milk molar, 26in. Width, 1:4in. Length of third milk molar, 3°65 in.
Width, 1:7 in.

Figs. 14 and 14 a.—WMastodon Andium. Fragment of lower jaw
with last true molar, exhibiting four ridges and a complicated heel.
The three anterior ridges are touched by wear.

Length of tooth, 7°8in. Width, 3-2 in.

Figs. 15 and 15 a.—M. Andium. Greater portion of lower jaw, left
side, with the first and second true molars (penultimate and ante-
penultimate) 7m situ, showing also the empty alveolus of the last true
grinder and of the third milk tooth. The ternary number holds in
the ridges, there being three collines in each tooth, with an aggregate
nest of tubercles in the intervals. There are no remains of cement.
The anterior tooth is somewhat worn, the discs taking the form of a
quadrifoil. This specimen is from Chili, and was presented to the
Museum at Canterbury by General Miller; there is a cast of it in the

' Memorandum of head of M. Andium | between teeth in front, 4°4in.; interval
in British Museum, not figured. Shows | between teeth behind, 3°4in. Length
palate with molars. of last molar, left side, 8°75in.; width

Length of the anterior or inter-maxil- | in front, 3°75 in.; width behind, 3-25 in.
lary portion of palate, 15-in. Interval |

474 FAUNA ANTIQUA SIVALENSIS.

British Museum. It is of an age intermediate between the specimens
figured in Plate XXXY. 3, and Plate XL. 13. It is very valuable,
and is believed to be at present unique in Europe. (Reproduced in
Plate VIII. fig. 2.)

Length of fragment of jaw, 15°in. Breadth, 4:3in. Height, 4:1in. Length of
first molar, 4*in. Breadth, 2°6in. Length of second molar, 5-5in. Breadth,
2-7 in.

Fig. 16.—Mastodon Ohioticus. Fragment of upper jaw, with three
ridges and fangs of last upper molar, also empty cavity in jaw for fang
of fourth ridge. —B.M.

Length, 5‘lin. Width, 4:in. Height of crown and fang, 7: in.

Fig. 16 a.—M. Ohioticus. Last true molar, upper jaw, with four
main ridges and a heel; the first ridge only very slightly touched by
wear. In Mastodon Ohioticus the upper teeth are distinctly cleft
lengthwise into two divisions, each division being indistinctly composed
of a pair of confluent points. The plane of the tooth is oblique, sloping
from the outside, which is higher, to the inside, which is lower, and
this relation continues during the wear, the inside being the most worn.
The inner division, both anteriorly and posteriorly, throws off the
decurrent talon crests, but in the first two milk teeth the inner division
is smaller than the outer. Precisely the reverse is seen in lower jaw,
the inner ridge being the highest and remaining so during detrition,
while the outer is the lowest but least complex.—B.M.

Length, 7'3in. Width, 4° in.

PuatTe XLI.

Tusks of Proboscidea.! Fragments and sections.

Figs. 1 and 1 a.— Twisted fragment.— B.M.

Length measured along great curvature, 40-in. Direct length or chord of cur-
vature, 28°in. Circumference at proximal end, 12°7in. Circumference at distal
end, 13°2 in.

Fig. 2.—B.M.

Length, 56:in. Circumference at proximal end, 138°5 in.

Figs. 3 and 3 a.—B.M.

Length of fragment, 12°2in. Circumference, 12°2in. Greatest diameter, 3-7 in.

Figs. 4, 4 a, and 4 6.—B.M.

Length of fragment, 12°7in. Greatest diameter, 7°7in. Smallest diameter,
5 in.

Fig. 5.—B.M.

Greatest diameter, 9°8 in. Smallest diameter, 7:2 in.

Figs. 6 and 6 a.—Fragment of tusk in socket.

Length of socket, 13°5in. Circumference of tusk, 6°2in. Breadth of incisive
alveolar margin, 6:2 in.

Figs. 7 and 7 a.—B.M.

Length, 68:in. Circumference at proximal end, 11: in.

Figs. 8 and 8 a.—B.M.

Length of fragment, 83°in. Circumference at smaller end, 16:5 in.

1 There is no evidence of any attempt | the plate (with the exception of figs. 6

made by Dr. Falconer to determine the | and 25) are collected in one place in the |

species to which any of these specimens | British Museum,
belonged. All the specimens figured in

oe

ee ee ee eee eee

DESCRIPTION OF PLATES. 475

Figs. 9 and 9 a.—B.M. |
* Length, 11-6in. Great diameter, 74in. Small diameter, 5:9 in.
Figs. 10 and 10 «.—B.M.
Length of fragment, 12°4in. Circumference, 16°6 in.
Figs. 11 and 11 a.—B.M.
Length of fragment, 14:7in. Circumference at upper end, 21:2 in.

Figs. 12 and 12 a.—B.M.

_Length along great curvature, 23'in. Greatest circumference, 12°in. Least
circumference at tip, 4°8in.

Figs. 13 and 13 a.—B.M.
Length, 20°in. Greatest circumference, 17: in,
Figs. 14 and 14 a.—B.M.

Length of fragment, 12°7 in. Great diameter of section, 5'in. Smaller diameter
of section, 4°4 in.

Figs: 15 and 15 a.—B.M.
Length of fragment, 11:8in. Greatest circumference, 11°65 in.
Figs. 16 and 16 a.—B.M.
Length of fragment, 7°S8in. Greatest circumference, 9°2 in.
Figs. 17 and 17 a.—B.M.
Length, 9:6 in. Greatest circumference, 7°8 in.
Figs. 18 and 18 a.—B.M.

Length, 10°3in. Greatest circumference, 13°5 in.

Figs. 19 and 19 a.—B.M.

Length, 9:°2in. Circumference, 13°5 in.
Figs. 20 and 20 a.—B.M.
Length, 9*in. Circumference, 19:2 in.
Figs. 21 and 21 a.—Matrix with fragments of two tusks.—B.M.
Length of fragment, 22°5 in. Proximal end of left tusk, 5-1 by 4-1 in. Proximal
end of right tusk, 5"1 by 48in. Distal end of left tusk, 3°9 by 8°in. Distal end
of right tusk, 4°38 by 4°2 in.
Figs. 22 and 22 a.—B.M.
Length, 14:5 in. Circumference, 13°6 in.
Figs. 23 and 23 a.—Fragment of jaw with alveolus of left tusk, and
part of right tusk —B.M.

Length of fragment of jaw, 13°7in. Between external alyeolar margins of
incisors, 20°4in. Great diameter of tusk, 6‘7in. Lesser diameter of tusk, 5°7 in,

Fig. 24.—B.M.
Length of fragment of upper jaw, 14:2 in.

Figs. 25 and 25 a.—Unequal fragments of two tusks joined together.

Length of great fragment, 10°2in. Length of small fragment, 78in. Circum-
ference of great fragment, 16-in. Circumference of smaller fragment, 15°3 in.

Prates XLII. and XLUI.
Anterior views of skulls of Proboscidea, restored, in outline.
Prates XLIV. and XLV.

Profile views of skulls of Proboscidea, restored, in outline.

476

FAUNA ANTIQUA SIVALENSIS.

In these four plates! the skulls are classified, and are arranged in a
series forming a transition from one to the other, as follows: ?

I. TRILOPHODONTES.

1.—Dinotherium giganteum (after Kaup), with two large deflected

tusks in lower jaw.

2.—Dinotherium Indicum (not figured).

5.—Mastodon Tapiroides (not figured).

4.—Mastodon Ohioticus * (copied from American Phil. Transactions,
1838, vol. viii. Plate II., adding a tusk to lower jaw).

5.—Mastodon angustidens (De Blainville’s Ostéographie, Plate III.).

6.—Mastodon Andium (British Museum specimen).

II. TETRALOPHODONTES.
7.—Mastodon Perimensis (Indian collection, see Plates XXXVIII.

and XXXIX.).

8.—Mastodon Sivalensis (Indian collection, see Plates XXXII,

XXXIII., and XXXIV.).

9.—Mastodon Arvernensis (after Nesti, imperfect).
10.—Mastodon longirostris (after Kaup, imperfect).
11.—Mastodon latidens (not figured).

III. Srecopon tes.

12.—Elephas Cliftit, Clift’s specimen, very imperfect.
13.—Elephas bombifrons (Indian collection, see Plate XXVII.).
14.—Elephas Ganesa (Col. Baker’s huge cranium in British

Museum, see Plate XXI.).

1 Reproduced in vol. ii.

? Note by Dr. F. in 1857.—‘ The views
which we entertain were fully elucidated
in 1847 in the four plates of outline-
heads, from Plate xlii. to xlv. of the
“Fauna Antiqua Sivalensis,” where a
synopsis is given of all the species, fossil
and recent, then known. The forms
included under the nominal species of
M. angustidens of Cuvier, are there
ranged as four distinct species, viz. :—
M. (Triloph.) angustidens, M. (Triloph.)
Andium, M. (TLetraloph.) longirostris,
and M. (Tetraloph.) Arvernensis. The
only change which subsequent investiga-
tion on fresh materials has led us to
make is to transfer M. Andium from the
subgenus Trilophodon into that of
Tetralophodon.’ In 1863, however, Dr.
F. expressed the opinion that M. Andiwm
would, after all, prove to belong to the
Trilophodon group (Memoir on Elephas
Columbi in vol. ii.).

$ Memorandum by Dr. F. on broken
head of Mastodon Ohioticus—‘ The
occiput forms a vertical plane, the con-
dyles being right under the base, and
not projecting behind. The crista galli
is not very large, and the ligamentary
depression is shallow with divergent

across occipital condyles, 8-7 in.

lobes, broad above and narrow below.
The pterygoid ale of the sphenoid,
instead of overlapping the maxillaries
by a conical lamina, rise up in a rough
rugous stem, and are much behind the
last teeth, which are very divergent.
The pterygoids are low, but not more so
than (if so much as) in M. Perimensis.

‘Extreme length of cranium from
occiput to incisive tips, 34-2in. Width
of brow at post orbitaries, 19-6 in. Width
at contraction of muzzle near sub-orbitary
foramen, 15-2in. Width of nasal open-
ing, 5‘4in. Antero-posterior diameter
of orbit, 5-6in. From anterior margin
of orbit to occipital plane, 22-in. Width
From
anterior margin of occipital hole to niche
of palate, 11‘5in. Height of the ptery-
goid alae from Vidian hole, 8-5in. From
back of molar to edge of pterygoid, 4:2
in. Length of molar with four ridges
and a heel, 6-7 in. Width of palate in
front of penultimate teeth, 3-9 in. Width
behind, 3:9 in. Length of palate from
niche to diasteme, 13-in. Interval be-
tween diastemal ridges at commence-
ment, 4°5in. Interval between tips of
divergence, 5-6 in.!’

DESCRIPTION OF PLATES.

477

15 a.—Elephas insignis, old (Indian collection, see Plate XV.).
15 b.—Elephas insignis (Indian collection, see Plate XVII., fig 1).
15 c.—Elephas insignis, young (Indian collection, see Plate XVIII.

fig. 3).

TV. Loxoponres.

16.—Elephas planifrons (Indian collection, see Plate IX.).
17.—Elephas Africanus (recent head).
18.—Hlephas priscus (not figured).

V. ELAsMopontEs.!

19.—Elephas meridionalis ? (after Nesti).
20 a.—Elephas Hysudricus (Indian collection, see Plate IV.).

20 6.—Hlephas Hysudricus, young

VI).

(Indian collection, see Plate

21.—Elephas antiquus* (not figured).
22.—Elephas Namadicus (Indian collection, see Plate XII. A.).
23 a.—LElephas Indicus * (Dauntela var.).

1 The designation of H/asmodus haying
been preoccupied by Sir Philip Egerton
for a series of fossil fish, Dr. F., in 1857
substituted the term welephas for
Elasmodon.

?This is erroneously designated F.
antiquus in the plate (see note, page 4438).
The illustration is taken from Nesti’s
figure. The skull is nearly perfect in
the frontal and occipital regions, con-
dyles, maxillaries, and molars, but im-

perfect in the facial portion, the border
of the nasal opening being broken, to-
gether with the terminal portion of the
incisive alveoli and zygomatic arches.
The line formed by the posterior border
of the vertex is transverse, the fossa
being overarched by a produced fold of
the vertex.

3 Hlephas meridionalis in Plate (see
note, page 443),

4 Comparison between Mukna and Dauntela varieties of Elephas Indicus.

Extreme length of cranium
Width between zygomatics
Ditto

Greatest width of occiput

Width of nasal opening

Depth of ditto. a

Width of tusk-sheaths

Narrow width of brow

Depth of orbit ;
Height from condyles to occiput 0
Across condyles .

From condyles to tip of tusk-sheath ;
From ditto to anterior margin of molar alveolus .
From anterior margin occipital hole to posterior

border palate .
Length of palate

post-orbitary processes :
Length from niche of occiput to tips of nasals

Depth of head from condyles to frontal surface at

middle, opposite nasal opening

Height from diastemal surface to bulge of occiput

Length of condyloid surface
From ear-hole to top of occiput .

Length of anterior tooth, upper jaw

Width ditto ditto
Number of plates, about ten

Mukna (big head). Dauntela.
Inches Inches
40°5 41°5
- r ¢ 29°5 29°75
22°5 24-0
22°0 21:0
0 . 0 31:0 z 30°5
. . » 14:0 . 15:0
5°78 5°75
14:25 17°5
13°5 13°25
9 6°5 6°5
22°25 22°0
8°75 8°5
33°0 39°25
22°5 22°25
. . 6 12°0 é 11°5
8:0 A 9°5
23°5 ; 23°5
30°75 31°25
. 5 6°5 A 6°25
¢ 19:0 19:0
: : 8°37
5 t 5 30

} dropped out

478 FAUNA ANTIQUA SIVALENSIS.

23 b.—Elephas Indicus (Mukna var.).
23 c.—Elephas Indicus (young).
24.—Elephas primigenius (Fischer’s drawing).

Puate XLVI.!

Figs. 1 to 11.—Atlases of Proboscidea.

Figs. 1, 1 a, and 1 0.

Between extreme points of transverse processes, 16°7 in. Between extreme
points of anterior articular surfaces, 8°5 in. Height, 9°3 in. Length of inferior
arch, under surface, 2°8 in.

Figs 2, 2 a, and 2 b.

Between extreme point of anterior articular ‘surfaces, 8°6 in. Height, 8°7 in.
Length of inferior arch (inferiorly), 2°7 in. Height of orifice, 43 in. Breadth
of spinal canal, 3:2 in. Narrowest part, 2°1 in. Breadth of fossa for odontoid
process, 2°5 in.

Figs. 3, 3 a, and 3 b.

Between. extreme points of anterior articular surfaces, 9:0 in. Height, 8°1in.
Height of orifice, 8°8 in. Breadth of odontoid fossa, 2°5 in. Breadth of spinal
canal, 3°8 in.

Figs. 4, 4 a, and 4 b.

Between extreme points of transverse processes, 14°8 in. Between extreme
points of anterior articular surfaces, 7°5 in. Height, 7-7 in. Length of inferior
arch (below) antero-posterior, 2°5 in. Great diameter of vertebral foramen, 1:2 in.
Length of superior arch (antero-posterior), 3:2 in,

Fig. 5.
Height, 7°7 in.

Fig. 6.

Between extreme points of anterior articular surfaces, 7°8 in. Height, 6°6 in.

Figs. 7, 7 a, and 7 b.

Between extreme points of transverse processes, 14°7 in. Between extreme
points of anterior articular surfaces, 8-7 in. Height, 7-6 in. Antero-posterior
of lower arch (inferiorly), 2°6 in.

Fig. 8.

Between extreme points of transverse processes, 17°4 in. Height, 8°6 in.

Comparison between Mukna and Dauntela—continued.
Mukna (big head). Dauntela.

i ae 4 hd : esis . Inches
ength of anterior tooth, lowerjaw . A 2 :
Width ditto ditto : : 2 : 3°37 f daopped out
Length of lower jaw from tip to eg edge at

middle of height . Z : - 22°5 25°
Height of jaw to alveolus . : : 6 ; 6:5 : 65
Thickness of jaw (diam.) . c . : : 637 =. 6:
Height of condyles : : : : : 5 20° . 18°
Height to coronoids . : < F A 15°5 7 12°5
Antero-posterior diameter leaf. a : : 13°5 3 10°25
Interval between condyles, inside . j : 14: 5 14:
Ditto ditto outside : : : : 21°5

The plates of teeth in the Mukna| the British Museum, except where the
variety slope greatly backwards and are | contrary is stated. Except in a few in-
excessively and finely crimped ; those of | stances, no attempt was made by Dr. F. to
Dauntela are much less crimped. determine the species of Proboscidean to

1The specimens figured in Plates | which the bone belonged.
xlyi. to lvi. inclusive, are mostly in

DESCRIPTION OF PLATES. 479

Fig. 9.

Between extreme points of anterior articular surfaces, 7°8 in. Height, 7°7 in.

Fig. 10.

Between extreme points of anterior articular surfaces, 10:4 in. Height, 10° in.
Length antero-posterior of superior arch, 3°6 in. Length antero-posterior of
inferior arch, 3°1 in.

Fig. 11.

Between extreme points of anterior articular surfaces, 9°2 in. Height, 6:9 in.
Antero-posterior of superior arch, 2°8 in. Antero-posterior of inferior arch,
2:1 in.

Fig. 12.—Basilar process of occipital bone with condyles and foramen
magnum of a Proboscidean.

Between extreme points of occipital condyles, 10:0 in. Vertical diameter of

condyle, 4°5 in. Transverse diameter of foramen magnum, 3°5 in. Vertical
diameter of foramen magnum, 3:1 in.

Puate XLVI.
Axes and other vertebree of Proboscidea.
Figs. 1 and 1 a.—Axis.

Length of body inferiorly, including the odontoid, 6:0 in. Breadth of body
posteriorly, 6°5 in. Height of the posterior surface of body, 5-1 in. Height
of spinal canal, 2°6 in. Breadth of spinal canal, 2:2 in. Height of upper surface
of spine from inferior surface of the body, posteriorly, 10°6 in. Antero-posterior
diameter of spinal platform, 4°5 in.

Figs. 2 and 2 a.—Axis.

Length of body inferiorly, including the odontoid, 7-4 in. Breadth of body
posteriorly, 6°8 in. Height of posterior surface of body, 6°3 in. Height of spinal
canal, 2°8 in. Breadth of spinal canal, 2°8 in. Height of upper surface of spine
from inferior surface of the body posteriorly, 13:5 in. Antero-posterior diameter
of spinal platform, 5°3 in. Between extreme points of transverse processes, 12: in.

Figs. 3 and 3 a.— Axis.

Length of body, 4-6 in. Breadth of body posteriorly, 4:5in. Height of body,
4°5 in.

Figs. 4 and 4 a.—Axis.

Length of body, 6:8 in. Breadth of body posteriorly, 6°3 in. Height of body,
5°7 in.

Figs. 5 and 5 a.— Axis.

Length of body, 6-4 in. Breadth of body posteriorly, 5:5 in. Height of body,
5°3 in.

Figs. 6 and 6 a.—Axis.

Length of body, 4:0 in. Breadth of body posteriorly, 4-4in. Height of body,
3°6 in.

Figs. 7 and 7 a.— Axis.

Length of body, 6°3 in. Breadth of body posteriorly, 7-0 in. Height of body,
6:0 in.

Figs. 8 and 8 a.—Axis.

Length of body, 56 in. Breadth of body posteriorly, 6-4 in. Height of body,
5°5 in.

Figs. 9 and 9 a.—Axis.

Length of body, 6:0 in. Breadth of body posteriorly, 6.3 in.

Fig. 10.—Eight vertebre conjoined. They are the posterior cervical

and anterior dorsal.

480 FAUNA ANTIQUA SIVALENSIS.

Fig. 11.—Fragment of dorsal vertebra.

Height of posterior surface of body, 5-6 in. Breadth of posterior surface at
inferior angles of costal pits, 56 in. Length of body inferiorly, 2:7 in.

Fig. 12.—Dorsal vertebra.

Between extreme points of transverse processes, 12°7 in. Height of body
posteriorly, 5°5 in. Breadth of body posteriorly, 6°7 in. Length of body
inferiorly, 2°8 in.

Fig. 138.—Dorsal vertebra.

Between extreme points of transverse processes, 13°2 in. Height of body
posteriorly, 56 in. Breadth of body posteriorly, 6-2 in. Length of body
inferiorly, 3°1 in.

Fig. 13 a.—Dorsal vertebra.

Between extreme points of transverse processes, 14° in. Height of body
posteriorly, 6°7 in. Length of body inferiorly, 3-7 in.

Fig. 14.—Lumbar vertebra.

Height of body posteriorly, 5: in. Breadth of body posteriorly, 5-5 in. Length
of body inferiorly, 3: in. Between extreme points of transverse processes, 8°5 in.

Fig. 15.—Lumbar vertebra.

Height of body posteriorly, 3°7 in. Breadth of body posteriorly, 4°2 in. Length
of body inferiorly, 3°2 in. Between extreme points of transverse processes, 6°5 in.

Fig. 16.—Portion of sacrum, comprising three upper sacral vertebra
and portion of a fourth.

Length of fragment, 13-3 m. Length of three upper sacral vertebre, 10°2 in,
Between extreme points of transverse processes, 10°4 in. Height of body (upper,
right), 3°5 in. Transverse of body (upper), 6:1 in.

Puate XLVUI.

Bones of anterior extremity of Proboscidea.

Figs. 1, 1 a, and 1 6.—Elephas Namadicus, from the Nerbudda.
Upper end of shaft, and articulating extremity of left humerus.

Length of fragment, 29°6 in. Transverse diameter of upper extremity, 14-4 in,
Antero-posterior diameter of great tuberosity, 12°2in. Antero-posterior diameter of
head (articular surface), 11°8 in. Transverse diameter, 8° in. Smallest trans-
verse diameter of shaft in centre, 10°3 in. Smallest antero-posterior diameter of
shaft, in centre, 3°7 in.

Figs. 2, 2 a, and 2 b.—E. Namadicus. Fragment of shaft and
upper articulating extremity of united radius and ulna, left side.
Specimen formerly in United Service Museum.

Extreme length, 40:in. Extreme width below sigmoid cavity in fig. 2 a, 10: in.
Depth from upper and back part of olecranon to anterior angle of sigmoid cavity,
fig. 26, 14°5 in.

Figs. 3, 3 a, and 8 6.—Fragment of shaft and upper end of left
humerus of a Proboscidean.

Length of fragment, 18-8 in. Transverse diameter of upper extremity, 12-2 in.
Antero-posterior diameter of great tuberosity, 11- in.

Fig. 4.—Upper articulating end of left humerus.

Length, 13°5 in. Transverse diameter, 10°5 in. Antero-posterior diameter of
great tuberosity, 11: in.

Fig. 5.—Upper articulating end of left humerus.

Length of fragment, 18-2 in. Transverse diameter of upper end, 13-in. Antero-
posterior diameter of great tuberosity, 10°5in. Antero-posterior diameter of articular
surface of head, 9°8 in. ‘Transverse diameter of articular surface of head, 6°4 in.

DESCRIPTION OF PLATES. 481

- Fig. 6.—Upper articulating end of left humerus.

Length, 13:in. Transverse diameter of upper end, 12° in, Antero-posterior
diameter of great tuberosity, 10-5 in.

Fig. 7.—Upper articulating end of right humerus.

Length, 15"in. Transverse diameter of upper end, 12°7 in. Antero-posterior
diameter of articular surface, 9:2 in. Transverse diameter of articular surface, 7°5 in.

Fig. 8—Upper articulating end of left humerus.

Length, 11-6 in. Transverse diameter of upper end, 9° in. Antero-posterior
diameter of great tuberosity, 8°5 in.

Fig. 9.—Upper articulating end of right humerus.

Length, 9°5 in. Transverse diameter of upper end, 9°5 in. Antero-posterior
diameter of great tuberosity, 9°5 in. Antero-posterior diameter of articular
surface, 7-9 in. Transverse diameter of articular surface, 6:2 in.

Fig. 10.—Upper articulating end of left humerus.

Length, 11:4 in. Transverse diameter of upper end, 9:4 in. Antero-posterior
diameter of great tuberosity, 7°7 in. Antero-posterior diameter of articular surface,
68 in. Transverse diameter of articular surface, 5-5 in.

Figs. 11, 11 a, and 11 6.—Lower end of shaft and articular surface of
left humerus.

Length of fragment, 21:5 in. Breadth of inferior extremity, including external
eondyloid ridge, 11:in. Breadth of trochlear surface, 9°5 in. Smallest antero-
posterior diameter of trochlear surface, 4:4 in. Circumference at broken extremity,
18-7 in.

Figs. 12, 12 a, and 12 b.—Lower end of shaft and articular surface
of left humerus.

Length, 14:3 in. Breadth of inferior extremity, including external condyloid
ridge, 10:2 in. Breadth of trochlear surface, 8-6 in. Smallest antero-posterior
diameter of trochlear surface, 4°1 in.

Figs. 15, 13 a, and 15 b.—Lower end of shaft and articular surface
of right humerus.

Length, 17-3 in. Breadth of inferior extremity, including external condyloid
ridge, 12° in. Breadth of trochlear surface, 9:°3 in. Smallest antero-posterior
diameter of trochlear surface, 4°7 in.

Figs. 14, 14 a, and 14 b.—Lower end of shaft and articular surface
of left humerus.

Length, 14: in. Breadth of inferior extremity, including external condyloid
ridge, 14-7 in. Breadth of trochlear surface, 11:9 in. Smallest antero-posterior
diameter of trochlear surface, 4-4 in.

Fig. 15.—Lower articulating end of left humerus.

Breadth of inferior extremity, including external condyloid ridge, 13: in.
_ Breadth of trochlear surface, 10: in. Smallest antero-posterior diameter of troch-
lear surface, 4°5 in.

Fig. 16.—Lower articulating end of left humerus.

_ Breadth of inferior extremity, including external condyloid ridge, 9: in. Breadth
of trochlear surface, 7°3 in. Smallest antero-posterior diameter of trochlear sur-

face, 3°3 in.

Fig. 17.—Lower articulating end of right humerus.

Length, 135 in. Breadth of inferior extremity, including external condyloid
ridge, 11: in. Breadth of trochlear surface, 9°8 in. Smallest antero-posterior dia-
meter of trochlear surface, 5: in.

Fig. 18.—Lower articulating end of left humerus.
Breadth of inferior extremity, including external condyloid ridge, 9-7 in.

KOL. I. II

452 FAUNA ANTIQUA SIVALENSIS.

Breadth of trochlear surface, 8°5 in. Smallest antero-posterior diameter of troch-
lear surface, 3°5 in.

Fig. 19.—Lower articulating end of right humerus.

_ Length, 15: in. Breadth of inferior extremity, including external condyloid
ridge, 12-4 in, Breadth of trochlear surface, 9°7 in. Smallest antero-posterior
diameter of trochlear surface, 4° in.

Fig. 20.—Lower articulating end of right humerus.

Breadth of inferior extremity, including external condyloid ridge, 12° in.
Breadth of trochlear surface, 9°6 in. Smallest antero-posterior diameter of troch-
lear surface, 4°7 in.

Fig. 21.—Lower articulating end of right humerus.

_ Length, 15:7 in. Breadth of inferior extremity, including external condyloid
ridge, 12-4 in. Breadth of trochlear surface, 10°1 in. Smallest antero-posterior
diameter of trochlear surface, 4:3 in.

Figs. 22, 22 a, and 22 b.—Upper end of shaft and articular surface
of united radius and ulna, left side.

Length, 22-4 in. Width of upper end or head (fig. 22 a), 7-4in. Depth (fig. 220),
10°5 in. :

Figs. 23 and 23 6.—Upper end of shaft and articular surface of
united radius and ulna, right side.

Length, 17: in. Width of upper end or head, 7°5 in. Depth (0), 12°2 in.

Fig. 24.—Upper articular surface of radius and ulna, right side.

Width of upper end or head, 9°6 in. Depth, 12-7 in.
Fig. 25.—Upper articular surface of radius and ulna, left side.
Width of upper end or head, 7° in. Depth, 10-8 in.

ig. 26.—Upper articular surface of radius and ulna, left side.
Width, 7°Sin. Depth, 10-2 in.

Fig. 27.—Upper articular surface of radius and ulna, left side.
Width, 6-8 in. Depth, 10: in.
Fig. 28.—Upper articular surface of radius and ulna, left side.
Depth to internal angle of ulna, 10-7 in.

Figs. 29 and 29 a.—Upper articular surface of radius and ulna,
right side.
Length, 12°5 in. Depth to internal angle of ulna, 11°3 in.

Puate XLIX.

Bones of anterior extremity of Proboscidea.
Fig. 1.—Fragment of right scapula, showing spine, glenoid
cavity, &e.

Length of fragment, 22°7 in. Breadth of fragment (greatest), 13°in. Greatest
height of spine above infra-spinous fossa, 7-2 in. Greatest diameter of glenoid
cavity, 7°5 in. Lesser diameter of glenoid cavity, 5° in.

Figs. 2 and 2 a.—Fragment of right scapula, showing spine and
glenoid cavity.

Length of scapula, 31‘in. Breadth of fragment, 11:2 in, Height of spine,
75 in. :
Fig. 3.—Fragment of right scapula, including glenoid cavity.

Length of fragment, 14: in. Greatest diameter of glenoid cavity, 6°6 in. Lesser
diameter of glenoid cavity, 3°6 in.

DESCRIPTION OF PLATES. 483

Figs. 4 and 4 6,—Fragment of right scapula, including glenoid
cavity.

Length of fragment, 14°8 in. Greatest diameter of glenoid cavity, 9:4 in.
Lesser diameter of glenoid cavity, 5°5 in.

Figs. 5, 5 a, and 5 6.—Upper end of shaft and articular head of
right humerus.

Extreme length, 22: in. Transverse diameter of upper extremity, 9°7 in. An-
tero-posterior of great tuberosity, 8°5 in. Antero-posterior of articular surface of
head, 7°3 in. Transverse surface of head, 4:2 in. Smallest transverse diameter of
shaft, 4-7 in. Smallest antero-posterior diameter of shaft, 3-2 in.

Figs. 6, 6 a, and 6 6.—Lower end of shaft and articular surface of
left humerus.

Length, 17-6 in. Breadth of inferior extremity, including condyloid ridge,

12: in. Transverse diameter of trochlea, 9:3 in. Smallest antero-posterior diam,
of trochlea, 4-2 in.

Figs. 7, 7 a, and 7 6.—Lower end of shaft and articular surface of

- left humerus.

Length, 14: in. Breadth of inferior extremity, including condyloid ridge, 10: in.
Transverse diameter of trochlea, 86 in. Smallest antero-posterior diameter of
trochlea, 4°3 in.

Figs. 8, 8a, and 8 b.—Upper articulating extremity of left radius and

na.

Length, 19°2 in. Breadth of sigmoid cavity, 9°7 in. Depth or greatest oblique
diameter from before backwards (a), 129 in.

Fig. 9.—Lower end of right radius and ulna.

Length, 14 in. Width of united extremities, 11: in. Width of ulna, 7°3 in.

Figs. 10 and 10 a.—Lower end of left radius and ulna.
Length, 12:5 in. Width of united extremities, 11:4 in. Width of ulna, 6°2 in.
Width of radius, 4°6 in.
Figs. 11 and 11 a.—Lower end of right radius and ulna. This
specimen is from Perim Island.
Extreme length, 14:2 in. Width of conjoined ends, 10: in. Width of ulna, 4-5 in.
Width of radius, 6: in.
Figs. 12 and 12 a.—Lower end of right radius and ulna.
Length, 10°4 in. Breadth of inferior extremity, 6°3 in.

Figs. 15 and 13 a.—Lower end of left ulna.
Length, 13:5 in. Breadth of inferior extremity, 5°9 in. Antero-posterior
diameter of inferior extremity, 53 in.
Figs. 14 and 14 a.—Lower end of right ulna.
Length, 85 in. Breadth of inferior end, 6-4 in. Antero-posterior diameter of
inferior end, 56 in.
Fig. 15.—Lower end of right radius and ulna. The radius is to
right of figure.
Length, 12:2 in. Breadth of conjoined extremities, 10°5 in.
Fig. 16.—Lower end of right radius and ulna.
Breadth of conjoined extremities, 9°4 in. Breadth of ulna, 5:5 in. Breadth of
radius, 4°3 in.
Fig. 17.—Lower end of right ulna.
Breadth, 5:3 in.
Fig. 18.—Lower end of right ulna.
Breadth, 6-2 in.
II 2

484 FAUNA ANTIQUA SIVALENSIS.

Fig. 19.—Lower end of right radius.
Length, 11°8 in. Breadth of inferior extremity, 6°4 in.
Fig. 20.—Lower end of left radius.
Breadth, 4-2 in. Antero-posterior diameter, 5:8 in.
g. 21.—Lower end of right radius.
Breadth, 5°5 in.
Fig. 22.—Lower end of left ulna.
Breadth, 5°6 in.
Fig. 23.—Lower end of right radius.
Breadth, 3°9 in.
Fig. 24.—Lower end of right radius.

_Breadth of inferior extremity, 4°3 in. Extreme length, 8:8 in. Antero-posterior
diameter of inferior extremity, 5-2 in.

Fig. 25.—Lower end of left ulna.

Length, 10°5in. Breadth of inferior extremity, 5-8 in. Antero-posterior diam.
of inferior extremity, 5°4 in.

Fig. 26.—Lower end of left radius.
Breadth, 3°4 in.
Fig. 27.—Lower end of left radius.
Length, 17°7 in. Breadth of inferior extremity, 4° in.
Fig. 28.—Lower end of left radius.
Breadth, 4°8 in.
Fig. 29.—Lower end of left ulna.
Breadth of inferior extremity, 5°3in. Antero-posterior of inferior extremity, 4°3 in.
Fig. 30.—Lower end of left ulna.
Breadth, 4°8 in.
Fig, 31.—Lower end of left ulna.
Breadth, 5°5 in. Antero-posterior diameter, 5:5 in.
Fig. 32.—Lower end of right ulna. J
Breadth, 4° in.
Fig. 33.—Lower end of left ulna.
Breadth, 7:1 in. Antero-posterior diameter, 4°4 in.
Fig. 834.—Lower end of right ulna.
Breadth, 7: in. Antero-posterior diameter, 6: in.
Fig. 35.—Lower end of left ulna.

Breadth, 7:1 in. Antero-posterior diameter, 674 in.

Fig. 86.—Lower end of right ulna.

Breadth, 7-4 in. Antero-posterior diameter, 6°8 in.

Puate L.
Bones of anterior extremity of Proboscidea.

Fig. 1.—Lower end of right radius and ulna, with bones of carpus
(semilunar, trapezoid, os magnum and unciform) and metacarpus
(second, third, and fourth).

Length of fragment of ulna, 10°9 in. Length of fragment of radius, 9:2 in.
Breadth of semilunar, 4°7 in. Vertical diameter of semilunar in centre, 2-5 in,
Vertical diameter of trapezoid, 26 in. Transverse diameter of trapezoid, 3:2 in.
Transverse diameter of os magnum, 3° in. Vertical diameter of os magnum, 2-9 in,

a Oe Se ee, ae |

eS

DESCRIPTION OF PLATES. 485

Transverse diameter of unciform, 4° in. Vertical diameter of unciform, 3° in.
Length of second metacarpal, 4:5 in. Breadth of second metacarpal, 2°65 in.
Length of third metacarpal, 42 in. Breadth of third metacarpal, 3:3 in. Length
of fourth metacarpal, 2°5 in. Breadth of fourth metacarpal, 2°5 in. Breadth of
inferior extremity of radius, 4°8 in. Breadth of inferior extremity of ulna, 7° in.
Figs. 2, 2a, and 26.—Lower end of left ulna (a), with bones of
carpus (scaphoid (1), semilunar (2), cuneiform (3), pisiform (4), tra-

_pezium (5), trapezoid (6), os magnum (7), and unciform (8) ), and

middle metacarpal (9).

Breadth of lower extremity of ulna, 5°5in. Vertical diameter of scaphoid,
51in. Antero-posterior diameter of scaphoid, 3-9 in. Transverse diameter of
scaphoid, 2°1 in. Vertical diameter of semilunar, 2°9 in. Antero-posterior dia-
meter of semilunar, 5:2 in. Transverse diameter of semilunar, 5°6 in. Vertical
diameter of cuneiform, 2°8 in. Antero-posterior diameter of cuneiform, 4°7 in.
Transverse diameter of cuneiform, 7-in. Vertical diameter of pisiform, 5:3 in.
Transverse diameter of pisiform, 3°4in. Antero-posterior diameter of pisiform,
21in. Vertical diameter of trapezium, 3°7in. Transverse diameter of trape-
zium, 4°6in. Antero-posterior diameter of trapezium, 3:3 in. Vertical diameter
of trapezoid, 2°3 in. Transverse diameter of trapezoid, 3-3 in. Antero-
posterior diameter of trapezoid, 3°5in. Vertical diameter of os magnum, 3°1 in.
Transverse diameter of os magnum, 3°2in. Antero-posterior diameter of os mag-
num, 4‘8in. Vertical diameter of unciform, 3°3in. Transverse diameter of unci-
form, 5-in. Antero-posterior diameter of unciform, 4:3in. Transverse diameter of
medius metacarpal, 2°7in. Antero-posterior diameter of medius metacarpal, 4° in.

Figs. 3 and 3 a.—Right scaphoid.
Figs. 4 and 4 a.—Left scaphoid.

Length, 5:6 in. Breadth, 3°8 in. Thickness, 2°6in.
Figs. 5 and 5 a.—Left scaphoid.
Length, 4°9 in. Breadth, 4:1 in. Thickness, 2°3 in.

Figs. 6, 6 a, and 6 6.—Right semilunar.

Height, 2°8in. Breadth, 5:3 in. Antero-posterior diameter, 5:in.
Figs. 7, 7 a, and 7 6.—Left semilunar.

Height, 28in. Breadth, 4-2in. Antero-posterior diameter, 4:6 in.
Figs. 8, 8 a, and 8 b.—Right semilunar.

Height, 2°5in. Breadth, 4-4in. Antero-posterior diameter, 4:5 in.
Figs. 9, 9a, and 9 d.—Right semilunar.

Height, 2°7 in. Breadth, 4:4in. | Antero-posterior diameter, 4°5 in.
Figs. 10, 10 a, and 10 6.—Right semilunar.

Height, 3: in. Breadth, 5-4in. Antero-posterior diameter, 5-8 in.
Figs. 11, 11 a, and 11 b.—Right semilunar.

Height, 2°7in. Breadth, 4-4in. Antero-posterior diameter, 4:8 in.
Figs. 12, 12 a, and 12 6.—Left semilunar.

Height, 2-7in. Breadth, 46in. | Antero-posterior diameter, 4:2 in.
Figs. 13, 13 a, and 18 6.—Right semilunar.

Height, 3° in. Breadth, 5-5in. Antero-posterior diameter, 5: in.
Figs. 14, 14 a, and 14 6.— Left semilunar.

Height, 2°2in. Breadth, 4°3in. Antero-posterior diameter, 4:2 in.
Figs. 15, 15 a, and 15 6.—Right semilunar.

Height, 26in. Breadth, 48in. Antero-posterior diameter, 5-1 in.
Figs. 16, 16 a, and 16 6.—Right cuneiform.

Height, 2:1in. Transverse diameter, 53 in. Antero-posterior diameter, 4°3 in.

486 FAUNA ANTIQUA SIVALENSIS.

Figs. 17, 17 a, and 17 b.—Right cuneiform.

Height, 1:7in. Transverse diameter, 4°8in. Antero-posterior diameter, 3-3 in.
Figs. 18, 18 a, and 18 6.—Right cuneiform.

Height, 26in. Transverse diameter, 5-Sin. Antero-posterior diameter, 4-4 in.
Figs. 19, 19 a, and 19 6.—Right cuneiform.

Height, 28in. Transverse diameter, 6°4in. Antero-posterior diameter, 4-4 in.
Figs. 20, 20 a, and 20 6.—Right cuneiform.

Figs. 21, 21 a, and 21 6.—Left cuneiform.

Height, 2°5in. Transverse diameter, 5°2in. Antero-posterior diameter, 4°3 in.

Figs. 22 and 22 a.—Left pisiform.

Height, 5:3 in. Breadth, 3°5in. Thickness, 2°3 in.
Figs. 23 and 23 a.—Left pisiform.

Height, 5:1 in. Breadth, 2°8 in. Thickness, 2° in.
Figs. 24 and 24 a.—Left pisiform.

Height, 4°6 in. Breadth, 3° in. Thickness, 2° in.

Figs. 25 and 25 a.— Figured by mistake. Ought to have been
erased.’—[H.F. ]

Figs. 26, 26 a, 26 b, and 26 c.—-Left trapezoid.
Height, 2°5 in. Breadth, 3° in. Antero-posterior diameter, 4°1 in.

Figs. 27, 27 a, 27 b, and 27 c.—Right trapezoid.
Height, 2°65 in. Breadth, 3°2in. Antero-posterior diameter, 4°3 in.

Puate LI.

Bones of anterior extremity of Proboscidea.

Figs. 1, la, 16, le, and 1 d.—Left os magnum.
Figs. 2, 2 a, 26, 2c, and 2 d.—Left os magnum.

Antero-posterior diameter, 5°3in. Transverse, 36 in. Vertical, 4:4 in.
Figs. 38, 3a, 36, 3c, and 3 d.—Left os magnum.

Antero-posterior diameter, 5‘lin. Transverse, 3°8in. Vertical, 4°4 in.
Figs. 4, 4a, 4b, 4c, and 4 d.—Left os magnum.

Antero-posterior diameter, 5'2in. Transverse, 3°3in. Vertical, 4:4 in.
Figs. 5,5 a, 5b, 5c, and 5 d.—Right os magnum.

Antero-posterior diameter, 5°6 in. Transverse, 4°3 in.

Figs. 6, 6a, 66, 6c, and 6 d.—Left os magnum.

Antero-posterior diameter, 6-4in. Transverse, 5‘8in. Vertical, 5°7 in.
Figs. 7, 7a, 7 b, and 7 c.—Right unciform.

Antero-posterior diameter, 4°7in. Transverse, 44in. Vertical, 4°7 in.
Figs. 8, 8a, 8b, and 8 c.—Left unciform.

Antero-posterior diameter, 5'6in. Transverse, 5: in. Vertical, 5*1 in.

Figs. 9, 9a, 9b, and 9 c.—Left unciform.

Antero-posterior diameter, 5: in. Transverse, 4°6 in. Vertical, 3°8.in.
Figs. 10, 10 a, 10 6, and 10 c.—Right unciform.
Antero-posterior diameter, 5:2 in. Transverse, 5° in. Vertical, 4°7 in.

Figs. 11, 11 a, 11 0, and 11 c.—Left unciform.

Antero-posterior diameter, 4°7 in. Transverse, 4°5in. Vertical, 4°2 in.

DESCRIPTION OF PLATES. 487

Figs. 12, 12 a, 12 6, and 12 c.—Left unciform.

Antero-posterior diameter, 4° in. Transverse, 8°6in. _—-~ Vertical, 3°5in.
Figs. 13, 138 a, and 18 6.—Left pollex metacarpal.
Length, 5-2in. © Height of posterior surface, 3°8 in. Breadth ditto, 2°3 in.
Figs. 14, 14 a, and 14 ’.—Right index metacarpal.

Length, 8°3 in. Breadth of posterior surface, 2°9 in.

Figs. 15, 15 a, and 15 b.— Left index metacarpal.
Length, 6°8in. Height of posterior surface, 3:°3in. Breadth ditto, 2°5 in.
Figs. 16, 16 a, and 16 6.—Left index metacarpal.

Length, 6-9in. Height of posterior surface, 4-3in. Breadth ditto, 2°6 in.
Figs. 17, 17 a, and 17 6.—Right medius metacarpal.

Length, 7°8 in. Height of posterior surface, 4° in. Breadth ditto, 2°8 in.
Figs. 18, 18 a, and 18 b.—Right medius metacarpal.
Length, 8:5in. Height of posterior surface, 4:9 in. Breadth of ditto, 3°8 in.
Figs. 19, 19 a, and 19 6.—Right medius metacarpal.
Length, 8'6in. Height of posterior surface, 4°8in. Breadth of ditto, 3-4 in.
Figs. 20, 20 a, and 20 6.—Right medius metacarpal.

Length, 10°in. Height of posterior surface, 5'in. Breadth of ditto, 4°5 in.
Figs. 21, 21 a, and 21 b.—Left annular metacarpal.
Length, 7-1in. Height of posterior surface, 3°9in. Breadth of ditto, 3-4 in.
Figs. 22, 22 a, and 22 6.—Left annular metacarpal.
Length, 8-in. Height of posterior surface, 3°8in. Breadth of ditto, 3°5 in.
Figs. 23, 23 a, and 23 b.—Left annular metacarpal.
Length, 73in. Height of posterior surface, 3°6in. Breadth of ditto, 3° in.
Figs. 24, 24 a, and 24 6.—Right minimus metacarpal.

Length, 6-in. Height of posterior surface, 2°9in. Breadth of ditto, 2°6 in.
Figs. 25, 25 a, and 25 6.—Right minimus metacarpal.
Length, 7-4in. Height of posterior surface, 4:1in. Breadth of ditto, 4-4 in.
Figs. 26, 26 a, and 26 b.—Left minimus metacarpal.

Length, 7'3in. Height of posterior surface, 36 in. Breadth of ditto, 3-7 in.
Figs. 27, 27 a, and 27 b.—Right medius first phalanx.

Length, 4:3in. Height of posterior surface, 2°6in. Breadth of ditto, 3:9 in.
Figs. 28, 28 a, and 28 b.—Left annular first phalanx.

Length, 4-in. Height of posterior surface, 2°6in. Breadth of ditto, 3°6 in.
Figs. 29, 29 a, and 29 6.—Right medius metacarpal.
Length, 3-7in. Height of posterior surface, 2°3in. Breadth of ditto, 3°3 in.
Figs. 30, 80 a, and 30 b.—Left medius metacarpal.

