FALCONER’S

PALEZONTOLOGICAL MEMOIRS

&e.

VOLUME I.

‘
*
ISWOODE AND Co.

LOND
tow
| NEW-STRELT SQUARE

PRINTED BY SPO’

Photographed by Erm
‘Printed in Pig

PALAONTOLOGICAL MEMOIRS AND NOTES

OF THE LATE

HUGH FALCONER, A.M, MLD.

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 CALOUTTA.

WITH A

Hiographical Sketch of the Author.

COMPILED AND EDITED BY

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

FELLOW OF THE ROYAL COLLEGE OF PHYSICIANS OF LONDON,

VOL. II.

MASTODON, ELEPHANT, RHINOCEROS,
OSSIFEROUS CAVES,
PRIMEVAL MAN AND HIS COTEMPORARIES.

LONDON :
PSROBERT HARDWICKE, 192 PICCADELLY.
1868.

a 7
= Jae 4
—"s.

aos i> 2
i.

: ia seals et" 5+
P

by

oe Bey

*

This Volume is Dedicuted to

M. ED. LARTET,

FOREIGN MEMBER OF THE GEOLOGICAL SOCIETY OF LONDON,

GEORGE BUSK, FERS.
THE REV. JOHN GUNN, F.G.S,
JOSEPH PRESTWICH, F.RS.,
CAPT. SPRATT, BN., CB,

AND
COLONEL WOOD, F.G:S.,
THE FREQUENT MENTION OF WHOSE NAMES IN ITS PAGES
TESTIFIES TO THEIR HAVING BEEN

ALIKE THE FRIENDS AND FELLOW-LABOURERS

OF

ITS AUTHOR.

,
ree
eat ay

‘~" wee. Sty ;

i

CONTENTS

OF

THE SECOND VOLUME.

PAGE
I, On the Species of Mastodon and Elephant occurring in the
Fossil state in Great Britain. Part I. Mastodon : Fa il

AppEenDIx.—1. Note on J. F. Brandt’s Memoir on the Skeleton
of a Mastodon, discovered near Nicolaieff in Southern
Russia. By Dr. Falconer and Mr. T. Rupert Jones. . 65

2. Notes on specimens of Mastodon angustidens and IZ. Borsoni
in the Museums of Zurich and Winterthur : . 68

38. Note on Cast of a Lower Jaw of an American Mastodon at
Genoa . ; ; ; 5 ; ’ 5 : ee:

II. On the Species of Mastodon and Elephant occurring in the
Fossil state in Great Britain, Part II. Elephant . g xhe

III, On the American Fossil Elephant of the regions bordering the
Gulf of Mexico (Z, Columbi, Fale.), with general observa-
tions on the’living and extinct Species of Elephants . . 212

IV. On the Fossil Remains of Elephas Melitensis, an extinct pigmy
Species of Elephant, and of other Mammalia &e. from the

Ossiferous Caves of Malta ; ; : . 292
APPENDIX.—1. Extracts of Letters from Dr. Falconer to Capt.

Spratt, C.B. . : : ; ‘ : ; E . 299

2. Extracts from Dr. Falconer’s Note Books . F , . 801

3. Abstract of Communication to the British Association at
Cambridge, on Oct. cp 1862, on the Ossiferous Caves of
Malta . 3 ; : : ; 4 J . 807

V. Onthe European Pliocene and Post-Pliocene Species of the
Genus Rhinoceros. ; : . . awidve . 809

vill

XU.

XVI.

XVII.

XVIII.

CONTENTS OF

PAGE

1. On Rhinoceros hemiteechus (Fale.). An extinct Species pre-
yailing in the Gower Caves, South Wales : ; . dll
AppEnpIx.—(Extracts from Dr. Falconer’s Note Books) . 349
2. Notes on Rhinoceros Etruscus (Falc.) ° . . . 34

3. Notes on Fhinoceros leptorhinus saa pro par si RR. mega-
rhinus (Christol) . > 5 . 368

4, Notes on Rhinoceros antiquitatis cae ) R. tichorhinus
)Fisch.& Cuv.) +. . eK a Nee , 399

Notes on Dentition of existing Species of Rhinoceros.
Rh. Keitloa, R. camus, and R. bicornis . . - . 402

Note on the existing Hippopotamus Liberiensis (Morton), with
a synopsis of the Hippopotamide, Fossil and Recent . . 404

Description of two Species of the Fossil Mammalian Genus,
Plagiaulax, from Purbeck . : : . . : . 408

On the disputed affinity of the Mammalian Genus Plagiaulax,
from the Purbeck Beds. : ; ; 4 : . 430

Note on the occurrence of Spermophilus in the Cave Fauna
of England : . 5 : , d . : . 452

Note on the occurrence of Felis spelea in the Mendip Caverns
and elsewhere, and on a Species of Felis found in one of

the Gower Caves c ; . Pee . : . 455
Note on the Remains of Drepanodon or Machairodus of reputed

British origin. 5 . : ; : : : . 459
Note on the Remains of a cok wed Wolf, from Spyritsail-Tor

Cave . : : : : ‘ ; < . 462
Notes on Hyzena . F 5 i : : F 2 . 464.
Notes on Fossil Species of Ursus. : : ( : . 466

Notes on Fossil Species of Cervus, including a description of a
remarkable Fossil Antler of a large Species of extinct Cer-
vus, C. (Eucladoceros) Sedgwickit, in the collection of the

Rey John Gunn, Irstead. : ; : : . 471
Note on an undescribed Species of Bos in the Florence Museum

(Bos Etruscus) . é , c é - 2 : . 481
Note on Crania of Crocodilus cataphractus and C. marginatus,

in the Belfast Museum c : ‘ , ; i . 482
On the Ossiferous Cave of Brixham, near Torquay ¢ . 486

1. Letter from Dr. Falconer to the Secretary of the Geological
Society . : ; . : ; - : - . 487

XX.

XXI.

XXII.

XXII.

XXIV.

XXV.

XXVI.

XXVII.

XXVIII.

THE SECOND VOLUME. 1X

PAGE
2. Report of progress in the Brixham Cave, Sept. 9, 1858 . 491
3. Note by the Editor . : : ‘ : 5 . 497
On the Ossiferous Caves of Gower, South Wales ; . 498

ApprNnDIx.—1l. Memorandum by Mr. Prestwich on the
Boulders and Gravels of the Gower Cave District 5 pXo

2. Note on the occurrence of Wrought Flints in association
with two extinct Species of Rhinoceros, &c. in ‘Long

Hole’ Cave, Gower. : : : ; : . 588

On the Fossil Remains found in Cefn Cave, near Bryn
Elwy, North Wales : : : ; ; : . 541
On the Ossiferous Grotta di Maccagnone, near Palermo . 548

AppENDIx.—1, Notice of the discovery of two Bone Caves
in Northern Sicily. By Frangois Anca, Baron de Man-

galaviti 561
2. Memorandum by Dy. Falconer on the former connection
by Land of Sicily with Malta and Africa. , . 552
On the Fossil contents of the Genista Cave, Gibraltar, By
H. Falconer, M.D., and G. Busk, Esq. , ; . 554
Notes on a collection of Fossil Bones discovered in a section
of gravel in excavating the Folkestone Battery ; . 564
Primeval Man and his Cotemporaries . ; : ‘ . 570
On the Evidence in the Case of the Controverted Human
Jaw and Flint-Implements of Moulin-Quignon é . 601
Works of Art by Primeval Man in Europe: . : : . 626
On the asserted Occurrence of Human Bones in the Ancient
Fluviatile Deposits of the Nile and Ganges; with Com-
parative Remarks on the Alluvial Formations of the two
Valleys . ; : 2 ; : ; > : . 632
On the Glacier-Erosion Theory of Lake-Basins . : . 648

INDEX . - ; - ; 5 , . - . 661

PLATE

Ii.

LIST OF ILLUSTRATIONS

IN

THE SECOND VOLUME.