Length, 2°7in. Height of posterior surface, 1‘8in. Breadth of ditto, 2°6 in.

Puate LI.

Bones of posterior extremity of Proboscidea.

Fig. 1.—Head, neck, and great trochanter of left femur.
Length of fragment, 20-in. Breadth of upper extremity, including great tro-

chanter, 157 in. Antero-posterior diameter of head, 7°5in. Transverse diameter
of broken extremity, 7-5in. Antero-posterior diameter of broken extremity, 3°6 in.

488 FAUNA ANTIQUA SIVALENSIS.

Fig. 2.—Head, neck, and great trochanter of right femur.

Length, 17-5in, Breadth of upper end, 15:in. Antero-posterior diameter of
head, 7: in.

Fig. 3.—Head, neck, and upper part of shaft of right femur.

Length, 17:in. Breadth of upper end, 12-in. Antero-posterior diameter of
head, 6° in.

Fig. 4.—Head, neck, and upper part of shaft of right femur.

Length, 18:3in. Breadth of upper end, 13:in. Antero-posterior diameter of
head, 6°6 in.

Fig. 5.—Head, neck, and upper part of shaft of left femur.

Length, 15:in. Breadth of upper end, 11:3in. Antero-posterior diameter of
head (imperfect), 6:in.

Fig. 6.—Head, neck, and upper part of shaft of right femur.

Length, 14:3in, Breadth of upper end, 14:7 in. Antero-posterior diameter of
head, 6°4 in.

Fig. 7.—Head, neck, and upper part of shaft of right femur.

Length, 18:in. Breadth of upper end, 12'in. Antero-posterior diameter of
head, 6-4 in.

Fig. 8.—Head, neck, and great trochanter of left femur.

Length, 13'in. Breadth of upper end, 15:in. Antero-posterior diameter of
head, 7°2 in.

Figs. 9, 9 a, 9 6, and 9 c.—Lower end of right femur with articu-
lating surface.

Length, 17°6in. Transverse diameter of lower end, 9:in. Antero-posterior
diameter internally, 8°5in. Ditto, externally, 8°5 in.

Figs. 10, 10 a, 10 6, and 10 c.—Lower end of right femur with arti-
culating surface.

Length, 21'4in. Transverse diameter of lower end, 8°3in. Antero-posterior
diameter internally, 8°6in. Circumference at fractured end, 13°7 in.

Figs. 11, 11 a, 11 6, and 11 c.—Lower end of right femur with arti-
culating surface.
Length, 13: in. Transverse diameter of lower end, 9°4 in.

Figs. 12, 12 a, 12 6, and 12 c.—Lower end of left femur with arti-
culating surface.

Length, 13°7in. Transverse diameter of lower end, 9°2in. Antero-posterior
diameter internally, 9°6in. Ditto, externally, 8°5in. Transverse diameter of
rotular surface, 4°2in. Height in centre of ditto, 4:1 in.

Figs. 13, 13 a, 13 6, and 18 c.—Lower end of right femur with arti-
culating surface.

Length, 11‘5in. Transverse diameter of lower end, 7:9in. Antero-posterior
diameter externally, 7-1in. Transverse diameter of rotular surface, 3°7 in.
Height in centre of rotular surface, 3°6 in.

Figs. 14, 14a, and 14 6.— Lower end of left femur, with articulating
surface.
Length, 14° in. Transverse diameter of lower end, 7:7 in.
Fig. 15.—Lower end of right femur with articulating surface.

Length, 17:in, Transverse diameter of lower end, 9°6in. Antero-posterior —
diameter internally, 10-4 in. a

2

“*

ee ee

-eF;

e-em

DESCRIPTION OF PLATES. 489

Fig. 16.—Lower end of left femur with articulating surface.

Length, 8-6in. Transverse diameter of lower end, 8:°3in. Antero-posterior
diameter internally, 7°8in. Ditto, externally, 7-1in. Transverse diameter of
rotular surface, 3°8in. Height in centre of rotular surface, 3°8 in.

Fig. 17.—Lower epiphysis of left femur.

Transverse diameter, 8:in. Transverse diameter of rotular surface, 3°7 in.
Vertical diameter of rotular surface in centre, 3°8 in.

Fig. 18.—Lower articulating surface of right femur.

Length, 9°3 in. Transverse diameter, 8-2 in. Antero-posterior diameter in-
ternally, 9‘8in. Antero-posterior diameter externally, 8-4in. Transverse diameter
of rotular surface, 3°8in. Vertical diameter of rotular surface in centre, 4:1 in.

Fig. 19.—Lower articulating surface of right femur.

Length, 13:in. Transverse diameter of lower end, 9: in. Antero-posterior
diameter, externally, 8: in.

Fig. 20.—Lower end of left femur.

Length, 11:in. Transverse diameter of lower end, 7: in. tero-posterior
diameter internally, 7-1in. Antero-posterior diameter externally, 6-4in. Trans-
verse diameter of rotular surface, 8‘lin. Vertical diameter of rotular surface in
centre, 3°8 in.

Fig. 21.—Lower end of left femur.

Transverse diameter, 8°7 in. Antero-posterior diameter internally, 8°7 in.
Antero-posterior diameter externally, 7°9 in. Transverse diameter of rotular sur-
face, 4:1in. Vertical diameter of rotular surface in centre, 41 in.

Fig. 22.—Lower end of right femur.
Length, 15°4in. Transverse diameter of lower end, 10°3 in.
Fig. 23.—Lower end of right femur.

Length, 11°5in. Transverse diameter of lower end, 9:2 in. Antero-posterior
diameter internally, 9°4in. Antero-posterior diameter externally, 8°3in. Trans-
verse diameter of rotular surface, 4-in. Vertical diameter of rotular surface,
4-6in.

Fig. 24.—Lower end of right femur.

Length, 10°8 in. ‘Transverse diameter of lower end, 8-1 in. Antero-posterior
diameter internally, 8-3in. Antero-posterior diameter externally, 7-5in. Trans-
verse diameter of rotular surface, 4:2 in. Vertical diameter of rotular sur-
face, 41 in.

Fig 25.—Lower end of right femur.

Length, 11:in. Transverse diameter of lower end, 8° in. Antero-posterior
diameter internally, 8:3in. Antero-posterior diameter externally, 7-6in. Height
of rotular surface, 3°7 in.

Fig. 26.—Lower end of right femur.

Transverse diameter of lower end, 9'5in. Antero-posterior diameter internally,
98in. Ditto, externally, 8-7 in. .

Fig. 27._Lower end of left femur.

Length, 11:2 in. Transverse diameter of lower end, 10-in. Antero-posterior
diameter internally, 10:1 in. Ditto externally, 9°in. Transverse diameter of
rotular surface, 5°"1 in. Vertical diameter of rotular surface, 5: in.

Fig. 28.—Lower end of right femur.

Length, 13°5in. Transverse diameter of lower end, 9°4in. Antero-posterior
diameter internally, 10°6in. Ditto externally, 9:2 in.

490 FAUNA ANTIQUA SIVALENSIS.

Fig. 29.—Lower end of left femur.

Length, 11'4in. Transverse diameter of lower end, 8°8in. Antero-posterior
diameter internally, 9°2in. Ditto externally, 7-6in. Transverse diameter of
rotular surface, 3°7in. Vertical diameter of rotular surface in centre, 4° in.

Fig. 80.—Lower end of left femur.

Length, 15‘in. Transverse diameter of lower end, 7°65 in.

Puate LIII.

Bones of trunk and posterior extremity of Proboscidea.

Fig. 1.—Fragment of pelvis, right side, showing acetabulum.

Length of fragment, 19°in. Great diameter of acetabulum 7-in. Lesser ditto,
6°5 in.

Figs. 2 and 2 a.—Fragment of pelvis showing acetabulum.

Length of fragment, 17: in. Great diameter of acetabulum, 6: in.

Fig. 8.—Fragment of pelvis, showing acetabulum.

Length of fragment, 15-in. Great diameter of acetabulum, 7-6in. Lesser
ditto, 7°4 in.

Fig. 4.—Fragment of pelvis, showing acetabulum.

Length of fragment, 17:in. Great diameter of acetabulum, 8-2in. Lesser
ditto, 7°6 in.

Fig. 5.—Fragment of pelvis, showing acetabulum.

Length of fragment, 13‘7in. Great diameter of acetabulum, 6:in. Lesser
ditto, 5-4 in.

Fig. 6.—Fragment of pelvis, showing acetabulum.

Length, 14"in. Great diameter of acetabulum, 7:in. _— Lesser ditto, 6°8 in.

Fig. 7.—Fragment of pelvis, showing acetabulum.

Length, 16°2in. Great diameter of acetabulum, 7°5in. Lesser ditto, 7: in.

Fig. 8.—Fragment of pelvis, showing acetabulum.

Length, 13:in. Great diameter of acetabulum, 7°3in. Lesser ditto, 6°9 in.

Fig. 9.—Head, neck, and upper part of shaft of right femur.

Length of fragment, 35°5in. Breadth of upper extremity, including great tro-
chanter, 18'in. Antero-posterior diameter of articular surface, 7-1in. Smallest
transverse diameter, 5°6in. Antero-posterior diameter of shaft, 3°5in.

Figs. 10 and 10 a.—Head of femur.

Greater diameter, 8°5 in. Lesser ditto, 7°8 in.

Figs. 11 and 11 a.—Lower end of right femur and articular surface
of Elephas primigenius.

Length of fragment (longer than the figure), 34: in. Breadth of lower end, 7:2 in.
Antero-posterior diameter internally, 7°9in. Ditto externally, 66in. Height of
rotular surface in centre, 3°2in. Breadth of ditto, 3:1 in.

Figs. 12 and 12 a.—Lower articulating end of right femur.

Breadth of lower extremity, 10°7 in.

Fig. 13.—Lower end of right femur of Elephas antiquus, from Walton
in Essex.

Breadth, 10°in. Antero-posterior diameter internally, 9:5in. Ditto externally,
88in. Height of rotular surface in centre, 4*1in. Breadth of ditto, 4°5 in.

Figs. 14 and 14 a.—Upper end of left tibia.

Length of fragment, 10°7 in. ‘Transverse diameter of upper end, 9°65 in.
Greatest antero-posterior diameter of ditto, 5:6 in.

DESCRIPTION OF PLATES. 491

Figs. 15 and 15 a.—Upper end of right tibia.
Length, 13°5in. Transverse diameter of upper end, 9:7in. Greatest antero-
posterior diameter of ditto, 7°3 in.
Figs. 16 and 16 a.—Upper end of right tibia.
Length, 156in. Transverse diameter of upper extremity (imperfect), 9°3 in.
Greatest antero-posterior diameter of ditto, 7: in.
Figs. 17 and 17 a.—Upper end of right tibia.
Length, 12-7 in. Transverse diameter of upper extremity, 8°2 in.
Fig. 18.—Upper articulating surface of right tibia.
Transverse diameter, 9°7 in. Greatest antero-posterior diameter, 7°3 in.
Fig. 19.—Upper articulating surface of right tibia.
Transverse diameter, 9°6 in. Greatest antero-posterior ditto, 6°8 in.
» Fig. 20.—Upper articulating surface of right tibia.
Transverse diameter, 6°8 in. Greatest antero-posterior ditto, 5° in.
Fig. 21.—Upper articulating surface of right tibia.

Transverse diameter, 7'4in. Greatest antero-posterior ditto, 5-2 in.

Fig. 22.—Upper articulating surface of right tibia.

Transverse diameter, 7°3 in. Greatest antero-posterior ditto, 5:7 in.
Fig. 23.—Upper articulating surface of left tibia.
Transverse diameter, 8° in. Greatest antero-posterior ditto, 6°5 in.
Fig. 24.—Upper articulating surface of right tibia.
Transverse diameter, 7°8in. Greatest antero-posterior ditto, 6° in.
Fig. 25.— Upper articulating surface of left tibia.
Transyerse diameter, 8: in. Greatest antero-posterior ditto, 5°5 in.
Fig. 26.—Upper articulating surface of left tibia.
Transverse diameter, 10: in. Greatest antero-posterior ditto, 7°2 in.

Figs. 27 and 27 a.—Upper end of left tibia.
Length, 9°6in. Transverse diameter of upper surface, 9'4in. Greatest antero-
posterior diameter of ditto, 7: in.

Fig. 28.—Upper articulating surface of right tibia.

Transverse diameter, 7°4 in. Greatest antero-posterior diameter, 6° in.
Fig. 29.—Upper articulating surface of left tibia.
Transverse diameter, 8-7 in. Greatest antero-posterior diameter, 5°6 in.

Fig. 30.—Upper articulating surface of left tibia.

Transverse diameter, 10°in. Greatest antero-posterior diameter, 6°5 in.

Fig. 31.—Lower articulating surface of left tibia.
Transverse diameter, 8°7in. Greatest antero-posterior diameter, 6°7 in.

Figs. 82 and 32 a.—Lower end of left tibia.

Length of fragment, 9°5in. Transverse diameter of lower end, 6-9in. Antero-
posterior diameter of ditto, 5°3 in.

Figs. 83 and 33 a.—Lower end of right tibia.

Length, 7-7in. Transverse diameter of lower end, 7:7in. Antero-posterior
diameter of ditto, 6:1 in.

Figs. 34 and 34 a.—Lower end of left tibia.

Length, 10°7in. Transverse diameter of lower end, 65in. Antero-posterior
diameter of ditto, 4°8 in,

492 FAUNA ANTIQUA SIVALENSIS.

Figs. 55 and 35 a.—Lower end of right tibia.

Length, 7:5in. Transverse diameter of lower end, 7:7 in. Antero-posterior
diameter of ditto, 6: in.

Figs. 36 and 86 a.—Lower end of right tibia.
Length, 10°7in. Transverse diameter of lower end, 5°6in. Antero-posterior
diameter of ditto, 4° in.

Fig. 87.—Lower articulating surface of right tibia.

Transverse diameter, 7°3 in. Antero-posterior diameter, 5°4 in.
Fig. 38.—Lower articulating surface of right tibia.

Transverse diameter, 6:1 in. Antero-posterior diameter, 5°3 in.
Fig. 39.—Lower articulating surface of left tibia.

Transverse diameter, 6°4 in. Antero-posterior diameter, 4°5 in.
Fig. 40.—Lower articulating surface of left tibia.

Transverse diameter, 6°2 in. Antero-posterior diameter, 5-4 in.
Fig. 41.—Lower articulating surface of right tibia.

Transverse diameter, 6°5 in. Antero-posterior diameter, 4°9 in.
Fig. 42.—Lower articulating surface of right tibia.

Transverse diameter, 7*in.

Puate LIV.

Bones of posterior extremity of Proboscidea.

Figs. 1, 1 a, and 1 6.—Right calcaneum.

Length, 1l‘in. Height of cuboid surface, 26in. Breadth of ditto, 4:2 in.

Figs. 2, 2a, and 2 6.—Left calcaneum.

Length, 10-1in. Height of cuboid surface, 2°6in. Projection of caleaneum,
5'2in. Breadth of cuboid surface, 4:2in. Length of fibular surface, 4:2 in.
Breadth of ditto, 2°3 in.

Figs. 3, 3 a, and 3 6.—Left calcaneum.

Length, 9'8in. Height of cuboid surface, 2°8in. Breadth of ditto, 3-9 in.
Projection of heel, 4:9 in.

Figs. 4, 4a, and 4 6.—Right calcaneum.

Length, 9°5in. Height of cuboid surface, 3‘lin. Breadth of ditto, 3-9 in.
Projection of heel, 5:3 in.

Figs. 5, 5 a, and 5 6.—Right calcaneum.

Length, 8°8 in. Projection of heel, 5°5 in.

Figs. 6, 6 a, and 6 6.—Right calcaneum.

Length, 9‘in. Height of cuboid surface, 2:7in. Breadth of ditto, 3‘5in. Pro-
jection of heel, 5-5 in.

Figs. 7, 7 a, and 7 6.—Left calcaneum, imperfect.

Length of fragment, 8:lin. Height of cuboid surface, 2°7in. Breadth of
ditto, 3°3 in.

Figs. 8, 8 a, and 8 6.—Left calcaneum.'

Length, 76in. Height of cuboid surface, 3:2in. Breadth of ditto, 2°2in. Pro-
jection of heel, 4°9 in.

Figs. 9, 9a, and 9 6.—Right caleaneum.

Length, 7:7 in. Projection of heel, 4°8 in.
Figs. 10, 10 a, and 10 6.—Left calcaneum.
Length, 8:4in. Height of cuboid surface, 2°lin. Breadth of ditto, 3-2 in.

DESCRIPTION OF PLATES. 493

Figs. 11, 11 a, and 11 6.—Right calcaneum.

Length, 85in. Height of cuboid surface, 2‘5in. Breadth of ditto, 3-4 in.
Projection of heel, 4°7 in.

Figs. 12, 12 a, and 12 6.—Left calcaneum.

Length, 7:8 in. Projection of heel, 4:5 in,

Figs. 13, 13 a, and 13 6.—Right caleaneum.

Length, 8°7in. Height of cuboid surface, 1'8in. Breadth of ditto, 3:4 in.
Projection of heel, 51 in.

Figs. 14, 14 a, and 14 6.—Left calcaneum.

Length, 8:-lin. Height of cuboid surface, 2°3in. Breadth of ditto, 3-4 in,
Projection of heel, 4°7 in.

Figs. 15, 15 a, and 15 6.—Right caleaneum.

Length, 83in. Height of cuboid surface, 2:°2in. Breadth of ditto, 3-lin.
P¥ojection of heel, 4°8 in.

Figs. 16, 16a, and 16 6.—Right calcaneum of Mastodon Ohioticus,
imperfect.

Length of fragment, 7‘7in. Height of cuboid surface, 3-lin. Breadth of ditto,
3°6in. Projection of heel, wanting epiphysis, 3-6in. Length of fibular surface,
3°8in. Breadth of ditto, 2:1 in.

Figs. 17, 17 a, and 17 6.—Right calcaneum of Dinotherium.

Length, 13-in. Projection of heel, 6°Sin. Length of fibular surface, 4°6 in.
Breadth of ditto, 2°8 in.

Figs. 18 and 18 a.—Left astragalus.
Figs. 19 and 19 a.—Right astragalus.
Length, 6-3 in. Breadth, 7: in. Thickness, 4:4 in.

Figs. 20 and 20 a.—Left astragalus.

Length, 4°6 in. Breadth, 4:7 in. Thickness, 3°5 in.
Figs. 21 and 21 a.—Left astragalus.

Length, 4°4 in. Breadth, 4°7 in, Thickness, 2°8 in.
Figs. 22 and 22 a.—Left astragalus:

Length, 5:5 in. Breadth, 6:4 in. Thickness, 3°8 in.
Figs. 25 and 23 a.—Right astragalus.

Length, 5°5 in. Breadth, 6°4 in. Thickness, 4:4 in. -
Figs. 24 and 24 a.—Left astragalus.

Length, 4°7 in. Breadth, 5:7 in. Thickness, 3°4 in,
Figs. 25 and 25 a.—Right astragalus.

Length, 5:2 in. Breadth, 5:4in, Thickness, 3°6in.
Figs. 26 and 26 a.—Right astragalus.

Length, 5: in. Breadth, 6°3 in. Thickness, 3°5 in.
Figs. 27 and 27 a.—Left astragalus.

Length, 5:3 in. Breadth, 6:8 in. Thickness, 4: in.
Figs. 28 and 28 a.—Right astragalus.

Length, 4°3 in, Breadth, 4°8 in. Thickness, 3° in,
Figs. 29 and 29 a.—Left astragalus.

Length, 5:1 in. Breadth, 5:5 in. Thickness, 3°5 in.

494 FAUNA ANTIQUA SIVALENSIS.

Figs. 30 and 30 a.—Right astragalus.

Length, 5°5 in. Breadth, 6°5 in. Thickness, 4° in,
Figs. 31 and 31 a.—Left astragalus.

Length, 5:3 in. Breadth, 6°4 in. Thickness, 4°1 in.

Priate LV.

Bones of posterior extremity of Proboscidea.

Figs. 1, 1 a, 1 6, and 1 ¢.—Left calcaneum.

Length, 9°5in. Height of cuboid surface, 2-2in, Breadth of cuboid surface,
4in. Projection of heel (from posterior border of surface for astragalus to most
projecting part of calcaneal tuberosity) oblique, 6-2in. Projection direct, 3° in.
Height from lower surface to astragalar surface (ext.),6°3in. Breadth of fibular
surface, 2°5in. Length of fibular surface, 4°2in. Length of astragalar surface,
4-7in. Breadth of astragalar surface 2°7 in.

Figs. 2, 2a, 2 6, and 2 c.—Left caleaneum of Elephas antiquus, from
Grays in Essex.
Figs. 3, 3. a, 36, and 3 c.—Right calcaneum.

Length, 8°in. Height of cuboid surface, 2°2in. Breadth of ditto, 3-4in. Pro-
jection of heel (direct), 4°2in. Height externally, 4°9 in.

Figs. 4 and 4 a.—Patella of Elephas antiquus from Grays in Essex.
Figs. 5 and 5 a.—Patella.
Length, 6°6 in. Breadth, 5:5 in. Thickness, 4:1 in.
Figs. 6 and 6 a.—Patella.
Length, 5:in. Breadth, 4°5 in. Thickness, 4: in.
Figs. 7 and 7 a.—Patella.
Length, 5:2 in. Breadth, 4°4 in. Thickness, 3-3 in.
Figs. 8 and 8 a.—Patella.
Length, 5°6 in. Breadth, 4°4in. Thickness, 3°4in.
Figs. 9 and 9 a.—Patella.
Length, 4°6 in. Breadth, 3:7 in. Thickness, 2-9 in.
Figs. 10, 10 a, and 10 6.—Left ecto-cuneiform bone.
Height, 4°3 in. Breadth, 4°5 in, Thickness, 2: in.
Figs. 11; 11a, and 11 6.—Left ecto-cuneiform bone.
Height, 4°4 in. Breadth, 4°8 in. Thickness, 1°8 in,
Figs. 12, 12 a, and 12 b.—Left ecto-cuneiform bone.
Height, 4° in. Breadth, 3°8 in. Thickness, 1-9 in.
Figs. 13, 13 a, and 13 6.—Left ecto-cuneiform bone.
Height, 3-8 in. Breadth, 4°7 in. Thickness, 1°7 in.
Figs. 14, 14 a, and 14 b.—Right ecto-cuneiform bone.
Height, 3°9 in. Breadth, 4°4 in. Thickness, 2-1 in,
Figs. 15, 15 a, and 15 6.—Right ecto-cuneiform bone.
Height, 3°8 in. Breadth, 42 in. Thickness, 2°1 in,
Figs. 16, 16 a, and 16 6.—Right index metatarsal.
Length, 5'1in. Height of posterior surface, 3°in. Breadth of ditto, 2-1 in.
Figs. 17, 17 a, and 17 6.—Right index metatarsal.
Length, 5'in. Height of posterior surface, 2°5in. Breadth of ditto, 1-7 in.

hLepeed.\.

pee

DESCRIPTION OF PLATES. 495

Figs. 18, 18 a, and 18 6.—Left medius metatarsal.

Length, 4-5in. Height of posterior surface, 1°8in. Breadth of ditto, 2:3 in.
Figs. 19, 19 a, and 19 b.—Right index metatarsal.

Length, 5-2in. Height of posterior surface, 2°6 in. Breadth of ditto, 1:7 in.
Figs. 20, 20 a, and 20 b.—Left medius metatarsal.
Length, 5°6in, Height of anterior articular surface, 2‘6in. Breadth of ditto, 2:5in.
Figs. 21, 21 a, and 21 6.—Left medius metatarsal.

Length, 5°5in. Height of posterior surface, 2°4in. Breadth of ditto, 2-2 in.
Figs. 22, 22 a, and 22 6.—Left medius metatarsal.

Length, 5:in. Height of posterior surface, 2°7in. Breadth of ditto, 2-1 in.
Figs. 23, 25 a, and 23 b.—Right medius metatarsal.

Length, 5°5in. Height of posterior surface, 3:lin. Breadth of ditto, 2°4 in.
Figs. 24, 24 a, and 24 b.—Left annularis metatarsal.

Length, 4-3in. Height of posterior surface, 2°4in. Breadth of ditto, 1:7 in.
Figs. 25, 25 a, and 25 6.—Left annularis metatarsal.

Length, 5'in. Height of posterior surface, 2°7in. Breadth of ditto, 2°5 in.
Figs. 26, 26 a, and 26 b.—Right index first phalanx.

Length, 3°5in. Height of posterior surface, 2°8in. Breadth of ditto, 3-3 in,
Figs. 27, 27 a, and 27 6.—Right index first phalanx.

Length, 2°8 in.

Figs. 28, 28 a, and 28 b.—Right index first phalanx.

Length, 2°2in. Height of posterior surface, 1:7in. Breadth of ditto, 1-9 in.
Figs. 29, 29 a, and 29 6.—Right index first phalanx.

Length, 2°9in. Height of posterior surface, 2°lin. Breadth of ditto, 2-8 in.
Figs. 30, 30 a, and 30 6.—Left index first phalanx.

Length, 3°3in. Height of posterior surface, 2°5in. Breadth of ditto, 3-2 in.
Figs. 31, 31 a, and 31 6.—Right index first phalanx.

Length, 2°5in. Height of posterior surface, 2:in. Breadth of ditto, 2-4 in.
Figs 32, 32 a, and 32 b.—Right medius first phalanx.

Length, 2°3in. Height of posterior surface, 1‘6in. Breadth of ditto, 2°1 in.
Figs. 33, 33 a, and 33 6.—Left annularis first phalanx.

Length, 2°7in. Height of posterior surface, 2-lin. Breadth of ditto, 2°5 in.
Figs. 34, 34a, and 34 b.—Right annularis first phalanx.

Length; 3:lin. Height of posterior surface, 2°2in. Breadth of ditto, 2:4 in.
Figs. 35, 35 a, and 35 b.—Left annularis first phalanx.

Length, 3°5in. Height of posterior surface, 2°3 in. Breadth of ditto, 2°9 in.
Figs. 36, 36 a, and 36 6—Right annularis first phalanx.

Length, 4*in, Height of posterior surface, 2°9in. Breadth of ditto, 3-in.

Prats LVI.

Figs. 1, 1 a, 1b, and 1e.—Elephas Namadicus. Lower end of right
femur with articulating surface. From the Valley of the Nerbudda
River. This is the specimen figured by Dr. Spilsbury, in Journ. As.
Soc., vol. x. p. 626, Plate A. fig. 3—B.M.

Transverse diameter of inferior extremity, 10°7 in. Antero-posterior diameter of
inner surface of ditto, 11-4in. Antero-posterior diameter of outer surface of ditto,

496 FAUNA ANTIQUA SIVALENSIS.

9-7 in. (from posterior margins of respective condyles to anterior margins of rotular
surface.) Length of rotular surface (in centre), 5-5in. Breadth of rotular surface
(in centre), 5°6 in. Circumference at commencement of fractured portion 13 in. above
inferior articular surface, 22° in.

Figs. 2 and 2a.—Elephas Namadicus. Lower end of right tibia,
with articular surface ; from the Nerbudda.— B.M.

Greatest transverse diameter of inferior extremity, 7°4in. Greatest antero-
posterior diameter of ditto, 6"1in. Projection downwards of the internal mal-
leolus, 1:4in. Circumference at broken end 9 in. above inferior surface, 12°6 in.

Figs. 3 and 3 a.—Elephas Namadicus. Upper end of right radius ;
from the Nerbudda.—B.M.

Greatest transverse diameter, 6-2in. Greatest antero-posterior diameter, 3°8 in.

Figs. 4, 4a, and 4 6.—Elephas Namadicus. Dorsal vertebra ; from
the Nerbudda.—B.M.

Height of body, anteriorly, 5‘8in. Breadth of body, anteriorly, 56in. Length
of body, inferiorly, 2°4in. Spinal canal, height, anteriorly, 2° in. Spinal canal,
breadth, anteriorly, 3in. Between extreme points of transverse processes, 12°8 in.
Anterior costal surface, height, 2°6in. Anterior costal surface, breadth, 1-8, in.

Fig. 5.—Elephas Namadicus. Left femur in three fragments. This
figure is copied from an illustration of Mr. J. Prinsep’s description of
a fossil found by Dr. Spilsbury in the Nerbudda Valley, near Nar-
sinhpoor (Journ. As. Soc., Aug. 1834, vol. ill. p. 396, Plate XXIV.
fig.3). The dimensions of the femur, while it remained whole and at-
tached to the rocky matrix, were as follows:—

Greatest length, 63:in. Circumference of the head, 27-in.; diameter of ditto,
8-75in. Breadth from tip of great trochanter to inner edge of head, 18-in, Cir-
cumference of shaft at centre, 19°in. Breadth of condyles, 11-in.

Fig. 6.—Elephas Namadicus. Fragments of right femur. The epi-
physis of the head is lost, but its place is shown by a dotted line. This
figure is copied from the same source as the last, (Journ. As. Soc., vol. iii.
Plate XXIV. figs. 9 and 10. Fig. 6a. is lower end of the left femur
represented in fig. 5, and not of that in fig. 6, as might be inferred from
dotted line in Plate.

Fig. 7.—Elephus Namadicus. Humerus; copied from the illustra-
tion of a paper by Dr. Spilsbury in Journ. As. Soc., June 1837, vol.
vi. p. 487, Plate XXX. fig. 1.

Fig. 8—Elephas Namadicus. Bones of pelvis with acetabulum, copied
from same source as fig. 7. (Plate XXX. figs. 5 and 6.)

Fig. 9.—Elephas Namadicus. Lower jaw, with molar incomplete at
left side. This specimen is also copied from an illustration of Mr. J.
Prinsep’s description of a fossil found by Dr. Spilsbury in the Nerbudda
Valley (Journ. As. Soc., Nov. 1833, vol. ii. p. 583, Plate KX. fig. 1).
The jaw is inverted in the drawing.

Length of tooth, 11°5in. Breadth, 3-5in. Length of grinding surface, 8°5 in.
Girth of jaw-bone, 24-in. Probable length from chin to condyle, 26-in.

A description of the locality where the Nerbudda fossils were found, with sec-
tional drawings, will be found in the paper above referred to.

Figs. 10 and 10a.—Elephas insignis. Fragment of molar showing
four plates. The specimen is remarkable as coming from the Valley
of the Nerbudda.—B.M.

Length of molar, 4°7in. Greatest breadth, 3-1in. Height of fragment, 6: in.

ee ee ee

DESCRIPTION OF PLATES. 497

Figs. 11 and 11 a.—Elephas insignis. Fragment of molar, remark-
able as coming from the Valley of the Nerbudda.—B.M.

Length of fragment of grinding surface, 2°5in. Breadth of tooth at centre of
that fragment, 3° in.

Figs. 12 and 12 a—Elephas ? Fragment of molar containing
about nine plates; from the Nerbudda.—B.M.

Length, 5-2in. Breadth, 2°3 in.
Figs. 13 and 13 a.—EHlephas ? Fragment of lower jaw, con-

taining portion of molar with about seven plates; from the Nerbudda.
—B.M.

Length of fragment of jaw, 11°3in. Greatest breadth of ditto, 5-2in. Height
opposite posterior border of molar, 4°5in. Length of tooth, 5%in. Greatest
breadth of ditto, 2°7 in.

Figs, 14, 14 a, and 14 b.— Elephas

? Fragment of molar with

six plates; from the Nerbudda.—B.M.

Length, 6-4in. Breadth, 4:1 in.

Puate LVIL.

Fig. 1.—Hippopotamus (Tetraprotodon) Palewindicus. (Fale. and
Caut.) Almost entire skull, viewed from above, with zygomatic arches
complete. Specimen from the Nerbudda in B.M. It shows well the
great saliency of the sagittal crest. There is a finer specimen in the
Museum of the Asiatic Society of Bengal (See antea, p. 147).

Fig. 1 a.—Lateral view of same skull, showing cavity of orbit, &c.
The great projection of the orbit above the plane of the frontal, charac-
teristic of the species, is well seen.

Fig. 1 6.—Same skull, palatine view, showing three molars, the fur-
thest back intact, the two next ground down; also the three premolars
on one side, and two on the other.

Fig. 1 ¢.—Occipital view of same skull, showing condyles and
foramen magnum, and the great saliency of the occipital crest.

Figs. 2 and 2 a.—H. Paleindicus. Portion of upper jaw, right side,
with three molars.—B.M.

Figs. 3 and 3 a.—H. Paleindicus. Portion of lower jaw, with
teeth.—B.M.

Fig. 4.—H. Paleindicus. Fragment of canine.—B.M.

Figs. 5 and 5 a.—H. Paleindicus. Anterior portion of jaw with
alveoli of four incisors. The diameter of the alveoli of the central in-
cisors is much less than that of the external incisors—a fact which
refutes De Blainville’s opinion that the Nerbudda Tetaprotodon is.

identical with the living African species. In the latter the middle in-

cisors are the largest.

Figs. 6 and 6 a.—H. Paleindicus. Portion of lower jaw, with
grinders (three molars and two premolars) worn down.—B.M.

Figs. 7 and 7 a.—H. Paleindicus. Portion of lower jaw, with
two molars.—-B.M.

VOL. I. KK

498 FAUNA ANTIQUA SIVALENSIS.

Fig. 8.—H. Paleindicus. Lower canine.—B.M.

Fig. 8 a.—H. Paleindicus. Transverse section of canine, of ovoid
shape.

Fig. 9.—H. Paleindicus. Last lumbar vertebra. Anterior view.—B.M.
Fig. 9a.—H. Paleindicus. Last lumbar vertebra. Viewed from above.
Fig. 9b.—H. Paleindicus. Last lumbar vertebra. Lateral view.

Figs. 10, 10 a, 10 6, and 10 c.—Hippopotamus (Hexaprotodon) Irava-
ticus (Fale. and Caut.), from Ava. Anterior portion of the lower jaw,
different views. Shows the six incisor teeth characteristic of the sub-
genus.—B.M.

Figs. 11 and 11 a.—d. Iravaticus. Fragment of jaw with alveoli, &c.

Figs. 12 and 12 a.—Hippopotamus (Hexaprotodon) Namadicus. From
the Nerbudda. Anterior portion of lower jaw, showing the six
incisors.

Puate LVIII.

Fig. 1.—Hippopotamus (Hexaprotodon) Namadicus (Fale.and Caut.),
from the Nerbudda. Lower jaw, viewed from above, showing molars,
canines, and s¢x incisors.—B.M.

Fig. 1 a.—Hippopotamus (Hexaprotodon) Namadicus. Right lateral
view of same specimen.

Fig. 2.—H. Namadicus. Lower jaw, viewed from above. Showing
molars and premolars complete on left side, with portions of both canines
and of right outer incisor; also alveoli of five remaining incisors.—
B.M.

Fig. 2 b.—H. Namadicus. Lateral view of same specimen.

Fig. 3.—H. Namadicus. Fragment of lower jaw viewed from above,
showing molars and premolars on one side, and portions of left canine
and all six incisors.—B.M.

Fig. 3 b.—H. Namadicus. lateral view of same specimen.

Fig. 4.—Hippopotamus (Tetraprotodon) Paleindicus. Fine specimen
of skull, incomplete; upper surface showing the zygomatic arches, and
the great prominence of the sagittal ridge. There is a still more perfect
specimen in the Museum of the Asiatic Society of Bengal (See antea,
p- 147).—B.M.

Fig. 4 a.—Hippopotamus ( Tetraprotodon) Paleindicus. Lateral view
of same specimen, showing the great projection of the orbit above the
plane of the frontal.

Fig. 4 6.— Hippopotamus (Tetraprotodon) Paleindicus. Palatine view
. of same specimen, showing the three molars on both sides, well ground
down, and the left posterior premolar.

Figs. 5, 5 a, 5 6, and 5 c.—Hippopotamus Paleindicus. Four —
different views of first dorsal vertebra.—B.M.

Fig. 6.—H. Paleindicus. Head, neck, and upper portion of shaft —
of left femur, anterior view.—B.M.

Fig. 6 a.—H. Paleindicus. Posterior view of same specimen.

DESCRIPTION OF PLATES. 499

Fig. 66.—H. Paleindicus. Same specimen viewed from above.

Fig. 7.—H. Paleindicus. Lower end of shaft, with articulating
extremity, of femur, posterior view.—B.M.

Fis. 7 a.—H. Paleindicus. Posterior view of same specimen.
Fig. 7 b..—H. Paleindicus. Same specimen viewed from below.

Figs. 8, 8a,and 86.—H. Palwindicus. Upper end of tibia. Anterior,
posterior, and upper views.—B.M.

Figs. 9, 9a, 9b, and 9c.—H. Paleindicus. Entire tibia. Anterior,
posterior, upper and lower views.—B.M.

Figs. 10, 10a, 100, and 10 c.—H. Palwindicus. Entire radius. An-
terior, posterior, upper and lower views.—B.M.

*

Puate LIX.

Fig. 1.— Hippopotamus (Hexaprotodon) Sivalensis. (Fale. and Caut.)
Fine specimen of entire skull from the Sewalik hills. Upper surface,
showing the zygomatic arches, sagittal crest, and muzzle. The hollow
between the muzzle and the zygomatic arch is remarkably abrupt, and
the occipital crest is very elevated.—B.M. (See antea, p. 142.)

Fig. 1 a.—H. Sivalensis. Palatine view of same specimen, showing
the three molars and four premolars, well ground, and the alveoli of
the canines and szz incisors. ‘The line of molars is seen to curve
slightly outwards towards the front and also behind. The space be-
tween the most advanced molar and the canine is much shorter than
in the existing animal. There is a deep fissure in front between the
incisive bones.

Fig. 1b.—H. Sivalensis. Lateral view. The orbit projects but
slightly above the plane of the frontal, and in this respect the species
contrasts remarkably with that of the H. Paleindicus, Plate LVII. fig.
1a, and Plate LVIII. fig. 4a. The orbit is also much more advanced
than in the existing Hippopotamus, and this accounts for the abrupt
hollow between the muzzle and zygomatic arch. The incisors, drawn
in outline, are seen to curve downwards. (Reproduced in Pl. XI.
figs. 1, 2, and 3.)

Fig. 2.—H. Sivalensis. Another specimen of cranium, upper surface.
The right zygomatic arch is imperfect; the nasal sutures are more dis-
tinct than in fig. 1.—B.M.

Fig. 2a.—H. Sivalensis. Palate view of same specimen, showing

_ three molars and four premolars on either side. Posteriorly the molar
lines curve less out than in fig. 1, and the teeth are somewhat less

worn.

Fig. 3.—H. Sivalensis. Another specimen of cranium, upper sur-
face. Both zygomatic arches are imperfect.—B.M.

Fig. 3a.—H. Sivalensis. Palate view of same specimen, showing
three molars and four premolars on either side. The teeth are less
ground than in figs. 1 and 2. The trefoil wear of the coronals of each
pair of collines is well seen.

KK 2

500 FAUNA ANTIQUA SIVALENSIS.

Piate LX.
Fig. 1.—H. Sivalensis. Fragment of posterior portion of skull;
upper surface; showing occipital ridge and left zygomatic arch.—B.M.
Fig. 1 a.—Under surface of same specimen.
Fig. 1 6.—Lateral view of same specimen.

Fig. 2.—H. Sivalensis. Fine specimen of cranium, with occipital
ridge, and nasal sutures distinct, and both zygomatic arches intact.—

B.M

Fig. 2 a.—Lateral view of same specimen, showing lower jaw in
situ. The slight elevation of the orbit above the plane of the frontal
is also seen.

Fig. 2 b.—Posterior view of, same specimen, showing occipital ridge,
condyles, and foramen magnum.

Fig. 3.—H. Sivalensis. Fragment of posterior part of another skull
with both zygomatic arches.—B.M.

Fig. 3a.—Palatine view of same specimen, showing posterior and
part of middle molar, on either side.

Fig. 3 b.—Lateral view of same specimen.
Fig. 3 ¢.—Occipital, or posterior, view of same specimen.

Fig. 4.—H. Sivalensis. Another specimen of cranium deficient in
muzzle and right zygomatic arch ; upper surface.

Fig. 4 a.—Palatine view of same specimen, showing three molars and
one premolar, on either side, much worn.

Fig. 4 }.—Lateral view of same specimen, showing the slight ele-
vation of the orbit.

Fig. 4 c.—Posterior, or occipital, view of same specimen.

Puate LXI.

Fig. 1.—H. Sivalensis. Fragment of young skull, showing muzzle ;
under surface showing molars and canines and alveoli of premolars.—

B.M.

Fig. 1 a.—Upper surface of same specimen.

Fig. 1 6.—Same specimen; lateral view. ;

Fig. 2.—Hippopotamus Sivalensis. Skull, imperfect; palatal view.
.M.

Fig. 2 a.—Lateral view of same specimen.

Fig. 3.—H. Sivalensis. Fragment of lower jaw, right side, viewed
from above. The molar line is seen to curve outwards, both in front —
and behind, as in the upper jaw.

Fig. 3 a.—Lateral view of same fragment. The condyle, coronoid —
process, and the descending process are broken off. The lower mane
is straight.

Figs. 4 and 4 a.—H. Sivalensis. Fragment of lower jaw, upper and
lateral surface. The alveolar ridge on right side is very perfect, and

DESCRIPTION OF PLATES. 501

shows three molars and three premolars, with a portion of the canine.
The condyle, coronoid process, and descending process are wanting.
The lower margin is straight B.M.

Figs. 5 and 5 a.—H. Sivalensis. Lower jaw, more perfect ; viewed
from above and also laterally. The alveolar ridges on both sides are
perfect, and comprise three molars and three premolars, and also the
alveolus of a fourth premolar. Both canines are broken off; but the
right one is tolerably perfect, and is seen to curve back slightly at its
tip. The incisive ridge is perfect, but the teeth are wanting. The
width across the muzzle from the outer side of one canine alveolus to
that of the other is greater, and the width of the jaw over the penulti-
mate false molar is less, than in H. amphibius. The eondyle, coronoid
process, and descending ramus are wanting. The lower margin is

straight B.M.

Figs. 6 and 6 a.—H. Sivalensis. Lower jaw, viewed from above and
also laterally. The posterior molars on both sides are wanting; but
the two anterior molars and three premolars, on either side, and the two
canines and s7x incisors are present. The space between the anterior
premolar and the canine is very contracted. The right canine is very
perfect. The anterior angle of the jaw below the canines is more
abrupt, and the depth of the body of the jaw more regular, than in H.
amphibius. The coronoid process is present, but the condyle and
descending portion are wanting. The coronoid process is not projected
so much forward as in H. amphibius.

Figs. 7 and 7 a.—H. Sivalensis. Fragment showing symphysis of
lower jaw, with canines and incisors remarkably perfect. The hori-
zontal direction of the six incisors and the peculiar curve of the canines
upwards and slightly backwards are well seen. The incisors are of nearly
equal dimensions, and the two central ones are not larger, as in H.
amphibius; they are cylindrical, and inclined outwards at an obtuse
angle to the plane of the grinding surface; their ends are truncated.
They are much larger than in the specimen shown in fig. 6, so that the
animal was probably an adult male—B.M. (Reproduced in Pl. XII.
figs. 2 and 3.)

Fig. 8.—H. Sivalensis. Large descending process of ramus of lower
jaw, detached. This remarkable appendage for the attachment of the
masseter and temporal muscles, peculiar to the genus, is even more
developed than in H. amphibius; it is less tapering and more deep and
massive in its proportions; the posterior margin is more round, and
the anterior, which in H. amphibius is curved and pointed forwards, is
here blunt and unmarked by any peculiarity of form. The process is
inclined outwards, and its outer surface is as marked for the reception
of muscles as in the living Hippopotamus.—B.M.

Fig. 9.—H. Sivalensis. Another specimen of descending process of

ramus of lower jaw. (Reproduced in Pl. XII. fig. 4.)

Fig. 10.—Anterior portion of palate, with sz incisors, and with
three premolars on right side, and two on left.—B.M.

_ Figs. 11 and 11 a.—Anterior portion of palate with canine and two
premolars ; viewed from above, and also laterally.—B.M.

502 FAUNA ANTIQUA SIVALENSIS.

Puiate LXII.

Fig. 1.—Hippopotamus (Hexaprotodon) Sivalensis. Left side of
lower jaw, viewed from above, with three molars and the three pos-
terior premolars. The trefoil wear of the coronals of each pair of
collines is well seen. The fragment is broken off in front of the second
premolar. The drawing shows the descending process of the jaw.
(Reproduced in PI. XII. fig. 1.)

Fig. 2.—H. Sivalensis. Right side of lower jaw, with three molars
and four premolars. The specimen is remarkable as showing the first
or foremost premolar, which in most of the specimens is wanting. The
canine is absent, but the three right incisors are present. The teeth
are less worn than in fig. 1, the posterior molar being intact. —B.M.

Fig. 3.—H. Sivalensis. Left side of lower jaw, viewed from above,
showing the two anterior molars, and a portion of the third or posterior
molar ; also the four premolars, the fourth or posterior one being very
small. Both the molars and premolars are well ground, so that the
animal was probably old. The specimen also shows the canine tooth
broken off, and the alveoli of the three left mcisors.—B.M.

Fig. 4.—H. Sivalensis. Anterior margin of lower jaw, showing the
six incisors, all about the same size, with the two canines incomplete.

—B.M.

Fig. 4a.—H. Sivalensis. Vertical section from side to side through
anterior portion of lower jaw, with sections through the six incisors
and two canines, showing the relative position of their alveoli B.M.

Fig. 5.—H. Sivalensis. Premolar from lower jaw.—B.M.
Fig. 6.—H. Sivalensis. Premolar from lower jaw.—B.M.

Figs. 7, 7a, and 7 6.—H. Sivalensis. Upper canine deeply grooved
along posterior surface and obliquely truncated in front. The trans-
verse section (7 }) presents a reniform outline.—B.M.