A. PLATES.
Representations in outline of crania of Proboscidea,
Anterior views

Representations in outline of crania of Proboscidea. Profile
views

Molars of Mastodon (Caton longi and J,
(Trilophodon) angustidens

Molars of Mastodon he dena Aver ae the
Crag :

Molars of Hlephas COs Clif and £. (Stegodon) i msignis

Molars of Elephas (Loxodon) Africanus, E. (Loxodon) plani-
Srons, E. Ca) ase and £, ecesuee primi-
genius

Molars of Elephas (aoa priscus
Molars of Elephas (Loxodon) meridionalis
Molars of Elephas (Huelephas) antiquus

Molars of Elephas rae are Columbi and £. (Bucs

Armeniacus
Molars, &c. of Klephas (Lox nite Melitensis
Molars of Elephas (Loxodon) Melitensis
Vertebra and Pelvis of Elephas (Loxodon) Melitensis
Humerus and Femur of Elephas (Loxodon) Melitensis .
Cranium of Rhinoceros hemitechus, The ‘ Clacton Skull’
Upper Molars of Rhinoceros hemitechus, from the Gower
Caves : :
Upper Molars of Rhinoceros hemitechus, from the Gower
Caves ; : : : . ’

PAGE

16

16

22

26
86

90
100
132
184

222
294
298
304
304
318

324

326

XXY.

XXVI.
XXVII.

XXVIII.

XXIX.

XXX.

XXXII.

XXXII.

XXXII.

XXXIV.

XXXV.

XXXVI.

XXXVI.
XXXVIII.

LIST OF ILLUSTRATIONS

Upper Molars of Rhinoceros leptorhinus Ss mega i 2
R. hemitechus, and R. bicornis . :

Lower Jaw with Molars of Rhinoceros hottie folk
the Gower Caves.

Lower Jaw with Molars advanced in wear of Rhinoceros
hemitechus, from the Gower Caves

Lower Jaw with Molars, and Upper Jaw with Milk Den-
tition, of Rhinoceros hemitachus . . ; é

Lower Jaw and Molars of Rhinoceros Etruscus, from the
Norfolk Forest-bed and the Val d’Arno

Comparison of Skulls of Rhinoceros hemitachus, from the
Gower Caves and Northamptonshire. Basal views

Comparison of Skulls of Rhinoceros hemiteechus, from the
Gower Cayes and Northamptonshire. Profile and upper
views .

Milk Dentition of Rhinoceros hemitechus, from the Gower
Caves, and Molars of R. Etruscus, from Pisa

Cranium of Rhinoceros Etruscus in Florence Museum

Cranium, Lower Jaw and Molars of Rhinoceros Etruscus
in Florence Museum

Cranium of Rhinoceros Etruscus in Museum at Pisa, and
Lower Jaw of ditto at Leghorn

Cranium, showing Molars, of Rhinoceros Etruscus, in the
University Museum of Bologna

Lower Jaw of Zthinoceros leptorhinus (R. megarhinus) in
the Museum of Montpellier .

Molars and Cranium of Rhinoceros leptorhinus (R. megar-
hinus) at Imola and Lyons .

Molars of Rhinoceros leptorhinus oe megar: ore) at Stutt-
gart and Rome :

Lower Jaws and Dentition of Plagiaulax Becklesii .

Lower Jaws and Dentition of Plagiaulax Beckles, Plag.
minor, Microlestes antiquus, Thylacoleo carnifex, and
Cheiromys Madagascariensis

LowerJaws of various Species of Spermophilus and Carnas-
sier of Felis spelea

Lower Jaw of Hyznoid Wolf, from Spritsail-Tor ome
and of Ursus from ‘ Deborah’s Den’ ,

Antler of Cervus (Eucladoceros) Sedgwicki

Crania of Crocodilus cataphractus and C. marginatus

416

454

470

480
484

IN THE SECOND VOLUME. Xl

B. WOOD ENGRAVINGS.
ria. PAGE
1&2, Twoviews of Pelvis of Elephas Indicus (vay. dauntela), copied.
from Cuvier’s ‘Ossemens Fossiles,’ Eléph, Pl. VII. figs. 3
& 4, to illustrate measurements of Mr. Gunn’s Pelvis of Ele-

phas meridionalis . C : : 5 : ; i . 142
3. Ground Plan of Galleries in Brixham Cave . , . 492
4, Section of Raised Beach near Mewsladz, Gower, 8. Wales . , 597
5. Ground Plan of Cefn Cave A . ‘ é . ; . 541
6. Vertical Section of the Grotta di Maccagnone . : ; . 547
7. Ground Plan of the Grotta di Maccagnone . ; . 5 . 547
8. The Elephant victorious over the Tortoise, and supporting the

world, from a sketch by the late Professor Edward Forbes . 575

9. The Elephant borne by a Tortoise, and supporting the world,
from a sketch by the late Professor Edward Forbes : . 600

CORRIGENDA AND ADDENDA.

Page 52 line 17 of Note 3. For ereur, read erreur.
» 92 ,, 27. For mandibule, read mandible.

yy 1385 ,, 4& 38. Forchanneled, read channelled, and for channeling, read
channelling.

» 1387 ,, 387. For channeled, read channelled.

» 1388 ,, 82. For channeling, read channelling.

ee ef In notes 1 and 2 for Plate VIII. substitute Plate VI.
» 188, line 6. For Amasses? read Anapus.

» 188 ,, 21. For Folger’s, read Foulger’s.

» 199 ,, 15. For lagon, read lagoon.

» 207 ,, 14. For Liberensis, read Liberiensis.
» 208 ,, 40. For Sutton, read Stutton.

» 270 ,, 36. For continuous, read contiguous.

,, 3809, lastline. For tichorinus, read tichorhinus.*
» 310, lines 28 & 38, Jor tichorinus, read tichorhinus.

» 3811 ,, 3,10,&18. For tichorinus, read tichorhinus.

» 825 ,, 11. For left, substitute right.

» 830 ,, 38 from bottom. For fig. 5, read fig. 3.

» 801 ,, 12ofsecond paragraph. Wor B. M. 132,133, read B. M. 27,836.
» 3803 ,, 1. For Cesell’s, read Ceselli’s.

» 3805 ,, 7. For Sub-Appenines, read Sub-Apennines,
» 806 ,, 17. For about, read orbital,
» 809 ,, 19. For ou, read o.

I. ON THE SPECIES OF MASTODON AND ELEPHANT
OCCURRING IN THE FOSSIL STATE IN GREAT
BRITAIN.

BY H. FALCONER, M.D.

PART I. MASTODON.!

INTRODUCTION :—-GENERIC DISTINCTIONS AND NOMENCLATURE OF THE
PROBOSCIDEA—DINOTHERIUM—MASTODON AND ELEPHAS :—THE DIS-
TINCTIVE AND SPECIFIC CHARACTERS OF MASTODON AND ELEPHAS:
—THE BRITISH FOSSIL MASTODON, AND ITS COMPARISON WITH M.
ANGUSTIDENS, M. ARVERNENSIS, AND M. LONGIROSTRIS—M. ANGUSTI-
DENS—M. ARVERNENSIS AND M. LONGIROSTRIS :—BRITISH SPECIMENS
OF MASTODON—MOLARS—PREMOLARS—MILK MOLARS—LOWER JAW:
—GEOLOGICAL AGE OF THE MASTODONS—MASTODON ANGUSTIDENS,
M. LONGIROSTRIS, AND M. ARVERNENSIS—-MASTODON OF THE CRAG
IN PARTICULAR :—CONCLUSION.

Introduction.—It is of the highest importance to Geology,
that every mammal found in the fossil state should be de-
fined as regards, Ist, its specific distinctness, and, 2ndly, its
range of existence geographically and in time, with as much
exactitude as the available materials and the state of our
knowledge at the time will admit. Hvery form well ascer-
tained becomes a powerful exponent; while, ill-determined,
it is a fertile source of error. For the pure Geologist, in
_ most of his conclusions where age or climatal conditions are
in question, is more or less at the mercy of the Palzonto-
logist, since he must accept the paleontological evidence as
it is laid before him, and square his speculations to fit and
dovetail into the various mortises which the data inexorably
present to him. There is a subordination in the value of
the evidence: the higher the form in the scale of organiza-
tion, the more weighty is the import of its indications.