Figs. 8 and 8 a.—ZH. Sivalensis. Fragment of upper canine with
truncated anterior extremity and reniform outline on section.—B.M.

Figs. 9 and 9a.—H. Sivalensis. Fragment of upper canine, with
truncated anterior extremity and reniform outline on section.

Fig. 10.—H. Sivalensis. Lower canine, curved upwards and slightly
backwards at tip. The point obliquely truncated on its posterior
surface. The form of the tooth is such as to present a pyriform outline
when cut across.—B.M.

Fig. 11.—Hippopotamus (Tetraprotodon) Paleindicus. Fragment of
lower jaw, right side, with three molars and the three posterior pre-
molars. The hindmost molar is intact; those in front are moderately
worn. The fourth or hindmost premolar is very small—deciduous.
—B.M.

Fig. 12.—H. Paleindicus. Fragment of lower jaw, with three
molars and three posterior premolars. The teeth are more ground
than. in fig. 11.

Figs. 13 and 13 a.—Hippopotamus (Tetraprotodon) major. Upper
canine, obliquely truncated at front, with cordate outline on section.

DESCRIPTION OF PLATES. 503

Figs. 14 and 14a.—Hippopotamus (Tetraprotodon) amphibius.
Upper canine, truncated in front, with reniform outline on section.

Fig. 15.—WMerycopotamus dissimilis. (Fale. and Caut.) Lower jaw,
right side, with three molars, four premolars, portion of canine, and
alveoli of incisors. The teeth exhibit a ruminant-like pattern of wear
in the crown which is characteristic of the genus.

Fig. 16.—Merycopotamus dissimilis. Lower jaw, right side. Larger
specimen than fig. 15, with three molars, fourth molar, and canine.
The alveoli of three anterior premolars and three incisors are seen.
—B.M.

Figs. 17 and 17 a.—Merycopotamus dissimilis. Molars showing well
the rugous surface of the enamel, the basal cingulum, and the rumi-
nant-like pattern of wear characteristic of the genus—B.M.

Fig. 18.—WMerycopotamus dissimilis. Molar.

Figs. 19, 19 a, and 196.—This tooth was found in the Kalowala
Pass by Capt. (new Sir Proby) Cautley, and is figured by Royle in
‘Illustrations of Botany of the Himalayah Mountains’ (vol. ii. Plate
IIL. figs. 12, 18, 14, and 15), as the tooth of an Anthracotherium. In
several of Dr. Falconer’s published papers reference is made to the oc-
currence of Anthracotherium among the Sewalik fossils (See Synopsis
of Sewalik Fossils, in Journ. As. Soc., vol. iv. p. 706, and first paper
on Monkey, antea, p. 296; and also note, p. 508.) The specimen, how-
ever, is not named on the Plate, and differs from Anthracotheri'um.—
B.M.

PuateE LXITII.

Vertebre of Hippopotamus (Hexaprotodon) Sivalensis.

Figs. 1 and 1a.—Seven cervical vertebra in position, viewed an-
teriorly and laterally.

Figs. 2, 2a, and 26.—Atlas. Upper, lower, and anterior views.
—B.M.

Figs. 3, 3a, and 3b.—Atlas. Upper, lower, and lateral views.—
B.M.

Figs. 4, 4a, and 40.—Atlas. Upper, lower, and anterior views.—
B.M.

Figs. 5, 5a, 5b, and 5c.—Axis. Four different views.—B.M.
Figs. 6, 6a, 66, and 6¢.—Axis. Four different views.—B.M.
Figs. 7, 7a, 7b, and 7¢c.—Axis. Four different views.—B.M.
Figs. 8 and 8a.—Axis. Anterior and lateral views.—B.M.
Figs. 9 and 9 a.—Axis. Anterior and lateral views —B.M.
Figs. 10 and 10 a.—Axis? Upper and lateral views.—B.M.

Figs. 11, lla, and 110.—Sixth ? cervical vertebra. Upper, lower,
and lateral views.—B.M.

Figs. 12, 12a, and 120.—Third cervical vertebra. Upper lower,
and lateral views.—B.M.

504 FAUNA ANTIQUA SIVALENSIS.

Pirate LXIV.
Hippopotamus (Hexaprotodon) Sivalensis.

Figs. 1 to 12.—Vertebre, mainly dorsal and lumbar. Fig. 1 is a
cervical vertebra ; fig. 2 is the 5th cervical; fig. 3 is an anterior (1st?)
dorsal (see antea, p. 146); fig. 4 is the second dorsal; fig. 5 is the
fourth dorsal; fig. 6 is the second dorsal; fig. 7 is the fifth dorsal; fig.
8 is the seventh dorsal; fig. 9 is the fourth and fifth dorsals, united,
with a fragment of rib on each side; fig. 10 is the eighth dorsal; and
figs. 11 and 12 are lumbar vertebre.—B.M.

Figs. 18, 13 a, 14, and 14a.—Bones of sacrum, different views.—
B.M.

Figs. 15, 15a, and 150.—Sacrum. Anterior and lateral view.—
B.M.

Fig. 16.—Fragment of iium.—B.M.

Figs. 17 and 17 a.—Portion of pelvis, showing acetabulum. Two
different views.—B.M.

Figs. 18 and 18 a.—Portion of pelvis, showing acetabulum. Two

different views.—B.M.
Piate LXV.
Hippopotamus (Hexaprotodon) Sivalensis. Bones of the anterior
extremity. '

Figs. 1, 1 a, and 10.—Fragment of scapula, three different views,
showing spine, upper margin, and glenoid cavity.—B.M.

Figs. 2 and 2a.—Fragment of scapula, less perfect, showing spine
and glenoid cavity. —B.M.

Figs. 3 and 3 a—Fragment of scapula with spine and glenoid cavity.
—B.M.

Figs. 4 and 4. a.—Fragment of scapula with spine and glenoid cavity.
—B.M.

Fig. 5.—Glenoid cavity of scapula.—B.M.

Figs. 6 to 8.—Three specimens of upper articulating extremity of
humerus; three views of each.—B.M.

Figs. 9 and 9 a.—Shows the bones of the elbow joint, the lower end
of the humerus, and the upper end of the radius and ulna.—B.M.

Figs. 10 to 13.—Show four different specimens of lower end of
humerus; three views of each specimen.—B.M.

Figs. 14 and 15.—Two fragments showing upper articulating end of
ulna; two views of each.—B.M.

Figs. 16, 16 a, 16d, and 16 c.—Single bone of fore-arm, nearly per-
fect.—B.M.

Figs. 17 and 17 «.—Upper end of fore-arm.—B.M.
Fig. 18.—Lower end of fore-arm.—B.M.

Figs. 19 and 19 a.—Lower end of radius and ulna, with bones of
carpus :—a. scaphoid; 6. semilunar; c¢. cuneiform; d. pisiform in

~
al

%
4

:

DESCRIPTION OF PLATES. 505

outline ; é. trapezium in outline; /. trapezoid in outline; g. os magnum
in outline; h. cuneiform in outline.—B.M.

Figs. 20, 20 a, and 20 b.—Lower end of radius and ulna.—B.M.

Fig. 21.—Carpal bones in situ :—a. scaphoid; 6. semilunar in out-
line; c. cuneiform; /. trapezoid in outline; g.os magnum ; h. cuneiform.

Figs. 22 to 26.—Carpal bones detached; three views of each.
Fig. 22.—Right scaphoid.—B.M.

Fig. 23.—Right semilunar.

Fig. 24.—Right cuneiform.—B.M.

Fig. 25.—Right os magnum.—B.M.

Fig. 26.—Right cuneiform.

Figs. 27 to 82.—Metacarpal bones and phalanges.—B.M.

_ Figs. 33, 33 a, and 336.—Lower end of united radius and ulna,
right side. This specimen is remarkable, as being from Ava.—B.M.

Puate LXVI.
Hippopotamus (Hexaprotodon) Sivalensis. Bones of posterior
extremity.
Figs. 1, 1 a, 1%, and 1 c.—Left femur, entire —B.M.
Figs. 2, 2 a, and 2 b—Upper end of left femur.—B.M.
Figs. 3, 3 a, and 3 6.—Upper end of right femur.—B.M.

Figs. 4, 5, and 6.—Three fragments showing lower end of femur;
three views of each specimen.—B.M.

Figs. 7, 8, and 9.—Three patelle; two views of each_—B.M.

Figs. 10, 11, and 12. Fragments showing upper extremity of tibia;
three views of each.—B.M.

Figs. 13, 13 a, 13 b, and 13 c.—Entire tibia.—B.M.

Figs. 14, 15, 16, and 17.—Fragments showing lower end of tibia;
three views of each.—B.M.

Figs. 18 and 18 a.—Lower end of tibia and fibula, with bones of
tarsus :—a. astragalus; c. caleaneum; e. scaphoid.

Figs. 19 and 19 a.—Calcaneum.—B.M.
Figs. 20 to 25.—Bones of tarsus, detached.—B.M.

Figs. 20, 21, 22, 23, and 24. Represent different specimens of
astragalus; three views of each.

Figs. 25, 25 a, and 25 b.—Left cuboid bone.
Figs, 26, 26 a, 26 b, 27, and 27 a.—Metatarsal bones.—B.M.

Pirate LXVII.

Merycopotamus dissimilis. (Fale. and Caut.)

Fig. 1.—Var. major.! Upper surface of cranium, the anterior portion,
or muzzle, broken off.—B.M.

1 See page 138, note,

506 FAUNA ANTIQUA SIVALENSIS.

Fig. 1 a.—Lateral view of same specimen. The orbit is not elevated

above the plane of the frontal.

Fig. 1 5.—Palatine view of same specimen, showing three molars and
one premolar. The absence of the trefoil wear of the coronals is to be
noted. ach pair of collines takes a crescentic form outwards, not un-
like that of ruminants, and the grinding surface slopes outwards, as in
the description given by Cuvier of Hippopotamus minutus."

Fig. 1 ¢.—Posterior or occipital view of same specimen.

Figs. 2, 2 a, and 2b.—WM. dissimilis (var. major). Imperfect cranium
including muzzle. Lateral, upper, and palatal views.—B.M.

Figs. 3, 3a, and 3 6.—WM. dissimilis (var. major). Cranium ; upper,
palatine, and lateral views, showing three molars, four premolars, and
canines. The second left and first right premolars have dropped out.
The left canine is seen to be remarkably curved downwards, first out-
wards and forwards, and then slightly backwards—B.M. (Reproduced
in Pl. XIII. figs. 1 and 2.)

Figs. 4, 4a, 4b, and 4ce.-M. dissimilis (var. major). Lower jaw,
right side; outer, upper, and inner views. The alveoli of three in-
cisors and first three premolars are empty ; the three molars and fourth
premolar are present, but, excepting hindmost molar, are well worn ;
the canine is curved upwards and outwards and slightly backwards at
the tip; it is pear-shaped on section, as in Hipp. Sivalensis. 'The de-
scending process is well seen, and is separated from the horizontal ramus
by aconsiderable indentation. The anterior extremity of the horizontal
ramus is much more oblique than in H. Stvalensis, and the junction of
the lower with the anterior margin, corresponding to the lower end of
the symphysis, is marked by a distinct tuberosity or projection down-
wards (x). One large mentary foramen is seen on outer surface below
the fourth molar, and between this and the canine the bone is deeply
channelied ; the molar ridges are almost parallel, and there is very little
widening of the symphysial portion of the jaw. The great peculiarity
of the jaw is the general slenderness of its proportions and the inequality
of its depth. From the descending process it first becomes deeper, and
then it gradually diminishes towards the symphysis. In Hipp. Sivalensis
the jaw is straight, thick, and massive, as in Plate LXI. 3, 4, 5.—B.M.
(Reproduced in Pl. XIII. figs. 3 and 4.)

Figs. 5,5 a, and 5b.—WM. dissimilis (var. minor?). Cranium; upper,

1 «Jn the true molars of the Meryco-
potamus, the inner demi-cones are simply
convex, and the two grooves on the outer
ares form a deep external depression, at
the bottom of which is the convex ridge.
The antero-posterior cleft, instead of
being straight, as in the Hippopotamus,
forms two bends convex inwards, and
thus the symmetrical pattern of the
Hippopotamic molar is converted into
the double-crescentie are of the Rumi-
nant molar. The cement at the bottom
of the valleys is thinner than in the
Ruminants ; the enamel is as rugose as
in the Giraffe or Sivathere; but the

strong ragged ridge along the inner half
of the base of the crown forms the chief
distinction between the molars of the
Merycopotamus and those of the Rumi-
nant. The teeth in the lower jaw make
a similar approximation to the Rumi-
nant type, but the anterior and posterior
primary divisions are separated by a
wider cleft; the last molar has a third
hinder lobe; the lower molars are im-
planted by two roots. The forms, pro-
portions, and relative position of the
canines and incisors closely accord with
the Hippopotamic type of these teeth.’
Owen’s ‘Odontography,’ i. 566.

ra ee ee

DESCRIPTION OF PLATES. 507

palatine, and lateral views. The left zygomatic arch is almost complete;
the right is absent. The molars, premolars, and right canine are well
seen; the incisor ridge is mutilated. The molar ridges are parallel ;
the whole jaw tapers forward, and there is no widening of its anterior
extremity, and no abrupt angle between the line of the jaw and the
zygomatic arch.—B.M.

Fig. 6.—M. dissimilis (var. minor?). Lower jaw, right side, com-
prising the horizontal and part of the ascending ramus, with the ex-
panded disc below. The three molars are in situ. The premolars
have dropped out, but their alveoli are seen. The canine is also in
situ, but broken off. The molars in the original exhibit well the rugous
surface of the enamel, with the basal cingulum and the ruminant-like
pattern of wear on the crown which are characteristic of the genus,
which is nearly allied to Anthracotherium in the teeth. The colline
apices of the molars are more widely separated than in other Hippo-
potami. The specimen from which this figure is taken is in the
Museum of the Asiatic Society at Calcutta (Sewalik series, No. 246),
and is described by Dr. Falconer in the Catalogue of the Museum.
See antea, p. 146.—Cast in B.M.

Figs. 7, 7 a, and 7 6.—(Var. minor ?). Fragment of lower jaw, right
side, with molars and premolars in situ. The alveoli of the canine and
three incisors are seen in the broken surface in front. The ascending
ramus and descending process are broken off. The large mentary
foramen and deep channel in front are very distinct.—B.M.

Figs. 8 and 8 a.—WM. dissimilis. Fragment of anterior portion of lower
jaw, left side, with very perfect canine.

Piate LXVIII.
Figs. 1 to 18.—Merycopotamus dissimilis.

Figs. 1 and 2.—Two fragments of pelvis with acetabulum; two
views of each.—B.M.

Figs. 3, 3a, and 3 b.—Upper end of right femur.—B.M.
Figs. 4, 4a, 40, and 4 c.—Lower end of femur.—B.M.
Figs. 5, 5 a, and 5 6.—Upper end of tibia.—B.M.

Figs. 6, 6 a, 6b, and 6 c.—Fragment of caleaneum.—B.M.
Figs. 7, 7a, 7b, 7c, and 7 d.—Calcaneum.—B.M.

Figs. 8, 8a, 8 b, and 8 c.—Calcaneum.—B.M.

Figs. 9, 9a, 9b, 9c, and 9 d.—Astragalus.—B.M.

Fig. 10.—Calcaneum and astragalus zn situ.

Figs. 11 and 12.—T wo specimens of upper end of humerus; three
views.—B.M.

Figs. 13, 13 a, 13 6, 18.c, and 13 d.—Lower end of humerus.—B.M.

Figs. 14, 14a, 140, and 14 ¢.—Four different views of radius of
Merycopotamus dissimilis—B.M.

Fig. 15.—Fragment of occiput, showing condyles and occipital crest.
—B.M.

Fig. 16.—Fragment of lower jaw.

508 FAUNA ANTIQUA SIVALENSIS.
Figs. 17 and 17 a.—Lower jaw, anterior portion, both sides, showing
absence of any widening of symphysis.—B.M.

Figs. 18 and 18 a.—Incisive ridge.—B.M.

Figs. 19, 19a, 196, and 19c.—Hipp. Sivalensis. Calcaneum.—B.M.

Figs. 20, 20 a, and 20 6.—Hipp. Sivalensis. Astragalus.—B.M.

Fig. 21.—Hipp. Sivalensis. Calcaneum and astragalus, placed in
situ.

Figs. 22 and 22 a.—Anthracotherium Silistrense. Molars in Museum
Geol. Soc.

Figs. 23 and 23 a.—Anthracotherium Silistrense. Molars in Museum
Geol. Soc.

Fig. 24.—Anthracotherium Velaunum. Molars in Mus. Geol. Soc.
Fig. 25.—Anthracotherium Velaunum. Molars in Mus. Geol. Soc.

The Anthracotherium, like the closely allied Merycopotamus, formed
a link connecting the Hippopotamus with the Ruminants. The molars,
however, depart less from the Hippopotamic type than in Meryco-
potamus.*

Puate LXIX.

Figs. 1, 1a, 16, and 1c.—Sus giganteus (Falc. and Caut.). Upper,
palatal, lateral, and occipital views of skull. The zygomatic arches
are perfect. There are three molars on either side, and also the last
premolar. The specimen is broken off in front of the last premolar.
‘The extreme distance between the zygomata is much greater than in
Sus scrofa. The sub-orbital foramina are large, and the bone is deeply
channelled in front. From the Sewalik hills—B.M.

Length of fragment, 11:7in. Between the most distant points of the zygomata,
8:5 in. Between the post-orbital processes, 5-1in. Least breadth of cranium
between temporal fosse,1:lin. Height of occipital facet from lower border of
occipital foramen, 6°5in. Height of occipital foramen, ‘9in. Breadth of ditto,
l:in. Breadth of occipital condyle, 1°3in. From lower border of occipital foramen
to posterior border of palate, 3°8in. Height of cranium at sub-orbital foramen
from palate, 3°lin. Breadth of ditto superiorly, 2-4in. Least breadth of occipital
facet, 3°5in. Width of posterior nares, °7in. Length of three true molars, 3°2 in.
Of ditto, including last premolar, 3°7in. Width of palate, posteriorly, 1:7in. Of
ditto, anteriorly, 1‘Sin. Greatest breadth of alveoli, 1°3in. Height of posterior
nares, 1‘7in. Greatest diameter of orbit, 1°7 in.

Figs. 2, 2a, and 2 6.—Sus giganteus. Fragment showing anterior
portion of skull broken off about the line of the sub-orbital foramina.
Upper, lateral, and palatine views. The three molars and two last pre-
molars are well seen, and are less ground than in fig. 1.—B.M.

Length of fragment, 9° in. Width superiorly at sub-orbital foramen, 2-2 in.
Height, from palate, 4:in. Length of three true molars, 3°7in. Of ditto, including
two last premolars, 5"1in. Width of palate posteriorly, 1:6 in. Of ditto, anteriorly,
14in. Greatest width of alveoli, 1:3 in.

1 Dr. F. was at one time inclined to
regard the Merycopotamus as identical
in genus with the Anthracotherium Ve-
launum of Cuvier. On Dee. 6, 1843, he
wrote thus to Capt. Cautley: ‘What do
you think! Our Hippo. dissimilis is iden-
tical in genus with Cuvier’s Anthracothe-

rium Velaunum; that is to say, he mis-
named his specimen from imperfect ma-
terials. I have had the twoheadschiselled
out, and intend describing them under the
name of a new genus Merycopotamus
(merico, from the resemblance of the
teeth to those of a Ruminant),’

a

ee ea

first and second premolars, ‘9 in.

DESCRIPTION OF PLATES. 509

Figs. 3, 3 a, and 3 b.— Sus giganteus. Cranium. Upper, palatal, and
lateral views. The right zygoma is imperfect, and the left is almost
absent; but, in other respects, the cranium is more perfect than in figs.
1 or 2. There are three molars and three premolars on either side.
The incisive alveoli and the tuberosity and alveolus of the right
canine are also present.—B.M.

Length of fragment, 14°3in. Width of cranium superiorly at sub-orbital fora-
men, 1:‘8in. Height of ditto from palate, 3:3in. Length of canine tuberosity,
4-in. Length of molar series, 56in. Length of three true molars, 3°5in. From
posterior border of palate to anterior margin of incisive alveolus, 76in. From
posterior border of palate to posterior angle of incisive foramen, 8:‘7in. Diasteme
between canine and external incisor, 1-in. Width of palate posteriorly, 1-2 in.
Ditto between canines, 2°in. Greatest width of alveoli, 1:2 in.

Fig. 4.—Sus giganteus. Lower jaw, right side. The ascending
ramus is mostly absent. Shows three molars and three premolars, with
canine and incisive alveoli.

Length of fragment, 11°5in. Height of horizontal ramus, 2:in. Thickness of
ditto, 1-7in. Length of symphysis superiorly, 3°3in. Leugth of three true molars,
3:5in. Length of ditto, with three posterior premolars, 5:in. Interval between
Interval between second premolar and canine,
Between canine tuberosities, 3°9in. Width between molars posteriorly,
Between ditto anteriorly, 1°7 in.

15 in.
15 in.

Figs. 5 and 5 a.—Sus scrofa (var. Indicus).
lower jaw, not fossil. Upper and lateral views.
natural size.—B.M.

Entire skull, with
One-third of the

PLATE LXX.

Figs. 1 and 1a.—Sus (Hippohyus) Sivalensis (Fale. and Caut.).'
Cranium. Upper and palatal views. Except that the zygomatic arches
are absent, the specimen is very perfect. Shows three molars and two
last premolars on either side, with alveoli of canines and six incisors.
The sub-orbitary and incisive foramina are well marked.—B.M.

Extreme length of fragment, 9°2in. From post. plane of occipital condyles to
anterior margin of incisive alveolus, 9:in. From lower border of occipital foramen

to post. border of palate, 2°3in. From post. border of palate to posterior border of

incisive foramen, 5'4in. Width of palate between second molars, 1-1in. Width of
ditto between inner margins of canine alveoli, 1:05in. Width of ditto between
anterior angles of middle incisive alveoli, ‘7in. Greatest width of alveoli, -8 in.

Length of the molar series, 43 in. Length of the three true molars, 2°7 in. Dias-

1 «Tn this extinct genus of quadrupeds
from the Himalayan tertiary deposits,
the dental formula shows incisors =
and corresponds with that of the Chero-
potamus in the number of canines, pre-
molars and molars; but the true molars
have a more complex crown, approaching
nearer to those of the typical Swide in
the depth and number of the secondary
enamel folds. Each upper true molar
has its crown cleft by the common or

primary crucial valleys, the transverse

one passing somewhat obliquely from
within forwards and outwards. Each of
the four principal lobes is subdivided,
not by a vertical central depression,

but by a fold penetrating its anterior
and posterior margins. The enamel at
first shows additional minor plications,
but is worn down to the simpler pattern
above described; the outer lobes are
convex externally. The first premolar
is very small and simple, separated by
an interval of its own breadth from the
second; both this and the third have
transversely compressed crowns; the
fourth has a sub-trihedral crown. The
Hippohyus equalled in size the Chero-
potamus, but exhibits as strong a ten-
dency towards the Hippopotamoid family
as that does towards the plantigrade
Carnivora.’—Owen’s ‘Odontography, vol.
1. p. 562.

510 FAUNA ANTIQUA SIVALENSIS.

tema between first and second premolars, ‘2in. Length of incisor ridge on’ one
side, 1‘5in. Breadth of nasal ridge at sub-orbital foramina, 1°2in. Between post-
orbital processes, 2°65 in.

Figs. 2 and 2 a.—Sus Hysudricus (Fale. and Caut.). From Sewalik
hills. Anterior portion of skull broken off about sub-orbitary foramina.
Shows three molars and three premolars and canine in situ.—B.M.

Length of fragment, 5°7in. Height from palate opposite sub-orbital foramen,
2:2in. Length of three true molars, 1‘8in. Length of ditto and three posterior
premolars, 3°3in. Diastema between first and second premolars, ‘lin. Length of
whole molar series, 4" in.

Figs. 3 and 3 a—Sus Hysudricus. Lower jaw. The ascending
ramus and the incisive and canine alveoli are absent. Shows three
molars and four premolars.—B.M.

Extreme length of fragment, 6'4in. Height of jaw opposite second molar, 2° in.
Length of three molars, 3°in. Ditto of entire molar series, 5:3 in.

Figs. 4, 4a, and 4b.—Sus giganteus. Superior, palate, and side
views of cranium. The specimen is imperfect and mutilated. Pos-
teriorly it is broken off behind the orbit. Anteriorly it is also frac-
tured in front of the first molar, but the anterior fragment is joined on.
The specimen shows the two anterior molars and the third molar in
germ.—B.M.

Length of fragment, 8:in. Length of palate, 5°5in. Height from palate at sub-
orbital foramen, 1°8in. Width of cranium superiorly at ditto ditto, 1-in. Length
of exposed molar series, 3°in. Of first and second molars, 1-4in. Between first
premolar and external incisor, ‘6 in. Between first premolar and internal incisor,
1:'4in. Width of palate posteriorly, ‘8 in. Width of palate between canines,
1°3 in.

Fig. 5.—Sus giganteus.—Lower jaw, showing three molars and three
premolars on either side. The left canine and the alveolus are also
seen. The incisive alveoli are imperfect.—B.M.

Extreme length of fragment, 12:in. Length of three true molars, 3-8in. Of
ditto with two premolars, 5:in. Distance between canine and second premolar,
1:‘9in. Width of symphysis between external margins of canine alveoli, 3°4 in.
Length of symphysis, 5-2 in.

Fig. 6.—Sus giganteus. Imperfect specimen of lower jaw. Shows
the molars and premolars on right side, and the premolars and canine
alveolus on left; also the incisor alveoli.—B.M.

Length of fragment, 7-4 in. Height of ramus, 2°lin. Width of ramus, 1:4 in.
Length of symphysis, 2°8in. Width between external alveolarmargins of canines,
2°8in. Alveolar margin of four incisors, 1:2 in.

Figs. 7 and 7 a.—Sus giganteus. Lower jaw, showing three molars,
four premolars, canine, and three incisors. The last or posterior
molar is intact. The ascending ramus is absent.—B.M.

Length of fragment, 10°in. Depth of ramus at anterior margin of third molar,
2-in. Width of ditto, 14in. Length of symphysis, 2°8in. Between external
alyeolar margins of canines, 2°3 in. Length of three posterior premolars and three
true molars, 4:9in. Of three true molars, 3°2in. Between first and second pre-
molars, ‘5 in. Between first premolar and canine, ‘3in. Alveolar margin of three
incisors, 1°3 in.

Figs. 8 and 8 a.—Sus giganteus. Another specimen of lower jaw,
with molars, premolars, and canine on both sides; also the incisor
alveoli. The third or last molar is only partly ground. The ascend-
ing ramus is absent. From the Nerbudda.—B.M.

Length of fragment, 10°1,in. Depth of ramus at anterior margin of last molar,

in:

eT aa a

DESCRIPTION OF PLATES. 511

2-in. Width of ditto, 1:2in. Length of symphysis, 3°5in. Between external al-
veolar margins of canines, 2°7in. Length of three last premolars and three true
molars, 5°5in, Length of true molar series, 3:1 in. Between first and second pre-
molars, ‘6in. Between first premolar and canine, ‘Sin. From anterior margin of
canine alveolus to mesial line, 1°5in. Width between rami posteriorly, 1°d in.

PiaTe LXXI.
Fig. 1.—Sus (IHippohyus) Sivalensis. Left half of palate, natural

size, showing three molars, three posterior premolars, and the alveoli of
the first premolar, canine, and three incisors. The third or hindmost
molar is not at all ground.—B.M.

Extreme length of palate, 6-3 in. Length of first molar, ‘6 in.; of second, ‘95in. ;
of third, 1:15in. Width of first molar, ‘6in.; of second, ‘8in.; of third, ‘8in. °
Length of third premolar, *5in.; of fourth,*5in. Width of third premolar, °4 in. ;
of fourth, ‘5 in.; between canine and first premolar, 6 in.

Fig. 2.—Sus (Hippohyus) Sivalensis. Second and third molars,
imperfect.—B.M.

Length of fragment of second molar, *55 in.; of third, 1°45in. Greatest width of
third posteriorly, 73 in.

Fig. 3.—Sus (Hippohyus) Sivalensis. First and second true molar
and portion of fourth premolar.—B.M. :

Length of first true molar, ‘5 in.; of second ditto, 1*in. Width of first true molar,
45 in. ; of second, *63 in.

Fig. 4.—Sus (Hippohyus) Sivalensis. Fourth premolar and first and
second true molars.—B.M.

_ Length of first molar, ‘55 in. ; of second, ‘85 in.

Fig. 5.—Sus Hysudricus. Three true molars and third and fourth
premolars, upper jaw, left.—B.M.
Length cf third premolar, ‘5 in.; of fourth ditto, ‘5 in.; of first molar, 6 in. ; of

second, ‘Sin.; of third, 1:‘1in. Width of third premolar, °35 in.; of fourth, ‘53 in. ;
of first molar, ‘55 in.; of second, ‘66 in.; of third, -7 in.

Fig. 6.—Sus Hysudricus. Three molars and three posterior pre-
molars, lower jaw, left. The last molar is imperfect.—B.M.

Length of second premolar (at alveolar ridge), -45in.; of third, -45 in.; of fourth,
*bin.; of first molar, ‘55in.; of second, *7in. Width of fourth premolar, -35in. ;
of first molar, -43in.; of second, 53 in.

Fig. 7.—Sus Hysudricus. Second and third true molars, upper jaw,
right.—B.M.

Length of second molar, *65in.; of third, -85in. Width of second molar, °7 in.
Extreme width of third molar anteriorly, °7 in.

Fig. 8.—Sus Hysudricus. Second and third true molars, well worn.
—B.M.

Length of second molar, *73in.; of third, 1-4in. Width of second, ‘6in.; of
third anteriorly, -65 in.

Fig. 9.—Sus Hysudricus. Canine, four premolars, and first and
second molar, upper jaw, right.—B.M.

Length of first premolar, or retained milk molar, *4in.; of second, ‘din.; of
third, ‘5 in.; of fourth, :45 in. ; of first molar, ‘6 in. ; of second, ‘66 in,

512 FAUNA ANTIQUA SIVALENSIS.

Fig. 10.—Sus Hysudricus. Three molars, and two posterior pre-

molars, lower jaw.—B.M.

Length of last premolar, ‘5 in.; of first molar, ‘Sin. ; of second, ‘66 in. ; of third,
1-in. Width of last premolar, 35 in.; of first molar, *4in.; of second, ‘5in.; of
third ditto anteriorly, ‘5 in.

Fig. 11.—Sus Hysudricus. Symphysis of lower jaw, with four pre-
molars on left side, and second and third on right.—B.M.

Distance between inner margins of second premolars, 1°3in. Length of second
premolar, °45 in. ; of third, ‘din. ; of fourth, ‘din.

Fig. 12.—Sus giganteus. Three true molars, upper jaw, left. The
first is imperfect.—B.M.

Length of first molar, 65 in. ; of second, 1:in.; of third, 1-6in. Width of first
molar, *75 in; of second, ‘9in.; of third anteriorly, 1: in.

Fig. 13.—Sus giganteus. Second and third true molars, upper jaw,
right; large and perfect.—B.M.

Length of second molar, 1:4in.; of third, 2°2in. Width of second molar, ‘8 in. ;
of third ditto, 1*1 in.

Fig. 14.—Sus giganteus. Lower jaw, right side, showing canine,
four molars, and first two true molars. The second true molar is not
at all ground down, and the third has not appeared.

Distance between canine and first premolar, ‘36in. Length of first premolar,
+86 in.; of second, ‘5 in.; of third, ‘5 in.; of fourth, ‘5in.; of first molar, °85 in. ;
of second, 1:25in. Width of first molar, °7in.; of second, ‘8 in.

Fig. 15.—Sus giganteus. Lower jaw, right side, with first, second,
and third true molars, all well ground.—B.M.

Length of first molar, *55in.; of second, ‘95in.; of third, 1‘8in. Width of first
molar, ‘55in.; of second, *7 in.; of third, °7 in.

Fig. 16.—Sus giganteus. Last premolar and first molar, upper
jaw.—B.M.

Length of last premolar, ‘7in.; of first molar, ‘93 in. Width of last premolar,
*8in.; of first molar posteriorly, 9 in.

Figs. 17 and 17 a.—Sus giganteus. Anterior portion of lower jaw,
showing canine and incisive alveoli. The outer incisor on both sides
has dropped out; the two inner incisors and the canines are present;
the latter are broken off.—B.M.

Distance between outer margins of outer incisive alveoli, 2:1 in.; between outer
margins of middle ditto, 1°6in.; antero-posterior diameter of middle incisor, ‘36 in.;
of inner ditto, °4 in.

Figs. 18 and 18 a.— Sus giganteus. Symphysis of lower jaw, with
six incisors very perfect. The canines are broken off.—B.M.

Length of symphysis measured inferiorly, 2°6 in. Distance between outer mar-
gins of outer incisors, 2° in.; between outer margins of second premolars, 1°8 in.
Width of three incisors on one side,. 1°2 in.

Figs. 19, 19 a, and 19 6.—Sus giganteus. Fragment of canine,
slightly compressed and grooved on each side.—B.M.

Length of fragment, 2°4 in.; great diameter, 1:15in.; lesser diameter, °8 in.

1 Description by Dr. Falconer of Fossil | specimen, consisting of a mass of bones
Remains of Suide from the Sewalik | cemented together by clay-marl and
Hills, in the Museum of the Asiatic | crossing each other in every direction ;
Society of Bengal. the principal object being the lower jaw,

No. 317. Sus——? Conglomerated | nearly entire, of a Sus, exposed so as to

A

DESCRIPTION OF PLATES. 513

Puate LXXII.

Figs. 1, 1 a, and 1 6.—Rhinoceros platyrhinus. (Fale. and Caut.)
From the Sewalik hills. Mutilated cranium, anterior part, showing
lateral, upper, and palate surfaces. The specimen is so worn that the
teeth are scarcely distinguishable. The upper surface of the skull is
broad and flat—_ B.M. (See antea, p. 157, and Pl. XIV. fig. 3.)

Length of fragment, 17’in. ; height posteriorly, 9-7 in.; height anteriorly, 8: in. ;
greatest breadth at anterior angles of orbits, 10°6 in.; depth of nasal notch, 6° in. ;
height of nasal notch anteriorly, 5°5 in.

In 1847 Dr. Falconer noted that R. platyrhinus partakes of the
characters of both &. leptorhinus (sic) and R. tichorinus, and on the 9th
of August, 1860, he made the following note :—

‘Examined Baker’s large skull of the Sewalik Rhinoceros platyrhinus in B.M.
The molars are in fine condition, six on either side. The last true molar only just
touched by wear. The last t. m. exactly like Rh. hemitechus, in having a pos-
terior basal funnel-shaped pit! while the penultimate and antepenultimate t. m.
and the penultimate and antepenultimate milk m. have each three distinct fossettes,
as in Rhinoceros tichorhinus! the vertical ridges of the anterior side very well
pronounced in three valleys. Had two large incisors above and four below: of the
latter, the two outer big ; the two inner small, as in the existing Indian Rhinoceros.’

Figs. 2, 2 a, and 2 b.— Rhinoceros platyrhinus. Fragment showing
posterior part of cranium, with foramen magnum, occipital condyles
and crest, portion of right zygomatic arch, and condyle of lower jaw.

Length of fragment, 10°6in.; height of occipital facet from lower margin of
occipital foramen to summit of occipital crest, 12:in. Breadth of occipital facet
above, 84in.; ditto below, 13°2in. Height of occipital foramen, 2°5in.; breadth
of ditto, 2>in. Between extreme points of occip. condyles, 5‘3in. Least width of
cranium, 3°3in. Breadth of condyle of lower jaw, 6°7 in.; ditto of ascending ramus,
6-in. Between inner angles of glenoid facets, 2°5in. Depth of zygomatic process,

3°3 in.

foramen, 7°5 in.
_veoli, 2°8 in.

Fig. 3.—Rhinoceros platyrhinus.
with molar ridge, and large sub-orbital foramen.—B.M.

Length of fragment, 13-2in. From root of molar origin of zygoma to sub-orbital
Length of molar series, 10°8 in.

Fragment of skull, upper jaw,

Greatest breadth of molar al-

show the two horizontal rami with the
remains more or less of seven molars on
either side, the bases of both canines
and more or less of the six incisors.
_ The specimen is still much covered with
matrix; the four premolars on the left
side show part of their crowns; on the
right side the first premolar is close to

worn; the canine on the left side shows
a part of the tooth bending outwards,
but the apex broken off. The other
bones are so much covered by matrix as
to be undeterminable.

No. 318. Sus ? Fragment com-
prising the posterior part of upper max-
illa right side, containing the two last
teeth in situ; the penultimate is well

worn, showing a very complex pattern
_ of crown; the last molar is half worn.

VOL. I.

the canine; the true molars are well |

No. 319. Sus ? Fragment of
lower jaw, right side, comprising pos-
terior part of horizontal ramus, broken
across horizontally near the base of the
| teeth, and containing the last two mo-
| lars, the penultimate well worn with
very flexuous enamel ; the last molar in
| germ and of very large size.

No. 320. Mutilated fragment com-
prising part of the last true molar, much
| broken and cemented with matrix.

No, 321.. Fragment comprising the
posterior part of horizontal ramus lower
jaw right side, containing the two last
teeth im situ ; they are in the same con-
dition of wear as No. 319, but consi-
derably smaller.

No. 54 (from Perim Island). Lower
jaw, left side, fragment containing
merely the last molar of Sus Hysudricus ?

LL

514 FAUNA ANTIQUA SIVALENSIS.

Fig. 4, 4a, 4b, and 4.¢.—Rhinoceros platyrhinus. Fragment show-
ing anterior portion of lower jaw, with symphysis and four anterior
molars, and a portion of fifth; also a small inner and large outer
incisor on both sides. —B.M. (See antea, p. 157.)

Length of fragment, 13°5in. Breadth of symphysis, 5°7in. Length of sym-
physis inferiorly, 7-in. Depth of jaw,.4-7in. Thickness of jaw, 3°3in. Length of
four anterior molars, 7°-4in. Between anterior premolar and external incisive al-
veolus, 3:lin. Between incisive alveoli, ‘6in. Width between molars posteriorly,
4-in.; ditto anteriorly, 3°4 in.

Figs. 5, 5 a, and 5 6.—Rhinoceros platyrhinus. Small fragment of
lower jaw, with two molars.—B.M.

Length of fragment, 6°7in.; greatest depth, 5°7in.; Thickness, 3-2; length of
molar, 3°1in.; breadth, 1°7 in.

Figs. 6 and 6 a.—Rhinoceros platyrhinus. Fragment of molar.—B.M.

Length, 2°3in. Width, 3°4 in.

Figs. 7 and 7 a.—Rhinoceros platyrhinus. Molar.

Length, 3°2in.; breadth, 2°8in.; height of crown, 3°1 in.

Puate LXXIII.

Figs. 1, 1a, 1b, and 1¢e.—Rhinoceros Paleindicus. (Fale. and Caut.)
Mutilated specimen of cranium. The zygomatic arches and the anterior
portion of the palate are broken off. On the right side the three true
molars and three posterior premolars are present; on the left there are
three molars and one premolar. All the teeth are much worn. The
upper surface of the skull is very concave.—B.M. (See antea, p. 157.)

Length of fragment, 21°8in. Height of occiput (imperfect) from basilar process,
8:1in. From occipital surface to posterior border of palate (imperfect), 12° in.
Between mastoid angles, greatest diameter of occiput, 9'in. Transyerse diameter
of occipital foramen, 1°9 in. Vertical ditto, 1°3in. Breadth of cranium at anterior
orbital angles, 8°7in. Between anterior angles of orbital margin, 3°9in. Between
sub-orbital foramina (posterior border), 48 in. Chord of nasal notch, 4:5 in, Length
of three true molars, 6-1in. Length of three posterior premolars, 5°in. Width of
palate between posterior molars, 2°2in. Ditto between second premolars, 2°65 in.
Greatest width of alveolus, 2°8in. Length of palatine notch, 5-5in. Width of
ditto, 2°in.

Figs. 2, 2a, 2b, and 2¢.—Rhinoceros Sivalensis (Fale. and Caut.),
from the Sewalik hills. Tolerably perfect specimen of cranium. The ~
upper part of the occiput and the left zygoma are absent. The left —
maxilla shows three molars and three premolars, and also the alveolus ~
of the first premolar. The teeth are well worn; the palate is narrow.
The upper surface of the cranium is concave, and the tip of the nasal
shows the gibbosity of the base of a very large horn. The species was —
evidently unicorned—B.M. (See antea, p. 157, and Pl. XIV. fig. 1.)

Extreme length of fragment, 22°5in. From posterior plate of occipital condyles
to anterior margin of first premolar, 20-4in. From lower border of occipital fora-
men to posterior border of palate, 11‘9in. Length of molar series, 11‘1in. Length
of three true molars, 5°8in. Width of palate between posterior molars, 2°5 in.
Width of palate at anterior angle of first premolars, 2-2in. Greatest width of al-
veoli, 26in. Length of palatine notch, 5:3in. Width of ditto,1‘9in. Between
inner angles of articular surfaces for lower jaw, 3°2in. Between most distant
points of zygomatic processes, 13°7 in. Depth of zygomatic fossa, 38-lin. Height of —
occiput (imperfect) from lower border of occipital foramen, 9° in. Between outer
angles of occipital condyles, 4°8in. Between mastoid angles, or greatest transverse
diameter of occiput, 8°6in, Breadth of occipital foramen, 1°7in, Height of ditto,

——

————————— ee

es

DESCRIPTION OF PLATES. 515

1:‘9in. Breadth of cranium at anterior orbital angle, 8°3in. Between posterior
borders of sub-orbital foramina, 5-2in. Between anterior angle of orbital margin
and posterior border of sub-orbital foramen, 4°7 in. Breadth of rostrum, 2°8 in.

Figs. 3 and 3 a.—Rhinoceros Sivalensis. Fragment of skull, com-
prising upper jaw, portion of orbit and prolongation of nasals for
horn.—B.M.

Length of fragment, 14°8in. From anterior angle of orbit to tip of nasal pro-
tuberance, 9°8in. From concavity of nasal notch to tip of ditto, 6°8 in.

Pruate LXXIV.

Figs. 1, 1 a, 1 0, and 1 e.—Rhinoceros Paleindicus. Very perfect
specimen of cranium, with both zygomatic arches entire. Shows two
molars and two posterior premolars on either side. ‘The third molar is
stillin germ. ‘The palate is deficient in front—B.M.

Length of cranium (fragment), 18°2in. Between most projecting points of zy-
gomata, 9°8in. Breadth of occiput (behind the auditory foramina), 6-1in. Least
breadth of cranium (between the temporal fossz), 3-4in.. Breadth of cranium at
anterior orbital angles, 5:3 in. From anterior margin of second premolar to pos-
terior border of pterygoid process, 9°7in. Length of palatine fissure, 3°7 in. Dis-
tance between the internal angles of the glenoid facets, 3:in. Length of alveolar
margin of exposed molars, 6*lin. Between external alveolar margins of last ex-
posed molars, 6°4in. Between external alveolar margins of anterior molars, 3-7 in,
Height of cranium from alveolar margin at anterior margin of third molar, 6:1 in.
Width of palate anteriorly, 2°3 in. ; ditto, posteriorly, 2°3 in.

Figs. 2, 2 a, 2b, and 2 c.—Rhinoceros Paleindicus. Skull of a
larger and older animal than fig. 1. Both zygomatic arches are defi-
cient, and the portion in front of the fourth premolar is also broken
off.—B.M.

Length of fragment, 20°1 in. From lower margin of occipital foramen to posterior
border of palate, 12:2in. From ear (anterior margin) to-sub-orbital foramen,
18:5in. From ditto to anterior angle of orbit, 10°6in. Height of occipital facet
from lower margin of occipital foramen to occipital crest, 7-7in. Height of cra-
nium at anterior angle of orbit from alveolar border, 7:2in. Height of occipital
foramen, 1°2in. Breadth of ditto, 1:2in. Between internal angles of glenoid
facets, 3°3in. Width of palate posteriorly and anteriorly, 3-lin. Between ex-
treme points of external alveolar borders of molars, 10°in. Least breadth of cra-
nium (between temporal foss), 4°3in. Breadth of cranium at anterior orbital
angles, 8°5in. From centre of occipital crest to posterior border of nasal notch,
16°7in. Length of alveolar border of three true molars, 6°5in. Breadth of alveoli,
3°2 in. :
Figs. 8 and 3. a.—Rhinoceros Paleindicus. Fragment of lower jaw,
left side, with four posterior molars.—B.M.

Length of fragment, 15°8 in. Length of alveolar border of molars, 8:3 in.
Breadth of ascending ramus, 6'in. Depth of jaw anteriorly, 3-1in. Thickness of
ditto, 3:2 in.

Figs. 4 and 4 a.—Rhinoceros Paleindicus. Fragment of symphysis
of lower jaw, with incisive alveolar ridge and large outer left incisor.

—B.M. (See antea, p. 157.)

Between external alveolar borders of incisive alveoli, 4:5in. Length of existing
portion of symphysis, 4-4in. Interval between anterior premolar and incisive al-
yeolus, 2°6in. Greatest thickness of alveolus, 1-7in. Great diameter of incisor,
1:3in. Lesser diameter of ditto, 1:lin, Length of tusk (projection), 1-9 in.

Fig. 5.—Rhinoceros Sivalensis. Portion of cranium, showing palate
with molar ridges and nasal projections. The portion behind the

second molar is broken off—B.M.
DL 2

516 FAUNA ANTIQUA SIVALENSIS.

Length of fragment, 14°8in. Width of palate between second molars, 2°3 in.
Width of palate between first premolars, 1‘6in. Length of four premolars and
first and second molar series, 8‘in, Length of four premolar series, 4-9in. Sup-
posed depth of nasal notch, 6°6in. Breadth of cranium between anterior angles of
orbit, 77 in. Between external alveolar borders posteriorly, 7:3 in. Between ditto
anteriorly, 2°5 in.