1 This paper was communicated tothe | tions in Plates III. and IV. have been
Geological Society of London on April | reproduced from those which accompa-
8, 1857, and was published in the ‘ Quar- | nied the original memoir. The outline
terly Journal of the Geological Society’ | heads in Plates I. and II. have been
for November 1857, from which itis here | copied from Plates XLII. to XLV. of
reprinted with additions. The illustra- | the‘ Fauna Antiqua Sivalensis,—[Ep. ]

VOL. II. B

2 BRITISH AND EUROPEAN FOSSIL MASTODONS.

The difficulty with which the Mammalian paleontologist
has to contend in arriving at satisfactory results depends
doubtless, in many cases, on the imperfect nature and scanti-
ness of his materials. But it is deserving of remark, that
the fossil genera and species which are in the most unsatis-
factory and unsettled state, as to definition and nomencla-
ture, are not those that are the rarest, but often the reverse.
Take Mastodon or Rhinoceros, for example, in which the array
and confusion of specific names are signally perplexing.
The reason of this apparent anomaly would seem to be
this: when the remains are few and seldom met with, the
species are usually limited in number, and thus more easily
discriminated ; on the other hand, when the remains are
very abundant over wide areas, the species are at the same
time, as a general rule, numerous: and it is well known
among naturalists, that the genera which are the most
difficult to disentangle specifically are the most complete
and natural, where the species are many, and follow each
other with the least amount of difference in serial develop-
ment; or, in other words, where they are most closely allied
to one another.

Remains of either of the Proboscidean genera, Dinotheriwm,
Mastodon, and Elephas, abound in all the Tertiary Forma-
tions of Europe, Asia, and America, from the Miocene up to
the Post-Pliocene; they have been the subject of a vast
amount of observation, while it is hardly possible to conceive
anything more unsettled and opposed than the generally
received opinions respecting the species and their nomen-
clature in the standard works which are of the greatest
authority on the subject. . Cuvier, De Blainville, and Owen
are agreed in limiting the Elephants and narrow-toothed
Mastodons found fossil in Europe each to a single species ;
while other paleontologists consider that the latter group
comprises at least three well-marked specific forms, and the
former three or four. This paleontological uncertainty has
naturally been refiected in systematic works on Geology,
wherever the faunas of the Tertiary Formation are referred
to, in statements sufficiently startling, which are repeated at
the present day. Thus the Miocene Mastodon angustidens,
of the Faluns of Touraine, of the Molasse of Switzerland
and of the Sub-Pyrenees, as also the Miocene Mastodon longi-
rostris of Eppelsheim, are mentioned by Sir Charles Lyell,
in the fifth edition of his Manual,! under the comprehensive
name (on the authority of Owen) of Mastodon anqustidens, as
occurring in the so-called ‘ Older Pliocene’ Red Crag, and

1 Op. cit. p. 166,

INTRODUCTORY REMARKS. 3

in the ‘ Pleistocene’ Norwich Crag; while this English
species of Mastodon, wherever it has been met with, whether
in this country or on the Continent, has been almost in-
variably found in company with remains of a species of
Hlephant which Professor Owen has described as identical
with the Hlephas primigenius or Mammoth of the Post-
Pliocene Drift and the modern Siberian ice-fields.

The object of the present communication is, to endeavour
to ascertain what are the species of Mastodon and Elephant
found fossil in Britain ; what the specific names which ought
to be applied to them; and what the principal formations
and localities where they are elsewhere met with in Europe.
I am the more induced to attempt the task from the circum-
stance that Prof. Owen, in an important memoir ‘On some
Mammalian Fossils from the Red Crag of Suffolk,’ which
appeared in a late number of the Society’s Quarterly Journal,!
adheres to the opinion expressed in his Report to the
British Association for 1843, and subsequently discussed at
greater leneth in his ‘ British Fossil Mammalia’ in 1846,
that the Mastodon of the English Crag is identical with the
Mastodon angustidens of Cuvier, the Mastodon longirostris of
Kaup, and the Mastodon Arvernensis of Croizet and Jobert.
- Prof. Owen, on both the occasions here quoted, up to 1846,
has maintained the prevalent opinion, that all the Elephant
remains met with in England are referable to a single
species, namely Hlephas primigenius; and I am not aware
that he has altered his views upon this point in any subse-
quent publication. I have devoted much study to the sub-
ject, during the last fifteen years, in connexion with the
numerous fossil species of both genera which are met with
in India, with a view to a monograph of the Proboscidean
’ family, fossil and recent. The results to which I have been
conducted, as to the disputed European species, are different
from those arrived at by Prof. Owen. The most of those
results have been long exhibited, so far as figured evidence
goes, in the published illustrations of the ‘ Fauna Antiqua
Sivalensis’;? but, having devoted the last summer and
autumn to a Proboscidean examination, so to speak, of some
of the principal collections on the Continent, with special
reference to the European fossil species, I have been enabled
to confirm or correct previous conclusions on a wider field
of observation. In order to avoid needless repetition in
the sequel, I may mention that the tour here referred to
embraced a detailed study of the very extensive collection of

1 No. 47, vol. xii. part iii, (Aug. 1,| Plates xlii—xlv. 1847. (Reproduced
1856), p. 228. in Plates i. and ii. of this volume. Seo
? Fauna Antiq. Sival. Illustr. part y. | also vol. i. p. 476.—Ep.)
B2

4 BRITISH AND EUROPEAN FOSSIL MASTODONS

Val d’Arno Proboscidean remains contained in the Museum
at Florence; the collections of Turin, Milan, and Pavia; of
Geneva, Lausanne, Berne, Zurich, Basle, and Winterthur in
Switzerland; of Darmstadt, Mannheim, and Strasbourg on
the Rhine; of the Jardin des Plantes and Ficole des Mines
in Paris ; the Duc de Luynes’s fine collection of the Chartres
Fossil Elephant, in Chateau Dampierre ; and the surpassingly
rich and unrivalled collection made by my friend M. Lartet,
of the Sub-Pyrenean Proboscidea, at Seissan on the Ga-
ronne; together with some of the principal collections at
Toulouse. In one or other of these Museums I had oppor-
tunities of studying all the fossil species hitherto described
as having been met with in Europe, together with one fine
species of Mastodon discovered by M. Lartet, but which has
not yet been published.

Generic distinctions and nomenclature of the Proboscidea.—
Before entering on the special consideration of the British
fossil forms, it will be necessary to give some explanation of
the principles on which the genera have been limited, and
subdivided into subgeneric groups, in order to comprehend
the reasons for the nomenclature adopted in this communica-
tion. A detailed paleontological disquisition would be out of
place on the present occasion. Such salient points only will
be touched upon as are essential to the elucidation of the
subject.

The Proboscidean species, fossil and recent, constitute a
large group, embracing at least twenty-five distinct forms,
which are comprised under the three genera of Dinotheriwm,
Mastodon, and EHlephas. These genera, regarded in a
systematic view, are of very unequal value numerically; the
first being very limited in the number of ascertained species,
but defined by well-marked generic distinctions; while the
last two represent a large number of specific forms, which,
although their opposite extremes are widely separated, yet
are connected together through so complete and natural a
series of intermediate specific links, that it has proved
difficult to devise good generic characters to distinguish,
them. Putting aside all other considerations of structure and
form, the diagnostic marks will be regarded on the present
occasion solely as they are furnished by the teeth and jaws.

Dinotheriwum.—The adult dentition of Dimotheriwm' is cha-
racterized by two vertically succeeding premolars and three
true molars, five teeth in all, with transverse crenulated
ridges closely resembling those of the Tapir; and by two

1 Kaup, Akten der Urwelt (1841), pp. 22-40.

GENERIC DISTINCTIONS OF PROBOSCIDEA. 5

huge inferior recurved incisors, implanted in an enormously
thickened and deflected beak or prolongation of the sym-
physis of the lower jaw (Plate II. fig. 1). Most of the
molar teeth present the normal Tapir-like crown character
of two ridges; but, when the milk and permanent dentition
are taken together, Dinotheriwm differs from all the non-
elephantoid Pachydermata in the circumstance that the last
milk molar and antepenultimate true molar (being con-
tiguous teeth in the order of horizontal succession) present a
more complex development of three ridges, or, a ‘ternary-
ridged crown formula’ (to use a term which will be found of
importance in the sequel). Two species only of Dinotheriwm
have, I believe, hitherto been met with—the one in Europe,
and the other in India.!_ The European species, D. giganteum,
occurs in the Older Miocene formations, such as Eppelsheim,
the Faluns of Touraine, and the Molasse and lacustrine
strata of the Sub-Pyrenees ; it has nowhere been met with in
Britain.