Figs. 6 and 6 a.—Rhinoceros Sivalensis. Fragment of lower jaw,
with symphysis and five anterior molars.—B.M.

Length of fragment, 9'4in. Length of existing portion of symphysis, 3:1 in.
Length of molar series, 7°-in. Width between the posterior molars, 2:7 in. Be-

tween anterior ditto, 2:in. Greatest depth of jaw, 3°6in. Thickness of ditto,
27 in,

PuLate LXXYV.

Fig. 1— Rhinoceros Paleindicus. Fragment of upper jaw, left side,
with three true molars.

Length of first molar, 2:in.; of second, 2:15in.; of third along anterior edge,
3'lin. Width of first molar, 3*in.; of second, 3:2in.; of third along anterior
edge, 2°9 in.

Fig. 2.—Rhinoceros Paleindicus. Fragment of lower jaw, with
three true molars and fourth premolar.—B.M.

Length of fourth premolar, 1‘8 in. ; of first molar, 2°15 in.; of second, 2°2 in.; of
third, 2°2in. Width of fourth premolar, 1:35 in. ; of first molar, 1‘3 in.; of second,
1:4in.; of third, 1:3 in.

Fig. 3.—Rhinoceros Palwindicus—Fragment of lower jaw, with
four molars.—B.M.

Length of first tooth, *65in.; of second, 1:3in.; of third, 1:75in.; of fourth,
1°75 in.

Fig. 4.—Rhinoceros Paleindicus. Premolar tooth detached.—No.
39,648 B.M.

Length along outer edge, 2°5in, Width of grinding surface anteriorly, 2°5 in.

Fig. 5.—Rhinoceros Sivalensis. Fragment of upper jaw, right side,
with three true molars and third and fourth premolar.—B.M. (Repro-
duced in Pl. XIV. fig. 2.)

Length of third premolar, 1°6in.; of fourth ditto, 1:5 in.; of first molar, 1°75 in.;
of second ditto, 2-in.; of third ditto along outer edge, 2:3in. Width of third
premolar, 2°3in.; of fourth ditto, 2°6 in.; of first molar, 2‘6in.; of second ditto,
2°6in.; of third ditto along anterior edge, 2°5 in.

Fig. 6.—Rhinoceros Sivalensis. Lower jaw, right side, with second,
third, and fourth premolars, and first and second true molars.—B.M.

Length of second premolar along outer edge, 1-2in.; of third premolar at centre
of grinding surface, 1:2in.; of fourth ditto, 1:7in.; of first molar, 1:6 in. ; of
second ditto, 1°9in. Width of second premolar, -75in.; of third ditto, 1:0 in. ;
of fourth ditto, 1:15in.; of first molar, 1°3in.; of second ditto, 1°36in. Width
between anterior angles of second premolars, 2°4 in.

Fig. 7.—Rhinoceros Sondaicus (recent). Upper jaw, right side, with
fourth premolar and three true molars.

Length of fourth premolar, 1-55in.; of first molar, 1:6in.; of second, 1:9 in. ;
of third along outer edge, 2°1 in.

Fig. 8.—hinoceros Sondaicus (recent). Lower jaw, right side, with
third and fourth premolars and three true molars.

Length of third premolar, 1:2in,; of fourth, 1‘5in.; of first molar, 1°6in.; of
second, 1‘8in.; of third, 1°7 in.

DESCRIPTION OF PLATES. 517

Fig. 9.— Rhinoceros platyrhinus. Upper jaw, right side, with third
and fourth premolars and three true molars. The first true molar is
imperfect.—B.M.

Length of third premolar, 1°6 in. ; of fourth ditto, 1:8in.; of fragment of first
molar, extreme, 1'5in.; of second molar, 1°9in.; of third ditto along outer edge,
2-2in. Width of third premolar, 2°7in.; of fourth ditto, 2°8in.; of second molar,
2-9in.; of third ditto along anterior edge, 2°55 in.

Fig. 10.—Rhinoceros platyrhinus. Lower jaw, right side, and sym-
physis, containing very large outer and small inner incisor of both sides,
second, third, and fourth premolars, and first two true molars of right
side—B.M. (Reproduced in Plate XIV, fig. 4.)

Length of second premolar, -7in,; of third ditto, 1-4 in.; of fourth ditto,
1°65 in. ; of first molar, 1-46 in. ; of second ditto, 2:in. Width of second premolar,
45 in.; of third ditto, °85in.; of fourth ditto, 1‘lin.; of first molar, 1:05in.; of
second ditto, 1:°2in. Width between second premolars, 3°5 in. ; ditto between outer
margins of external incisors, 3°65in. Oblique width of external incisor, 1-5 in.
Thickness externally of ditto, oblique, ‘7in. Length of exserted portion along outer
edge, 2°1in.

Fig. 11.—Rhinoceros platyrhinus. Penultimate true molar upper
jaw, right side, detached, but shattered. Fig. 11 a.—Ditto, ditto, re-
stored.—B.M.

Fig. 12.—Last true molar upper jaw, right side—B.M.
Fig. 18.—Rhinoceros (Acerotherium?) Perimensis (from Perim Island).
Fragment of lower jaw, with three true molars and first premolar.
Length of first true molar, 1:15in.; of second, 1°4 in.; of third, 1°5 in.
Fig. 14.—Rhinoceros Perimensis. Premolar tooth, detached.
Fig. 15.—Rhinoceros Perimensis. Molar, detached and shattered,

Fig. 16.—Rhinoceros Perimensis. Molar, detached and shattered.

Puate LXXVI.

Divers Indian Fossil Species of Rhinoceros.

Figs. 1 and 1 a.—Fragment of left humerus, near upper end, from
the Niti Pass—B.M. (See antea, p. 173, and Plate XV.)

Length of fragment, 5°9 in. Breadth, 35in. Greatest thickness of fractured
surface, 2°3 in.

Fig. 2.—Fragment of left humerus, near upper end, from the Niti

Pass.
Length of fragment, 5-2in.; greatest breadth, 5:2 in.; thickness, 2°3 in.

Fig. 8.—Upper extremity and portion of shaft of left radius, from the
Niti Pass—B.M.

Length of fragment, 6'in.; greatest antero-posterior diameter of superior arti-
cular surface, 2°5in. Transverse diameter of perfect portion, 2-1in. Transverse
diameter of shaft at fractured portion, 2:1 in. Antero-posterior diameter of ditto,
1din.

Fig. 4.—Upper extremity and portion of shaft of tibia, from the
Niti Pass —B.M.

Length from anterior margin of the crista tibize to posterior border of articular
wurface, 4°8in. Breadth of inner condyloid fossa, 2°6in, Antero-posterior diam, of
inner condyloid fossa, centre, 2°3 in,

518 FAUNA ANTIQUA SIVALENSIS.

Figs. 5, 5a, 5b, and 5 c.—Scaphoid bone of carpus, left side, from
the Niti Pass. —B.M.

Greatest antero-posterior diameter, 2°2 in. Greatest transverse ditto, 3°3 in.
Greatest vertical, 2°6 in.

Figs. 6, 6 a, and 6 b.—Fragment of scapula, including glenoid cavity
and coracoid process, from the Niti Pass.—B.M.

Length of fragment, 5‘8in. Height of glenoid cavity, 2°9in. Greatest breadth
of ditto, 2-4in. Height of coracoid process above glenoid cavity, 1°6 in.

Figs. 7, 7a, 76, and 7¢.—First phalanx, from the Niti Pass.—
No. 39,654 B.M.

Length (superiorly), 1:3in. Transyerse diameter of posterior surface, 2° in.
Vertical ditto, 1:in. Transverse diameter of anterior surface, 1°7 in.

Figs. 8, 8a, 8b, and 8 c.—Second phalanx, from the Niti Pass.
—B.M.

Length between centres of articular surfaces, 1‘lin. Greatest breadth, 2°5 in.
Breadth of posterior articular surface, 1:8in. Breadth of anterior articular ditto,
1:6in. Height of posterior articular surface, °8 in.

Fig. 9.—Fragment of bone of Rhinoceros, from the Niti Pass.

Length of fragment, 1-7 in. Breadth, 1-2 in.

Figs. 10, 10 a, and 10 .—Fragment of lower end of femur, from the
Niti Pass.—B.M.

Antero-posterior diameter internally, 6Gin. Length of rotular surface ditto,
4-3in. Length of rotular surface in centre, 2°8in. Breadth of ditto in centre of
height, 2°8 in.

Figs. 11, 11 a, and 11 6.—Fragment of head of humerus from Beloo-
chistan (See p. 395).

Length of fragment, 64 in. Breadth of upper extremity, 3°5in. Smallest

antero-posterior diameter of ditto, 2°6in. Greatest diameter of head (articular sur-
face), 2°7in. Length of crest of great tuberosity, 5-9 in.

Figs. 12, 12 a, and 12 b,—Fragment of lower end of right radius,
from Beloochistan,
Breadth of inferior articular surface, 3-3in. Length of ridge dividing scaphoid

and semilunar surfaces, 1*4in. Breadth of scaphoid surface, 1-8 in. Breadth of
semilunar ditto, 1-5 in.

Figs. 18, 13 a, 1386, and 13 ¢c.—Scaphoid of right carpus, from
Beloochistan.

Antero-posterior diameter, 2-9 in. ; transverse, 3:4 in.; vertical, 2°8 in.

Figs. 14 and 14 a.—Fragment of adult lower jaw of Rhinoceros Peri-
mensis, horizontal ramus, containing three true molars.—Col. Fulljames.

Length of fragment, 15-9in, Depth of ramus, 4°5in. Thickness, 3-in. Length
of three true molars, 7:9 in.

Figs. 15 and 15 a.—Fragment of horizontal ramus of lower jaw of
Rhinoceros Perimensis, containing three true molars, which agree
closely with those of Kaup’s Acerotherium incisivum.—B.M.

Length of fragment, 12°5 in. Depth of ramus, 5:2 in. Thickness, 2-1 in.
Length of three true molars, 7° in.

Figs. 16, 16 a, and 16 6.—Upper articulating extremity of femur of
Rhinoceros Perimensis.—B.M.

Length of fragment, 84in. Breadth of upper extremity, including great tro-
chanter, 9°lin, Diameter of articular surface of head, 4° in,

DESCRIPTION OF PLATES. 519

Figs. 17, 17 a, and 17 b.—Metacarpal bone (medius) of Rhinoceros
Perimensis.

Extreme length, 7°7 in. Smallest transverse diameter of shaft, 2:2in. Breadth
of posterior articular surface, 1:°9in. Height of ditto, 1-9 in.

Figs. 18, 18 a, and 18 b.—Astragalus of Rhinoceros from the Ner-
budda Pass.—B.M.
Breadth of tibial surface, 3-in. Smallest antero-posterior diameter, 1°7 in.

Breadth of scaphoid surface, 1‘9in. Greatest breadth of cuboid surface, ‘9 in.
Height of seaphoid surface, 1-8in. Height of cuboid surface, 2° in.

Figs. 19,19 a, 19 6, and 19 c.—Scaphoid bone of carpus of Rhino-

ceros.
Figs. 20, 20 a, and 20 b.—Head of humerus.

Figs. 21, 21 a, and 21 6.—Lower extremity of right radius.

PuatTe LXXVII.

Bones of Anterior Extremity of divers Fossil Indian Species of
Rhinoceros.

Figs. 1, 1 a, 1b, and 1 c—Humerus, radius, and ulna zn situ. This
specimen was described and figured by Messrs. Baker and Durand in
the Journ. As. Soc. for August 1836, vol. v. p. 498, Plate XVII. figs.
land 2. The humerus is perfect, with the exception of the deltoid
erest. The length of the humerus exceeds that of any of the existing
species of Rhinoceros. Its thickness, in proportion to the length of
the bone and the development of the articulating pulley, are inter-
mediate between the Sumatra and Indian species. The breadth at the
condyles is nearly in the same proportion as that of the Indian Rhino-
ceros. The length of the radius in proportion to the femur is a little
less than in the Indian, and somewhat in excess of the small Sumatra
species —B.M. (See antea, p. 163.)

Length of humerus from upper articular surface to lower surface of inner con-
dyle,17:3in. Extreme length of humerus, 21-in. Greatest width of humerus at
termination of deltoid crest, 6-3 in. Greatest width of humerus at upper extre-
mity, 62 in. Greatest oblique diam. of humerus at lower extremity, 7°8 in.
Greatest ant. posterior diam. of upper extremity, 6-3in. Greatest ant. post. diam.
of lower extremity, 5°in. Circumference of shaft beneath deltoid crest, 11-5 in.
Diameter of upper articular surface, 4-2 in. Width of lower articular surface, 4°6
in. Length of radius, 15°5in. Width of upper extremity of ditto, 48in. Probable
width of lower extremity of ditto, 4’8in. Length of ulna (olecranon broken),
19°3in. Width of conjoined lower surfaces of radius and ulna, 6:6in. Cireum-
ference round centre of conjoined shafts, 11-7 in. .

Figs. 2, 2 a, 2 6, and 2 c.—Humerus, with strongly-developed deltoid
crest. This specimen also is described and figured by Messrs. Baker
and Durand, Journ. As. Soc., vol. v. p. 499, Plate XVII. fig. 5.—B.M.

Length of fragment, 12-2in. Width including deltoid crest (upper extremity),
8:in. Antero-posterior diameter of ditto, 61lin. Length of deltoid crest, 8-2 in.
Greatest width of lower extremity, 7:in. Antero-posterior diameter of ditto
internally, 46 in. Width of lower articular surface, 4-4 in.

Figs. 3, 3a, and 3 6,—Fragment of head of humerus.—B.M.

Length of fragment, 12‘5in. Width of upper extremity, including deltoid crest,
7-9in. Antero-posterior diameter of ditto, 5-in. Length of deltoid crest, 8°6 in.
Diameter of articular surface of head, 3°6 in.

520 FAUNA. ANTIQUA SIVALENSIS. .

Figs. 4, 4a, and 4 .—Fragment of lower end of humerus, with arti-
culating surface.

Length of fragment, 10-4in. Width of lower extremity, 5-4in. Antero-pos- —
terior of lower extremity internally, 4:7 in.

Figs. 5, 5a, and 5 b.—Fragment of lower end of humerus, with arti-
culating surface.—B.M.

Length of fragment, 9°2in. Width of lower extremity, 5-3in. Antero-posterior
diam. of lower extremity internally, 4°3 in.

Fig. 6.—Upper articulating surface of ulna, with upper end of
radius. The tip of the olecranon is broken off.—B.M.

Width of articulating surface, 4°3in. Chord of sigmoid cavity, 2-4 in.

Figs. 7, 7 a, 7b, and 7 c.—Upper end of ulna, with entire radius.—
B.M.
Extreme length of conjoined radius and ulna, 15°3in. Length of radius from
upper surface to styloid process, 11:3in. Width of upper extremity of radius,
4-in. Width of lower extremity of radius across epiphysial line, 4°in. Cireum-
ference of radius in centre of shaft, 5°5 in.

Figs. 8, 8a, and 8 b.—Fragment of radius, with lower articulating
surface.—B.M.

Length of fragment, 9°4 in. Circumference of shaft at fractured extremity, 6°7 in.
Width of lower articular surface, 3°6 in.

Figs. 9, 9a, 9b, and 9 c.—Fragment of ulna, with lower articulating
surface.—B.M.

Length of fragment, 12:2in. Greatest width of lower articular surface, 2° in.
Least transverse diameter of shaft of tibia, 2°7 in.

Pruate LXXVIII.

Bones of Posterior Extremity of divers Fossil Indian Species of
Rhinoceros.

Figs. 1, 1a, and 16.—Femur of fossil Rhinoceros from the Sewalik
hills. The figures are copied from drawings by Messrs. Baker and
Durand, in the Journ. Asiatic Soc. for Aug. 1836, vol. v. p. 499.
The specimen was found in close proximity to the humerus and radius,
Plate LXXVIL, fig. 1, so that there could be no doubt that it belonged
to the same animal. It is perfect except at the lower part of the great
trochanter. The fossil has a greater development in its anterior, and a
somewhat less development of its posterior, extremity, than in the Indian
Rhinoceros, but the difference is not excessive. The third trochanter
also differs from the existing species, as figured in Cuvier’s ‘ Oss. Foss.,’
in not possessing the double point, for it has a single well-defined
ascending process, without any sign of the bicuspid termination (See
antea, p. 164).

Length from head to bottom of inner condyle, 24:5 in.; from head to bottom
of third trochanter, 17°7in. Breadth from head to most salient point of great
trochanter, 10°6 in. Breadth across condyles, 6°82in. Diameter of articulating

head, 4:65in. Antero-posterior diameter of inner condyle, 8°45 in. ; antero-pos-
terior diameter of outer, 6°35 in.

Figs. 2 and 2 a.—Mutilated fragment of upper end of femur.—B.M.

Extreme length of fragment, 11-5in. Width across third trochanter, 6: in.
Circumference below third trochanter, 9°65 in.

DESCRIPTION OF PLATES. 521

Figs. 3, 3 a, 36, and 3 c.—Tibia and fibula conjoined.—B.M.

Extreme length of tibia, 16°9in. Extreme length of fibula, 16-in. Extreme
transverse diameter of upper extremity of tibia, 6-1in. Extreme antero-posterior
diameter of upper, including tuberosity, 7-in. Extreme width of lower articular
surface, 3°8in. Extreme antero-posterior diameter of ditto, 3-1in. Least cireum-
ference of shaft of tibia, 9°1in.

Figs. 4, 4a, 4b, and 4 c.—Fragment of tibia almost perfect.—B.M.

Extreme length, 13°6in. Least circumference of shaft, 8° in.

Figs. 5, 5a, and 5 b. Fragment of tibia, including lower articulating
surface.—B.M.

Length of fragment of tibia, 11°8in. Width of inferior articular surface, 3:4 in.
Least circumference of shaft, 8°1 in.

Figs. 6 and 6 a.—Patella.—B.M.

Height, 41 in.

Figs. 7 and 7 a.—Patella.

Height, 4°7in. Width of articulating surface, 4:in. Height of ditto, 2:9 in.

Figs. 8, 8 a, and 8 6.—Bones of tarsus (calcaneum, scaphoid, cuboid,
and three cuneiforms) with index and medius metatarsals. The cal-
caneum and scaphoid do not belong to the remainder.—B.M.

Greatest width of scaphoid, 3-lin. Greatest width of cuboid, 2-in. Greatest
width of external cuneiform, 2°3in. Greatest width of middle ditto, 1:3 in.
Greatest width of inner ditto, 1‘9in. Greatest width of upper articular surface
of medius, 2°35in. Greatest width of upper extremity of index, 1°75 in.

Figs. 9, 9 a, and 9 6.—Calcaneum.—B.M.

Extreme length, 5-5in. Height, 3°4in. Width, 2°8 in.

Figs. 10, 10 a, and 10 6.—Calcaneum.

Extreme length, 6°5in. Height, 2°9in. Width, 3:9 in,

Figs. 11 and 11 a.—Calcaneum.—B.M.

Extreme length, 5‘8in. Height, 3°lin. Width, 3-4in.

Figs. 12, 12 a, and 12 b.—Astragalus.—B.M.

Width of anterior articular surface, 3°in. Greatest width of cuboid segment of
ditto, 1-in. Width of trochlea, 3°6 in. Antero-posterior diam. of trochlea in centre,
2° in. Greatest height, 2°8 in. Greatest antero-posterior diameter internally,
3°6 in.

Figs. 13, 13 a, and 13 b.—Astragalus.

Width of anterior articular surface, 3-4in. Greatest width of cuboid segment
of ditto, 1-lin. Width of trochlea, 3°3in. Antero-posterior diam. of trochlea in
centre, 2°in. Greatest height, 26 in. Greatest antero-posterior diameter in-
ternally, 3:5 in.

Puate LXXIX.

Bones of Anterior and Posterior Hxtremities of divers Fossil Indian
Species of Rhinoceros.
Figs. 1, 1 a, 1b, and 1 c.—lLeft scaphoid bone of carpus.—B.M.

Height, 2°5in. Width of inferior articular surface, 2°9in. Greatest antero-
posterior diameter, 2°6 in.

Figs. 2, 2 a, 2 6, and 2 c—Left scaphoid of carpus—B.M.

Greatest height, 2°6in. Antero-posterior diameter, 2°35in. Width of inferior
articular surface, 2°7 in.

Figs. 3, 3a, and 3 6.—Unciform bone of carpus.— B.M.

Greatest antero-posterior diameter of upper surface, 2°] in.

522 FAUNA ANTIQUA SIVALENSIS.

Figs. 4, 4a, and 4b.—Unciform bone of carpus.—B.M.
Greatest antero-posterior diameter of upper surface, 2°2 in.
Figs. 5, 5 a, and 5 b.—Unciform bone of carpus.—B.M.
Greatest antero-posterior diameter of upper surface, 1°6 in.
Figs. 6, 6 a, and 6 6.—Trapezoid bone of carpus.
Antero-posterior diameter of upper articular surface, 1-4 in.
Figs. 7, 7 a, and 7 b.—Left index metacarpal bone.—B.M.
Extreme length, 6°5in. Width of shaft in centre, 1°7 in.
Figs. 8, 8 a, and 8 6.—Right index metacarpal bone.—B.M.
Extreme length, 6°5in. Width of shaft in centre, 1-5 in.
Figs. 9, 9 a, and 9 6.—Left index metacarpal bone.—B.M.

Extreme length, 6-6 in. Width of trapezoid surface, 1-1in. Width of shaft in
centre, 1°6 in.

Figs. 10, 10 a, and 10 b.—Left medius metacarpal bone.—B.M.
Extreme length, 7-2in. Greatest width of shaft, 2°3 in.

Figs. 11, 11 a, and 11 6.—Left medius metacarpal bone.—B.M.
Extreme length, 7°3in. Extreme width of shaft, 2°3 in.

Figs. 12, 12 a, and 12 b.—Left medius metacarpal bone.—No. 39,655
B.M.

Width of surface for os magnum, 2°in. Width of surface for os unciforme,
l‘in. Width of shaft, 2-in.

Figs. 13, 13 a, and 13 b.—Left annularis metacarpal bone.—B.M.
Extreme length, 6-5in. Width of shaft in centre, 1:7 in.
Figs. 14, 14a, and 14 6.—Left annularis metacarpal bone.—B.M.
Extreme length, 6°3 in. Width of facet for unciform, 1-5in. Width of shaft
in centre, 1°5in. Width of distal articular surtace, 1°65 in.
Figs. 15 and 15 a.—Index and medius metatarsal bones conjoined.
b. Index. c. Medius.—B.M.

Extreme length of index, 6°5in. Extreme length of medius, 7-5in. Width of
index shaft at centre, 1:3in. Width of medius shaft at centre, 2-3 in.

Figs. 16, 16 a, and 16 6.—Right index metatarsal bone.—B M.
Extreme length, 6-°95in. Width of shaft in centre, 1-265 in.

Figs. 17, 17 a, and 17 6.—Right medius metatarsal bone.—B.M.
Extreme length, 6-5in. Width of shaft in centre, 1°9 in.

Figs. 18, 18 a, and 18 6.—Left medius metatarsal bone.—B.M.
Extreme length, 6°7in. Width of shaft in centre, 2:06 in.

Figs. 19, 19 a, and 19 b.—Left annularis metatarsal bone.—B.M.
Extreme length, 5-6 in. Width of shaft in centre, 1-4 in.

Figs. 20, 20 a, 20 b, 20. c, and 20 d.—Medius metatarsal bone.—B.M.
Extreme length, 6-1in. Width of shaft in centre, 2-3 in.

Figs. 21, 21 a, and 21 b.—Calcaneum.—B.M.

Extreme length, 5-$in. Extreme height, 2°9in. Extreme width, 3°6 in.
Figs. 22, 22 a, 22 b, and 22 c.—Cuboid bone.—B.M.

Greatest antero-posterior diameter of upper surface, 1‘8in. Greatest width of
ditto, 2°35in. Greatest height, 3-in.

Figs. 25, 23 a, 23 6, and 23 c.—Cuboid bone.

_Greatest antero-posterior diameter of upper surface, 1-65in, Greatest width of
ditto, 2°-in. Greatest height, 2°75 in.

pepo ks

DESCRIPTION OF PLATES. 523

Pirate LXXX.

Chalicotherium Sivalense (Fale. and Caut.). See antea, page 208, and
Pl. XVII.

Figs. 1, 1 a,.1 6, and 1 c.—Anterior half of an adult head, with the
upper and lower jaws in natural apposition, and exhibiting the greatest
portion of the dental series of both jaws. The greater part of the
cranium proper is absent. The specimen demonstrates the very re-
markable fact that the Chalicotheriwm Sivalense was entirely destitute
of incisor teeth in either jaw. The intermaxillary bones are perfect to
their tips, and consist of slender slips of bone converging to a sharp
point; they show that no incisor teeth could have existed in the upper
jaw at any period of the animal’s age. The anterior portion of the
lower jaw is perfect to the alveolar edge. A detached canine is seen
on either side, but the intervening space is without a vestige of incisors,
and is contracted in correspondence with the convergence of the
intermaxillary bones, and sloped off to a fine edge. The upper jaw
is also destitute of canines, or of any trace of canine alveoli; but
the lower jaw contains two canines, as shown in figs. la@ and 1 8,
the crowns of which are thick, cuneiform, and somewhat triangular,
and slightly inclined forwards, with a blunt apex. The specimen
shows three premolars and the first true molar; the two back molars
are absent. The characters of the molars are better seen in figs. 3
and 4, and are described in great detail in the memoir on Chalico-
therium (Page 213).

This beautiful specimen was originally in the Dadoopoor collection
of Messrs. Baker and Durand, and is now in the Museum of Marischal
College, Aberdeen. Cast in B.M. Its dimensions are as follows :—

Length of intermaxillary bone of right side, 3-in. Greatest depth of ditto,
‘4in. Length of three premolars and first molar, 2°6in. Length of three premo-
lars, 1‘8in. Breadth of fragment opposite last premolar, 3-2in. Breadth of pa-
late “5 in. in front of anterior premolar, 1-5 in. Breadth of palate 1-2 in. in front of
anterior premolar, l‘lin. Height of fragment of maxillary bone from alveolar
border (right side), 2°5in. Length of fragment of maxillary bone on right side,
39in. Greatest breadth of anterior nares, 13in. Extreme length of fragment
of lower jaw, 5°5in. Length of symphysis, 3-lin. Depth of horizontal ramus at
posterior border of first molar, 1‘6in. Greatest thickness of ramus at ditto, ‘8 in.
Interval between the horizontal ramus at ditto, l-in. Breadth of lower jaw at
posterior border of symphysis, 2°2in. Least breadth of symphysis, 1-lin. From
posterior border of symphysis to narrowest part of symphysis, 1:‘9in. Between
alveolar border of canines, 1°2in. Breadth of incisive margin, -9in. Width of
palate posteriorly between first molars, 1-3in. Length of first premolar, upper
jaw, right side, Sin. Length of second premolar ditto, °55in. Length of third
premolar (greatest) ditto, -7in. Length of first molar ditto, -95in. Breadth
of first molar ditto, 1:03in. Breadth of third premolar ditto, -9in. Breadth of
second premolar ditto, ‘76in. Breadth of first premolar ditto, -53in. Length of
three premolars and first molar, 2°8in. Length of three premolars, 1:‘8in. Be-
tween opposed margins of canine and first premolar, -9in. Between anterior
margin of first premolar and incisive margin, 1°7in. Length of first premolar,
lower jaw, ‘5in. Length of second premolar, ditto, *6in. Length of third premo-
lar, ditto, Sin. Length of first molar, ditto, 1:0 in. Breadth of first molar, ditto,
“6in. Breadth of third premolar, ditto, Sin. Breadth of second premolar, ditto,
-43in, Breadth of first premolar, ditto, -3in. Length of crown of canine, ‘46 in.
Breadth of crown of canine, *3 in.

Fig. 2.—Chalicotherium Sivalense. Upper jaw, right side, with part
of orbit, three true molars and last premolar. The muzzle seems
to have fined off rather abruptly in front of the molar protuberances,

524 FAUNA ANTIQUA SIVALENSIS.

and the orbit to have been more forward on the face and more
depressed below the brow than in Anoplotherium commune. The
upper surface of the sub-orbitary canal is seen opening behind the
anterior angle of the orbit, the floor of which seems to have extended
behind the post-orbitary processes.—B.M.

This specimen is also figured as Anoplotherium Sivalense in the Pro-

ceedings Geol. Soc., No. 98, 18438, Plate II. fig. 2.

Figs. 8 and 8 a.—Chalicotherium Sivalense. Horizontal and lateral
view of left upper jaw, comprising the three true molars and three
premolars. The true molars, and especially the two last, are enormously
large in comparison with the other teeth, or with the dimensions of the
head. If found isolated, they would seem suitable to an animal
approaching the size of Rhinoceros, whereas the anterior part of the
lower jaw and the muzzle do not reach the dimensions of the Indian
Tapir. The outer surface of the molars presents both vertically and
horizontally the double chevron or W form of Anoplotherium, but
with this difference, that the surface of the re-entering angles is more
inclined inwards. The characters of the teeth in this specimen are
minutely described in the memoir on Chalicotherium (See page 213).

This specimen is also figured in the Proceedings Geol. Soc. No. 98,
1843, Plate II. fig. 1.--B.M.

Figs. 4 and 4a.—Chalicotherium Sivalense. Fragment comprising
the left half of the lower jaw from the angle on to the commencement
of the symphysis of an individual which was not quite full grown,
containing three true molars and the last premolars, with the empty
alveoli of the first two premolars. The last premolar is fully pro-
truded, but unworn; the last molar is in the germ state. The
characters of the teeth in this specimen are minutely described in the
memoir on Chalicotherium.—B.M.

The dimensions of the specimen are as follows :—

Extreme length of fragment, 6°8in. Greatest depth of ramus, 2-lin. Greatest
thickness (towards symphysis), l‘lin. Depth of ramus at anterior margin of
third premolar, 1‘5in. Length of alveolus of second premolar, -55in. Breadth
of alveolus of ditto, °35in. Length of third premolar, -7in. Breadth of ditto,
‘Sin. Length of first molar, ‘8in. Breadth of ditto, -5in. Length of second
molar, 1:2in. Breadth of ditto, 65in. Length of third molar, 1-5in. Breadth
of ditto, 65 in.

Puiate LXXXI.

Figs. 1, la, and 1b.—Kquus Sivalensis (Fale. and Caut.). Cra-
nium. Upper, palate, and lateral views. The specimen is broken off
transversely in front of the second premolar. The three true molars
and two back premolars on the right side are well preserved. The
left alveolar ridge is mostly deficient—B.M. (See antea, p. 186.)

Length of fragment, 15-in. Between extreme points of zygomata, 8-lin. Be-
tween anterior angles of the orbits, 6-2in. Breadth of nasal ridge at sub-orbital
foramen, 2°7 in. Height of cranium from palate at ditto, 3:3in. From anterior
angle of orbit to nasal notch, 6-in. Great diameter of orbit, 2-6in. Lesser dia-
meter of orbit, 1:9in. Depth of zygomatic fossa, 1°8in. Greatest width of ecra-
nium at root of zygomata, 4°5in. Height of cranium from base of occipital to
summit of sagittal crest, 3°6in, From lower border of occipital foramen to pos-
terior border of palate, 8°5in. Length of three true molars, 3:lin. Length of
two posterior premolars, 2°2in, Width of palatine notch, 1:9 in. Width of
palate posteriorly, 3°3in. Width of palate anteriorly, 2°5in. Width of alveoli,
1:2 in.

DESCRIPTION OF PLATES. 525

Figs. 2, 2a, and 26.—Equus Sivalensis. Fragment comprising
posterior portion of skull, broken off in front in a line with anterior
angles of zygomatic arches. Shows occipital foramen, crest, condyles,
and posterior roots of zygomata.—B.M.

Length of fragment, 7°5in. Height of occipital facet from lower border of
occipital foramen to summit of occipital crest, 48 in. Between inferior angles of
occipital crest, 4°7 in. Breadth of cranium between roots of zygomata, 3°7 in.
Length of ridge of occipital condyle, 1:7 in. Height of condyle (greatest), 2:1 in.
Between inner margins of condyles, 1‘5in. Height of occipital foramen, 1°6 in.

Fig. 3.—Equus Sivalensis. Fragment of upper jaw, with whole
series of six molars.—B.M.

Length of fragment, 9°3in. Height of fragment (length of molar), 4:1 in.
Length of molar series, 7-7 in. Length of three true molars, 3‘4in. Breadth of

’ alveoli, 1:3 in.

Fig. 4. Equus Sivalensis. Fragment of horizontal ramus of lower
jaw with whole series of six molars.—B.M.
Length of fragment, 11-in. Depth of jaw at anterior border of fourth premo-

lar, 8-6in. Width of ditto, 1:‘2in, Length of molar gories, 7:8in. Length of
three true molars, 37 in.

Figs. 5, 5a, 56, and 5¢—Equus Namadicus (Fale. and Caut.),
from the Nerbudda Valley. The occipital condyles and foramen and
the left zygomatic arch are very perfect; also the whole series of six
molars on left side. The specimen is broken off in front of first
(permanent) premolar on left side; from this the line of fracture passes
obliquely across the palate and through the middle of the hindmost
right premolar. The three right true molars are present. The right
zygomatic arch is absent.—B.M. (See antea, p. 186.)

Extreme length of fragment, 176 in. From lower border of occipital foramen
to posterior border of palate, 96 in. Greatest breadth of eranium at roots of Zy-
gomatic processes, 4.4 in. Between extreme points of zygomata, 7°9in. Between
anterior angles of orbits, 6'in. Height of cranium from palate at fractured
extremity, 3°6 in. Great diameter of orbit, 2°8 in. Lesser. diameter of orbit,
18in. Height of occipital facet from lower border of occipital foramen, 4:2 in.
Between inferior angles of occipital facet, 4:1 in. Depth of zygomatic fossa, 1°7 in.
Width of palatine notch, 1:7in. Width of palate posteriorly, 3-in. Width of
palate anteriorly, 2°7in. Width of alveoli, 1-1in. Length of molar series, 7: in.
Length of true molars, 3°3 in.

Fig. 6.— Equus Namadicus. Fragment of left upper jaw comprising
whole molar series.—B.M.

Length of fragment, 9°2in. Length of molar series, 7° in. Length of three
true molars, 3°2in. Breadth of alveoli, 1:2 in.

Fig. 7—Equus Namadicus. Fragment of left lower jaw with entire
molar series. The fracture exposes the fang of the last true molar.—

B.M.

Length of fragment, 12‘7in. Depth of jaw at anterior border of fourth pre-
molar, 3°8in. Width, 1-2in, Length of molar series, 8'in. Length of three
true molars, 3°8 in.

Puate LX XXII.

Fig. 1.—Hquus Sivalensis. Upper jaw, right side, with entire molar
series.—B.M.

Length of third molar,1:16in. Breadth of ditto, 1:08in. Length of second
molar, 1:16in. Breadth of ditto, 1:°2in, Length of first molar, 1:16in. Breadth

526 FAUNA ANTIQUA SIVALENSIS.

of ditto, 1:28 in. Length of third premolar, 1:23in. Breadth of ditto, 1:3 in.
Length of second premolar, 1°33in. Breadth of ditto, 1-3in. ‘Length of first
premolar, 1°7in. Breadth of ditto, 1:2 in.

Fig. 2.—Lower jaw, right side, with entire molar series.

Length of third molar, 1:°26in. Breadth of ditto, 6 in. Length of second
molar, 1:2in. Breadth of ditto, °7in. Length of first molar, 1:2in. Breadth of
ditto, ‘*8in. Length of third premolar, 1°3in. Breadth of ditto, ‘8in. Length
of second premolar, 1:26in. Breadth of ditto, -83in. Length of first premolar,
1:45in. Breadth of ditto, :73in. Length of molar series, 7°65in. Length of
three true molars, 3°55 in. ,

Fig. 3.—Equus Sivalensis. Fragment of upper jaw, right side, with
three true molars and two posterior premolars.—B.M.

Length of third molar, 1:°25in. Breadth of ditto, 1-05in. Length of second
molar, 1‘in. Breadth of ditto, 1:15in. Length of first molar, -9in. Breadth
of ditto, 1:13in: Length of third premolar, 1:15in. Breadth of ditto, 1:2in.
Length of second premolar, 1:2in. Breadth of ditto, 1-2 in. Length of molar
series (first premolar wanting), 5°45in. Length of three true molars, 3°16 in.

Fig. 4.—Equus Sivalensis. Fragment of lower jaw, including right
horizontal ramus, with three anterior molars, milk dentition, and sym-
physis with outer incisor and alveoli of middle and inner incisors on
either side.

Length of fragment, 9°3in. Length of milk molar series, 4:15in, Interval
between first milk molar and external incisive alveolus, 3°75in. Between central
points of inner alveolar border of external incisors, 1°3in. Length of third
milk molar, 1-45in. Breadth of ditto, ‘66in. Length of second milk molar,
1:25in. Breadth of ditto, °6in. Length of first milk molar, 1‘5in. Breadth of
ditto, °63in. Length of crown of external incisor, °53in. Breadth of crown of
ditto, *35in. Length of alveolus of middle incisor, *55in. Length ditto of internal
ditto, °55in. Between posterior angles of last milk molars, 2°16 in. Between
anterior angles of anterior milk molars, 1-2 in.

Figs. 5 and 5 a.— Equus Sivalensis. Fragment, comprising anterior
part of upper and lower jaws in almost natural apposition. Shows six
incisors and two small canines in both jaws; also the two front pre-
molars on one side of lower jaw.—B.M.

Between posterior angles of external incisors of lower figure in 5 a, 1:9in. -
Length of crown of three incisors of one side, 1:2in. Between anterior edge of
mental foramen and anterior edge of canine, 2:-4in. Between anterior edge of
mental for. and posterior edge’of ext. incisors, 2°9in. Diastema between canine and
exterior incisor, “4 in. Diastema between canine and anterior edge of first pre- ©
molar, 3°3in. Length of anterior premolar, 1°35in. Breadth of ditto, ‘6 in.
Length of external incisor, ‘5 in. Length of middle ditto, -45in. Length of
internal ditto, -35 in. Between posterior angles of exterior incisors of upper
jaw in fig. 5a, 2°2in. Length of three incisors of one side, 1‘8in. Diastema be-
tween anterior premolar and canine, 3-2 in. Height of first premolar, 3° in.
Length of external incisor, ‘8in. Length of middle ditto, 8in. Length of in-
ternal ditto, °75 in.

Figs. 6, 6 a, and 6 b.— Equus Sivalensis. Fragment comprising an-
terior portion of palate, with six upper incisors and two rudimentary
canines.—B.M.

Between posterior angles of external incisor alveoli, 2:45in. Between external
alveolar margins of canines, 2°lin. Between canine and external incisor, -4 in,
Breadth (extreme) of incisor series, 2‘8in. Breadth of three incisors (oblique),
1:75in. Length of crown of external incisor, -7in. Breadth of ditto, -45in.
Length of crown of middle incisor, °65in. Breadth of ditto, °46 in. Length of —
crown of internal incisor, ‘56in. Breadth of ditto, -5in.

= eee Se

DESCRIPTION OF PLATES. 527

Fig. 7.— Equus Namadicus, from the Nerbudda. Fragment of upper
jaw, right side, with entire molar series.—B.M.

Length of molar series, 7:05 in. Length of three true molars, 3:2in. Length
of third molar, 1:15in. Breadth of ditto, ‘9in. Length of second molar, 1:06 in.
Breadth of ditto, 1‘lin. Length of first molar, :96in. Breadth of ditto, 1:05 in.
Length of third premolar, 1:15in. Breadth of ditto, 1:15in, Length of second
premolar, 1:16in. Breadth of ditto, 1-lin. Length of first premolar, 1°55 in.
Breadth of ditto, 1:0 in.

Fig. 8.—Equus Namadicus. Fragment of lower jaw, left side, with
entire molar series—B.M.

Length of molar series, 7°93 in. Length of three true molars, 3°76in. Length
of third molar, 1:°2in. Breadth of ditto, Sin. Length of second molar, 1:2 in.
Breadth of ditto, ‘65in. Length of first molar, 1:°25in. Breadth of ditto, -7 in.
Length of third premolar, 1:2in. Breadth of ditto, -7in. Length of second pre-
molar, 1:3in. Breadth of ditto, -73 in. Length of first premolar, 1°55 in. Breadth
of ditto, 66 in.

Figs. 9, 9 a, and 9 b6.— Equus Paleonus (Fale. and Caut.), from
the Nerbudda. Fragment of anterior portion of palate, with six in-
cisors and two small canines. Presented by C. Frazer, Esq.—B.M.

Between inner alveolar margins of canines, 1-lin. Diastema between canine
and external incisor, °8in, Between posterior angles of external incisors, 1-7 in.
Length of three incisors of one side, 1:6in. Length of external incisor, ‘6in. Length
of middle ditto, °65im. Length of internal ditto, -56in. Length of alveolus of
canine on left side, -4in. Breath of ditto, °3 in.

Figs. 10, 10 a, and 10 b.—Equus Paleonus. Fragment of anterior
portion of lower jaw, with six incisors.—B.M.

Breadth across posterior margin of external incisive alveolus of left side,
2-4in. Length of three incisors of one side, 1-45in. Length of broken end of
external incisor, ‘45in. Length of crown of middle incisor, ‘Sin, Length of
crown of internal ditto, 4 in.

Figs. 11 and 11 a.—Kquus Paleonus. Fragment of lower jaw, right
side, milk dentition.—B.M. (See antea, p. 186, note.)

Length of third milk molar, grinding surface, 1‘3in. Breadth of ditto, 86 in.
Length of second milk molar, grinding surface, 1:15in. Breadth of ditto, -4 in.
Length of first milk molar, grinding surface, 1°2in. Breadth of ditto, -4 in.

Figs. 12, 12 a, and 12 6.—Fossil Hquus, from the Irrawaddi. Frag-
ment of lower jaw, comprising symphysis and six incisors.—B.M.

Length of fragment, 4-4 in. Between external angles of external incisors, 2-4 in.
Length of three incisive alveoli of one side, 1°5 in.

Fig. 13.—Hippotherium Antilopinum (Fale. and Caut.), from the
Sewalik hills. Fragment of upper jaw, left side, with entire series of
six molars.—B.M. (See antea, p. 186.)

Length of molar series, 5-3in. Length of three true molars, 2°36in. Length
of third molar, ‘8in. Breadth of ditto, -55in. Length of second molar, -85 in.
Breadth of ditto,*76in. Length of first molar, ‘8in. Breadth of ditto, -85in.
Length of third premolar, -93in. Breadth of ditto, °86in. Length of second pre-
molar, ‘96in. Breadth of ditto, -95 im. Length of first premolar, 1:25 in.
Breadth of ditto, -83in. Height of second premolar, 2°15in. Height of first pre-
molar to origin of fangs, 1°4 in.

Fig. 14.—Hippotherium Antilopinum. Fragment of lower jaw, right
side, with three premolars and portion of first true molar.—B.M.

Length of three premolars, 3-lin. Length of three premolars and first molar (frag-
ment), 4:1in, Remaining portion of diastema, 1°8 in. Length of first molar
(fragment), ‘8in. Breadth of ditto, 55 in. Length of third premolar, :96 in,
Breadth of ditto, ‘6in. Length of second premolar, ‘96in. Breadth of ditto,
*55in, Length of first premolar, 1:05in. Breadth of ditto, ° in,

528 FAUNA ANTIQUA SIVALENSIS.

Figs. 15, 15 a, and 15 b.—Hippotherium Antilopinum. Symphysis of
lower jaw, with fragments of six incisors.—B.M.

Length of fragment, 2°2in. Breadth at anterior angles of canines, 1°6 in.
Length of fragments of three incisors of one side, ‘95 in.

Fig 16.—Hippotherium Antilopinum. Portion of skull, with palate.
Shows three true molars and third (permanent) premolar on both sides,
and portion of second premolar on right side-—B.M.

Length of three true molars, 2°5in. Length of three true molars, including
last premolar, 3°36 in. Width of palate posteriorly, 2-in. Width of palate be-
tween third premolars, 1‘9in. Length of third molar, ‘85in. Breadth of ditto,
‘75in. Length of second molar, ‘8in, Breadth of ditto, *85in. Length of first
molar, ‘8in. Breadth of ditto, -9in. Length of third premolar, ‘9in. Breadth
of ditto, *9 in.

Fig. 17.—Hippotherium Antilopinum. Portion of molar, showing

plication of enamel.—B.M.
Length of fragment, ‘75in. Breadth, ‘45 in.

Fig. 18.—Hippotherium Antilopinum. Molar, with characteristic

plication of the enamel, like that shown in fig. 19.1
Length of grinding surface, ‘85in. Greatest breadth of ditto, -85 in.

Fig. 19.—Hippotherium gracile (of Europe). Molar, with charac-
teristic plication of enamel.

Greatest length of grinding surface, 1:03 in. Greatest breadth of ditto, -975 in.

Puate LXXXIII.

Figs. 1 to 11.—Equus and Hippotherium.
Figs. 1, 1 a, 1 6, and 1 c.—Atlas.—B.M.

Extreme width, 4°5in. Length of inferior arch, 1‘4in. Height of spinal canal
anteriorly, 1°5in. Width of ditto, 1-7 in.

Figs. 2, 2 a, 2 6, and 2 c.—Axis.—B.M.

Extreme length of body, 4-1in. Length of spinal platform in centre, 2° in.
Width across posterior articular processes, 2°8 in.

Figs. 3, 3 a, 5b, and 8 c.—Cervical vertebra.—B.M.

Extreme length of body, 4-4in. Between extremities of oblique processes,
4°5in. Width between posterior oblique processes, 2°6 in.

Figs. 4, 4a, 4b, and 4 c.—Cervical vertebra.—B.M.

Extreme length between extremities of oblique processes, 4°in. Width of
spinal platform in centre, 1°7 in.

Figs. 5, 5 a, 5 b, and 5 c.—Cervical vertebra.—B.M.
Extreme length of body, 4°3 in.
Figs. 6, 6 a, 66, and 6 c.—Dorsal vertebra.—B.M.

Length of spine (fractured), 6-5in. Length of body, 2-2in. Width between
transverse processes, 3°8 in.

Fig. 7.—Portion of pelvis, showing acetabulum.—B.M.
Chord of acetabulum, 2°4 in.

1 This specimen somewhat resembles | and described by Dr, Falconer in same
that figured by Messrs. Baker and Durand | volume, p. 58.
in Journ. As. Soc., vol. iv., Plate xly.,

DESCRIPTION OF PLATES. 529

Fig. 8.—Portion of pelvis, showing acetabulum and thyroid foramen.
—B.M.

Chord of acetabulum, 2:in. Diameter of thyroid foramen (anterior portior),
2°3 in.

Figs. 9, 9a, and 96.—Lower end of humerus and upper end of
radius and ulna in sitw.u—B.M.

Width of lower end of humerus, 3*1in. Width of upper end of radius, 3:2 in.
Antero-posterior diameter of lower end of humerus internally, 3-4 in.

Figs. 10, 10 a, 10 6, and 10 c.—Fragment of lower articulating extre-
mity of femur.—B.M. :
Width of lower extremity of femur, 2°9in. Antero-posterior diameter of femur
externally, 3-2in. Width of rotular surface, 1°5 in.
Figs. 11, 11 a, 11 4, and 11 ¢.—Lower end of femur.—B.M.
Width of lower end, 3°5 in.

Figs. 12, 12 a, and 12 6.— Hippopotamus (Hexaprotodon) Iravaticus.
Lower end of radius. —B.M.
Width of lower end, 2°8in. Antero-posterior diameter of ditto, 1°7 in.

Figs. 18, 13 a, 186, and 13 ¢.—Second cervical vertebra or axis.
Species undetermined. From Col. Baker’s collection.—B.M.

Extreme length of body (fractured), 5-5 in. Width of odontoid surface, 3°2 in.
Width between outer edges of posterior articular surfaces, 3°5 in.

Figs. 14, 14a, 146, and 14c.—Fragment of lower end of femur.
Species undetermined.—B.M.

Circumference aboye rotular surface, 8‘in. Width of rotular surface, 1°5 in.

Puate LXXXIV.

Figs. 1 and 1 a.— Equus Sivalensis. Upper articulating extremity,
and portion of shaft of ulna.—B.M.

Width of radial articular surface, 3°2in. Chord of sigmoid cavity, 1°55 in.
From apex of olecranon to anterior edge of sigmoid cavity, 5°7 in.

Figs. 2, 2a,26, and 2c.—Equus Sivalensis. Upper end and por-
tion of shaft of una.—B.M.

Width of radial articular surface, 3-lin. Cireumference at lower fractured ex-
tremity, 5:2 in.

Figs. 3, 3. a, and 86.—Hquus Sivalensis. Fragment of lower end of
radius. —B.M.

Width of inferior articulating surface, 2°8in. Greatest antero-posterior dia-
meter, 1°5 in.

Figs. 4, 4a, 40, and 4c.—Kquus Sivalensis. Metacarpal bone.
Entire shaft and lower articulating surface—B.M. ~

Extreme length, 10:in. Width of upper articular surface, 2-lin. Width of
lower, 1:‘8in. Circumference of shaft in centre, 4:1 in.

Figs. 5, 5a, and 5 b.—Hippotherium Antilopinum. Fragment of
upper end of radius.—B.M.
3 Width of upper articulating surface, 2°d in.

Figs. 6, 6a, and 66.—Hippotherium Antilopinum. Fragment of
lower end of radius.—B.M.

Width of lower articular surface, 1°85 in.
VOL. I. MM

530 FAUNA ANTIQUA SIVALENSIS.

Figs. 7, 7a, and 7 6.—Hippotherium Antilopinum. Fragment of

lower end of radius with bones of carpus.—B.M.
Width of inferior articulating surface of radius, 2:1 in.

Figs. 8 and 8 a.—Hippotherium Antilopinum. Lower end of radius,
with bones of carpus, and portion of metacarpus.

Figs. 9, 9a, 96, and 9¢.—Hippotherium Antilopinum. Metacarpal
bone.—B.M.

Extreme length, 8:8in. Width of upper articular surface, 1‘5in. Width of
lower articular surface fractured, 1°5in. Cireumference in centre of shaft, 3°2 in.

Figs. 10, 10 a, and 10 b.—Hippothertum Antilopinum. Fragment
of metacarpal bone; lower end broken off—B.M.

Length of fragment, 8-lin. Width of articular surface of middle metacarpal,
1:°65in. Width of articular surface of left metacarpal, ‘4in. Width of articular
surface of right ditto, °35 in.

Figs. 11 and 11 a—Hippotherium Antilopinum. First phalanx.—
B.M.

Length, 2°9in. Greatest width of upper articular surface, 1:7 in. Greatest
width of lower ditto, 1:3 in.

Figs. 12, 12 a, 12 b, and 12 e.—Hippotherium Antilopinum. Second
phalanx.—B.M.

Length, 1:5 in. Greatest width of upper articular surface, 1'4in. Greatest
width of lower ditto, 1:25 in.

Figs. 138, 13 a, 134, 13, and 13 d.—Radius of fossil Equus from
the Nerbudda, entire—B.M.

Greatest length of radius, 11°-4in. Width of upper articular surface, 2°7 in.
Width of lower articular surface, 2°15 in. Circumference of shaft in centre, 5: in.

Figs. 14, 14 a, and 14b.—Shaft of radius of fossil Equus from the
Nerbudda. The articulating extremities are imperfect.—B.M.

Length of fragment, 10°7in. Circumference of shaft in centre, 4°3 in.

Figs. 15, 15 a, and 15 6.—Equus from the Niti Pass. Upper end of

shaft with articulating extremity of radius —B.M.
Width of upper articulating surface, 2°45 in.

Figs. 16, 16 a, and 16 6.—Equus from the Niti Pass. Fragment of

lower end of tibia.
Width of lower surface, 1:9 in.

Figs. 17 and 17 a.—Kquus from the Niti Pass. Astragalus —B.M.

Width of trochlea, 1:45in. Antero-posterior diameter of ditto, 1:3in. Width
of scaphoid surface, 1°7 in.

Figs. 18, 18 a, and 186.—Equus from the Niti Pass. Os magnum
of carpus.—B.M.

Transverse diameter, 1°4in. Antero-posterior diameter, 1:15in. Thickness in
centre, ‘7 in.

Figs. 19, 19 a, and 19 b.—Equus from the Niti Pass. Third or un-
gual phalanx.—B.M.

Width of articular surface, 1°55in. Probable antero-posterior diameter, 2°1 in.

Figs. 20 and 20a.—Equus. Metatarsal bone from Sewalik hills.
—B.M. No. 17,828.

Extreme length, 11:1in. Width of upper articular surface, 2-in. Width of
lower ditto, 1‘8in. Circumference at middle of shaft, 4:4 in. ;

Fig. 21.—Equus. Metatarsal bone from Sewalik hills.—B.M.

Circumference in centre of shaft, 3-8 in.

;

DESCRIPTION OF PLATES. 551

Puate LXXXV.

Figs. 1, la, 1b, 1¢, and 1d.—Equus Stvalensis. Entire femur,
with both articulating extremities —B.M.

Extreme length, 15-6 in. Transverse diameter of upper extremity, including
trochlea, 4°6in. Antero-posterior diameter of posterior segment of great trochlea,
1-9 in. Transverse diameter of articular surface, 2°5in. Antero-posterior dia-
meter of articular surface, 2°lin. Smallest transverse diameter of shaft, 1-8 in.
Smallest antero-posterior diameter of shaft, 1:°9in. Transverse diameter of lower
extremity, 3°6 in., Antero-posterior diameter externally, 3°6in. Height of rotular
surface in centre, 2°4in. Height of external condyle above neck of femur, 1°8 in.

Figs. 2, 2a, and 2b.—Kquus Sivalensis. Upper end of shaft of

_ femur, with upper articular extremity.

Length of fragment, 6‘8in. Breadth of upper extremity, 4:°3in. Transverse
diameter of articular surface, 2°2in. Antero-posterior diameter of ditto, 1-9 in.

Figs. 3, 3a, 3b, and 3 c.—Equus Sivalensis——Entire tibia.—B.M.

Extreme length, 14-5 in. Transverse diameter of upper extremity, 3°5 in.
Transverse diameter of shaft (smallest), 1-7 in. Antero-posterior diameter of shaft
(smallest), 1:2.im. Transverse diameter of lower extremity, 3°in, Length of
ridge dividing articular fosse, 2°4 in.

Fig. 4.—Lower end of tibia and astragalus of Equus zn situ, resto-
ration.

Figs. 5, 5 a, and 5 6.—Calcaneum of Equus Sivalensis.—B.M.

Length, 5‘8in. Projection of heel, 3-°2in. Greatest breadth, 2-lin. Greatest
height, 1-9 in.

Figs. 6, 6 a, and 6 6.—Astragalus of Hquus Sivalensis.—B.M.

Length (greatest), 2°5in. Height (greatest), 1:9in. Breadth of scaphoid
surface, 2:lin. Breadth of trochlea (tibial surface), 1:°7in. Antero-posterior
diameter of trochlea in centre, 1°4 in.

Figs. 7, 7 a, 7 6, and 7 c.—Metatarsal bone of Equus Sivalensis.

Extreme length, 11°5in. Antero-posterior diam. of shaft in centre, 1:°2in. Trans-
verse of shaft ditto, 1:°3in. Transverse of upper extremity, 2:in. Antero-pos-
terior of ditto, 1:7in. Transverse of lower articular surface, 1:‘8in. Greatest
antero-posterior of lower articular surface (in centre), 1°5 in.

Figs. 8, 8 a, 8b, and 8 c.—First phalangeal bone posterior extremity
of Equus Sivalensis.—B.M.

Length superiorly, 3:lin. Transverse diam. of posterior extremity, 2:1 in.
Vertical of ditto, 1°3in. Transverse of anterior articular surface, 1:5in, Vertical
of ditto, *9 in.

Figs. 9, 9a, and 96.—Hippotherium Antilopinum. Fragment of
shaft of tibia with lower articulating extremity.—B.M.

Length, 5-in. Breadth of inferior articular surface, 2-4in. Length of ridge
dividing articular fosse, 1-8 in.

Figs. 10, 10 a, and 10 6.—Astragalus of Hippotherium Antilopinum.
—B.M.

Length, 2:in. Breadth of seaphoid surface, 1‘6in. Breadth of tibial surface,
13in. Length of tibia (in centre), 1:2 in.

Figs. 11, 11 a, and 11 6.—Tarsus, metatarsus, and phalanx of Hippo-
therium Antilopinum.—B.M.

Length of whole figure, 15:3 in.
MM 2

532 FAUNA ANTIQUA SIVALENSIS.

Figs. 12, 12a, 126, and 12 ¢.—Metatarsal bone of Hippotherium
Antilopinum.—B.M.

Length, 10°4in. Smallest transverse diameter of shaft, 1-in. Smallest antero-
posterior diameter of shaft, -9in. Transverse of upper extremity, 1‘6in. An-
tero-posterior of ditto, 1-4in. Transverse of lower extremity, 1-4in. Antero-
posterior of lower central ridge, 1-2 in.

Figs. 13, 13 a, 13 6, and 13 c.—Portion of metatarsal bone and first
two phalanges of Hippotherium Antilopinum.—B.M.

Transverse diameter of inferior extremity of metatarsal, 15in. Length of first
phalanx, 2°7in.; of second, 1-5in. Antero-posterior diameter of lower extremity
of metatarsal, 1: in. ?

Figs. 14, 14 a, and 14 6.—First phalanx of posterior extremity of
Hippotherium Antilopinum.—B.M.

Length, 3-lin. Transverse diameter of posterior extremity, 1°6in. Vertical
diameter of ditto, ‘9 in. Transverse diameter of anterior extremity, 1‘3in. Ver-
tical diameter of anterior extremity, 6 in.

Figs. 15, 15 a, and 15 6.—Second phalanx of posterior extremity of
Hippotherium Antilopinum.—B.M.

Length, l-in. Transverse of posterior extremity, 1:5 in. Vertical of ditto,
l-in. Transverse of anterior extremity, 1°3in. Vertical of ditto, °7 in.

Figs. 16, 16a, and 16.—Last phalanx of posterior extremity of
Hippotherium Antilopinum.—B.M.

Length of fragment, 1:6in. Greatest breadth, 1‘8in. Height, 1-in.

Fig. 17._-Lower end of tibia and astragalus of Wippotherium Anti-
lopinum, restored.
Figs. 18, 18 a, and 18 6.—Calcaneum of Hippotherium Antilopinum.
Greatest length, 4° in,

Prate LXXXVI.

Figs. 1, la, 1b, and 1e—Camelus Sivalensis. (Fale. and Caut.)
Mutilated fragment of cranium broken off in front through the first
true molar. The great elevation of the sagittal and occipital crests,
the development of the temporal fossze, and the advanced position and
prominence of the orbits, are to be noted. The orbits also are elon-
gated from before backwards, instead of being circular or elongated
vertically as in the existing Camel—B.M. (See Memoir on ‘ Camel,’
antea, p. 227.)

Height of occipital facet, 4-°2in. Width between extreme parts of occipital con-
dyles, 3°in. Height of for. magnum, 1°5in. Width of ditto, 1:3in. Between
pariet. occipital angles, 4°3in. From lower angle of for. magnum to posterior bor-
der of last molar, 6-1in. Width of palate between anterior angles of last molars,
2°8in. Width across widest part of cranial cavity, 4°7in. Between external aud.
canal and posterior border of orbit, 4:8in. Antero-posterior diameter of orbit,
2‘7in. Vertical diameter of ditto, 17 in. Width across at posterior extremity of
zygomatic arches, 8-4in. Width across at posterior angles of orbits, 9°d in.

Figs. 2,2 a, and 26.—Camelus Sivalensis. Fragment of cranium
showing palate with series of true molars on both sides. The specimen
also shows the extreme depth of the maxillary which leads to the
arched appearance in the nose of the Camel.—B.M.

Length of true molar series, 4°8in. Width of palate between posterior angles
of last molars, 3°2in. Width of palate between anterior angles of anterior mo-
lars, 2° in.

Fig. 3—Camelus Sivalensis. Skull and lower jaw. Both jaws are

DESCRIPTION OF PLATES. 533

locked together, but the anterior and posterior extremities with the
upper surface of the skull are wanting. The animal was young, its
last permanent molars not being completely developed, and the third
milk molar being still in position. The general character is that of the
present Camel; the form of maxillaries, thickness of lower jaw and ex-
ternal appearance of the teeth corresponding as closely as two skulls of
one species would do. The position of the sub-orbitary foramen, how-
ever, is rather higher up on the maxillary, and the tapering of the
lower jaw is less than in the existing Camel. This specimen is also,
figured in ‘ Asiatic Researches,’ vol. xix. Plate XX. fig. 3—B.M. (See
antea, p. 252, and Pl. XVIII. fig. 2.)

Length of molar series (including two last premolars), 6°2in. Length of three
true molars, 4°8in. Height of ramus of lower jaw opposite last molar, 2°7 in.
Thickness of ramus of lower jaw opposite last molar, 1°35in. Length of molar
series of lower jaw (including last premolar), 6°3 in. Length of true molar of lower
jaw, ditto, 5°6 in.

Figs. 4 and 4a.—Camelus Sivalensis. Cranium including occiput
and nasal bones. The great width and massiveness of the cranium as
compared with the muzzle are well seen, and also the antero-posterior
elongation of the orbit—B.M. (See antea, p. 232, and Pl. XVIIL. fig. 1.)

Antero-posterior diameter of orbit, 2°3in. Height of ditto, 1-6in. Between
anterior angle of orbit and sub-orbital foramen, 2°3in. Length of first and second
true molars, 2°6in. Widest part of cranial box, 3°7 in.

Figs. 5 and 5 a.—Camelus Sivalensis. Lower jaw, which on the
right side, with the exception of the condyle and coronoid process, is
almost perfect. Fragments containing molars of upper jaw are still in
apposition at some places. The specimen shows four incisors on the
left side; the third right incisor is wanting. The wear of the teeth
and the flattened surface of the fourth or pointed incisor show that the
animal must have been of considerable age. This specimen is also
figured in the ‘ Asiatic Researches,’ vol. xix. Plate XX. fig. 4, a larger
quantity of matrix containing remains of upper jaw being there still
adherent.—B.M. (See antea, p. 232, and Pl. XVIII. figs. 3 and 4.)

Between outer margins of canines, 2°5in. Between outer margins of first pre-
molars, 2"lin. Diastema between canine and first premolar, *7in. Length of
the molar series, 5‘9in. Diastema between first and last premolar, 3:in. Length
of the three true molars, 4°9in. Length of symphysis, 6-3in. Interval between
rami opposite last molars, 2°6 in.

Pirate LXXXVII.
Camelus Sivalensis.

Figs. 1 and 1 a.—Palate with molar series on both sides imperfect.
That on the right side is most complete, and contains the penultimate
and last deciduous molar and the two first true molars.—B.M.

Length of molar series, 4:9in. Length of penultimate milk molar, ‘din.
Length of last deciduous molar, 1°3in. Length of first true molar,1°6in. Length
of second true molar, 1°9 in.

Figs. 2 and 2 a.—Fragment of upper jaw, left side, containing three
true molars.—B.M.

Length of true molar series, 4-9 in. Length of first molar, 1:3in. Length of
second ditto, 16in. Length of third ditto, 1:‘9in. Width of grinding surface of
first true molar, 1:1 in.

Figs. 3 and 3 a.—Fragment of upper jaw, left side, showing the

534 FAUNA ANTIQUA SIVALENSIS.

second and third true molars. This specimen isalso figured in ‘ Asiatic
Researches,’ vol. xix. Plate XXI. figs. 12 and 13—B.M. (See antea,
p- 234, and Pl. XVIII. fig. 5.)

Length of second true molar, 1°65 in. ; of third ditto, 2: in.

Figs. 4 and 4 a.—Fragment of upper jaw, containing third and fourth
premolars.—B.M.

Length of third premolar, 85 in. ; of fourth, -95 in.

Figs. 5 and 5 a—Fragment of horizontal ramus of lower jaw, con-
taining three true molars and fourth premolar.—B.M.

Length of molar series of lower jaw, 5-7in. Length of true molar ditto, 4°7 in.
Length of fourth premolar, 0°9 in. Length of first molar, 1:°2in. Length of second
ditto, 1-4in. Length of third ditto, 2-1in. Length of last lobe of third molar, ‘6in.

Figs. 6 and 6 a.—Fragment of lower jaw, right side, with ascending
ramus, condyle, and coronoid process, and containing last molar. The
jaw exhibits remarkable differences from the jaw of the existing Camel.
It more resembles the lower jaw of Ox, Deer, or Antelope, but is shown
to be of Camel by the heel or step on the posterior ascending margin,
which is the generic mark of a Camel. In the existing Camel the
ascending ramus rises at nearly a right angle to the line of jaw; it has:
considerable breadth antero-posteriorly, and its coronoid process is
short, straight, and massive. In the’ fossil the ascending ramus is as
oblique as in the Ox; it has no excess of breadth antero-posteriorly,
and the coronoid process is long, slightly curved back, and slender.
The condyle also has a much longer transverse diameter, its proportions
are more slender, and the depression on its upper margin much deeper
than in the existing Camel. The condyles, however, are not nearly so
slight and narrow as in the Ox and Buffalo. This specimen is also
figured in ‘ Asiatic Researches,’ vol. xix. Plate XX. figs. 6 and 7—B.M.
(See antea, p. 235.)

Length of last molar, 2:25in, Length of last lobe of ditto, °7 in.

Figs. 7 and 7 a.—Fragment of horizontal ramus of lower jaw, con-
taining fourth premolar, and the two first and a fragment of third true
molars.—B.M.

Length of fourth premolar, ‘85in. Length of first molar, 1:4 in, ie of
second ditto, 1‘8in. Length of fragment of third ditto, 1°6 in.

Figs. 8 and 8 a.—Fragment of horizontal ramus of lower jaw, con-
taining penultimate and last milk molars and first true molar.—B.M.

Length of penultimate milk molar, ‘6in. Length of last milk molar, 1-7 in.
Length of last lobe of ditto, -7in. Length of first true molar, 1°6 in.

Figs. 9 and 9 a.—Symphysis of lower jaw with series of six incisors.
The fourth incisor, or canine, on left side, also seen.— B.M.

Chord of the incisor series, 2°65in. Length of first incisor, *7in. Length of
second ditto, -73in. Length of third ditto, -6in.

Figs. 10, 10 a, and 10 6.—Symphysis of lower jaw with alveoli of
six incisors and two canines (fourth incisors).—B.M.

Width between outer margin of external incisive alveoli, 1-°8in. Width between
outer margin of canine ditto, 1°7 in.

Figs. 11 and 11 a.—Symphysis of lower jaw with six incisors and
two canines (fourth incisors). —B.M

Chord of incisor series, 1°55 in. Width between canine alveoli, 1‘5in. Length

of first incisor, 0'5in. Length of second ditto, 0°55in. Length of third or ex-
ternal, 0°65 in.

DESCRIPTION OF PLATES. 535

PuateE LXXXVIII.
Vertebre of Camelus Sivalensis.

Figs. 1, 1 a, 1 6, and 1 c.—Atlas with portion of axis adherent to
lower end.—B.M.

Length of lower arch of atlas, 26in. Length of upper ditto, 3:1in. Extreme
length of atlas, 4:4in. Between outer margin of posterior articular processes,
3°6 in.

Figs. 2, 2a, 26, 2c, and 2 d.—Third cervical vertebra.—B.M.

Greatest length of body, 7:-7in. Height of spinal canal posteriorly, 1- in.
Width of spinal canal ditto, 1°3 in.

Figs. 3, 3.a, 3b, 3c, and 3d.—Fourth cervical vertebra.—B.M.

Length of body, 7°8in. Length of spinal platform, 5°5in. Length between
extremities of oblique processes, 8'5in. Between extremities of inferior trans-
verse processes anteriorly, 4°6 in.

Figs. 4, 4 a, 46, 4c, and 4d—Fifth cervical vertebra.—B.M.

Greatest length of body, 6-9in. Width of spinal canal anteriorly, 1-1 in.
Height of spinal canal anteriorly, 1-in. Between outer margins of vertebral fora-
mina, 1:°95in. Diameter of vertebral foramen, 0-4in. Probable width across
transverse processes, 3°8 in.

Figs. 5, 5a, 5 6, and 5 c.—Cervical vertebra imperfect.—B.M.

Length of fragment of body, 3-9in. Height of spinal canal posteriorly,
1-2in. Width of spinal canal ditto, 1:2 in.

Figs. 6, 6 a, 66, and 6 c.—Cervical vertebra.— B.M.

Greatest length of body, 4:65in. Height of spinal canal posteriorly, 1:3 in.
Width of spinal canal ditto, 1-5 in.

PuateE LXXXIX.

Bones of anterior extremity of Camelus Sivalensis.

Fig. 1.—Scapula, almost perfect.—B.M.
Length of scapula, 22in. Width at narrowest part, 3-2in. Height of coracoid
process, 2°3in. Greatest projection of spine, 1°6 in.

Fig. 2—Glenoid cavity of scapula.

Greater diameter, 2°6in. Lesser ditto, 2°3 in.

Figs. 3,3 a, and 3 b.—Head of humerus, with double bicipital groove.
—B.M.

Greatest antero-posterior diameter, 5'in. Greatest transverse ditto, 4°6 in.
Chord of double bicipital groove, 3° in.

Fig. 4. Head of humerus, with double bicipital groove-—B.M.

Greatest antero-posterior diameter, 46in. Greatest transverse ditto, 3°6 in.
Chord of double bicipital groove, 2°7 in.

Figs. 5, 5a, and 5 b.—Lower end of humerus, with articular surface.
—B.M.

Transverse diameter of inferior articular surface, 36 in. Antero-posterior dia-
meter internally of inferior extremity, 4:in. Length of fragment, 7:1 in.

Figs. 6, 6 a, and 66.—Lower end of humerus and upper end of
conjoined radius and ulna zm situ.—B.M.

- Width of inferior articular surface of humerus, 3°3in. Width of superior arti-
cular surface of radius, 3°45 in.

536 FAUNA ANTIQUA SIVALENSIS.

Figs. 7, 7 a, and 7 b.—Portion of shaft and lower articulating ex-
tremity of conjoined radius and ulna.—B.M.

Width of conjoined lower extremity of radius and ulna, 2°9in. Circumference
of conjoined shafts at fractured extremity, 5-9in. Length of fragment, 7°3 in.

Figs. 8, 8 a, 8 5, and 8 c.—Lower end of conjoined radius and ulna,
with carpal and metacarpal bones.—B.M.

Extreme width of lower conjoined articular surface of radius and ulna, 3°8in.
Greatest antero-posterior diameter of ditto, 2°lin. Extreme length of fragment of
metacarpal, 14°55 in. Width of upper articular surface of metacarpal, 3°3 in.
Greatest transverse diameter of shaft at centre, 1-8in. Least antero-posterior
diameter of shaft at centre, 1-4 in.

Fig. 9.—Lower end of radius and ulna with bones of carpus.—
B.M.

Greatest length of pisiform, 2‘lin. Height, 2°in.
Figs. 10, 10 a, 100, and 10 c.—Bones of carpus.—B.M.

Figs. 11 and 11 a.—Fragment of upper end of metacarpal bone.—B.M.

Length, 12°in. Width of upper articular surface of ditto, 2-9in. Greatest
transverse diameter at centre of shaft, 16 in. Least antero-posterior diameter at
centre of shaft, 1°3 in.

Figs. 12,12 a, 126, and 12c.—Lower articulating extremity of meta-
carpal bone, deeply fissured.—B.M.

Extreme length of fragment, 5-6in. Interval between articular surfaces, °d in.
Greatest width of articular surface, 1:85in. Greatest antero-posterior of articu-
lar surface in centre, 19in, Circumference of shaft at fractured extremity, 5-9 in.

Figs. 18, 13 a, 13 6, and 18 c.—First phalanx and sesamoid bone.—
B.M.

Length of first phalanx, 3°9in. Width of proximal articular surface, 1-6 in.
Greatest width of distal surface, 1‘5in. Length of articular surface of sesamoid
bone, l‘in. Width of articular surface of sesamoid bone, °7 in.

Figs. 14, 14 a, 14 6, and 14 c.—First phalanx.—B.M.

Length of first phalanx, 4:in. Width of upper articular surface, 1:9in. Width
of lower articular surface (greatest), 1°6 in.

Piate XC.
Bones of posterior extremity of Camelus Sivalensis.
Figs. 1, 1 a, and 1 6.—Fragment of head of femur.—B.M.
Antero-posterior diameter of great trochanter, 2°9 in.
Figs. 2 and 2 a.— Articulating surface of head of femur.—B.M.
Diameter of articulating surface, 2°5 in.

Figs. 3, 3a, 3b, and 3c.—Fragment of lower end of femur with
condyles.—B.M.
Length of fragment, 6°6in. Width of lower extremity, 4°8 in.

Figs. 4, 4a, and 4 6.—Fragment of lower end of femur.—B.M.

Transverse diameter of lower extremity, 3°8 in. Width of rotular surface, 2:2 in.
Height of rotular surface in centre, 2°7 in.

Figs. 5, 5a, and 5 6.—Fragment of lower end of femur.

Width of lower extremity, 46in. Width of rotular surface in centre, 1°6in.
Height of rotular surface in centre, 2°6 in.

DESCRIPTION OF PLATES. 537

Figs. 6, 6 a, and 6 6.—Patella—B.M.
Length, 3°7in. Width, 2°9in. Width of articular surface in centre, 1°4 in.

Figs. 7, 7 a, 76, and 7 c.—-Entire tibia.—B.M.

Extreme length of tibia, 16°7in. Extreme width of upper articular surface,
3'6in. Width of inferior articular surface posteriorly, 2°3in. Circumference of
shaft below crest, 6: in.

Figs. 8, 8 a,-8b, 8 c, and 8 d.—Calcaneum and cuboid.—B.M.

Extreme length of calecaneum, 6°7in. Projection of calcaneal process, 4°3 in.
Greatest height, 3-lin. Height of cuboid, 2-2in. Greatest width superiorly,
1:7in. Antero-posterior diameter, 1°d in.

Fig. 9.—Tarsal bones, conjoined. a. caleaneum; 6. astragalus; c.
seaphoid ; d. internal cuneiform ; e. external cuneiform.—B.M.

Figs. 10, 10 a, 10, 10 ¢, and 10 d.—Astragalus.—B.M.

Extreme antero-posterior diameter, 3°4in. Width of anterior articular surface,
2°3in. Width of cuboid segment of ditto, ‘9 in. Width of trochlea, 2:1in. Height,
1:95 in.

Fig. 11.—Astragalus—B.M.

Extreme height, 2‘8in. Width of anterior articular surface, 1‘8in. Width of
cuboid segment of ditto, ‘6in. Width of trochlea, 1:6in. Height of astragalus,
1°55 in.

Figs. 12 and 12 a.—Upper extremity of metatarsal bone.—B.M.

Extreme width of upper articular surface, 2°2in. Antero-posterior diameter in
centre of upper extremity, 1-8 in.

Figs. 13, 13 a, and 13 b.—Metatarsal bone, entire; lower end deeply
fissured—B.M.

Length, 16‘in. Width of upper articular surface, 2‘3in. Width of each of
lower articular surfaces, 1:3in. Interval between them, °35in. Antero-posterior
diameter of each, 1-5in. Greatest transverse diameter of shaft in centre, 1:45 in.
Antero-posterior diameter of ditto, 13 in.

Figs. 14, 14 a, 14.0, and 14c¢. First phalanx.—B.M.

Length, 4:in. Width of proximal surface, 1-55in. Width of distal articular
surface, 1°45 in.

Figs. 15, 15 a, 15 6, and 15 e—Second phalanx.—B.M.

Extreme length, 2°4in. Width of upper articular surface, l‘in, Extreme
width of distal ditto, 1:15 in,

Puate XCI.

Sivatherium giganteum (Fale. and Caut.), from the Sewalik hills,
(See antea, p. 247.) Splendid specimen of cranium, anterior view, from
Sir Proby Cautley’s collection in British Museum. A full description
of this specimen, with measurements, will be found in the Memoir on

Sivatherium (See p. 256).—B.M. (Reproduced in Pl. XIX.)

Piate XCII.
Sivatnerium giganteum.
Figs. 1, 1 a, 16, and 1 c.—Four different views of same cranium as
figured in Plate XC.—B.M. (Figs. 1 a and 1 6 reproduced in Pl. XX.)

Figs. 2 and 2 a.—Fragment of cranium showing orbit and temporal
fossa, basilar process of occipital, depressions for condyles of lower
jaw, &c. This is the same specimen as is figured in Plate A. fig. 3,
under which the dimensions will be found.—B.M.

538 FAUNA ANTIQUA SIVALENSIS.

[Plate XCII. completes the series of published plates. Copies
of the seventeen plates which follow, and which I have de-
signated by letters (A. to R.), were found among Dr. Falconer’s
papers, or have been furnished by Mr. Ford, the artist and litho-
graphic engraver. These plates had been executed on stone,
and proof impressions struck off, but the plates were never
published, and unfortunately the stones were destroyed during
Dr. Falconer’s absence in India. These seventeen plates have
been deposited in the library of the Geological Department
in the British Museum, and from them several of the specimens
in the Museum have been named. Through the kindness of
Mr. Davies, I am enabled to give the British Museum Catalogue
number for each of the specimens figured, so that there will
be no difficulty in referring to the originals. |

Pate A.

Figs. 1, la, 16, and 1e.—Sivatherium giganteum. Cranium of
female, with perfect series of six molars on either side. ‘The specimen
is broken off in front of the molar ridges.—B.M. No. 89,528.

From anterior margin of foramen magnum to alveolus of first molar, 16: in.
From anterior to posterior side of last molar, 8°25 in. Width of skull between
borders of auditory foramina, 9°3 in. From the anterior margin of auditory foramen
to the rear molar, 69 in. Extreme length of fragment, 19°7 in. Height of
occiput, 6°5 in. Breadth of occiput, 9°5 in. Length of molar series, 8°5 in. Length
of true molar series along alveolar border, 4:11 in. Length of premolar series
along alveolar border, 4°4 in. Between anterior premolars, 2°4 in. Between
posterior molars, 4°3 in. Length of palate in mesial line from anterior edge of
first premolar to palatine notch, 6°10 in. Length from lower border of foramen
magnum to palatine notch, 9°5 in. Probable width across external orbital angles,
12-in. Length of orbit, 3:2 in. Length between auditory process and posterior
border of orbit, 6°5 in. Probable height at posterior border of palate in mesial
line, 6°5 in.

Figs. 2 and 2 a.—Portion of cranium of Sivatherium giganteum, found
by Col. Colvin in the lower hills below and west of Nahun. The
specimen is valuable, though it has no teeth, from having the occiput
very entire, and from its proving the accuracy of Dr. Falconer’s
assumptions, made before the specimen was found, and based on exami-
nation of the original head (Plate XCI.), that the animal had four
horns with bony cores, as this has the offset of one of the back branched
horns very clearly marked, and suitable to which a large flat horn was
found in Capt. Cautley’s collection, fig. 4. The parts appear slightly
distorted from the occurrence of a shift. This specimen is figured and
described by Col. Colvin in the Jour. As. Soc., Feb. 1837, vol. vi.
p- 152. It is also figured in Royle’s ‘ Illustrations of the Botany of the
Himalayah Mountains,’ vol. ii., Plate VI. fig. le.

This specimen was presented by Col. Colvin to the Museum of the
University of Edinburgh, where it now is. Its dimensions are as
follows :— <

Length from occipital crest to anterior margin of base of anterior horn-core (on
right side), 14:2 in. Between extreme points of occipital crest (imperfect), 16-2

DESCRIPTION OF PLATES. 539

in. Between extreme points of occipital condyles (external angles), 6-lin. From
the basilar surface, between the occipital condyles to occipital crest, 7°5 in. Height
of occipital foramen, 1°9 in. Breadth of occipital foramen, 1:8 in. Greatest
breadth of upper surface of occipital condyle, 2° in. From occipital crest to
posterior border of posterior core, 1‘2 in. Breadth of cranium beneath posterior
core (distance between outer margins of roots of posterior cores), 12'1in. Extreme
distance between fractured extremities of posterior cores, 21:5 in. Transverse
diameter of posterior core before its expansion, 5°8 in. Transverse diameter of
root of posterior core, 8:4 in. Thickness of core before expansion, 31 in. Great
diameter (transverse) of posterior branch, 2°5 in. Thickness (vertical diameter)
of posterior branch, 1°6 in. Chord of arc between the origin of posterior branch
and occipital crest, 3:1 in. Thickness of stem (thickest part) of core, 3°6 in.
Circumference of core before expanding, 15:8 in. Antero-posterior diameter of
broken end of anterior core, 4:1in. Greatest transverse diameter of broken end
of anterior core, 3°2 in. Diameter of base of anterior core, 4°9 in. Breadth of
cranium beneath anterior cores (at foramen orbitale lacerum), 53 in. Between
inner margins of articular surfaces of occipital condyles superiorly, 1:7 in. Height
of fossa for insertion of ligam. nuche, 3°8 in. Base of fossa for insertion of ditto,
6-4 in. Groove between condyles, inferiorly, °45 in. Between outer margin of
condyles (inferiorly) immediately behind parieto-occipital crest, 2°7 in. From
grooves behind occipital condyles, inferiorly, to floor of fossa for ligam. nuche,
3°7 in.

This is the specimen referred to by Dr. Falconer at page 266, and
also figured in Plate XXI. fig. 2.

Fig. 3.—Sivatherium giganteum. Fragment of cranium, showing
forehead, orbits, and cores of anterior horns. This is the same specimen
as figured in Plate XCII. figs. 2 and 2 a.—B.M.

Antero-posterior diameter of orbit, 4:in. Height of orbit, 2°7 in. Width of
malar bone, 2°7 in. Width between anterior angles of orbit, 11:2 in. Width be-
tween outer margins of orbital cornua, 12° in. Long diameter of fractured end of
right horn, 5"18in. Short or transverse diameter of right horn, 3:10 in. Probable
width between mastoid processes, 16° in.

Fig. 4.—Sivatherium giganteum. Fragment from middle of posterior
horn. This specimen was in Sir Proby Cautley’s collection, and was
found to correspond to the posterior horn-core in Col. Colvin’s specimen.
It is also figured in Plate XXI. fig. 3, antea, p. 268.—B.M. No. 39,525.

Length following curvature, 21: in. Circumference at lower attachment, 18: in.
Breadth at offset, 8-1 in. Length of offset at base, 5:in. Width of offset at base,
46in. Breadth at fractured upper extremity, 6°8 in. Greatest thickness at upper
extremity, 2°3 in.

Fig. 5.—Sivatherium giganteum, Fragment from apex of posterior
flattened horn.—B.M. No. 39,524.

Extreme length, 10°6 in. Breadth at base, 8-4 in. Thickness (extreme),
1'8 in.

PuateE B.

Sivatherium giganteum.—Figs. 1, 1 a, and 16.—Fragment of atlas,
comprising lower arch.—B.M. No. 39,526.
Extreme length of fragment, 8°3 in. Antero-posterior diameter of lower arch,
371 in.
Fig. 2.—Atlas, very perfect.—B.M. No. 39,527.
Extreme breadth, 8°3 in.

Figs. 3, 3 a, 36, and 3 c.—Axis.—B.M. No. 39,528.

Extreme length, including odontoid process, 7°6 in. Extreme breadth of anterior
articular surface, 5°7 in. Length of spinal platform, 5°2 in. Height of spinal
canal, posteriorly, 1°7 in. Breadth of spinal canal, posteriorly, 1°7 in.

540 FAUNA ANTIQUA SIVALENSIS.

Figs. 4 and 4a.—Second, third, fourth, fifth, sixth, and seventh
cervical vertebra in situ.
Length of conjoined vertebrae, 27°4 in.

Figs. 5, 5a, 56, and 5c.—Sixth cervical vertebre.—B.M. No.
16,225.

Extreme length of body, 5°38 in. Height of spinal canal, posteriorly, 1:3 in.
Width of spinal canal, ditto, 1-6 in.

Figs. 6, 6 a, 6b, and 6 c.—Cervical vertebra.—B.M. No. 18,173.

Extreme length of body, 6:1 in. Between outer margins of anterior articular
processes, 7°6 in.

Figs. 7, 7a, 76, and 7¢.—Seventh cervical vertebra.—B.M. No.
15,707.

Extreme length of body, 5:1 in. Width of spinal canal, 2°5 in.

Figs. 8, 8 a, and 8 b.—First dorsal vertebra. It resembles that of the
British elk.

Extreme length of body, 4°5 in. Width across transverse processes, 8°] in.

Figs. 9, 9 a, and 9 b.—Dorsal vertebra.—B.M. No. 17,078.
Extreme length of body, 4°7 in. Extreme width across transverse processes,
7 in. ;
Fig. 10.—Four dorsal vertebra in situ.
Length of conjoined vertebre, 13°8 in.

Figs. 11 and 11 a. —Lower end of tibia, with bones of tarsus. a.
tibia; 6. fibular element; c. astragalus; d. calcaneum; e. cuboid; f.
scaphoid.—B.M. No. 39,529.

Length of fibular element, 2°6 in. Height of fibular element (a), 1°3 in.
Length of fragment of os calcis (lower edge), 5:in. Height of fragment to fibular
surface, 3°3 in. Width of trochlea of astragalus (¢), 3°lin. Length of astragalus
internally, 4°2 in. Length of scapho-euboid (inferiorly), 2-1 in. Length of ento-
cuneiform (superiorly), 1-1 in.

Figs. 12, 12 a, and 12 6.—Phalanx (first?) of posterior extremity.
—B.M. No. 39,530. ,

Extreme length, 5:1 in. Width of upper articular surface, 2°5 in. Width of
lower articular surface, 2°2 in.

Figs. 13, 138 a, 13 6, and 13 c.—Phalanx (second ?) of posterior ex-
tremity.—B.M. No. 15,805.

Extreme length, 2°6 in. Width of upper articular surface, 2:2 in. Width of
lower articular surface, in centre, 2° in.

Pate C,
Sivatherium giganteum. Bones of anterior extremity.

Figs. 1 and 1 a.—Fragment of sternum.—B.M.

Length of fragment, 14-8 in. Greatest depth, 5°5in. Greatest width, inferior
end, 4-2in. Greatest width, upper end, 3°3 in.

This specimen is also figured in Plate XXI. fig. 5. (See antea, p. 270.)

Figs. 2 and 2 a.—Fragment of scapula, showing glenoid cavity and
coracoid process.—B.M. No. 36,680.

Length of fragment, 11°8in. Great diameter of glenoid cavity, 4-4in. Lesser
diameter of glenoid cavity, 3°3in. Lleyation of coracoid process above glenoid
cavity, 1‘8in. Breadth of scapula towards neck, 44 in.

DESCRIPTION OF PLATES. 541

Figs. 8 and 3 a.—Glenoid cavity, and coracoid process of scapula.—
B.M. No. 39,531.

Length of fragment, 6-in. Great diameter of glenoid cavity, 4:8in. Lesser
diameter of glenoid cavity, 4:in. Elevation of coracoid process above glenoid
cavity, 1-8 in.

Figs. 4, 4a, 46, and 4c.—Humerus entire. The dimensions almost
(not quite) agree with B.M. No. 39,688.

Length of humerus, 20°-2in. Breadth of upper extremity, 8'4in. Greatest
ant.-posterior diam. of upper extremity, 7°7 in. Great diameter of articular surface
of head, 5-5in. Lesser diameter of articular surface of head, 5-°2in. Breadth of
inferior extremity, 7°2in. Antero-posterior diameter, internally, 55in. Breadth
of inferior articular surface, 6°2in. Smallest antero-posterior diameter of inferior
articular surface, 2°2in. Smallest transverse diameter of shaft, 3°8in. Smallest
antero-posterior diameter of shaft, 3°3 in.

Figs. 5, 5 a, and 5 6.—Upper half of humerus with double bicipital
groove.—B.M. No. 89,532.

Length of fragment, 10:in. Transverse diameter of upper extremity, 6°3 in.
Antero-posterior diameter greatest of ditto, 7°8in. Greatest diameter of articular
surface of head, 5-in. Lesser diameter of articular surface of head, 4-6in. Cir-
cumference at broken extremity, 11: in.

Figs. 6, 6 a, 6 6, and 6 c—Radius and ulna conjoined ; entire length.
—B.M. No. 39,534.

Length of united radius and ulna, 30-2in. Length of radius, 25-8in. Width
of radius above articular surface, 7-4in. Chord of sigmoid cavity, 3-lin. From
anterior edge of sigmoid cavity to posterior border of olecranon, 10-8in. Depth
of olecranon, 55in. Breadth of inferior extremity, 6:1in. Breadth of inferior
articular surface, 5°5 in. Greatest antero-posterior diameter of ditto, 3:6 in.
Transverse diameter of centre of shaft, 4°5in. Antero-posterior diameter of centre
of ditto, 2°5 in.

Fig. 7—Upper articular surface of conjoined radius and ulna.—
B.M. No. 39,535.

Length of radius, 10°8in. Width of articular surface above, 7:°2in. Width of
upper extremity, 8'3in. Greatest antero-posterior diameter of articular surface,
3°6 in.

Figs. 8 and 8 a.—Lower end of conjoined radius and ulna, with
bones of carpus and upper end of metacarpal bone.

Transverse diameter of scaphoid (1), 1°9in.; height of ditto, 1-6 in. Transverse
diameter of semilunar (2), 2°1in. ; height of ditto,1:8in. Transverse diameter of
cuneiform (3), 1:4 in.; height of ditto, 18 in. Transverse diameter of united
trapezium, trapezoid, and os magnum (5), 2°3 in.; height of ditto, 1:4 in.
Transverse diameter of unciform (6), 1°9in.; height of ditto, 1:4 in. Breadth
of inferior extremity of radius and ulna conjoined, 53 in. Breadth of upper
extremity of metacarpal bone, 4°65 in.

Fig. 9.—Carpal bones, viz., scaphoid (1), semi-lunar, cuneiform (3),
os magnum (5), and unciform (6).

Height of scaphoid, 2:lin.; breadth of ditto, 1°7in. Height of semilunar,
2-4in.; breadth of ditto, 23in. Height of cuneiform, 2-6in.; breadth of ditto,
24in. Height of os magnum, 1:7in.; breadth of ditto, 3:lin. Height of os
unciforme, 2*in.; breadth of ditto, 2:lin. Antero-posterior diameter of os mag-
num, 3'5in. Antero-posterior diameter of os unciforme, 2°9 in.

Figs. 10 and 12.—First row of carpal bones, viz., scaphoid (1), semi-
lunar (2), and cuneiform (3).

Antero-posterior of scaphoid, 3:4 in. Antero-posterior of semilunar, 2:2 in,
Antero-posterior of cuneiform, 2°5 in.

542 FAUNA ANTIQUA SIVALENSIS.

Figs. 11 and 13.—Second row of carpal bones, viz., os magnum (5)
and unciform (6).

Fig. 14.—Phalanges of anterior extremity, restored.

Figs. 15, 15 a, 15 6, and 15 c.—Right metacarpal bone.—B.M. No.
39,533.

Length of right metacarpal, 13°7in. Breadth of superior articular surface,
4:7in. Greatest antero-posterior of ditto, 2°7in. Breadth of inferior articular
surface, 4°7in. Breadth of a single trochlea, 2:2in. Antero-posterior diameter of
a single trochlea measured along ridge, 2°4 in.