Mastodon and Elephas.—Up to the date of the last 4to
edition of the ‘Ossemens Fossiles,’ published during the
author’s life, in 1825, the species of Mastodon and Elephas
then known were sufficiently well distinguished by the cha-
racters indicated by Cuvier,” the founder of the former genus ;
namely, that the molar teeth of Mastodon consisted of a com-
paratively simple crown, divided into mammillz or tubercles,
arranged in transverse ridges, more or less numerous, and
more or less prominent, with corresponding empty valleys or
hollows between them; while those of Hlephas were more
complex, consisting of numerous thin transverse plates, having
their intervals filled up with cement. The subsequent

1 As in the case of the Mastodon of
North America, numerous nominal species
have been founded by different authors
(Kaup, Von Meyer, Eichwald, &c.) upon
what would appear to have been merely
varieties of the same species, depending
on race, sex, &c., as evinced by the
comparative size of the teeth. Dr. Kaup
now entertains doubts of there being any
other European species than D. gigan-
teum (Akten der Urwelt, p. 49). The
difference of size between the teeth of
D. giganteum and D. Cuvieri is not
greater than is known to occur between
homologous teeth from different indi-
viduals of Mastodon (Tetraloph.) longi-
_ rostris, dug out of the same deposit at
Eppelsheim. The nominal species, D.
Kenigii of Kaup, is founded on a single
small tooth, and therefore doubtful. I
have lately seen well-marked specimens
of fossil teeth of aspecies of Dinotherium

from Attock in the Punjab, at no great
distance from the Sewalik hills, and
judging from the associated Mammalia,
out of beds of the same age with them.
The materials are not sufficient to es-
tablish whether the species is identical
with D. Indicum of Perim Island, or
distinct. In dimensions the teeth corres-
pond with medium-sized specimens of
D. giganteum. They were discovered by
Lieut. Garnett, of the Bengal Engineers,
and are now in the possession of Prof.
Oldham, Superintendent of the Geologi-
cal Survey of India, who has obligingly
communicated them to me.—July 1857,
H. F. (See vol. i. p. 414. A specimen
of the third lower premolar of this
species, from the ‘ Red Marl’ at Noorpoor,
found in Dr. Faleoner’s collection, is
labelled in his hand-writing, ‘ Dinothe-
rium Pentapotamicum, Fale.’—Ep.)
2 «Oss, Fossiles,’ tom. i. p. 205.

6 BRITISH AND EUROPEAN FOSSIL MASTODONS.

discovery of the Mastodon Elephantoides of Clift, in which
Cuvier’s characters of both genera are blended, and of
European and American forms with tusks in the lower jaw
(Mastodon longirostris and Mastodon Ohioticus), led to the
necessity of remodelling the technical diagnostic characters
of the genera. This was first attempted, so far as I am
aware, by Bronn of Heidelberg, in his ‘ Lethzea Geognostica,’
as far back as 1838. In his elaborate definition of the two
genera, he states (omitting other characters) that Mastodon is
characterized by lower incisors, and by molars which are
replaced from back to front, excepting, however, the most
anterior of these teeth, 7.e. one or more milk molars; while
in Elephant there are no inferior incisors, and all the molars
are replaced in a horizontal direction.!. In his remarks upon
the species, he mentions that ‘Tetracaulodon (i.e. Mastodon
longirostris), according to Kaup, has premolars in the upper
jaw, which are very similar to the back molars of Hippopota-
mus and are very caducous,”? and in regard to inferior tusks,
that ‘Mastodon giganteus, M. angustidens, and M. 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.”
The same characters, i.e. of premolars and inferior incisors,
were several years afterwards advanced by Prof. Owen (who
must obviously have overlooked the previous remarks of
Bronn) in his ‘British Fossil Mammalia” as being dis-
tinctive of Mastodon from Elephant in a well-marked and
unequivocal manner. But they are assuredly neither absolute
nor constant, whether regarded in a positive or negative view,
as generic distinctions. For on the one hand, premolars have
not yet been met with in M. (Trilophodon) Ohioticus, in place
in the jaws, although made a subject of special research by
Dr. Warren,* Dr. Jackson, and myself, upon a large quantity
of materials, up to a very late date; nor have they been yet
met with in certain species of the Tetralophodon group;
while, so far from being restricted to species of Mastodon,
they have been detected by us in a typical fossil Elephant
from India, H. (Loxodon) planifrons, both in the upper and
lower jaws, in as great a number asin any known Mastodon.°®
And as regards inferior tusks, although these have been ob-
served in three species of the Trilophodon group, and in two
of Tetralophodon, there are other species in which, among
abundant materials, they have not been noticed up to the

1 Bronn, ‘ Lethea Geognostica,’ Band 4 ¢ Brit. Foss. Mam.,’ p. 274.

ii. pp. 1233 and 1239 (1st ed.). 5 Warren, On Mastodon giganteus,
2 Id. loc. cit. p. 1218. p. 80.
8 Id. doe. cit. p. 1233. 5 See yol. i, p. 483.—[Eb.]