Figs. 16, 16 a, 16 b, and 16 c.—First phalanx.—B.M. No. 39,541.

Length between articular surfaces, 46in. Transverse diameter of posterior
articular surface, 2°3in. Height of posterior articular surface in centre, 1°6in.
Transverse diameter of anterior articular surface, 2°3in. Height of anterior arti-
cular surface in centre, 1°4 in.

Figs. 17; 17 a, 17 6, and 17 c.—Second phalanx—B.M. No. 39,542.

Length, 2-4 in. Transverse diameter of posterior articular surface, 2°4 in.
Height of posterior articular surface measured along ridge, 1-5in. Transverse
diameter of anterior articular surface, 2°5in. Height of anterior articular surface,
2°5 in.

Figs. 18, 18 a, and 18 6.—Hoof-bone.

Length, 5-2in. Height, 2°5in. Breadth, 2°3in. Length of articular surface,
19in. Breadth of articular surface, 1°9 in.

Piate D.
Bones of posterior extremity of Sivatherium giganteum.

Figs. 1, 1 a, and 1 b.—Fragment of upper end of femur, with articular
surface.—B.M. No. 39,545.

Length of fragment, 7-4in. Breadth of upper extremity including the troch-
anter, 7°5in. Antero-posterior diameter of greater trochanter, 3°6in. Antero-
posterior diameter of articular surface of head, 3-in. Transverse diameter of arti-
cular surface of head, 3°8 in.

Figs. 2, 2 a, and 2 6.—Fragment of upper end of femur with articular
surface.

Length of fragment, 8°4 in. Antero-posterior diameter of great trochanter,
4:in.

Figs. 3, 3a, and 36.—Lower end of femur, with condyles and
articular surface.—B.M. No. 39,546.

Length of fragment, 9°6in. Breadth of inferior extremity, 64in, Antero-
posterior diameter internally, 8-6 in. Antero-posterior diameter externally, 6° in,
Height of rotular surface in centre, 4*11in. Breadth of rotular surface in centre,
2:9 in.

Figs. 4, 4a, and 46.—Fragment of lower end of femur, showing
articular surface.-—B.M. No. 39,547.

Breadth of inferior extremity, 7-6in. Height of rotular surface in centre,
4-7 in.
Figs. 5, 5 a, 5 6, and 5 c.—Entire tibia.—B.M. No. 17,072.

Extreme length of tibia, 20°3in. Breadth of upper extremity, 6°in. Antero-
posterior diameter of upper extremity, 3°5in. Breadth of inferior extremity,
4-6in. Antero-posterior diameter of ditto, 3-2in. Breadth of shaft (smallest),
2°7in. Antero-posterior diameter of shaft (ditto), 1°9 in.

“by

DESCRIPTION OF PLATES. 543

Figs. 6, 6 a, and 6 6.—Proximal end of tibia, with articular surface.
—B.M. No. 18,452.

Length of fragment, 9°5 in. Breadth of upper extremity, 6:8 in. Antero-
posterior diameter of ditto, 3°8 in.

Figs. 7, 7 a, and 7 6.—Proximal end of tibia, with articular surface.
—B.M. No. 16,611.

Breadth of upper extremity, 7-1in. Antero-posterior diameter of ditto, 4: in.

Figs. 8, 8 a, and 8 6.—Distal half of tibia, with lower articular sur-
face.—B.M. No. 39,548.

Length of fragment, 9°5in. Breadth of inferior extremity; 5°in. Antero-
posterior diameter of ditto, 3° in.

Figs. 9, 9 a, and 9 6.—Distal end of tibia, with lower articular sur-
face.—B.M. No. 39,549.

Length of fragment, 7° in. Breadth of inferior extremity, 5:in. Antero-
posterior diameter of ditto, 2°6 in.

Fig. 10.—Calcaneum and astragalus, in situ, restored.

Figs. 11, 11 a, and 11 6.—Astragalus.—B.M. No. 16,998.

Length of astragalus, extreme, 4°9in. Breadth of ditto, 3-4in. Height of ditto,
2°4 in.

Figs. 12, 12 a, and 12 6.—Calcaneum, imperfect.—B.M. No. 39,543.

Length of fragment, 8:2 in. Projection of heel, 6°2in. Breadth of calcaneal
tuberosity, 2°9in. Height of calcaneal tuberosity, 2°8 in.

Figs. 13, 18 a, 13 6, 138 ¢, and 18 d.—Scapho-cuboid bone. —B.M.
No. 39,544.
* Breadth of scapho-cuboid bone, 4-9in. Greatest antero-posterior diameter,
5-in. Greatest height, 3-4in. Breadth of astragalar surface, 3:°6in. Breadth of

ealeaneal ditto, 1:3in. Breadth of cuneiform ditto, 1‘5in. Breadth of metatarsal
ditto, 18 in.

Puate E.
Camelopardalis Sivalensis (Fale. and Caut.).

Figs. 1,1 a, 1 6,1 ¢, and 1 d.—Third cervical vertebra of fossil giraffe,
from the Sewalik hills. The elongated character of the vertebra shows
that the animal had a columnar neck, and the fact that the transverse
processes are provided with foramina for the vertebral arteries shows
that it was not a camel. The complete synostosis of the upper and
lower articulating surfaces, the strong relief of the ridges, and the depth
of the muscular depressions, indicate that the animal was an adult,
which had long attained its full size.

A note of this specimen, by Captain (now Sir Proby T.) Cautley,
appeared in the Journ. As. Soc. for July 1838, vol vil. p. 658, and a
detailed account was afterwards communicated to the Geological So-
ciety of London by Dr. Falconer and Captain Cautley, an abstract of
which appeared in the ‘Proceedings,’ No. 98. In the latter com-
munication the measurements and drawings of the specimen are given
(See antea, p. 197, and Pl. XVI. figs. 1, 2, 38, and 4).—B.M. No. 39,747.

Figs. 2 and 2 a.—Fragment of second cervical vertebra of Camelo-
pardalis Sivalensis, from Perim Island. The right margin of the drawing
shows the mesial longitudinal ridge under the side of the body, and the
left margin is the ridge of the spinous process. The process pointing

544 FAUNA ANTIQUA SIVALENSIS.

downwards on the left side is the inferior oblique process. The cup-
shaped articulating surface for the head of the third cervical vertebra
is well seen.

This specimen was in the collection of fossils brought from Perim
Island by Captain Fulljames, and was described and figured by Dr.
Falconer in the Quarterly Journal of the Geol. Soc., vol. 1. Plate XIV.
fig. 5 (See antea, p. 398).—B.M. No. 39,748.

Length of fragment, 4:9 in. Height of body posteriorly, 2°5in. Greatest
breadth posteriorly between remains of transverse processes, 3'lin. Height of
the spinal canal, 1-4in, Height of the broken surface of the spine above inferior
margin of body, 6‘4in, Vertical diameter of articulating cup, 2;>1in. Transverse
diameter of ditto, 2°1 in.

Figs. 3 and 3 a.—R. humerus, head wanting.—-B.M. No. 39,749.

Length of humerus wanting upper head, 17-7in. Breadth of inferior extremity,
5:2in. Antero-posterior diameter of inferior extremity, 4°3in. Breadth of arti-
cular surface of inferior extremity, 4°8in. Breadth of upper extremity, 4*lin.
Circumference at smallest part of shaft, 7-9 in.

Figs. 4, 4a, and 4b.—Fragment of shaft of left radius and ulna.
B.M. No. 17,130.

Length of fragment, 8°5 in. Greatest diameter, 3° in. Smaller ditto, 2:1 in,
Great diameter of ulna at upper extremity, °7 in. Thickness of ditto, °5 in.

Figs. 5, 5 a, and 5 6.—Radius and ulna, restored.

Figs. 6, 6 a, and 6 6.—Metacarpal bone, fragment including upper
end.—B.M. No. 389,750.

Length of fragment, 18°7 in. Transverse diameter of upper extremity, 3°7 in.
Antero-posterior diameter of upper extremity, 2°3 in. Transverse diameter of
centre of shaft, 2-4 in. Antero-posterior diameter of ditto, 1:8 in.

Figs. 7 and 7 a.—Fragment of shaft of metacarpal bone.—B.M. No.
39,751.

Length of fragment, 6°8 in. Transverse diameter of shaft, 2°3 in. Antero-
posterior diameter of ditto, 1°3 in.

Figs. 8 and 8 a.—Fragment of shaft of metacarpal bone.—B.M. No.
17,129.

Length of fragment, 4:°2in. Transverse diameter of shaft, 2° in. Antero-posterior
diameter of ditto, 2° in.

Figs. 9 and 9 a.—Fragment of shaft of metacarpal bone, near lower
end.—B.M. No. 17,181.

Length of fragment, 3-9 in. Transverse diameter of shaft at lower extremity,
2°8 in. Antero-posterior ditto at upper ditto, 1:5 in.

Figs. 10, 10 a, and 10 6.—Entire metacarpal bone, restored.

Figs. 11, 11 a, and 11 6.—First cervical vertebra, imperfect.—B.M.
No. 39,746.

Length of fragment, 4:2 in. Height of body posteriorly, 3-3 in.; breadth of
ditto, 2°8 in. Between extremities of transverse processes, 6° in. Between inner
angles of posterior articular processes, 2°in. Length of posterior articular surface,
2:1 in.; breadth of ditto, 1:2in. Height of spine above inferior margin of body,
64in. Height of spinal canal, 1:4 in.; breadth of ditto, 1:6 in.

Figs. 12, 12 a, 12 6, and 12 c.—Left metatarsal bone of Sivatherium
giganteum.—B.M. No. 39,752.

Length of left metatarsal, 16-4 in. Transverse diameter of upper extremity,
38 in. Antero-posterior diameter of ditto, 8°38 in. Transverse diameter of lower

DESCRIPTION OF PLATES. 545

extremity, 4"in. Transverse diameter of a single trochlea, 1-9 in. Antero-posterior
diameter measured alongst ridge of trochlea, 2-2 in. Transverse diameter of shaft,
2°1in. Antero-posterior ditto, 2°1 in.

Figs. 13, 13 a, and 13 6.—Fragment of upper end of metatarsal bone
of Sivatherium giganteum.—B.M. No. 39,753.

Length of fragment, 7-2 in. Transverse diameter of upper extremity, 4:1 in.
Antero-posterior diameter of ditto, 3°8 in.

Puate F.

Bramatherium Perimense (Falc.), from Perim Island. A large and
peculiar ruminant, nearly equalling the Sivatherium in size, but essen-
tially different. The plate represents fragments of the bones of the
anterior and posterior extremities. A description of two fragments
of the left upper jaw, including the entire series of upper grinders,
will be found in the memoir on Perim Island fossils. (See antea,
p- 399). The specimens figured in this plate were brought from Perim
Island by Captain Fulljames.

Figs. 1, 1a, and 16.—Fragment of lower end of humerus, with
articular surface.

Figs. 2 and 2 a.—Fragment of upper end of ulna, with olecranon and
sigmoid cavity.

Figs. 3, 3 a, and 3 6.—Fragment of lower end of radius and ulna.

Figs. 4, 4 a, and 4 b.—Fragment of lower end of radius and ulna.

Figs. 5 and 5 a.—Fragment comprising portion of shaft and distal
extremity of metacarpal bone.

Figs. 6 and 6 a.—Fragment comprising distal articulating extremity
of metacarpal bone.

Figs. 7, 7 a, and 7 b.—Fragment of upper end of femur.

Figs. 8, 8 a, 8b, 8c, and 8 d.—Calcaneum.

Figs. 9, 9 a, 9b, 9c, and 9 d.—Astragalus.

Figs. 10, 10 a, 10 b, 10 c, 10 d, and 10 e.—Astragalus.

PLATE G.

Bos Namadicus (Falc. and Caut.), from the Nerbudda.

Figs. 1, 1 a, and 1 6.—Fragment of cranium, showing forehead, occiput,
occipital condyles, and foramen magnum ; portion of right horn and core
of left horn. The specimen shows well the flat square forehead, the height
being about equal to the breadth. The horns are attached to the ex-
tremity of the highest salient line of the head. The horn-cores spread
out horizontally, with a slight arch upwards and concavity below. The
section of the horn-core shown in fig. 1 0. is much more circular than
in the Gour or Gayal or than in Bos Paleindicus.

This specimen, which is in the British Museum (No. 39,760), is also
figured in Plate XXII. (See also antea, p. 286.)

Figs. 2, 2a, 2b, and 2¢.—Bos Namadicus. Four different views of
VOL. I. NWN

546 FAUNA ANTIQUA SIVALENSIS.

cranium, including orbit, nasal bones, and palate, and four posterior
molars on either side. ‘The forehead is mutila‘ed and the horn-cores

are broken off.—B.M. No. 39,758.
Bos Paleindicus (Fale. and Caut.), from the Nerbudda.

Figs. 3 and 3 a.—Fragment of cranium, including orbits, horn-cores,
frontal and occipital, on both sides (See antea, p. 284)—B.M. No.
39,716.

Fig. 4.—Fragment of cranium, showing occiput, foramen magnum,
condyles, and horn-cores.—B.M. No. 39,717.

Figs. 5 and 5 a.—Fine specimen of cranium, showing occiput, condyles,
and foramen magnum, portion of right horn and left horn-core, both
orbits, palate, and four posterior molars. The upper surface of the
frontal is arched. The horn-cores spread out more horizontally, and
with a less inclination upwards than in the existing wild buffalo, and
are slightly concave anteriorly and convex behind. In these respects
it differs from the existing wild buffalo, and so far as the horizontal
offset is concerned, it approximates to the Gayal, from which, however,
it differs in the flattened form of the horns and in every other respect.
These characters are so constant that there can be little doubt that the
species is distinct from the existing wild buffalo.—B.M. No. 39,759.

Fig. 6 —Fragment showing anterior portion of upper jaw, with in-
termaxillary bones.—B.M. No. 39,715.

Figs. 7 and 7 a.—Fragment of horn, broken off at tip, and of a com-
pressed form. Fig. 7 a shows the flattened form of the horn, as seen
at section. This, as well as the fragment represented in fig. 6, have
been found to belong to a skull in the British Museum (No. 39,715), to
which they are now attached. This skull is also represented in Plate
XXII. (See also postea, p. 354.)

Length, 33°5 in. Greatest diameter, 6°5 in. Least ditto, 3:25 in.

Pruate H.

Hemibos triquitriceras. (Fale. and Caut.), from the Sewalik hills.

Figs. 1 and 1 a.—Cranium, including both orbits and horn-cores,
occiput, nasals, palate, and entire molar series on both sides.—B.M.
No. 39,584.

Length of fragment, 18°2 in. Great diameter of core, 4 in. Breadth of cranium
at post. angles of orbits, 8°7 in. Great diameter of orbit, 2°7 in. Length of molar
series, 5-4 in, Length of three premolars, 23 in. Width of palate anteriorly and

‘posteriorly, 2°9 in. Height of cranium from middle of palate, 4-2 in. Breadth
of ditto in front of premolars, 4° in.

Figs. 2 and 2 a.—Another fine specimen of cranium, showing occipital
condyles and foramen, palate and five back molars on either side, both
horn-cores, and a portion of right horn. The remarkable triangular
form of the horn-core is well shown.—B.M. No. 16,411.

Length of fragment, 14-4 in. Height of occipital facet from lower border of
occipital foramen to summit of occipital crest, 4*8 in. Between extreme points of
occipital crest, 8°8 in. Breadth of cranium beneath cores, 4-6 in. Breadth of
cranium at post. angles of orbits, 8°8 in. Length of core from roughness on frontal
bone, 11°3 in, Base of the triangular core ant. surface (at origin), 38-9 in. Base of
the triangular core ant. surface (at broken end), 3° in. Diameter of occipital

DESCRIPTION OF PLATES. 547

condyle, 2°6 in. Height of occipital foramen, 1:5 in. Breadth of ditto, 1:2 in.
Greatest diameter of orbits, 2°5 in. From lower border of occipital foramen to pos-
terior border of palate, 6-2 in. Length of true molar series, 3:2 in. Length of
two posterior premolars, 1°3 in. Width of palate posteriorly and anteriorly,
2°8in. Width of alveoli, 1:3 in.

Figs. 8 and 3a.—Fine specimen of cranium, showing orbits and
horn-cores, but no horns, nasals, palate with entire molar series (on
left side), nasal and intermaxillary bones almost complete. Presented
by Colonel Colvin.—B.M. No. 23,109.

Length of fragment, 17- in. Breadth of cranium between posterior angles of
orbit, 8°8 in. Between most projecting points of maxillary bones, 5-7 in. Breadth
at intermaxillaries, 3:5in. Breadth of two nasal bones, 1-9 in. Length of molar
series, 5°3 in. Length of three true molars, 3°3 in. Breadth of alveoli, 1:2 in.
Width of palate, 3-1 in.

Figs. 4, 4 a, and 4b.—Very perfect specimen of cranium, showing
horn-cores, orbits, occipital foramen and condyles, palate, and entire
molar series on either side. Probably a female.—B.M. No. 16,173.

Length of fragment, 13-7 in. Height of occipital facet from lower border of
occipital foramen to occipital crest, 3°8 in. Breadth of cranium beneath cores,
38 in. Breadth of ditto at posterior angles of orbits, 7°6 in. Height of occipital
foramen, 1-4 in. Breadth of ditto, 1:2 in. Diameter of occipital condyle, 2° in.
From lower border of occipital foramen to posterior border of palate, 5-9 in. Length
of three true molars, 3:2 in. Length of two posterior premolars, 1:4 in. Width
of palate, 8: in. Breadth of alveoli, 1:2 in. Great diameter of orbits, 2°8 in.
Lesser diameter of ditto, 2°5 in. Diameter of core at root, 2°3 in.

Prare i.
Bos (Amphibos) acuticornis (Fale. and Caut.), from the Sewalik hills.

Figs. 1, 1 a, 1 6, and 1 c.—Fine specimen of cranium, showing palate
and molar series, occiput, orbits, and horn-cores, but without horns.

Figs. 2 and 2 a.—Fragment of cranium with both horns, perfect to
the tips. The direction of the horns is more upright and less hori-
zontal than in Hemibos triquitriceras. The horns are rounded on their
‘ anterior surface and flattened behind, and taper to a point.—B.M.
No. 39,560.

Between mastoid angles, 8°5in. Height of occipital facet from lower border of
occipital foramen, 4:°4in. Between external anglesof condyles, 4°3in. Diameter of
condyle, 2-°2in. Height of occipital foramen, 1°4in. Breadth of occipital foramen,
1:3in. Breadth of cranium beneath the cores, 4*5in. Breadth of cranium in
front of the cores, 7°5 in. Length of cores along great curvature, 27-in. Between
tips of cores, 33°6in. Antero-posterior diameter of core (inner surface), 3°4 in.
Cireumference of core at root, 12‘in. Between nearest points of base of cores,
2°8 in.

Figs. 5, 3a, 3b, and 3 c—Another specimen of cranium, with palate
and entire series of six molars, occiput, both orbits, and horn-cores.—
B.M. No, 39,564.

Length of fragment, 17-4 in. From posterior plane of occipital condyles to anterior
margin of molar series, 13°7in. Length of molar series, 54in. Length of three
true molar teeth, 3:3in. Breadth of alveoli, 1lin. Width of palate posteriorly,
2°8in. Between external angles of occipital condyles, 4°5in. Height of occipital
facet from lower border of foramen magnum, 4:3in. Height of occipital foramen,
14in. Between mastoid angles, 7°2in. Width of cranium beneath cores, 3°7 in.
Between most projecting points of maxillary bones, 5°8in. Breadth opposite
sub-orbital foramina, 4*1in. Least width of nasal bones, 1-4in. Great diameter of
orbit, 2°8 in.

NN 2

548 FAUNA ANTIQUA SIVALENSIS.

Puate K,

Felis cristata (Fale. and Caut.), from the Sewalik hills. This fossil
Tiger forms the subject of a special memoir (See antea, page 315).

Figs. 1, la, 16, and 1¢.—Four different views of an imperfect
specimen of the cranium, from Mr. W. Ewer’s collection. The left
maxillary bone with the teeth is absent, but this portion was found
after the drawing was made, and has been added to the specimen in the
British Museum (No. 15,902). The specimen shows well the great
prominence of the sagittal crest, whence the specific name is derived,
also the relative shortness of the facial portion of the head, the great
height of the occipital, and the horizontal outline of upper surface of
cranium.

Figs. 2, 2a, 2b, and 2e—Another specimen of the cranium with
the alveolar ridges almost perfect. The anterior portion of the palate
with the incisors is broken off. The canine, two false molars, the car-
nassier, and tuberculous teeth well seen. This is the specimen which
is also figured in Plate XXV. figs. 1 to 4.—B.M. No. 37,133.

Figs. 3, 3a, and 3b.—Mutilated fragment of posterior portion of
cranium.—B.M. No. 37,134.

Figs. 4,4 a, and 46.—Mutilated fragment of anterior portion of
cranium and face, showing the left orbit entire, the palate, and the series
of teeth on both sides.—B M. No. 87,135.

Puate L.

Hyena Sivalensis (Fale. and Caut.)—Fossil Hyena from the Se-
walik hills. Unfortunately no description of this fossil was ever
published, and no account of it is to be found among Dr. Falconer’s
notes. This species, however, is no doubt that designated Hyena
Stvalensis by Messrs. Baker and Durand in the brief description given
by them in the Journal of the Asiatic Society for October 1835, vol. iv.
p- 569. Their description is accompanied by drawings of a remark-
ably perfect specimen of the skull, with the lower jaw in situ.

Figs. 1 and 1 a.—Fragment of anterior portion of right side of palate,
with canine, and incisor teeth.—B.M. No. 39,718.

Figs. 2, 2a, and 26—Fragment of anterior portion of cranium,
showing canine and incisors in very perfect state—B.M. No. 16,583.

Figs. 3, 3a, and 3 6.—Another fragment of anterior portion of palate
with three incisors.—B.M. No. 89,719.

Fig. 4.—Dental series, right side, consisting of canine anteriorly, _
three false molars, the two posterior of which are very large. A very
large carnivorous tooth with a small tubercle within and in front,
and a small back or fifth molar, placed transversely at the back of the
palate.

Figs. 5 and 5 a.—Fragment of upper jaw, left side, containing hind-
most false (or third) molar, large carnivorous tooth or fourth molar,
and small back or fifth molar, placed transversely at the back of the
palate—B.M. No. 34,140.

DESCRIPTION OF PLATES. 549

Figs. 6 and 6 a.—Fragment of upper jaw, right side, containing three
false molars, large carnivorous tooth with its internal tubercle, and a
portion of the fifth or small back molar.—B.M. No. 37,1387.

Figs. 7 and 7 a.—Fragment of upper jaw, right side, with hindmost
false or third molar, and large carnivorous tooth with its internal

tubercle.—B.M. No. 37,139.

Figs. 8 and 8 a.—Fragement of upper jaw, right side, with two pos-
terior false molars, and large carnivorous tooth with internal tubercle.
—B.M. No. 37,138.

Figs. 9 and 9 a.—Large carnivorous tooth with internal tubercle
detached.—B.M. No. 15,413.

Fig. 10.—Dental series of Hyana, left side, comprising canine,
three false molars (two back ones large), large carnivorous tooth with
internal tubercle in front, and small back molar placed transversely at
back of palate.

Puate M.

Figs. 1 and 1 a—Hyena Sivalensis. Fragment of lower jaw, right
side, very perfect, containing carnivorous molar and two backmost false
molars.—B.M. No. 16,565.

Figs. 2 and 2 a.—Hyena Sivalensis? Fragment of lower jaw, left
side, containing three incisors, one canine, three false molars, and one
large carnivorous molar. Belongs to an undetermined feline animal

(Hyena ?).— B.M. No. 16,555.

Figs. 3 and 3 a—Hyena Sivalensis. Fragment of lower jaw, left
side, with canine and three false molars.—B.M. No. 16,584.

Figs. 4 and 4 a.—Hyena Sivalensis. Imperfect fragment, lower jaw,
right side, with two incisors, canine, and three false molars.

Figs. 5 and 5 a—Hyena Sivalensis. Imperfect fragment, lower jaw,
left side, containing canine, three false molars, and large carnivorous
molar. Large mental foramen corresponding to front false molar.—
B.M. No. 39,7381. .

Figs. 6 and 6 a.—Hyena Sivalensis? Fragment of anterior portion,
lower jaw, right side, containing two incisors, canine, and two anterior

false molars. (An. Felis ?)—B.M. No. 16,585.

Figs. 7 and7 a.—Hyena Sivalensis? Crushed fragment, lower jaw,
left side, containing carnivorous molar and two back false molars.
(An. Felis?)—B.M. No. 37,140.

Figs. 8 and 8 a.—Hyena Sivalensis? Fragment of lower jaw, right
side, containing two posterior false molars, and large bicuspid carnivo-
rous false molar.—B.M. No. 16,578.

Fig. 9.—Hyena Sivalensis. Fragment of lower jaw, right side, con-
taining three false molars and large canine. Large mentary foramen
corresponding to front false molar.

550 FAUNA ANTIQUA SIVALENSIS.

Puate N.

Drepanodon (Machairodus) Sivalensis (Falc.and Caut.), or fossil Dre-
panodon, from the Sewalik hills. No description of this fossil was ever
published, but the Sewalik specimens are referred to by Professor Owen
in ‘ British Fossil Mammalia,’ pp. 178,179, and also in ‘ Odontography,’
vol. i. p. 491. A brief description of it is also given by Dr. Falconer
in his ‘Notes on the Fossil Felis spelea of the Mendip Hills.’ See

vol. ii.

Figs. 1, 1 a, 1 6, and 1 e.—Drepanodon Sivalensis. Fragment of pos-
terior portion of skull, showing occipital condyles, foramen magnum,
and prominent sagittal crest.—B.M. No. 39,278.

Fig. 2.—Drepanodon Sivalensis. Mutilated specimen of cranium,
including facial portion, but no distinct evidence of teeth_—B.M. No.
39,729.

Figs. 3 and 3 a.—Drepanodon Sivalensis. Fine fragment of upper
jaw, right side, with apparently the first or deciduous dentition. The
crown of the canine is broken off, but what remains is seen to be flat,
and very finely serrated along the posterior edge, like a shark’s tooth.
The tooth evidently bore the same proportion to the molar series as
does the canine of the Felis megantereon of Bravard (Vide Owen,
Brit. Fos. Mam. p. 178).—B.M. No. 16,350.

The following note, from Dr. Falconer’s Note-book, dated October 2,
1858, probably referred to this specimen and to that represented
in fig. 5 :—

‘In the Sewalik Machairodus the right upper carnassier is formed with a very
thin blade. The anterior lobe is damaged, but judging from what remains it
would seem to haye been two-lobed. The middle lobe is thin and pointed; but
neither the anterior lobe nor the middle one bears the slightest indication of
an internal tubercle. If ever there, it is gone. Owen describes it as being there,
“‘but less developed than in the normal species of Felide.” The posterior lobe is
nearly horizontal and very trenchant; in fact, the tooth is compressed and sharp-
edged. All the points rise. The length of the crown is-75 inch. There is an
interval between the carnassier and canine of 0°8in., part of which has been arti-
ficially rubbed down, but there is not the least indication of a fang-pit or fang.
(Owen says there is, and that it is single-fanged and simple!) There is a distinet
show of a double fang, ‘fore and aft, of a tubercular in a line with the sectorial,
behind it. The breadth of the canine atits base is 0‘Sin. It is very compressed.
The posterior concave edge is finely serrated. (Owen says that both edges are
distinctly serrated.)’

This specimen is also figured in Plate XXV. fig. 5.

Figs. 4 and 4 a.—Drepanodon Sivalensis. Fragment of lower jaw with
three premolars, the last being the sectorial. Professor Owen refers to
this specimen in the following description :—

‘A portion of the lower jaw of a larger Machairodus, from the Sewalik range,
shows the beginning of the characteristic downward extension of the symphysis, and
the depression on the outside of the ramus for the lodgment of the long upper
canine. The molar series, which consists, as in the typical Felines, of three pre-
molars, the last being the sectorial tooth, has a longitudinal extent of two inches ;
the second molar slightly overlaps the third, which has an antero-posterior extent
of eleven lines. This portion of jaw indicates a species of Machairodus as large
as the Jaguar; it most probably belongs to an adult of the same species as the
one indicated by the instructive portion of the upper jaw.’ (Fig. 3). (Owen, Brit.
Foss. Mam. p. 179).

This specimen is also figured in Pl, XXV. fig. 6.—B.M. No. 16,557.

DESCRIPTION OF PLATES. 551

Figs. 5 and 5 a.—Drepanodon Sivalensis. Fragment of upper jaw,
containing two anterior molars. The first is simple, singled-fanged
and very small. The second is the carnassial or sectorial tooth. Its
crown is more compressed, its trenchant margins sharp. See descrip-

tion of fig. 4. —B.M. No. 39,730.

Figs. 6 and 6 a.—Drepanodon Sivalensis. Lower jaw, right side,
more perfect than fig. 4, and containing the incisors as well as the
canine and three molars. The downward projection of the symphysis,

and the depression for the upper canine, are well seen.—B.M. No.
16,573.

Figs. 7 and 7 a.—Drepanodon Sivalensis. Another specimen of
lower jaw, right side, containing three molars and alveolus of large

‘canine.—B.M. No. 16,537.

Figs. 8 and 8 a.—Drepanodon Sivalensis. Fragment of lower jaw,
with three molars.—B.M, No. 16,554.

Puate O.

Ursus (Hyenarectos) Sivalensis (Fale. and Caut.), from the Sewalik
hills. The fossil Bear of the Sewalik hills forms the subject of a dis-
tinct memoir (see antea, page 321). Its chief peculiarities are to be
found in the teeth, which are constructed more after the type of the
higher Carnivora than any other described species of the genus.

Figs. 1, 1a, 1 6, and 1 c.—Superb specimen of cranium. The three
rear molars are perfect on one side, and but little damaged on the other.
Both canines are present, and that of the right side is entire. The
alveoli of the two false molars and three incisors on either side are dis-
tinct, although the teeth are wanting. The only considerable deficien-
cies are in the posterior and lower parts of the occiput, both zygomatic
arches, and in the lower end of the nasals, where a fissure extends
across the face, on both sides towards the orbits. Fig. 1 a shows the
dental series on right side, of natural size.

This specimen is described in detail in the memoir already referred
to, and is also represented in Plate XXVI. figs. 1 and 2.—B.M. No.
39,721.

Figs. 2 and 2 a.—Ursus Sivalensis. Greater part of the body of the
lower jaw, broken off where the canine protrudes, and also deficient in
the coronoid and articulating processes. There are indications of six
molars, of which the two first premolars and the rear tubercular molar
have dropped out. The third premolar is distinctly three-lobed. The
antepenultimate or carnassier is chiefly remarkable for its length. The
penultimate or first tubercular is broader for its length and less com-
plicated with tubercles than what is general in the genus. Fig. 2a
shows the dental series of the natural size.

Further details of this specimen are given in the memoir on Ursus
(See antea, p. 8321).—B.M. No. 39,722.

The specimen is also represented in Plate XXVI. figs. 3 and 4.

Figs. 3, 3 a, 36, 3c, and 8d.—Ursus Sivalensis. Second cervical
vertebra or axis.—B.M. No. 37,143.

552 FAUNA ANTIQUA SIVALENSIS.

Figs. 4, 4a, and 4 b.—Ursus Sivalensis. Radiusand ulna. Greater
portion of shafts and lower articulating extremity. From Messrs.
Baker’s and Durand’s -collection.—B.M. Nos. 39,725-6.

Figs. 5, 5a, 5b, 5¢e, and 5d.—Ursus Sivalensis. Specimen of
femur, very perfect.—B.M. No. 39,723.

Figs. 6, 6a, 66, 6c, and 6d.—Ursus Sivalensis. Distal end of
metacarpal bones.—B.M. No. 87,147.

Figs. 7, 7 a, and 7 6.—Ursus Sivalensis. Fragment of phalanx.

Fig. 8.—Ursus Namadicus (Fale. and Caut.). Portion of upper
jaw with four molars of a smaller species of Bear, from the Nerbudda,
represented of the natural size. The rear molar is much more elon-
gated from before backwards than in the Sewalik species—B.M. No.
39,720. .

Figs. 9 and 9 a.—Ursus Namadicus. Tibia of Bear, from the Ner-
budda, presented by C. Frazer, Esq.—B.M. No. 39,727.

Fig. 10.—Right femur of Ursus spelwus, from College of Surgeons,
figured for comparison.

Puate P.
Fossil Otters, from the Sewalik hills.

Figs. 1, la, 16, and 1¢.—Zwutra Paleindica (Fale. and Caut.).
Beautiful specimen of cranium with alveolar ridges very perfect. The
zygomatic arches are absent. Shows the alveoli of three incisors on
either side, the outer one being slightly larger than the two inner ones.
Outside the three incisors is the alveolus of a large canine, followed by
the alveoli of four small molars, and last of all by the carnassier and
tubercular, the latter greatly developed.

This specimen is also represented in Plate XXVII. figs. 6 and 7.
—B.M. No. 37,151.

Figs. 2 and 2 a.—Lutra Paleindica. Beautiful specimen of lower
jaw, left side, including ascending ramus. Shows a portion of canine
and of three small molars, the crowns of which are broken off. Behind
there is a large carnassier, very perfect; and last of all is the alveolus
of the tubercular, which is small in comparison to that of the upper
jaw.

This specimen is also represented in Plate XXVII. fig. 8.—B.M. No.
37,152.

Figs. 3 and 8a.—Lutra Indica. Two views of skull, upper and
lateral, of existing Indian Otter.

Figs. 4, 4a, and 4b.—Enhydriodon ferox.1 A new fossil genus of
otter from the Sewalik hills. Three views of cranium, probably female,
much mutilated. Shows on right side the posterior of the two false
molars, the carnassier and the tubercular. The anterior premolar,
which is deciduous, is wanting. The remarkably square form of the
carnassier is well seen.

This specimen is also represented in Plate XXVII. figs. 3 and 4.
—B.M. No. 37,153.

1 Subsequently designated Enhydriodon Sivalensis. See memoir at p. 331,

DESCRIPTION OF PLATES. 553

Figs. 5 and 5.a.—Enhydriodon ferox. Fine specimen of anterior
portion of cranium of an old individual, with very perfect alveolar
ridges. Shows on either side the alveolus of a large outer incisor,
which evidently served as a subsidiary canine. The middle incisors
are not only wanting, but the alveoli are completely filled up and
obliterated. The canines, which are broken across, are seen to be very
large, and were evidently of great strength and massiveness. All trace
of the first molar, which is very small and deciduous, has disappeared.
The second molar has two fangs, and is encircled by a rugged basal
ridge. The carnassier is very remarkable, and presents the prominent
feature of the genus. It is nearly square, instead of being triangular,
as in both Lutra and Enhydra; and instead of the cusps and trenchant
ridges of Lutra, or the flattened inequalities of Enhydra, the coronal
lobes are developed into cervical mammille, somewhat like those of the
Mastodon. A more detailed description of this tooth will be found in
the memoir on ‘ Enhydriodon’ (See p. 335). Behind the carnassier,
the tubercular is seen zn situ.

This specimen is also represented in Plate XXVII. fig. 5—B.M. No.
37,155.

Figs. 6, 6 a, and 6b.—Enhydriodon ferox. Fine specimen of ante-
rior portion of cranium of a young and probably female individual.
Shows three incisors on either side, the two inner of which are very
much compressed laterally, so that their antero-posterior diameter is
three times that of their width. The outer incisors are remarkably
large, as are also the canines. ‘The left canine has dropped out, leaving
_a large oval alveolus; the right canine is seen in section. Behind the
canine, on either side, is an extremely small empty alveolus of the first
deciduous molar. ‘Then comes the bicuspid second molar, the peculiar,
square, mammillated carnassier, and the tubercular.

This specimen is also represented in Plate XXVII. figs. 1 and 2.
—B.M. No. 87,154.

PLATE Q.
Carnivora, from the Sewalik hills.

Figs. 1, 1 a, and 1 6.—Skull, showing palate and teeth, of a fossil
species of Canis? from the Sewalik hills.1\—B.M. No. 40,188.

Figs. 2, 2 a, and 2 6.—Skull, showing palate and alveoli of entire
dental series of a fossil species of Canis? from the Sewalik Hills—B.M.
No. 37,150.

Fig. 5.—Fragment of palate, right side, with two posterior molars of
fossil Canis ?—B.M. No. 40,180.

Figs. 4, 4a, 46, and 4¢—Ursitaxus Sivalensis. Very perfect
cranium and face of a species of Ratel from the Sewalik hills. This
appears to be the specimen of fossil ‘ Gulo’ described by Messrs. Baker
and Durand, in the Journ. As. Soc., vol. v. p. 582, and figured in Plate
XXVII. fig. 5, of the same volume (See antea, p. 851).—B.M. No.
40,184.

1 Another specimen belonging to Dr. | lection in the British Museum. Along
Falconer, and labelled by him ‘Skull of | with the skull are portions of the femur,
fossil Canis, from Sewalik hills, has, | tibia, and bones of the foot.
since his death, been added to the col-

554 FAUNA ANTIQUA SIVALENSIS.

Pruate R.

Fossil Remains of Birds from the Sewalik hills.

Figs. 1 and 1 a.—Cervical vertebre of a bird.—B.M. No. 28,105.

Figs. 2, 2a, 2b, 2c, and 2 d.—Lower end of tibia.—B.M. No.
89,732.

Figs. 3, 3 a, 36, and 3 c.—Lower end of ditto.—B.M. No. 39,735.

Figs. 4, 4 a, and 4 b.—Lower end of ditto.—B.M. No. 39,737.

Figs. 5, 5a, 5 6, and 5 c.—Lower end of ditto.—B.M. No. 39,734.

Fig. 6.—Lower end of tibia of a bird, recent.

Figs. 7, 7a, and 7 6.—Fragment of wing-bone.—B.M. No. 39,740.

Figs. 8, 8a, 8 b, and 8 c.—Fragment of diito.—B.M. No. 39,738.

Figs. 9 and 9 a.—Upper end of metatarsal bone.—B.M. No. 39,741.

Figs. 10, 10 a, and 10 b.—Fragment of wing-bone.—B.M. No. 39,744.

Figs. 11, 11 a, and 11 b.—Fragment of ditto.—B.M. No. 39,743.

Figs. 12 and 12 a.—Fragment of ditto.—B.M. No. 39,739.

Figs. 13, 13 a, 18 6.—Fraginent of ditto.—B.M. No. 39,742.

Figs. 14, 14a, and 14 6.—Fragment of tarso-metatarsus.—B.M. No.
39,736.

Figs. 15, 15 a, 15 6, 15 c, and.15 d.—Phalanx of large bird.—B.M.
No. 39,783.

Among Dr. Falconer’s papers were also found numerous outline
sketches on tracing paper of Sewalik fossils, &c., belonging to the Rumi-
nantia and Reptilia, intended for the Fauna, but which had never been
engraved. These tracings have been pasted on sheets of paper, and
deposited with the seventeen unpublished plates in the Library of the
Geological department of the British Museum. They are as follows :—

A. RUMINANTIA.

Sheet 1.—Four views of skull of European bison, recent.
Sheet 2.—Four views of skull of Indian bison, recent.

Sheet 8.—Four views of skull of wild Indian buffalo, recent.
Sheet 4.—Four views of skull of Bubalus brachycerus, recent.
Sheet 5.—Skull of Bos primigenius, for comparison.

Sheet 6.—Restoration of head of Bos Paleindicus, including anterior
portion of face and horn, figured in Plate G. figs. 6 and 7.

Sheet 7.—Fragment of horn of Amphibos acuticornis.

Length of fragment along great curvature, 18°2 in. Chord of lesser curvature,
11:8 in. Great diameter at larger end, 3°7 in. Thickness at ditto, 2°7 in. Cir-
cumference at smaller end, 8° in.

Sheet 8.—Skull of Amphibos acuticornis (See antea, p. 547), including

fragments of both horns and both orbits, but greater part of face defi-

cient.—B.M. No. 36,666.

Length of fragment, 10°1in. Between mastoid angles, 6°5 in. Between ex-
ternal angles of condyles, 3°8 in. Breadth of cranium beneath cores, 4:1 in.

Tig Se ee

DESCRIPTION OF PLATES. 555

Breadth of cranium in front of cores, 6'4in. Breadth of cranium between posterior
angles of orbits, 7°7 in. Height of occipital facet from lower border of occipital
foramen, 3°7 in. Diameter of condyle, 2° in. Height of occipital foramen, 1:2 in.
Breadth of ditto, 1-2 in. Diameter of orbits, 23 in. Length of three true molars,
3°2in. Width of alveoli, 1:1 in. Width of palate, 2-9 in. Diameter of core at
root (greatest), 2°7 in. Circumference of ditto, 8° in.

Sheet 9.—Figs. 1 to 3, and fig. 4. Two specimens of portion of
skull of Hemibos triquitriceras (See antea, p. 546). Fig. 4 includes
the orbit, palate, and molars, but the posterior portion of the skull is
broken off. Its dimensions are as follows :—

Length of fragment, 14:3 in. Length of three true molars, 3-3 in. Length of
three true molars and last premolar, 4° in. Breadth of alveoli, 1:2 in. Width of
palate posteriorly, 2°3 in. Between mastoid angles, 6:2 in. Breadth of cranium
beneath cores, 4° in. Breadth of cranium between posterior angles of orbits, 7°3 in.
Breadth of cranium between most projecting points of maxillary bones, 5:3 in.
Height of occipital facet from lower border of foramen magnum, 4:1 in. Great
diameter of orbit, 2-4 in. Great diameter of root of core, 3°1 in. Thickness of
ditto, 2°7in. Length of fragment of core, 6°2 in.

Sheet 10.—Portion of skull of Bos Sivalensis- (See antea, p. 280).

Length of fragment, 12°7 in. Breadth of cranium behind orbits, 8°7in. Breadth
of cranium between posterior angles of orbits, 10°-4in. Width of alveoli, 1:1 in.
Width of palate posteriorly, 3-5 in. Diameter of orbit, 2°5 in. Height of cranium
from palate in front, 7-4 in.

Sheet 11.—Antilopide. Figs. 1, 2, and 3 represent the skull of
Antilope Paleindica (Fale.), No. 89,594 in B.M. This is the skull
described and figured by Captain Baker in the Journ. As. Soc., vol.
xii. p. 770, and which is also represented in Plate XXIII. (See antea,
p- 290.)

Sheets 12 to 18.—Represent numerous fossil remains of Antilopide,
Cervida, &c.

B. ReEptivia.

Sheet 1.—Crocodilus bombifrons. Fossil Crocodile, from the Sewalik
hills (See antea, p. 855). Ten figures, illustrating different portions of
the skull. Figs. 1 and 2 correspond to British Museum Spec., No.
39,795; figs. 5 and 6 to B.M., 39,796; and figs. 8 and 9 to B.M.
39,097.

Sheet 2.—Crocodilus bombifrons. Other specimens of the skull.
Figs. 1 and 2 correspond to British Museum Spec., No. 39,799; figs.
3 and 4 to B.M., 39,800; figs. 5, 6, and 7 to B.M., 39,801; and figs.
8, 9, and 10 to B.M. 39,798.

Sheet 3.—Crocodilus (Leptorhynchus) crassidens. (Fale. and Caut.)
Fossil Crocodile from the Sewalik hills. Five specimens are figured of
different portions of the skull; four of which correspond to specimens

in the British Museum, Nos. 16,218, 39,802, 89,8038, and 39,804.

Sheet 4.—Crocodilus (Leptorhynchus) Leptodus. (Fale. and Caut.)
Fossil Crocodile from the Sewalik hills. Four different specimens of
cranium, which correspond to the specimens in the British Museum,

Nos. 7,453, 39,805, 39,806, and 39,807.

Sheet 5.—Crocodilus (Leptorhynchus) Gangeticus. Fossil Crocodile
from the Sewalik hills, identical with modern Gavial. Three specimens
of different portions of skull, which correspond to the following catalogue
numbers in the British Museum, viz., 36,647, 39,809, and 39,810.

556 FAUNA ANTIQUA SIVALENSIS.

Sheet 6.—Crocodilus (Leptorhynchus) Gangeticus. Five specimens
of different portions of cranium and jaws, four of which correspond to
the following catalogue numbers in the British Museum, viz., 36,726,
39,808, 39,811, and 39,812.

Sheet 7.—Crocodilus (Leptorhynchus) Gangeticus. Portion of cra-
nium, vertebre, and other parts of skeleton. Correspond to British
Museum numbers, 17,006, 39,811 a, 39,818, 39,814, 39,815, 39,816,
and 59,818.

Sheet 8.—Skull and jaws of Crocodile.

Sheet 9.—Colossochelys Atlas (Fale. and Caut.), from the Sewalik
hills. Different specimens of episternal bones.

Sheet 10.—Colossochelys Atlas. Entosternal and xiphosternal bones,
margin of carapace, &c.

Sheet 11.—Colossochelys Atlas. Portions of carapace.

Sheet 12.—Colossochelys Atlas. Humerus, fragments showing upper
and lower ends, &c.

Sheet 18.— Colossochelys Atlas. Specimens showing ulna, pubes,
ischium, femur, &c.

Sheet 14.—Colossochelys Atlas. Foot bones.

Sheet 15.—Fossil Hmyda, from the Sewalik hills. Portions of carapace.
Sheet 16.—Fossil Hmyde, from the Sewalik hills. Carapace.

Sheet 17.—Fossil Emyde, from the Sewalik hills. Carapace.

Sheet 18.—Fossil mys tecta, from the Sewalik hills, identical with
existing species (See antea, p. 384). Two specimens are figured corres-
ponding to Nos. 39,837 and 17,435 in the British Museum catalogue.

Sheet 19.—Fossil Trionyx, from the Sewalik hills and the Nerbudda.
Sheet 20.—Fossil Zrionyx, from the Sewalik hills.

OO EEE

EXPEDITION TO CASHMEER AND LITTLE TIBET. 557

XXVIII. OFFICIAL REPORT OF EXPEDITION TO
CASHMEER AND LITTLE TIBET IN
1837-38.

BY H. FALCONER, M.D.}

To F. Currie, Esquire, Secretary to the Governor-General,
North Western Provinces.

Srr,—In compliance with the orders of the Right Honour-
able the Governor-General, I have the~ honour to submit a
summary of my proceedings whilst on the deputation on
which I was ordered to Cashmeer, and of the results of my
journey.

2. Having received the orders of the Government, communi-
cated in a letter from Mr. Secretary McNaghten, addressed
to Captain Wade, political agent, Loodianah, dated the 22nd
May, 1837, to proceed on the mission with Lieutenant
Mackeson; and, having provided the materials and neces-
saries required for a collection in Natural History, I started
from Suharunpoor on the 29th June, and joined Captain
Wade on the banks of the Sutlej on the Ist July.

3. His Highness the Maharajah having granted permission
for my proceeding to Lahore, and thence through the Pun-
jab, to join and accompany Lieutenant Mackeson in the
capacity of a medical attendant, I separated from Captain
Wade on the 6th July, after having had the benefit of every
assistance which that officer could afford towards the comfort
and facility of my journey, and of advice regarding my con-
duct in the countries which I was likely to visit. I started
from Loodianah on the 9th July, and arrived at Lahore, vid
Kupoorthulla and Amritsir, on the morning of the 15th.
I was most kindly and hospitably received by General
Ventura, the only European officer then at the Court, the
others, with the principal Sikh Sirdars, having been ordered
to Peshawur with the army after the affair of Jumrood. I
was admitted to an audience with His Highness on the 15th,
and received by the Maharajah with the usual marks of atten-
tion and consideration which he exhibits in his intercourse
with British officers. I learned on this occasion that Captain

1This report is now for the first | years have elapsed since it was written,
time published. Although nearly thirty | it will still be read with interest.—[ Ep. |

558 REPORT OF EXPEDITION

Burnes had already arrived at Kala Bagh, and was en route
to Attock, which information induced me to urge upon His
Highness the favour of an early departure from Lahore, so as
to give me the chance of joining the Caubul Mission party
before they crossed the Indus. My request was acceded to,
and on the 17th I was admitted to an audience of leave.
On the 18th I set out for Attock. ‘

4. It had been my intention, originally founded on the
information which I got from Captain Wade regarding the
progress of Captain Burnes up the Indus, to proceed from
Lahore to Pind Dadun Khan, and thence to Kala Bagh
with the object of examining the structure and productions
of the salt range, so remarkable a feature in the physical
geography of the Punjab; but the advance of the mission
compelled me to forego this object, and to follow the direct
route to Attock. I reached Vazeerabad on the Chenab in
three marches, and thence I went to Rotas! vid Buerwal
and Jhelum; from Rotas I proceeded by Bukrala Bish-
endour and Maukejalla to Rawul Pindee, and thence by
Kale Sura to Hussein Abdul, where I had the satisfac-
tion of joing Captain Burnes’ party, and of meeting with
Lieutenant Mackeson, the companion of the journey which I
had in prospect.

5. My further movements were now dependant on those of
Lieutenant Mackeson, and I found that that officer, in pur-
suance of instructions from Captain Wade, purposed to proceed
to Peshawur. This decision caused me some perplexity. Hus-
sein Abdul is a point where the road diverges respectively to
Cashmeer and to Peshawur. Cashmeer was the destination
which the Government pointed out for my investigations in
Natural History ; and the season for collecting with advan-
tage in botany was already far advanced, and, as regarded the
hill tracts around Cashmeer, it was rapidly passing by. If
I went on to Peshawur, and subsequently got to Cashmeer, I
could, at the best, but expect to arrive there about the end

1 «T beg to inform you that one of the
most interesting results at which I have
arrived has been the finding of fossil
bones and other organic remains along
all the points of the Sub-Himalayan
range on which I have touched, from
Jhelum on to near Rawul Pindee, and
that these fossil bones have proved iden-
tical with those which are found in the
Sewalik or Sub-Himalayan range be-
tween the Jumna and Sutlej, thus estab-
lishing the identity of the tertiary for-
mations along the foot of the Himalayahs
from the Ganges to near the Indus, a
point of much interest in the physical

history of the mountain range. —Evtract
from Official Report by Dr. F. to Captain
Wade, July, 1837.

‘I got our Sewalik fossils on beyond
the Jhelum, proving the protraction of
the range as far nearly as the Indus,
and I got fossil shells, apparently recent,
like our Sewalik casts, in a superficial
bed of the salt formation near Rotas,
in a sandstone associated with crocodi-
lean remains. I had only half an hour
one evening to give to the inquiry.’—
Letter from Dr. F. to Captain Cauticy,
Cashmeer, January 11, 1838.—[Ep.]

ase: rely A

TO CASHMEER AND LITTLE TIBET. . 559

of autumn, when a great part of the flora would have dis-
appeared, or waned into the fall of the leaf. I had the option
of two alternatives—either to proced to Peshawur or to con-
tinue at Hussein Abdul, till such time as the Maharajah
should concede his permission to my going on to Cashmeer.
For it had not hitherto been intimated to His Highness, in
the communications made by Captain Wade, that Cashmeer
was the direction in which it was intended that I should be
sent. I had been mentioned on all the occasions on which
my name had been referred to simply in the capacity of a
medical attendant with Lieutenant Mackeson, and that offi-
cer had as yet received no reply to an application which he
had made to the Maharajah from Kala Bagh for permis-
sion to pay a visit to Cashmeer, accompanied by me. Under
these circumstances, I deemed it expedient, with the con-
eurrence of Lieutenant Mackeson, to proceed with him to
Peshawur. This determination afforded an opportunity of
examining the productions of a country which had not
hitherto been botanically surveyed.

6. From Hussein Abdul the Mission moved on to Attock, a
halt at which place and at Khyrabad led to the detection of
numerous plants new to the flora of the North of India. On
arrival at Peshawur, it appearing probable that the Mission
would remain some time there, I applied to Captain Burnes
for his sanction and assistance to go on to Kohat, and
the hills around the valley of Bunguish, with the object of
examining the salt range and the district reported to contain
coal. That distinguished officer entered zealously into the
design, and made arrangements which enabled Dr. Lord and
myself to proceed, General Avitabile, the Governor of Pesha-
wur, and Sirdar Sultan Mohamed Khan, having. readily
acceded, and given their best aid to our undertaking the
journey. In consequence of the then recent affair of Jum-
rood the country was in an excited and disturbed state,
which the arrival of the British Mission at Peshawur did not
tend to diminish ; and we were impressed by General Avita-
bile with the necessity of some caution in our procedure.
We started from Peshawur with an escort, supplied by
Sultan Mohamed Khan, on the 15th August, and halted
at Andavre; from that we moved by a long night’ march to
Kohat; thence to Hangoo, at the upper end of the Bunguish
valley, with the purpose of proceeding to Nuryoob, one of
the reported coal localities. But the country was so dis-
turbed that we were not permitted to proceed, and we had
to return to Kohat. By the assistance of Sirdar Sultan
Mohamed Khan, who had by this time arrived at Kohat,
we again set out; and proceeded in the direction of Lachee,

560 REPORT OF EXPEDITION

near which we had an opportunity of examining one of the
beds reported to be coal, and the great salt bed at Jutta.
The coal proved to be a subordinate bed of bituminous shale,
included between the arenaceous strata of the salt formation
and an overlying stratum of fibrous gypsum. Although an
anxious search was made both by Dr. Lord and myself,
neither of us came upon any organic remains, vegetable or
animal, either in the sandstone or in the shale.! Neither our
time nor the state of the country admitted of our examin-
ing the other reputed localities near Khoorpa and Nuryoob ;
but as the geological features of the country were exhibited
on a great scale, and displayed much uniformity of character
throughout, it is probable that no considerable difference
could have been found in these. I arrived at the opinion
that there are no grounds to believe that any considerable
coal deposit exists in the salt range, the specimens dis-
covered proving to be either mineral pitch or bituminous
shale,’ and that shale containing not a trace of organic
remains, or anything resembling a vegetable impression. It
appears probable to me that the carboniferous bed discovered
by Lieutenant Wood near Kala Bagh is of a relatively
modern era, so to speak, compared with any great system of
coal measures, and that the mineral does not occur there or
at any point in the salt range near the Indus, of a description
and in a quantity fit to be worked for fuel. In connection with
this point, I may here add that the ‘coal’ of the Sewalik
hills, mentioned by Dr. M‘Clelland and in the Report of the
Coal Committee, as found near Hurdwar, is a tertiary
‘lignite,’ occurring in thin seams or in detached masses,
formed probably by the bituminization of single portions of
wood washed into the alluvium during the deposition of the
strata; and after an extensive examination of the Sub-
Himalayah Hills, from the Ganges on to the Sutlej, and at
several points between that river and the Indus, I am of
opinion that there are no grounds to believe that a coal
formation anywhere exists along the tract, or that the
mineral is to be found in any other state than that of isolated
masses or thin seams of lignite.

7. From Jutta we returned to Kohat, where we had an
interview with Sultan Mohamed Khan, who had displayed
every inclination to further the objects of our journey to the
utmost of his ability. He possesses a certificate from Mr.

‘ In a letter to Captain Cautley, how- | pitch coal, compact and very splendent,
ever, dated Cashmeer, January 11, 1838, | and so inflammable as to catch fire in
Dr. F. observes: ‘I got fossil remains | the flame of a candle, and burn freely,
in a limestone near the salt range close | caking and leaving a spongy coke.—
to Kohat.’—[ Ep. ] Letter to Captain Cautley, 1838.—

* It is a most highly bituminized | [Ep.]

TO CASHMEER AND LITTLE TIBET. 561

Moorcroft, expressing in the strongest terms the sense enter-
tained by that lamented traveller of his friendly services, and
of his being better disposed towards Huropeans, and those of
the British nation in particular, than any of the other
Barukzye chiefs. At our last interview, he exhibited another
example of the lengths to which Afghan noblemen carry
their feuds, and of the sacrifice of every other consideration
to personal feeling, in requesting Dr. Lord to impress upon
Captain Burnes that if he retained any kindly recollection
of his (Sultan Mohamed’s) former friendship, he (Captain
Burnes) must not allow the Sikhs to come to terms with his
hated brother, Dost Mohamed, but thwart this object by
every means in his power. We moved from Kohat on the
24th, and returned to Peshawur, where we found Captain
Burnes preparing to start for Caubul. This journey to the
valley of Bunguish enabled me to make a considerable col-
lection of the plants which characterize the lower parts of
Afghanistan bordering on the Indus.

8. Lieutenant Mackeson had by this time received a
favourable answer from his Highness the Maharajah to his
application for permission to go to Cashmeer, and determined
to proceed on the departure of the Mission for Caubul. On
the 30th we embarked in a boat and floated down the
Lundye, or Caubul river, to Attock, where we arrived on the
dist. We halted some days to make the necessary arrange-
ments for carriage, and in expectation of the arrival of the
attendant appointed by the Maharajah to escort us, we
received a visit from Rajah Goolab Sing, of Jummoo, who
displayed not a little anxiety about the objects of our visit
to Cashmeer, and the direction in which we should move on
leaving it. Having completed our equipment, it appeared
desirable, in furtherance of the objects for which we were
deputed, to follow different routes if possible. Lieutenant
Mackeson determined to go by Drumtour and Mazufurabad,
with the design of making a map of the military road by
which the Afghans had generally entered the valley, and by
which the tratiic with Caubul was still pursued. In concur-
rence with that officer, I resolved on making an attempt to
run up along the eastern bank of the Indus as far as Der-
bend, in the first instance, and thence to be guided by cir-
cumstances, either in pushing further up if practicable along
the valley of the river, which had not been.explored at this
portion of its course, higher than Derbend, or, failing that,
to turn eastward across the mountains, through Thunaolee
to Mazufurabad, and thence rejoin Lieutenant Mackeson.
The road lay through a wild tract, occupied by fierce and
lawless tribes—at deadly enmity, and engaged in constant

VOL. I. 00

562 REPORT OF EXPEDITION

struggles with the Sikhs. But it appeared probable that if
Poynda Khan, the chief of Thanaolee, should prove favour-
ably inclined, the difficulties of the journey would be materi-
ally diminished, and the friendly reception which he had
lately given to Lieutenant Leach afforded every countenance
to the idea that he would not act otherwise. While at
Huzroo, an incident occurred exhibiting in a strong light
the barbarous kind of rule under which the country is
governed. Two Mussulmans were suspected to have mur-
dered a Hindoo muleteer. The Sikh officer at Huzroo, instead
of awarding punishment himself, made them over to the caste
of the murdered man, to be dealt with as they thought proper ;
and the unfortunate wretches were burnt alive under circum-
stances of the most aggravated torture.

9. We started in company from Attock for Huzroo, where,
having left my baggage and collections to proceed along with
Lieutenant Mackeson, I separated from that officer on the
7th September, and set out for Torbela with a few attendants,
and few articles besides paper for plants and some other
necessaries. The beautiful plain of Chuch, which appears to
have been formerly a lake of the Indus, abounds in the
richest capabilities of soil, but being the debateable ground
where the mutually embittered Hindoo and Mussulman
races meet, it is ina great measure uncultivated and in waste.
We reached Gazee about midday, but considering it unsafe to
remain there, pushed on for Torbela, which we reached in
the evening. A little beyond Gazee, the plain of Chuch
ceases, being cut off by a spur from the ‘ Ghodreh’ ridge,
above which the course of the river is entirely within moun-
tains, and its valley greatly contracted. We passed a few
scattered huts, and had some difficulty in pushing on and
escaping from the civilities which the rough-mannered Pathan
inhabitants pressed upon our small party. As we advanced,
the valley of the river became more and more confined, the
road leading along or upon its alluvial sand. The character
of the vegetation was unexpected, consisting of many of the
plants which are found in the Dhoons or Sub-Himalayan
valleys of Hindostan, such as species of Grislea, Rettlera,
Hastingia, Ehretia, Acacia, Colebrookia, &c.; and the
Dodonea Burmanniana, supposed to be peculiar in India to
the Peninsula, was seen in great abundance. As we ap-
proached Torbela, the valley gradually expanded till we
reached the broad and level tract upon which the town is
built. We were conducted to an open garden, where the
former chief, Munsoor Alli-Khan, waited on us, and showed
us every civility. The next day I had to halt, to arrange for

Se ee

a guide to conduct us to Derbend. The Torbela chief held ;

TO CASHMEER AND LITTLE TIBET. 563

back at first from supplying one, professing to be afraid of
giving offence to the Sikhs, whose hard rule he had expe-
rienced during a five years’ imprisonment at Lahore, the Sikhs
in Torbela fort having evinced by some marks of rudeness
that they were not pleased at our advent. Having gained,
by the promise of a reward, a person to accompany us, I
started on the morning of the 9th for Derbend. The path
lay along the banks of the river through a rugged and
broken tract. We saw but one small hamlet, and some Sikh
watch-towers, the cultivation having disappeared in the
struggle between the Sikhs and the Thanaolee chief. Before
noon we got to near the fort of Deyra, along a belt of level
ground which skirts the river. The Castle of Umba, a hold
of Poynda Khans, was seen on a rock overhanging the river,
on its further bank, with the Sikh fort on the eastern side.
Nearly opposite to it, in passing, we were hailed by the
Sikhs and desired to go up the hill. We found ourselves
in the midst of a large party of soldiers. They were greatly
excited, and used rude language and threats, and averring
that we had come on with an unfriendly purpose towards
them, and with the design of visiting Poynda Khan, which
they were resolved to prevent. The unlucky guide, Hussein
Alli, was seized, pinioned, and beaten, and some of the more
violent of them proposed the same treatment for the whole
party. No leader was distinguishable among them, the
band consisting ofa rabble of soldiers. After some hours’ de-
tention, and much quiet remonstrance, the guide was
liberated, and we were allowed to return without further
molestation, permission to go to Derbend, about three miles
higher up, being peremptorily refused. Two alternatives
remained, either to cross the river below Umba, and meet
Poynda Khan, or to turn across through Huzara on the route
followed by Lieutenant Mackeson. The former course had
been pressed upon me at Torbela, but I declined it, being
unwilling to go openly to a chief who was at declared enmity
with his Highness the Maharajah, through whose country
and under whose protection we were moving as favoured
guests. “After what had occurred at Deyra, this considera-
tion had additional force ; and however disinclined to forego
a journey higher up an unexplored part of the Indus, I
resolved to take the Huzara route, and rejoin Lieutenant
Mackeson. We followed the course of the river about half
way back to Torbela, when we turned eastward from the
valley along the summit of a low ridge, and about sunset
reached a village held by a family of Seyads, to whom our
guide explained the circumstances of our journey, and we
were hospitably entertained by them. It proved afterwards

oo 2

564 REPORT OF EXPEDITION

that the interruption which we had met with was owing to
secret orders to the purpose given by Maun Sing, the Sikh
Sirdar of Huzara. Recent events had created a ferment
among all the Pathan tribes then at enmity with the Sikhs,
and it did not appear to him desirable that a Huropean
should be permitted to pass through so unsettled a tract of
country at such a juncture.

10. Poynda Khan is the only chief who has been able to
maintain his ground against the Sikhs, in the Hill States be-
tween the Indus and Cashmeer, and he has, in consequence,
a great local celebrity. His father, Nawab Khan, was famous
for infesting the lower part of the Cashmeer road, which his
country of Thanaolee commanded. On the return of Mo-
hamed Azeem Khan from Cashmeer, after repelling the
first unsuccessful attack made upon the valley by the Sikhs,
he followed the Thanaolee road to Derbend, and inveigled
the chief to an interview. Nawab Khan was received with
distinction, but on a signal from the Afghan Sirdar he was
treacherously seized, and thrown, with a stone round his
neck, into the Indus. His son Poynda Khan, then a strip-
ling, who accompanied him to the interview, rushed out,
threw himself into the river, swam across, and escaped. He
soon after came into collision with the Sikhs, whom he suc-
cessfully resisted for a long time; but he was gradually
driven back, and lost the lower part of Thanaolee and the
fort of Derbend. The rest of the country he still holds,
keeping the Sikhs near him in a constant state of alarm.
Thanaolee is the highest Pathan country upon the east
bank of the Indus, the tract between it and Little Tibet
being occupied by numerous tribes of Dardohs, a distinct
race, who have a peculiar language called Dangree, and do
not speak Pushtoo.

11. On the morning of the 10th we parted from the hos-
pitable Seyads, and descended into the valley of Huzara,
where our guide was dismissed. We came upon the high road
to Cashmeer at Haripoor, whence we moved to Chumba,
and rejoined Lieutenant Mackeson there. From Chumba our
road lay through Drumtour, Mohunsa, and Saadut Khan Ke
Ghuree to Mazufurabad, where we entered the country of the
tribes called ‘ Kukas’ and ‘ Bhumbas.’ Here the province of
the Governor of Cashmeer commences, and we found an officer
with an escort waiting by his orders to receive us. From Ma-
zufurabad there are three roads leading to Cashmeer—Ist,
the eastern or Doarbid road, in the valley of the Jhelum
river, by which commerce is conducted; 2nd, the Sungur or
middle road along the ridge which separates the Kishen
Gunga from the Jhelum; 8rd, the Luchraut or western road,

TO CASHMEER AND LITTLE TIBET. 565

along the valley of the Kishen Gunga, by which Mr. Forster
travelled in 1783. Lieutenant Mackeson followed the first,
and I took the second. Mazufurabad, my route, lay by a
steep and severe ascent from the valley of the Kishen
Gunga to a small village called Patlee; thence along the
ridge of one of its shoulders to an elevated mountain, called
Peer Choonassee, richly covered with forest and alpine vege-
tation, and which yielded a large accession of specimens to
the herbarium. We passed the night a little below the
summit, and descended by a long and steep slope to Gun-
guni in the district of Punjcote. From Gunjunee the road
again touched on the Kishen Gunga, and conducted us to
the village of Teetival in Kurnah. From Teetival we passed
up through the fertile and highly cultivated valley of Kur-
nah to Tungtar. We had here arrived at the foot of the pass
which leads into Cashmeer on the morning of the 23rd. We
ascended by an easy slope through a dense forest of pines,
sycamores, and horse-chestnuts. On emerging from the upper
limit of the forest the ascent was seen to be steep for a short
distance, but not so much as to be impracticable for a
mounted horse. From the summit of the pass, which is
called Nutthoo Chunna, we had our first view of Cashmeer,
as confined to a small portion of the north-western end of
the valley. We descended and reached Ummur Ghur, in
the Pergunnah of Oontur. From Ummur Ghur I moved on
through the plain of Kamraj, passed down the river from
Sopur, and rejoined Lieutenant Mackeson at Baramula, on
September 26th. Thence we moved up towards the city of
Sopur, and thence to the city of Cashmeer, which we entered
on the 30th. By the orders of the Maharajah we were re-
ceived with every mark of consideration by the Governor,
Colonel Meean Sing.

12. Lieutenant Mackeson, soon after our arrival, received
instructions from Captain Wade to return to Peshawur with
all practicable speed, and made his arrangements for leaving
the valley immediately. It became necessary for me to decide
on the course which I shouldadopt. The season for collect-
ing plants had nearly gone by before we reached Cashmeer,
and in consequence I had been able to gather but a small
portion of the flora. There was nothing that required me to
accompany Lieutenant Mackeson to Peshawur, and were if
to return through the hills to Hindostan, the object for
which I had been sent would have been very imperfectly
accomplished. I therefore resolved to winter in Cashmeer,
and complete my collections in the spring and summer. iF
communicated my resolution to Captain Wade, with the
reasons that had led me to form it, which met with the con-

566 REPORT OF EXPEDITION
currence of that officer. The Governor of Cashmeer received
the intimation with courteous professions of satisfaction, and
there was no reason to believe that it would meet with any
marks of disapprobation from the Maharajah, after the ready
compliance with which His Highness had been pleased to
accede to our journey.

13. My proceedings during the next eight months which
I passed in Cashmeer' must be embraced here in a very
brief summary. On the departure of Lieutenant Mackeson
I was occupied in that portion of the valley which lies in the
vicinity of the city, in company with Mr. G. T. Vigne, well
known in India as an enterprising traveller. This gentleman
had returned from an attempt to penetrate to the sources of
the Noorba river, one of the two great branches of the Indus,
in which object he had been defeated by the ill-disguised
opposition of Rajah Goolab Sing of Jummoo. Before the
winter set in with severity I made an excursion to the Mut-

1 The following account of Cashmeer,
taken from a letter addressed to Captain
Cautley by Dr. F., in January 1838, will
be read with interest :

‘A glance at any map will show you
the figure of Cashmeer—a deep valley
surrounded on all sides by lofty snowy
mountains. The yalley consists of two
levels, one along the basin of the river
Jhelum, and another called Krewa, or
dry land, like the high land at Kheeri.
The river traverses the whole length of
Cashmeer, and its course is so very slow
that it looks more like a huge canal than
anything else, with countless serpentine
bends. All round the valley tributary
streams fall into the river, and here and
there are lakes, one of large size. At
intervals all along the foot of the sur-
rounding mountains there are low hills
of traps, which break out from the level
of the valley. The city is situated be-
tween two such hills, and it is so com-
pletely surrounded by lakes that it
appears to rise out of the water. These
trap hills add vastly to the beauty of
Cashmeer. Conceive this romantic situ-
ation—the lakes covered with water-
lilies, cotton (a poor sort, certainly)
growing on the banks, deodars and pines
sheeting the sides of the hills, noble
plane trees loaded with verdure, apple-
trees covered with red-cheeked English-
looking apples and as good as they look,
pear trees with pears, the most delicious
grapes, air the picture of transparency,
a heavenly temperature, and everything
looking divine. In my idea, all the

praises bestowed on Cashmeer are amply
merited.’
* * * *

‘The oppression of the Sikhs is be-
yond bounds. Nothing from the unborn
babe upwards escapes it. They accuse
the Cashmeerees of being a false and
treacherous race, and as deserving no-
thing better than Zu/m; but if they are
so, it is oppression that has made them
what they are. Excepting the Governor,
and two religious families, there is not
a native in the valley to whom you
would give a chair, so completely have
the respectable ranks disappeared. The
only Hindoos in the valley are the famous
Pundits of Cashmeer. They amount to
many thousand families, and they are
the scorpions of the valley. They en-
gage in no occupation save the collection
of the revenue. They are quartered to
an oppressive extent on every village,
and are the only servants of the Govern-
ment, to the almost utter exclusion of
the Mussulmans. They are a fine race
of men—sleek and well to do with the
world in appearance, even the meanest
of them. Their women are exceedingly
fair—almost European—and certainly
the beauties of the valley, with red cheeks
and white complexions. The Mussul-
man women have a coppery-yellow tinge
—they say from the excessive use of
tea, which is made in the Moghul fashion,
boiled up with salt, soda, and butter—
an abominable mess. The butter floats
in a thick cake when the tea cools. Tea
is universal with all ranks, and at all
hours of the day..—[Ep. |

‘tries.

TO CASHMEER AND LITTLE TIBET. 567

tayen Pass, which separates Cashmeer from Ladakh, for the
purpose of examining the rock formations to the north of the
valley.' I then returned to the city, where I housed myself
and party for the winter, during which I was occupied on the
collections which I had made, and in gathering general in-
formation regarding Cashmeer and the neighbouring coun-
Early in December I was seized with the preliminary
symptoms of illness, which from the 17th assumed an alarm-
ing form, and from that date till the middle of January I
was reduced to a state of extreme prostration, from which I
was long in recovering. arly in March I started on a
systematic tour around the valley, occupied on the spring
flora, which yielded a rich harvest. On June 3rd a con-
fidential servant arrived with a letter from Rajah Ahmud
Shah, conveying expressions of friendship and inviting me to
pay a visit to Little Tibet. I had previously, during the
winter, received a similar communication, and in anticipa-
tion of the ability to visit Iskardoh, had applied for the
Maharajah’s permission to proceed, which his Highness
readily granted. As the botany of this region, which forms
a connecting link between the Himalayahs and Hastern
Turkistan had never been explored, and the present occasion
appeared favourable for the accomplishment of so desirable
an object, I determined on taking advantage of the Rajah’s
invitation. I left plant collectors in Cashmeer, so that none
of the summer flora might be lost during my absence. There
are two great routes by which Little Tibet may be entered
from Cashmeer—the Drass road, which leads circuitously
through the western end of Ladakh, and the Kooeehamoo
road, turning north from the great lake across the Kishen
Gunga river. I followed the latter. We crossed the Kooee-
hamoo Puntsal, and descended upon Goress, on the Kishen
Gunga, in two severe marches. The transition takes
place here between the productions of Cashmeer and Tibet.
The famous ‘ Prangos’ of Moorcroft, so long and fruitlessly

sought after, was seen growing in great abundance.

1 ¢J have met with a most remarkable
volcanic tract in Cashmeer, and, so far
as my reading goes, without example
elsewhere ; a tract of alluvium with the
strata elevated at a slight angle, and
torrefied up to the surface to the
condition of a well-burnt brick; but
there is no outpouring of lava, and the
tract is very circumscribed. Thirty-
three years ago the ground was so hot
that the Hindoos of Cashmeer, simply
by digging a few inches, were enabled
to boil rice by the heat of the under

T had

strata. There must have been a layer
of incandescent fluid igneous matter un-
derneath; but strange, is it not, that it
nowhere reached the surface? ’—Letter
from Dr. F. to Captain Cautley, dated
Cashmeer, October 12, 1887.

‘Further in Cashmeer I haye got from
a mountain limestone countless remains
of Enerinites, some Corals, and some
obscure shells like Ttrebratule.’—ietter
from Dr. F. to Captain Cautley, dated
Cashmeer, January 11, 1888,—| Ep. ]

568 REPORT OF EXPEDITION

previously detected it at numerous points around Cashmeer,
but neither there nor at Goress could I find that it was col-
lected as a winter fodder for cattle. My inquiries led me to
the belief that this umbelliferous plant is only prized where
other herbage is scarce or wanting, as is the case at Drass.
A traveller advancing through the sterile regions of Ladakh
towards Cashmeer sees in the Prangos a highly valued agri-
cultural product, while another moving from Cashmeer into
Tibet finds it neglected.! Goress is inhabited by a tribe of
Dardohs, a race distinct in manners and language from the
people of Cashmeer and of Tibet. The Kishen Gunga is
here formed by two streams flowing from the eastward, the
one taking its rise on the borders of Kurtakhcheh, a province
of Little Tibet, and the other from the hills towards Him-
bass; but neither of them received any portion of the
drainage of Drass, as has been stated by one of the latest
authorities. From Goress we ran up the Kurtakhcheh branch
in an easterly direction to the village of Zeean, the last place
where cultivation. is seen, thence to Naigoon, Mukbun-i-
blangsa and Runtakhcheh. From this point, after passing
through the maidan of Noyah, we returned off from the
Kishen Gunga in a northerly direction to the foot of Boor-
zilla-bal Mountain, part of the lofty ridge which separates
the drainage of the Kishen Gunga from that of the Indus.
We crossed the pass over extensive beds of snow, and de-
scended through a steep and narrow gorge to Sookarun Mur-
bul, a bleak and rocky tract abounding in the Tibetian
Marmot, which we here saw for the first time ; thence to the
valde undulato-flexuosas, basi ab utroque
latere verrucosas expansis. Vitte plu-
rime, tenues,semen omninocireumdantes.
Semen lineare-oblongum, albumine mar-
ginibus involuto. Carpophorum bipar-

titum. Umbelle umbelluleque multira-
diate. Involucri utriusque foliola sim-

1 MS. Description of the Prangos |
pabularia, Lind., by Dr. F.

‘No detailed description of this cele- |
brated plant has yet appeared; those |
given by Wallich and Lindley haying |
been taken from imperfect dry specimens.
I had frequent opportunities of seeing |

the Prangos, in its native state, in all
stages of its growth, in Tibet and
Cashmeer. The following is the de- |
scription which I then made of it. For |
the sake of precision, I shall first give |
the characters of the fower and fruit in |
generic fashion, as if the genus consisted |
of the P. pabularia alone.
‘ Prangos Lind. Flora Medica, p. 56, |
D.C. Prod. p. 239.—Calices: margo 5- |
dentatus. Petala ovata, integra, apice
involuta. Stylopodia lata, convexa, cre-
nata, demum depressa, in fructu vix pro-
minula. Sty/é elongati, recurvi. Fruc- |
tus elliptico-oblongus, sectione trans- |
versa subteres, commissura lata. Meri- |
carpia a dorso subeompressa, suberosa, |

plicia.—Herba perennis, glabra, foliis
tripinnatim-supradecompositis, segmen-
tis linearibus sub divaricatis supra ca-
naliculatis, involucri utriusque foliolis
abbreviatis, floribus luteis polygamis :
radialibus cujusque umbellule plerum-
que androgynis, centralibus masculis:
umbellis inferioribus nune omnino steri-
libus.

‘Prangos pabularia, Lind. loc. cit,—
Hab. passim ad colles apertos Cashmeer
et Tibet. In Cashmeer floret Maio: in
valle fluminis Kishen Gunga prope
Goress florentem legi sub finem Julii.
Fructus maturescitAugusto et Septembre.

‘In lingua Tibetana “ Prangos” velrite
“ Pranhgos ;” sed proprius secund. A.C.

jugis primariis 5 crassis in alas latas | de Kirés “ Akron” nuneupatur.’—[ Ep. ]

TO CASHMEER AND LITTLE TIBET. 569

site of a deserted village called Bulleh Tseerchoo. We were
now close upon the inhabited territory of the Rajah, a village
called Dass-i-Kurum, being but two miles off. It had been
arranged by Ahmud Shah that I should descend from Boor-
zillah into Astore, a tributary state upon the Indus opposite
Gilghit, held by a nephew of his, called Jubbar Khan. With
this intention I had got to Bulleh Tseerchoo ; but we found
messengers there with intelligence that the Astore chief had
exhibited symptoms of disaffection to the Tibet Rajah, and
that he was not disposed to give the English stranger any as-
sistance in passing through his country. Soon after a letter
from Ahmud Shah apprized me that it was advisable that I
should avoid Astore and follow the route of Bearsoh. Relin-
quishing the Astore direction, we ran up an eastern arm of
the valley leading to Deosoh. This great plain, called
Bearsoh or Deosoh, constitutes one of the principal features
of the Tibetian region. Near Cashmeer, elevated 13,100 feet
above the sea, and surrounded by lofty snowy peaks, it forms
a nearly level plateau about fifty miles long and forty in
breadth, occupying the interval between the Indus and the
Kishen Gunga. High above the forest or ‘birch’ region,
its vegetation is restricted to herbaceous species, and a few
dwarf willows; but these are so abundantly produced as to
clothe it with verdure. We halted on the plain at Chukinun-
Kurroo, proceeding thence in an easterly direction to Alli
Mullik-i-Mat, where we turned north, and ran up a lateral
branch of Deosoh, which gradually contracted into a ravine,
and conducted us to the foot of the Boorgee Pass. A nearly
perpendicular ascent of 600 feet brought us to the summit,
which I found to be elevated 15,600 feet above the sea.

14. The Boorgee Pass commands an extensive view of the
mountains which stretch beyond the Indus towards Tur-
kistan, and of part of the immediate vicinity of Iskardoh.
The verdure of Cashmeer was still fresh in our recollections,
and the plain of Bearsoh, which we had just left, after jour-
neying three days along it, although destitute of trees was
sheeted over with herbaceous vegetation. The prospect
which now opened upon us was contrasted in every respect.
Looking down through a gloomy gorge, we saw a level tract
below, and countless lines of mountains running off into the
remote distances. But there were no signs of vegetation
visible; sterile sand, and naked rugged rocks, met the view
on every side. We descended the pass by a steep and most
difficult path to Kunkera Marphoh. Next day we continued
our descent, the gorge contracting as we advanced till it was
reduced to a narrow and nearly vertical cleft, across which a
military wall is built. Here, on the 8th of August, I met

570 REPORT OF EXPEDITION

Rajah Ahmud Shah, who had come a day’s journey from
Iskardoh to welcome me to his country. After a short halt,
we continued our descent, and reached the village of Khur-
peetah in the plain of Iskardoh. Next day we moved on to
the castle.

15. I remained in the neighbourhood of Iskardoh till the
14th, when I crossed the Indus with the object of pushing
north in the direction of the Mooztagh range (Kuenlun of
Humboldt), which separates the Valley of the Indus from
Chinese Tartary. My journey was readily acceded to by
Ahmud Shah. I went first to the Castle of Shiggur, on
the river of that name, a large feeder of the Indus, which
is joined by it opposite Iskardoh. I had the satisfaction
of meeting Mr. Vigne here. From Shiggur we crossed to
the western bank of the river, to a village called Bondoh ;
thence to Goma Ghondoo, Keyok, and Chookoogoh. From
Chookoogoh, we moved on to Chitrone, remarkable for its hot
spring and baths, and thence up the Basha river to Sezkoh,
and from there to Betsul, where there are considerable sulphu-
reous hot springs. From Betsul we proceeded to Arindoh, the
extreme limit of cultivation in the Basha Valley, and the
point where the river of that name bursts out from under
a great ice field! From Betsul we retraced our steps to
Sezkoh, and thence to Tongo, descending along the eastern
bank to the point where the Basha and Braldoh streams
unite to form the river of Shiggur. We now turned up the
the Braldoh Valley, which forms the route of communication
between Little Tibet and Yarkund, passing through Myeet,
Byanoh, Hotoh, and Thongul to Askolee, in as many marches.
The rugged impracticable nature of the road beggars deserip-
tion; the utmost that can be said of it is that we were able to
getalong. At Askolee, in lat. of about 35° 30’, we reached the
limit of cultivation; all beyond, on to the pass across the Mooz-

1 «JT have just returned from Iskardoh, | ney along it. Within a few hundred
where I pushed on to the Mooztagh. It | yards are fields of cultivation. The ice
is the most singular country I ever saw; | field rises to some hundreds of feet in
nothing but sand in the valleys below, | height, and is strewed on its surface
and burnt sterility above to the region | with countless millions of blocks of
of snow, under which a faint attempt at | granite, &c., and here and there, where
vegetation exists. The country is all | soil has formed, there is an attempt at
primitive, and such rocks and such moun- | cultivation, with willows, &c. A large
tains never were seen. They seem to | river comes out at a cow mouth’s open-
have been created to try the ingenuity ing, and the ice rises clear and trans-
of man in getting over them. The ice | parent above the arch. Some huge
scenery is magnificent. Above there is masses of ice tumbled down while I was
no snow saye a thin coating, but bound- | enjoying the view, with the redound of
less sheets of ice with fearful cracks and | ten thousand artillery. —Letter from
crevices. At two places an ocean of ice | Dr. I’. to Captain Cautley, dated Cash-
bas tumbled down into a valley and | mecr, October 12, 1838.
formed a mer de glace of two days’ jour-

TO CASHMEER AND LITTLE TIBET. 571

tagh, was reported to be a region of ice. It was now the 29th
of August. I found that the vegetation, both in the valley and
on the mountains above it, maintained a very uniform cha-
racter as we advanced. I could gather no information that
afforded me the least hope of finding the real Rhubarb of
commerce, which was an object of constant inquiry and
research with me during my journey; the species which I
daily saw being identical with those growing on the moun-
tains between Cashmeer and Tibet. To reach Hindostan
I had a long journey before me over several ranges of snowy
mountains, and the season when snow begins to fall was
close at hand. These reasons led me to determine on return-
ing from Askolee without attempting to proceed further
north. On the 30th we crossed the Braldoh river to Koor-
peh, whence our road led over the lofty mountains of Skora,
which we crossed at 16,200 feet above the level of the sea. We
turned down upon the valley of Shiggur, and got back to
Iskardoh on the 6th of Sept., when T communicated my in-
tention to the Rajah of starting immediately for Cashmeer.

16. Before leaving the country, I went to see the junction
of the Shayook and Ladakh rivers, regarding the position of
which, and the relative size of the streams, late authorities
are much at variance. The confluence occurs at a place
called Chundoo, a day’s journey above Iskardoh, and about
two miles above the village of Gol. The course of the river
is here almost due H. due W.; we ran up the southern bank.
The valley of the great river (called here Rgemtsoh), was
observed to be prolonged in an ENE. direction into the
Shayook or Noobra branch, which above the junction bends
round to the east. The Ladakh river comes by an abrupt
- turn from due south, through a narrow gorge, running
across the direction of the great river. Advancing as we
did, on the side from which it flows, its position was com-
pletely hid till we arrived on the brink of the stream. The
Shayook river (unknown by that name in Little Tibet,
where it is called Noobrasoh), was about 250 yards wide at
the junction, but rapid and shallow, while the Ladakh river
forms a deep and sluggish stream of hardly 120 yards. The
Tibetians consider the Ladakh branch as discharging the
greatest volume of water; but the physical configuration of
the valley would indicate the Shayook to be eastern pro-
longation of the Indus.

17. On the 11th of September, we set out on our return to
Cashmeer, and ran down the southern bank of the river to the
village of Kuchora. We were accompanied thus far by the
Rajah as a parting mark of attention on his part. It had been
my wish to avoid the route by Astore, in consequence of the

572 REPORT OF EXPEDITION

obstruction which I had met with in my journey to Iskardoh.
But Ahmud Shah had decided that I was to return by that
way, and he assured me that matters had been arranged be-
tween him and Jubbar Khan. On the 12th I parted from
Ahmud Shah, who at our last interview attacked me on poli-
tical matters, which I had studiously avoided on all previous
occasions. The purport of his conversation was to engage
the friendly services of the British Government in India to
protect him from aggression by the Sikhs on the side of
Cashmeer and Ladakh. I stated to him that I had come
into his country with no political functions, but merely as a
naturalist ; and that besides an interest in his welfare very
generally entertained by the British community in India, I
could assure him of nothing else. We separated, after mu-
tual expressions of good will, and I ran up the valley of
Kuchora to Taukhchund. Thence we proceeded along the
western branch of the valley to near Dokus, below the Pass of
Rhunnuk, which lay between us and Astore. In the evening
it commenced snowing, to the alarm of our Tibetian guides,
Rhunnuk being dangerous at all times, and often the scene
of dreadful disasters when attempted to be crossed in cloudy
weather. It cleared up in the morning, and after a severe
and most difficult ascent over extensive fields of snow, we
reached the summit in the afternoon. I found the elevation
to be 15,822 feet, considerably less than I imagined, as many
of our party were attacked with the symptoms of the distress
about the head which extreme altitude brings on. We des-
cended to the forest below, and next day followed down the
rich but uncultivated valley of Acherope to the village of that
name, in the Astore territory. We were received with sus-
picion and distrust; and matters soon took a turn which
gave me considerable anxiety. The few Tibetians sent by
Abhmud Shah to escort me were forcibly driven back; and I
was left to the chances of such treatment as the Astore chief
might be pleased to offer me. To my remonstrances it was
answered that I must go on to Astore; and that Jubbar
Khan would not allow a dog of Ahmud Shah’s to come into
his country. After a halt at Acherope, we ran down the
valley to near the castle ef Astore, and in the evening I had
an interview with the Rajahat Phenee. I tried to conciliate
him by means of such presents as I had at my disposal, and
by a display of satisfaction and confidence which I was far
from feeling. We were kept three days at Phenee, during
which time I could get no satisfactory answer as to when and
by what route I should be allowed to proceed. I was treated
civilly, but a good deal of stormy discussion went on in the
Rajah’s durbar, where a party of Gilghit soldiers, as I had

TO CASHMEER AND LITTLE TIBET. 573

reason to believe, recommended him to seize our party and
property. The inhabitants of Gilghit are notorious slave-
dealers. My impression, however, was a good deal conjec-
tural, as the conversation was carried on in the Dardoh lan-
guage, which none of our party understood. On the fourth
day the Rajah dismissed the Gilghit men, and we moved up
the Astore river to Gorekote, where we crossed it. There was
now a favourable change in his manner, and he yielded to
my urgent applications to be allowed to depart; on the 20th
we separated, and I pursued my route up to the valley to
Bosthone. I have dwelt so much on this part of my journey
as having been the only occasion during my absence from
Hindostan in which I had any cause for anxiety. I may
mention that Jubbar Khan took a fancy to have a European
dress, and that in the interview before we parted he was
equipped from top to toe in a suit of my clothes.

18. From Bosthone we continued to ascend the valley in
a SH. direction along the Astore river to Muenchah and
Durloh. The country is rich in natural capabilities for cul-
tivation, but entirely uninhabited, in consequence of long
protracted feuds between the Astore and Little Tibet Rajahs,
and frequent incursions from the side of Cashmeer. Among
other interesting plants, I found one of the Assafcetida-bearing
species of Ferula, growing in great abundance near Durloh.
Although rich in the gum resin, it is turned to no account
for the production of this valuable article. Great quantities
of the Prangos plant were also observed, but it was nowhere
collected for use. From Durloh we continued ascending by
the Kummeree valley to Ourle Surtey, and thence to Joojoo-
bassa, at the foot of the Kummeree Mountain, which we
crossed at an elevation of 13,080 feet above the sea, and
descended to the valley of the Kishen Gunga, where we got
at Zeean into the province of Cashmeer. From Zeean we
proceeded by Goress, and thence across the Kooeehamoo
Puntsal into Cashmeer, where I arrived on the 30th Septem-
ber. I made a detour round a portion of the western end of
the valley, and reached the city, preparatory to starting for
Hindostan, on the 10th October. On the 22nd I set out
upon my journey back by the main road through Rajouree
and Bimber. I crossed the Peerpungal mountain on the
26th, got to Rajouree on the 30th, and thence by Bimber,!

1 <The hills between the Punjab and | behind Nahun, and outwards as at Mo-
Cashmeer surprised me. In three | ginnund. I spent two days at Bimber
marches from the west of Peerpungal | hunting for fossils, and came upon an
you get out of the primary mountain | embedded mastodon’s head, with a cart-
formations, and then for nearly five | load of ivory, but I have only carried off
marches you pass across the Sewalik | four or five sorted specimens. They are
formation. Further up, you have it as | well known to the Mussulman popula-

574 REPORT OF EXPEDITION

and along the Punjab, through Vazeerabad and Amritsir,
and arrived in the territory at Loodianah on the 28rd
November, 1838.

19. To this brief summary of my proceedings I shall add
a few remarks regarding the reception which I met with in
the different States through which I passed, and mention
what I observed regarding the disposition evinced towards
the British in India.

20. My best acknowledgments are due for the treatment
which I experienced during the whole time that I remained in
the Maharajah Runjeet Sing’s territories. On all occasions,
where I was concerned, a strong desire was evinced in the
arrangements ordered by His Highness to mark, by the con-
sideration shown to one of its officers, his respect for the British
Government. In my tour around Cashmeer I had the unre-
strained power to go where I pleased ; and I readily received
from the Sikh authorities every assistance that I required to
facilitate my objects. I was never able to detect any indication
of that jealous concern with which the Maharajah has been
supposed to regard the visits of Huropeans to Cashmeer. If
such existed, it was studiously concealed.

21. The feelings displayed by the Sikh nation indicate no
friendly disposition towards the British in India. It would
be surprising, perhaps, if the case were otherwise. A highly
military race—rude and uninformed—who have hitherto had
a career of success, could hardly be expected to look with
favour on a power which appears to them to have stood in
their way to more extensive conquests.

22. The Mussulman tribes occupying the hill tracts along
the Punjab display a strong disposition in favour of the
British, and whenever a European goes among them he is
sure to be courted and caressed. The oppression to which
they are subjected by the Sikhs leads them to an ardent
desire for the extension of the British rule over them; and
there is a very general impression among them that it will
soon reach to the Indus. A similar belief is generally pre-
valent in Cashmeer, where the hateful misrule and tyranny
of the Sikhs have given rise to an idea that they themselves
consider their interest in the Valley to be but temporary, and
try to make the most of it while they may.