GENERIC DISTINCTIONS OF PROBOSCIDEA. 7
present time, even in young individuals, where they might
have been with most confidence looked for. This remark
applies with especial force to the forms here called M. (Trilo-
phodon) Humboldtii, and to M. (Tetralophodon) Sivalensis and
M. (Tetralophodon) Arvernensis.

Swayed by considerations of this nature, and struck more
particularly with the identity of general characters and close
similarity of form running throughout the whole of the
osteology of the species of Mastodon and Elephant, with the
exception of the molars and inferior incisors, De Blainville'
abandoned the idea of there being any sufficient generic
difference between the two, and made a retrograde step,
arranging all the forms in two divisions, Lamellidontes and
Mastodontes, under the common designation of ‘ Hlephas.’
This proposal has deservedly met with little favour among
palzontologists and zoologists.

There are characters however, which, when once recog-
nized, are happily of an obvious and readily applicable nature,
to distinguish Mastodon from Elephant, and which further
enable the paleontologist to break up an unwieldy mass of
species into subgeneric groups, that are at the same time
natural and convenient. Putting aside for the moment, as
extraneous, the consideration of incisors and premolars, and,
as in the case of Dinotherium, taking the milk and permanent
dentition together, the species of both Mastodon and Elephant
ordinarily present six molar teeth from first to last, in the
order of horizontal succession—7.e., three deciduous or milk
molars, and three true molars. It was stated above, that in
the Dinotherium the last milk molar and the antepenultimate
or first true molar are invariably characterized by a ternary-
ridged formula, or, in other words, that their crowns are
divided into three ridges. Applying this criterion in a similar
manner to Mastodon, we have found, that not only the last
milk molar and first or antepenultimate true molar, but in
addition the second or penultimate true molar, being three
teeth in immediate contiguity, in all the species (with one re-
markable exception) are severally characterized in both jaws
by an isomerous division of the crown into either 3 or 4 ridges.
These three isomerous-ridged teeth may, for convenience of
description, be referred to in the aggregate as ‘the inter-
mediate molars,’ a term which has been applied to them from
their position by Fischer and by Laurillard.? To the species
which present the ternary-ridged formula we have assigned
the subgeneric name of Trilophodon ;3 and to the quaternary-

1 De Blainville, Ostéographie: Des | Naturelle, tom. viii. p. 29.

Eléphants. 3 From tpeis and Adgos, three-ridged.
? Dictionnaire Universel d'Histoire

8 BRITISH AND EUROPEAN FOSSIL MASTODONS.
ridged species, Tetralophodon.! In citing the various forms
under discussion in the sequel, these subgeneric terms will, in
every case, be used for convenience in designating the species ;
and the same rule will be followed with the subgeneric
divisions of Hlephas. This will be of obvious use on the pre-
sent occasion, both as a help to the memory in dealing with
a large number of specific names, and as suggestive of broad
points of distinction, when referring to the disputed species.
The ternary and quaternary formule are, I believe, never
found mingled in the intermediate molars of the same
species ;? 7.e., a ternary-ridged molar of this series does not
occur in the species belonging to Tetralophodon, nor a qua-
ternary in Trilophodon. The ridge-formula indicates also,
with unerring certainty, the composition of the crown of the
tooth which is immediately in front of, and of that which is
immediately behind, the three intermediate molars: the
former showing invariably one ridge less, and the latter one
ridge more; that is to say, the penultimate or second milk
molar in all the species of Trilophodon is invariably two-
ridged, and the last true molar four-ridged; while in Tetra-
lophodon, in like manner, the former is three-ridged, and the
latter five-ridged—making due allowance in the last