23. My observation was limited in the countries north of
Cashmeer to Little Tibet. The friendly disposition of
Rajah Ahmud Shah has been long evinced in his communi-
cations with Lieutenant-Colonel Wade, and in the kind

tion as Rakass ki har. The Sewalik | —Eztract from letter from Dr. F. to
formation, you see, is more developed in | Captain Cautley, dated Ningul, November
this direction than eastward in width.’ | 12, 1838.—[Ep.]

TO CASHMEER AND LITTLE TIBET. 575

reception which he has given to English travellers who have
visited him. The treatment which I met with was marked
by every attention, but tempered with some cautious jealousy
in preventing my intercourse with such foreign residents at
Iskardoh as could give me information about the country.
Iskardoh will always be open to Europeans while Ahmud
Shah lives ; for he is fully alive to the benefits which he has
derived from the little intercourse he has had with us, and to
the advantages of the reputation which he has, in the neigh-
bouring countries, of being an object of interest with the
British in India.

‘24. With regard to the results of my journey and the
collections formed, I must at present confine myself to a
very general statement. In botany, the collection comprises
about 3,500 species, with a very numerous series of dupli-
cates ; the total number being estimated te amount to 150,000
specimens. It consists of the plants met with in two journies
through the Punjab ; of a series of specimens from the banks
of the Indus near Attock, from Peshawur, the valley of
Bunguish, and the salt range in Afghanistan, from the moun-
tains between Attock and Cashmeer, from Cashmeer into
Little Tibet, from the Valley of the Indus as high as the
confluence of the Shayook and Ladakh River, and from the
mountains between Little Tibet and Yarkund. The plants
of Cashmeer compose the largest part of the collection,
forming a very complete flora of that valley and of the
mountains around it. There is, besides, a valuable series of
specimens collected by Dr. Lord on the banks of the Oxus,
near Koondooz.

25. The details of the collection will be communicated
hereafter as they are worked out. I may at present state
that it comprises a large number of new species; and that,
when taken in conjunction with the collections made by Drs.
Hamilton and Wallich, and Messrs. Griffiths and Royle, it
will complete a nearly connected series of the flora of the
Himalayahs from the Indus down to Assam.

26. With regard to practical results, 653 grafted plants,
comprising all the most valuable varieties of fruits found in
Cashmeer, were dispatched on three separate occasions to the
Botanical Garden at Suharunpoor, where a large portion of
them are now in a flourishing condition. The plants which
yield Assafcetida' and Koost,’ articles of considerable com-

1 See ‘ Description of the Assafcetida | genus of Composite, believed to produce
Plant of Central Asia,’ by H. Falconer, | the Costus of Dioscorides. By H. Fal-
M.D., in ‘Trans. Linnean Soe.’ vol. xx. | coner, M.D., in ‘Trans. Linnean Soe.’
p. 285, 1846.—[ Ep. ] ° yol. xix. p. 23.

2 See ‘Account of Aucklandia, a new |

576 REPORT OF EXPEDITION

mercial value, detected in course of the journey, have also been
introduced. Of seeds, 587 sorts were collected in Cashmeer
and Tibet, in 1838, sets of which have been forwarded to
different destinations in England and the Continent of
Europe.’

27. In zoology, the collection comprises 234 skins of birds,
chiefly aquatic; 30 specimens mammalia, chiefly ruminants,
among which the following are believed to be new or unde-
scribed, viz. one species of Cervus, two of Capra, one of
Moschus.?

28. In geology, specimens illustrative of the geognostic
structure of the salt range in Afghanistan, of the valley of
the Indus near Attock, and of Cashmeer and the surround-
ing mountains. Of general results in this department I may
mention the detection in Cashmeer, at the north-western end
of the valley, of proofs of a limited volcanic region of a
modern date.’

I have, &e.,

H. Fatconer, M.D.
SuHARUNPOOR: April 18, 1839.

APPENDIX TO REPORT OF EXPEDITION TO CASHMEER.

(MANUSCRIPT MEMORANDA BY DR. FALCONER).
I.—On THE CERVUS CASHMEERIANUS‘ (Fauc.) oR HANGOOL.

Characters of Male.—Horns round, with three anterior antlers,
directed to the front and curving upwards; summit terminating in a
fork, with the snags pointing right and left; broad muzzle; large sub-
orbital sinus; canines in both sexes; tail considerable; colour dark-
brown and grey; neck shaggy but not maned; insides of the thighs
white; a white edge on the buttocks and fringing the tail; no disc
on the rump ; stature large (equal to, that of Cervus Hippelaphus).

1-To botany I have made great addi- | conditions of climate, are.’—Jetter to
tions, but it is the distribution of the | Captain Cautley, Cashmecr, January 11,
vegetable kingdom, or the philosophy | 1838.—[Ep.]
of botany, that has yielded me most 2 See Appendix.—[Eb. ]
fruit. Conceive there not being an oak, 3 See page 567, note.—[Ep.]
or a rhododendron, &c., in Cashmeer! 4 Skulls of the male and female Han-
There are thousands of them on the | gool were presented to the British Mu-
ascent up to Peerpungal, but the moment | seum by Dr. Falconer. In the letter
you cross the crest, the botany becomes | acknowledging their receipt, dated Sep-
purely Tibetian, and loses its exclusive | tember 22, 1846, the species was erro-
Indian or rather Himalayan character. | neously designated Cirvus Casperianus,
Then as you advance towards Tibet, it | and this error has been perpetuated in the
becomes more and more Tartaric till it | British Museum Catalogue of Mammalia,
gradually merges into that of the. Altai | where (at p. 199) the species is desig-
range and Siberia. You can conceive | nated Cervus Casperianus, and C. Cash- -
how interesting the development of | merensis. (Falconer, MSS.)—[Eb.]
general laws of this sort, dependent on

TO CASHMEER AND LITTLE TIBET. 577

Dimensions of a Male Hangool.

(First Specimen.)

Total length from muzzle to tip of tail, 7ft.1im. Height from tip of toe to
shoulder, 4ft. 7in. Height from tip of toe to croup, 4ft.4in. Length of head
from tip of muzzle to occipital crest, 1 ft. 6in. Width of brow over orbit, 7in. In-
terval between ears, Sin. Length of ears, 9in. Length of fleshy tail, 5 in.
Length of horns along curve, 2ft. lin. Interval between inner upper snags,
1ft. 5in. Interval between outer upper snags, 2ff. 2in. Interval between tips of
broad antlers, 8in. Interval between horns at base, 2 in.

(Second Specimen.—Cashmeer, Jan. 18, 1838.)

From muzzle to occipital crest between ears, 1 ft. 4 in. Interval between orbits,
5in. Length of ears, 9in. Width of leaf of ears, 6in. Girth of head behind
eyes, 2ft. Length of fleshy tail, 6in.

(Third Specimen.—Cashmeer, April 30, 1838.)

Extreme length, 7ft.4in. Length of head to occipital crest, 1ft. 9in. In-
terval between eyes, 5in. Interval between ears, 7in. Length of ears, 9 in.
Height at shoulder from exterior of toe, 4 ft. 10in. Height at croup, 4 ft. 7 in.

(Fourth Specimen. )

Length of head from extremity of muzzle to occipital crest between pedicles,
1ft. 3in. Width of forehead between the eyes, 7in. Girth of head over the
eyes, 2ft.4in. Length of ears, 9 in. Length of tail, 6in. Length of hoof in
foreleg, 5 in.

2. Characters of Female Hangool (Pampoor, October 25, 1837).—
Sub-orbital sinus well marked, but short and deep, being about two-thirds
of an inch in both dimensions, or a little less; canines in the upper jaw;
muzzle rather small, extending to about two-thirds of height of external
nostrils, and reticularly studded with scattered stiff-pencilled hairs;
a nearly circular patch of black hair, about an inch in diameter, on the
side of the mouth, over lower jaw, at about the middle of the diasteme,
and opposite the canine.

General colour dark-leaden grey, with hairs tipped with reddish-
brown, passing along the back towards the tail into brown black, and
under the belly into light-leaden or ash-grey, and running into light
tan down the legs to the toes; ridge of black running from the occiput
along the back to the tail, towards which it increases in width and in-
tensity ; tail short, black in the middle, reddish at the tip and tufted,
and bordered with reddish-white at the sides; hind aspect of the
buttocks marked by a very defined patch of white, running up to and
terminating at the angle of junction of the tail, and passing down on the
inside of the buttocks towards the thighs; no disc of white on the
rump.

Length from extremity of tail to tip of muzzle, 6 ft. 6in. Length of head from
tip of muzzle, to occiput, between and behind ears, 1 ft. 44in. Interval between
the osseous orbits, 5in. Length of neck, 1ft.4in. Height from tip of toe to
withers, 4ft. 44in. Height from hind ditto to top of croup, 4ft. 6in. Length
of tail (fleshy part), 4 in. Circumference of chest behind shoulder, 4 ft. 2 in.
Length of ears, 7in. Interval between ears, 5in. Width of leafy portion of ears,
5in. Circumference of neck at middle, 1 ft. 11 in.

General Description.—Legs long; height greater at croup than
withers; entire build light and strong; head long; profile straight
from between the eyes downwards, but with great gibbosity in the
brow between the line of the eyes and ears; nasal arch well defined,
and side of the face fined off under it and expanding from the angle of
the mouth to the tip of muzzle; muzzle broad; naked portion rather

VOL. I. PP.

578 REPORT OF EXPEDITION

small, and with the inter-nasal patch of hair running down into an
angle on it; sub-orbital sinus short, its direction deflected slightly
downwards from the line of the long axis of the eye.

Fie. 15. Fy

CERVUS CASHMEERIANUS. FROM A DRAWING EXECUTED FOR DR. FALCONER BY
G. T. VIGNE, ESQ., AT CASHMEER, OCTOBER, 1837. .

Colour of head dark-ash or leaden-grey, braided with reddish-brown ;
tan-brown on bridge of nose, between the sub-orbital sinuses, passing
into, and mottled with, light-reddish brown on the frontal gibbosity and

Fic. 16.

ee

SSM ue

CERVUS CASHMEERIANUS. FROM A DRAWING EXECUTED FOR DR. FALCONER BY
*

; G. T, VIGNE, ESQ., AT CASHMEER, JUNE 7, 1838. *%

between the ears, but with a crescentic patch of black to the rear of
the gibbosity, with the horns of the crescent downwards; ears long and
acuminated, white at the base anteriorly, greyish white and woolly
inside, and blackish-tan and rather naked outside.

TO CASHMEER AND LITTLE TIBET. 579

Inferior surface of the head and face, inferior half of the neck, inter-
scapular, pectoral, and ventral regions, leaden or ash-grey, with the
hairs discoloured at the tips, the same colour running down on the inside
of the thighs and fore legs.

Upper half of the sides of the neck and flanks of a reddish-brown, as
on the brow.

Scapular and femoral regions of a dark-leaden hue, mixed from the
discolouration of tips of the hairs, and passing down the outer aspect of
the legs into tan-brown.

A ridge of brownish-black, commencing between the ears and running
down the crest of the neck along the back to the tip of the tail, gradu-
ally spreading out behind the shoulders so as to form the predominat-
ing colour of the back and the entire colour of the rump and lumbar
region generally, and running gradually into the other colours where
the different regions meet; the tail black in the middle, and bordered
on the sides by reddish-white, a very defined patch of white descending
from the angle of the tail down on the rear aspect of the buttocks, so
as to give a border of white when the animal is viewed laterally ; tail
broad, and of coarse longish tufted hair, but very short. Mamme
4, inguinal; no inguinal glands; length of toes, 4 inches; circumference
of ditto, 8 inches; succentorials large and well defined.

II.—Norte on A vaRiETY OF Musk-Deerr, or Kustora.

[Mr. Ogilby, in his ‘ Mammalogy of the Himalayahs,’ published in
Royle’s ‘ Illustrations of the Botany of the Himalayahs,’ writes as fol-
lows: ‘ Whether the second species of Musk mentioned by Dr. Falconer
as having been discoverd during his recent journey into Cashmeer and
Little Tibet be founded on more important characters, we have no
means of ascertaining, that naturalist not having yet published his
observations.’—Ep. ]

Male Kustora, killed near Thaneon or Chundunwaree, Valley of Duchen-
parah, May 19, 1888.

Extreme length from muzzle to tip of tail, 3 ft. 2in. Length of tail, 2in. Length
of head, 3in. Length of ears, 6in. Interval between ears, lin. Interval be-
tween eyes, 2in. Girth, at middle of sternum, 2 in. Height at shoulder, when
extended on side from tip of toe, 2ft. lin. Height at croup, 2ft.9in. Length
of toe of foreleg, 14 in. Length of succentorial toe of foreleg, 1} in. Length of
canines, 24 in.

This specimen, an aged male, differs very remarkably from the speci-
mens I have hitherto seen in colour, size, magnitude of musk-bag, and
head.!

General colour of neck, back, sides, and whole body, a very light ochrey
buff, like the Indian jackal, with a tinge of red, and some very faint wavy
traces of a darker colour over the shoulder and towards the hips. Head
of a reddish brown, particularly above the eyes to between the ears and
for a slight way above the opening of the nostrils, brown mixed with
grey from thence to between the eyes, and a rufous (or orange-red)
patch under the anterior angle of the eyes, extending to along half the
opening of the eyelids and crescent-shaped. The same colour or. the

~-

1 «Tt is certainly a new species, if not ter from Dr. F.to Captain Cautley, dated
included in Hodgson’s catalogue’—Let- | Cashmeer, March 20, 1838.—[ Ep. ]

PP 2

580 REPORT OF EXPEDITION

canine flap at the angle of the mouth. Outside of the ears, rufous dark
brown slightly tipped with grey; inside of ditto, at base, dirty white ;
upwards on the leaf, grey, which also constitutes the margins. Out-
side breast, rufous ochrey tipped with grey, passing, on the underside
of belly, breast, and down the legs, into slate-blue, mixed with grey ;
legs pepper-and-salt grey, with some admixture of rufous on front;
toes, dirty greyish white.

Another remarkable character is a crescentic or semicircular flap
(canine flap) descending from skin of upper lip over the canine, so as
to reach the edge of the chin. This flap of skin is not observed in other
specimens; no bristly hairs at angle of mouth, nor white patch, as in
the common Cashmeer variety.

This individual, an aged male, with canines 21 inches long, is much
smaller than the common dark variety. Musk-bag much larger; all
the Shikarees who have seen it declare the specimen to be rare, and
only found in the neighbourhood of Tibet, where it was killed, and
where we now are, and the peculiarities are so marked that I consider
it a distinct species.

The specimen is moulting; the hair on the brow coming off in
patches.

TIIl.—Te Goats or CasHMEeEr.

[In a letter from Cashmeer in 1838, to Captain Cautley, Dr. Falconer
remarks:—‘ There are no fewer than three species of wild goat near
Cashmeer, the Ibex of the Snowy Range the same as I got from
Everest, the Tehr (Capra quadrimammis), and the Markhor,! a noble
goat, with horns twisted exactly like the blade of a sword, and as long
and diverging as those of the antelope. I went after the Ibex for five
days, and saw numbers, but only got a young male.’

The following is a description, by Dr. Falconer, of the Rev. R.
Everest’s specimen of the Ibex, taken from a letter written to Mr.
Hodgson, about 1835, but probably never sent.—Ep. |

I have never noticed in any of the synopses published in the Journal
of the Asiatic Society or elsewhere, by you, any mention of the Ibex
of the Snowy Range. Does it occur in the Nepaul snow frontier, or is
it included in the drawings of the illustrations of Nepaul zoology ?
Your series is so complete otherwise, that it were a pity if the Ibex
should be wanting. The Rev. Mr. Everest has just returned from a
trip nearly as far as the Chinese frontier, on the Spiti River in Upper
Kunaor, and brought back with him two specimens of the heads of the
male, and an entire stuffed skin, from all which a very tolerable idea of
the animal could be made out. It abounds in the snowy ridges of Kunaor,
along with the Bhurroor; but it is an infinitely larger animal, and I
imagine by far the largest caprine species known. I suspect also it is
specifically distinct from the Ibex of the Alps, and very markedly so ;
but I have no description of the latter at hand to refer to. The intra-
orbital portion of the forehead is more concave, and the nasal portion
of the chaffron more convex and prominent than in any ruminant I
have ever seen described. The stuffed specimen is now in my possession,
and here are some of the dimensions.

Extreme length of shrivelled skin from extremity of the tail to tip of muzzle,
7ft.4in. From tip of toe to the croup, 3 ft. 9in. From ditto to withers, 3 ft. 9 in.

! This is the Capra Falconeri of Hugel ; the Aigoceros Falconeri of Wagner.—[Eb. |

TO CASHMEER AND LITTLE TIBET. 581

Length of tail (exclusive of hair), 44in. Length of horns along the curve, 42 in.
Length of chord, 243in. Versed sine (from chord to upper curvature), 11} in.
Divergence at the tips, 29in. Greatest divergence, 30in. Interval between the
horns at their base, lin. Circumference at the base, 10in. Antero-posterior dia-
meter at the base, 385in. Transverse diameter at the base, 2in. Length of
hoof, 4 in.

The animal is singularly party-coloured. Horns start from the
head, on the same plane with the brow, and then gradually curve up-
wards, backwards, and outwards, with an increase of curvature and
sudden intlexion near the tips, and with eighteen transverse distant
ridges (incomplete rings) along the upper curve.

Muzzle hairy, tan-brown passing upwards into light hazel-brown on
the forehead and behind the ears. Chin, cheeks, under surface and
sides of the neck, shoulder, and hips, liver-brown, with the hair tipped
greyish, gradually passing down the fore and hind legs into deep reddish
brown. A large roundish patch of dirty greyish white on the upper
side of the neck, extending from behind the occipital to the top of
the shoulders. Shoulder patches meeting at the croup and sending off
a narrow ridge of long, tufted, reddish-brown hair, along the spine of
the back to the tail. Back, sides, and flanks, entirely of a dirty greyish
white, passing downwards into hazel brown. Under side of the chest
and belly, dark slate-grey, mottled and mixed with tan-brown. Tail
short and black, 44 inches long; ears erect, hazel-brown outside, tan
inside. Mamme four. No trace of sub-orbitary sinus. Beard full, dark
liver-brown, tipped with reddish-brown. Hair very thick, erect, stiff,
and rather short, with a very thick coat of shawl-fleece at the roots.

IV.—Nore on THE Mute Breeps BETWEEN THE YAK AND THE Cow,
IN TIBET.

For agricultural purposes the Tibetians find a cross between the
Yak (Bos grunniens) and the common Cow of the mountains to be more
serviceable than either of the pure stocks—for the labour of the plough
especially. The crosses are chiefly made with the male Yak as the sire,
and the common Cow for the dam. The mule produced in this instance
is called: the male, ‘ Bsoh,’ the female, ‘ Bsohn.’ The ‘ Bsohn’ is
then again crossed with the male Yak, and the produce called : the male,
‘ Gtir, and the female, ‘ Gurmoh.’ The Gurmoh is again crossed with
the Yak and has a produce which returns to the Yak ;! the male of this
breed is fertile and capable of propagation, the males of all the inter-
mediate breeds being sterile.

The Tibetians do not practice raising a cross between the Cow Yak
and the common Bull. The reason given is that the experience of the
country, as derived traditionally from their fathers, is against it, and
that where the cross has accidentally taken place the produce is in-
ferior.

The Yak, as compared with common Oxen, is an unstable, uncertain-
tempered animal, impatient of long fatigue, and does not take kindly
to the plough.

The ‘ Bsoh’? is a larger animal than the Yak, and more robust.

1 Tn stating the cross to return to the 2 The mules which were seen of the
Yak, it is merely meant to convey the im- | Bsohn were larger than the generality
pression of the people, probably derived | of the common cattle observed in the
from the fact of fertility in the third cross. | country.

582 REPORT OF EXPEDITION

The Gurmoh is smaller than the Bsoh; each of these crosses takes after
the sire in the Bisonic form of the head, but has shorter hair, and the
characteristic chowry tail much less bushy. The Bsoh shows some-
thing of the dewlap, and has longer limbs and is of less compact build
than the Ox. The temper improves, and the eye, which is fiery and
restless in the Yak, becomes quieter.

One reason of the breed is perhaps that the cross affords an animal
more tolerant of heat on the one hand than the Yak, which suffers ex-
tremely at a low altitude in summer, and on the other hand hardier than
the common Cow, and able to sustain the severest cold, partaking thus
of the good qualities of both originals. 'The Yak browses freely on
species of Artemisia, which are abundant on the dry hills of Tibet, form-
ing in many instances the only vegetable coat which the soil has.

The Yak is chiefly used as a beast of burthen, and in the upper parts
of Tibet it is ridden, but travels at a very slow pace. The mules are
kept-for draught or the plough, and the common cattle for their milk.

The above particulars are from notes taken at the time in the country.

V.—Tue CasHmMerr Martin.

3—3 1—1
Mustela.—Sub-genus Martes.—Incis. 33) canine Toe cheek
io os 1—1 —
teeth 6—6 (viz. false mol. 44 camass. 77) tubercular (=) =.=

=58.

Martes Macroura (Falc.), Long-tailed Cashmeer Martin —Head, face,
and back of neck, blackish brown ; lower jaw and a salient cheek patch
under the ear, dirty white ; throat, sides of neck and breast, yellowish ;
body dirty yellow, graduating backwards into dirty brown; crupper,
flanks, hind legs, and lower portion of forearms, dark sepia, passing
downwards into blackish brown: ears blunt, concolorous with the
head ; tail black ; body vermiform. Length of head and body 25 inches;
tail 17 inches (fleshy portion 16).

Described from a full grown female, March 28, 1838.

VI.—Tue Trpetian Hare.

Extreme length from tip of fleshy tail to muzzle, 1ft. 1lin. Length of fleshy
tail, 4in. Length of hairy tail, 5 in. Length of ears, 5in. Width of leaf of
ears, 3 in.

General colour mixed grey, rather light, with a lead-bluish tinge on
buttocks, and slightly rusty on neck; head greyish brown; ears long,
tipped on the outside with black; tail very long, black above, white
underneath and at sides; breast, belly, and inner side of the legs white ;
mamme, 4, ventral. It is a rusty red, long-legged, lanky hare, weighing
about 4 Ibs.!

" Dr. Hooker, in his account of the | hares with white rumps (Lepus giosto-
Cholamoo Lakes in Tibet, observes: lus.)’—Himalayan Journals, ii. 157.
“There were also many slate-coloured

TO CASHMEER AND LITTLE TIBET. 583

TiseTiAN Marwmot.!

(Tibetian ‘Pfyah’; Dardoh ‘ Tréoshion’; Cashmeer ‘ Drin.
Tibetian Marmot of Moorcroft).

Characters.—General colour rufous yellow; brownish black on the
muzzle, crown of the head, back, tip and under side of the tail ; throat,
under part of the body, and legs of a bright rusty red; ears incon-
spicuous, short, thin, and pointed; tail half the length of the body, grey
above, flattened. Upwards of three feet long ; tail 13 inches.

Inhabits the elevated plains and valleys of Tibet, Muttayen near
Drass, Deosoh, and Sookarun Murbul, between Boorzillah and Bulleh
Teerchoo.

Dimensions of a full grown female :—

Extreme length from tip of tail to point of the muzzle, 38 in. Length of fleshy

portion of tail, 11}in. Total length of tail, 13 in. Length of hind paw, 4 in.
Girth at the belly, 22 in.

Vil.—Arctomys TIBETANA.

The Tibetian Marmot belongs to the section of the true or social
Marmots, ‘ Arctomys,’ of F. Cuvier, as distinguished from ‘ Spermo-
philus’ or the solitary Marmots. It is perhaps the largest species of the
genus. The general colour of the body is a reddish yellow, with
brownish black from the back of the head along the back, passing on
the rump into yellowish grey. The sides are of a light yellow, the
underside of the neck, breast, belly, and the legs of a bright rusty red ;
the muzzle, chin, angles of the mouth, and a patch around the eyes are
black; forehead reddish yellow, mixed with black ; tail yellowish grey
above from the rump, brownish black at the extremity, and mostly so
underneath.

The body, as is characteristic in the genus, is bulky, loose, and
heavy, the head large, the muzzle short and thick. The ears are hardly
apparent till the head is closely looked at, and are pressed to its sides,
hairy on both sides, obliquely sloped off behind, and rather pointed.
There are no cheek-pouches, but a broad fold of loose skin, with long
erect hair, is continued from the sides of the neck under the ears and
eyes to the angle of the mouth, giving the appearance of a sort of ruff to
the face, and confounding the head, neck, and body together. The
whiskers scanty, and the upper bristles only of considerable length ;
the teeth are, as is normal in the genus, two incisors above and below, and
five cheek-teeth above and four below, on each side. The incisors are
thick and stout, rounded in front, and compressed on the sides. The upper
cheek-teeth have two transverse sharp ridges meeting on the inner side
in a connecting tubercle with a subordinate ridge on the anterior and

posterior sides—the anterior tooth

1 Mr. Ogilby, in Royle’s ‘ Illustrations
of the Botany of the Himalayahs’ (1839)
refers to Dr. Falconer having mentioned
the Tibet Marmot in his report of his
journey to Cashmeer, but adds that
there was no further knowledge of its
~ characters (i. 67). This Marmot was,
however, described and figured by Mr.
B. H. Hodgson in the Journ. As. Soc.

for 1841 (vol. x. p. 777), as Arctomys |

consisting of a simple projecting

Himalayanus. Two years later (vol. xii.
p. 409), he described another smaller
species of Marmot as inhabiting the
Himalayahs, and suggested that the
term A. Tibetensis would be more ap-
propriate to the larger species. The
Tibetian Marmot is also figured by Dr.
Hooker, in his ‘Himalayan Journals,
ii. 93.—[Ep.]

584 REPORT OF EXPEDITION

tubercle, surrounded on the inner side by a sweep of enamel. The
lower cheek-teeth have four tubercles at the angles, connected by longi-
tudinal and transverse low ridges. The legs are short and wrapped in
loose skin; the paws large and strong, furnished below with a thick
elastic pad under the skin; the toes long and distinct to the second
joint, the forepaws with four toes and a knobbed rudiment of a thumb,
the hind paws 4 inches long, with five toes; nails stout, compressed, and
slightly curved. The tail long, equalling half the length of the head
and body, the fleshy part being 114 inches, flattened and furnished
with long dense hair. The fur on the back and sides is long, close, and
rather coarse, but not harsh; very scanty underneath on the chest and
belly. The ruff on the sides of the head composed of erect stiff bushy
hair. The eyes are large and black.

This Marmot forms an entirely distinct species, characterized by its
large size, brilliant colour, and long tail. It was first observed by Mr.
Moorcroft in his journey from Tibet to Cashmeer, and described by
him as resembling a young hare.!_ The animal has in consequence been
conjectured to be a Lagomys. To me there is nothing in its appear-
ance or habits that conveys the idea of a young hare. I learned in
Cashmeer, from a person who accompanied him, that M. Jacquemont
shot numerous specimens at Muttayen, between Drass in Tibet and
Sir-i-cottul, the mountain pass which separates Cashmeer from Tibet.

The animal is entirely confined to the Tibetian region, and chiefly
to the northern face of the Himalayahs. Mr. Hodgson, in his Cata-
logue of the Nepaulese Zoology inserts an Arctomys Himalayanus ;? but
as the notice is unaccompanied by any details, I know not whether his
animal is the same with the Tibetian Marmot, or whether it was found
on the northern or on the southern side of the Himalayahs. But cer-
tainly neither this nor any other species of Marmot exists on the
southern front of the range from the river Gogra to the Indus. Nor
does it appear to be found in Kunaor, which partakes of the climate
and productions of Tibet, and has a language in common with it.
Excellent observers, such as the Rev. Mr. Everest, Captain. Her-
bert, and many others, have ascended high up the valley of the Spiti
without having noticed it. In the Tibetian regions it is found all
along from near the lake Manasarovara to the banks of the Indus, where
it first bends to the south. In Cashmeer I was informed of its existence
towards the heads of the valleys of Duchenparah and Shalimar, which
are couterminous with Ladakh, but nowhere along the Puntsal or
snowy range forming the mountain boundary of Cashmeer on the side
of Hindostan.

The Drin is a beautiful animal, and interesting both in its habits and
on account of the dreary regions which it inhabits. There, on the con-
fines of eternal snow, it is the almost solitary type of animated existence,
I first observed it on descending the northern side of the Boorzillah range,
which intervenes between Little Tibet and the valley of the Kishen
Gunga, early in the month of August, but 300 or 400 feet below the
belt of perpetual snow, and high above the region of the birch, the
rhododendron, or juniper. This is the favourite habitat of the animal,
where but a few grasses, the Tibetian Rumex, Polygona, Pediculares,

? Royle, Illustrations of Botany of | on wrapper of Journal of the Asiatie
Himalayahs, 1839, Introd. i. p. 37. Society, No. 65, for May 1837. (See
? Catalogue of the Nepaulese Zoology | p. 583, note 2.—Ep.)

TO CASHMEER AND LITTLE TIBET. 585

and Primroses have their upper limit ; and it may be said that its
Alpine propensities are only restricted by the absence of vegetation
and the incapability of the soil for being burrowed in. It inhabits
alike the elevated plateau of Tibet, or the heads of its most alpine
valleys. It is eminently social in its habits. The traveller, after crossing
a lofty ridge and traversing a snow field, descends into a bleak tract
where there is nothing to remind him of an animated creation besides the
companions or attendants who may be with him, and suddenly sees
below him a large community of animals which look golden-coloured in
the distance, young and old collected together within a small space.
On the top of a large stone or rock is seen a prominent figure on the
watch, who, by the vivacity of his movements and the intentness of his
gaze, shows the interest he takes in the approach of the stranger. He
raises himself erect on his hind legs, turns one side of the head and
then the other in quick succession towards the object of his suspicion, and,
on assuring himself of its nature, utters a loud clear and shrill ery; the
other members of the community instantly take the alarm, the young
make at once for the burrows, and the old endeavour to get a sight of
the object of alarm, and if the nature of the ground prevents this, they
also betake themselves to the burrows. The sentinel meanwhile keeps
quickly repeating his cry, and maintains his ground till the advance of
the stranger to within a few yards endangers his personal safety, and he
rapidly retreats. After the lapse of a few minutes a head is seen
cautiously protruded out of one of the holes, making a rapid circuit of
inspection. If any cause of alarm is observed it is speedily withdrawn,
and no cry uttered; if nothing is seen the body comes out with extreme
caution, and if the stranger has gone on he is soon apprized by a shrill
ery from behind him that the sentinel has again taken to the post where
he first saw him, and the community is informed that the object of
alarm has passed by, and the members again issue forth to bask in the
sun. The station for the burrows is usually near the base of the
sloping ridge of a valley, or, ifthe ground is level, the most elevated
part of it is selected, so as to guard against flooding from melted snow
or rain. With regard to soil the animal appears to be indifferent. At
Sookarun Murbul, below Boorzillah, the ground is entirely composed
of fragments of granite, covered by a scanty stratum of clayey. soil.
Here the burrows were numerous, and their excavation must have been
attended with difficulties which would have defeated a less skilful
pioneer. The excavated matter at the mouth of the burrow con-
sisted entirely of granite, pebbles, and gravel, and its direction must
have been devious to a degree, from the intervening blocks of granite
met with in the course of digging. The burrow at its mouth goes off
downwards at a considerable angle. Ifa raised block of stone happens
to be near, the excavated materials are thrown up into a careless
mound. But if not, the ejected gravel is dressed up into an elevated
broad glacis, which slopes rapidly into the mouth of the burrow. This
spot is the favourite station of the Drin, where he may be observed
basking for hours in the sun. Near the same place I saw several indi-
viduals perched on blocks of stone forming the talus of a granite cliff,
and uttering their shrill ery on my approach. Here the habitation of
the Drin appeared to be merely the casual crevices left between the
blocks of stone. The burrows, where the ground admits, are close
together, and as much so as in the most crowded rabbit warren in

586 EXPEDITION TO CASHMEER AND LITTLE TIBET.

England. The slope at the mouth is tidy and clean, and filth or drop-
pings are rarely seen near it. Judging from the number of burrows
on one spot, and the individuals seen about them, a community varies
from twenty to fifty members, young and old.

The Tibetian Marmot is a diurnal animal; from sunrise to sunset
young and old may be seen about the warren, gnawing plants or bask-
ing in the sun. During the day it never appears to retreat to its
burrow, except on the approach of a suspicious object. It is rarely
seen more than a few yards off from the warren, and if suddenly come
upon at a greater distance it is easily outrun and caught by the hand.
But this seldom happens. Although so abundant where its station is,
a specimen is not easily got. On the approach of man they retreat to
the mouth of the burrow, and atter receiving a ball or a charge of shot,
if a spark of life remains, they wriggle into their den and slide down
out of reach. The cry is peculiarly clear, loud, and shrill, and it some-
what resembles the sound of ‘chééoo’ raised to an octave. It has also
something ventriloquial about it. When the eye does not guide the
ear to where it is uttered, one looks a few yards ahead, expecting the
animal to be close under his feet, while the little trumpeter is soon
detected seated perhaps a hundred yards off. On uttering the ery the
Drin stretches out its neck, and bends the head down like a cock at the
finishing note of his crow, and the movement is such as to convey the
idea that the animal is about to take to flight.

The Drin exists im vast numbers in the elevated tracts of Tibet; it
has no enemy in man, nor any predaceous foe to control its prolific
production. The Tibetians do not eat its flesh, nor do they hunt it
for its fur, although in other countries the latter would be an article of
commerce. Its natural enemies, such as the eagle and the Mustelide,
are nowhere seen. ‘The latter could not exist the whole year in the
region of the Drin, and were they to invade it in the summer months
they would first have to pass through an intermediate lower belt,
where they would find nothing to feed upon; and this circumstance
is perhaps the security of the Marmot. Were the Mustelide to reach
so high, their production, from the vast abundance of food, would soon
be so great that numerous traces of them would be seen about the
‘haunts of the Drin, and the latter would soon be suppressed in number.
But nothing of either sort is observed. The only controlling agency
upon the excessive production of the Marmot is probably the ava-
lanche, or an unusually heavy fall of snow during the winter, either of
which might extend the period of hybernation to a whole community
longer than the’powers of life could sustain.

NOTES ON INDIAN ZOOLOGY. 587

XXIX. MISCELLANEOUS NOTES ON INDIAN
ZOOLOGY.

BY H. FALCONER, M.D.

I.—Nore on Cervus DuvanceLtu. (Cuv.)

Tur deer and antelopes of India are still but imperfectly known, and
much doubt exists about the number and character of the species. Much
has been done by Indian naturalists lately, but a great deal remains to
be accomplished. The number of species in the group of the genus
Cervus, called Rusa by Hamilton Smith, is by no means settled, and
the characters of the species are ill-defined. Mr. Hodgson has thrown
much light on the Rusine deer of Nepaul, and the tract of Terai below
it, as well as on every point connected with the zoology of the Hima-
layahs to which he has given his attention. I entirely agree with him
that the characters derived from the horns, as restricted to the sub-
genus Rusa by Major H. Smith, are insufficient. The Kalo Jarai figured
by Mr. Hodgson in the 1st volume of the Journal As. Soc. (p. 115) has
no terminal subdivision of the beam, and no branch from the horn but
the brow antler. The animal now to be noticed forms another exception
from the group, in having the beam divided into four subterminal
branches, but in other respects possessing all the characters of the Rusa,
I believe, in so far as could be determined from the fresh head.

Subgenus Rusa.
Cervus Eucladoceros. (Nob.)}

Stature very large; horns round, slender for the size, roughly ridged
and furrowed; basal antler anterior, long, thrown off at a large angle ;
beam reclining, divergent, quadrifurcate, giving off above the middle
an internal bifurcate branch with snags of unequal length, and termi-
nating in a fork with nearly equal processes; sub-orbital sinus large ;
muzzle broad; canines.

Head large ; ears large, lined with long whitish hair. Horns massive
near the burr, 7 inches in circumference, 2} inches apart on the brow.
Bez antler low, long, anterior and erect, slender, slightly recurved, roughly
ridged to about three-quarters of the length, and tapering into a very
smooth and sharp ivory white point, thrown off at an angle of above

1 This note was written in 1835 or | Dr.F.’s writing, Cervus Duvancellii. The
1836, but was never published. In a] species appears to be the Cervus Kla- .
drawing of the specimen described, the | photdes, or the Bahraiyaof Hodgson. (See
species was subsequently designated in | Cat. Mam. in Brit. Mus? p. 204.)—[ Ep. ]

588 NOTES ON INDIAN ZOOLOGY.

80° from the beam with which it forms a broad-ridged or mammillated
fork. Beam reclining, slightly curved forwards, divergent, round,
_bluntly ridged, slender for the size, 55 inches in girth, divided above
the middle into four prongs, viz. a branch thrown off inwards and
backwards, bifurcated laterally ; the inner snag very short, the outer
three times longer ; the beam then rising some distance, and terminat-
ing in a perch ; the processes nearly equal, anterior and posterior; the
anterior smooth, pointed and slender, like the bez antler; the anterior
incurved, forming the summit of the beam, the posterior nearly erect.

There is some slight difference between the horns of the opposite
sides. The above description is of the left one; on the right side,
where the subdivision of the branches commences, the beam is flattened,
and the offset of the branches approaches palmation. In other respects
the horns are alike. A plane on the anterior surface of the terminal
prongs would form a curve.

Length of the head from base of the horns to tip of the muzzle, 1 ft. 2 in. ; ditto
from the occipital ridge to ditto, 1 ft. 8in. Width from tip to tip of the ears 2 ft.
Length of the horns along the curve, 2 ft. 11 in.; ditto, straight line along the
chord, 2 ft. 5 in.; ditto of the intercornual axis, 2 ft. 1} in. Interval between
the horns on the brow, 24in. Divergence of the tips, 2 ft.5in. Greatest interval
between the horns, 2 ft. 8} in. Length of the brow antler along the curve, 1 ft.
45 in.; ditto straight from perch to tip, 1 ft. 2} in. Offset of ditto from the burr,
24 in. Interval between the points of ditto, 1 ft. 6 in.; ditto, from ditto to the
middle of the beam, 1 ft. 4 in. Length from buir to lower terminal fork on the
beam, 1 ft. 5 in.; ditto from the lower to the upper terminal fork, 64 in.; ditto of
the anterior terminal branch process, 1 ft. 11 in.; ditto, of the posterior ditto,
10in.; ditto of the outer snag of the lower fork, 6 in.; ditto of the inner snag
ditto, 24 in. Divergence between the tips of ditto, 1 ft. 3 in.

I had no means of ascertaining the dimensions of the body, as only
the fresh head was sent to me. Mr. Money, C.S., from whom I received
it, described the animal as large. He was one of a party of three
who shot twenty-one of the deer in one day, large and small. The deer
had a long shaggy mane on the neck, and long hair on the throat. The
colour of the head was a dusky greyish brown, but tawny above the
muzzle; that of the body was described as of a dark brown. It was
shot in a jhil on the west bank of the Ganges, south of Hurdwar. It
may be considered as inhabiting the Sub-Himalayan terai.

I apprehend there can be little or no doubt that this deer belongs to the
Rusa group. [have anearly full-grown female Jurao, and a young male
that has not yet thrown out horns. Along with the head I have described
I received the head of a young deer, with the sample horns of the first
year’s shoot, shot at the same time with the large one. The heads
respectively exactly resembled the living young male and the adult
female; and the expression of the heads was precisely alike in both.
The heavy mane on the neck, the rise and sigmoid flexure of the sub-
orbital sinuses, and the large muzzle, added to the other points of
similarity, leave no doubt that the species is a Rusa.

I am also ‘of opinion that the species is distinct from the animals
described by Mr. Hodgson, the Kalo, Rato, and Phusro Jarais of
Nepaul, and the Rusas of Major H. Smith. If any importance
attaches to characters derived from the horns, it cannot be regarded as

_avariety. The slender form of the horns, the great length and large
angle at which the bez antler is thrown off, and the four-pronged ter-
mination of the beam, although either might have existed singly in a

NOTES ON INDIAN ZOOLOGY. 589

variety, in conjunction must be regarded as indicating a distinct
species. Perhaps the Rusas might be conveniently distributed into
three sections :—

I. Those with brow antler and simple beam: Kalo Jarai, Hodgson.

II. Those with bifurcated horns. The Rusas of Hamilton Smith.

III. Those with slender horns and four-pronged beam, including the
animal above described.

I am not aware of any native name for this deer. On this point
there is much confusion, which has added greatly to the difficulty of
discriminating Indian deer. The Saumur of the lower provinces is
called the Maha in the upper part of the Doab; and these names
appear to be applied to more than one species. The Rusas of Nepaul are
called Jarai; in Kumaon, and the Jumna-Gangetic tract of the
Himalayahs, they are called Juraos. What is the Bara Singha?
Hamilton Smith says it is the C. Hippelaphus; and Mr. Hodgson says
it is a splendid variety of the common stag, or C. Hlaphus. Should
this notice reach the eye of Mr. Hodgson, I would suggest his recon-
sidering this opinion, or favouring Indian naturalists with his reasons
for thinking so. In the upper part of the Doab, Bara Singha is applied
to any large deer by the natives, particularly to such as have any
remarkable development of the antler. Many sportsmen say it is a
deer with six branches on each horn; but I have never met with any-
one who could specifically say what the deer was. It would be well
if the term Bara Singha were settled, and applied specially to some one
of the large deer.

Il.— Nore on An Inpran Species ofr Hsox.

Esox (Belone) Hindostanicus (Falc.).

Malacopterygiens abdominaux, Cuv.; Gen. Esox.—Muzzle long and
attenuated, with a large gape; upper jaw with its sides composed of the
intermaxillaries, which have a posterior pedicle, the maxillaries being
concealed under the tips and not entering into the composition of the
jaw; teeth in both jaws, some of them long, pointed, and distant, the
intervals occupied by a row of smaller teeth; no teeth in the palate.
Lower jaw projecting beyond the upper; no adipose fin or karbille ;
opereula and branchial openings large; body and tail elongated, the
latter compressed at thé sides; scales cycloidal, small, hard; dorsal and
anal fins short ; one dorsal fin placed right above the anal.

Subgen. Belone, Cuv.—Dorsal fin with 16 rays, anal fin 18 rayed ;
branchial membrane 14 rayed; caudal fin, with a nearly rectilineal
termination (no échancrure), 19 rayed; head rather small; lower jaw
extending beyond the upper, both being straight, and twice as long as
the head proper; body and tail cylindrical and serpentiform; two
longitudinal ridges, with an intermediate ridge, and bounding furrows
along the back; a mesial lateral line from the tail to the opercula, and
a longitudinal coronated line of scales on either side towards the belly.
Pectoral fins 9 rayed; ventral fins 6 rayed.

The muzzle long and needle-shaped; total length of body and
muzzle about 84 inches; diameter of the body half an inch; the teeth
are of two sorts—one very long and needle-shaped, rather distant, the
other shorter and occupying the intervals; eyes very large; head flattened

590 NOTES ON INDIAN ZOOLOGY.

above, with a deepish and broad longitudinal furrow ; a membranous
pouch under-the:fork of the inferior maxillaries; body green above,
white underneath ; intestine short, straight, and thick; no ccecum.,

Called by the natives of Hindostan ‘the Cowah Muchee,’ or Crow-
billed fish. Sometimes attains 18 inches.

Forms a distinct species, nearly allied to the marine Hsox Belone of
Cuvier, but differing in number of rays to the fins, and in having a
straight ended, not forked, tail.

Inhabits nullahs and stagnant waters at Suharunpoor; common.

Swims horizontally, with a zig-zag motion, like an eel. Is eaten by
the natives.

II].—MeEmMoRANDUM REGARDING THE PREDACEOUS Habits oF INDIAN
FRroGs IN AN INSTANCE OBSERVED BY Mr. J. WRIGHT.

About the end of August, 1840, Wright one evening was seated
outside the house on the terrace, and saw one of the large yellow rain
frogs of Hindostan quietly crouched under a raised piece of timber,
close to the terrace. There happened to be a quantity of chaff and
grain strewed over the adjoining ground near the terrace, left there
after feeding fowls. Several sparrows were attracted by the sight of
the grain, and settled upon the spot. The movements of the birds
hopping about and picking the grain soon appeared to arouse the atten-
tion of‘the frog, who evinced his interest by raising himself on his hind
legs, and vibrating his body rapidly backwards and forwards without
leaving his cover under the wood. At length one of the sparrows
hopped to within four or five feet of him, when, in one spring, he threw
himself most accurately on the bird, and seized it in an instant, taking
the head, neck, and body at once into his gape. He then sprang back
to his cover, and was vigorously endeavouring to swallow the bird,
when Wright, who was attentively watching what was going on, got
up, pushed the frog into a corner, where he was able to lay hold of the
reptile, and seizing the sparrow’s legs, compelled the frog, after a deter-
mined resistance, to disgorge his prey. The sparrow had a spark of
life remaining when drawn out. The correctness of these particulars
is hereunder authenticated by Wright.

‘The above statement is strictly correct.
‘J. H. S. Wrigart.’

Suparunpoor: July 15, 1841.

END OF THE FIRST VOLUME.

LONDON

PRINTED BY SPOTTISWOODE AND CO.

NEW-STREET SQUARE

FAUNA ANTIQUA SIVALENSIS ;

OR,

ILLUSTRATIONS OF THE FOSSIL FAUNA
OF THE SEWALIK HILLS.

BY

HUGH FALCONER, M.D. F.RS.

AND

PROBY T. CAUTLEY, F.R.S.

M® HARDWICKE has made arrangements for the sale of the

few remaining copies of this magnificent work at a greatly
reduced price. The work consists of nine parts, each of which,
containing twelve folio Plates executed in the first style of

lithography, was originally published at the price of One Guinea.

The Nine Parts, with descriptive Letterpress, are now offered for

Four Guineas.

Original Subscribers to the work can obtain a copy of the

descriptive letterpress on application to Mr. Hardwicke.

Te Bde iy

co © on) a
ate Ba
eS ae: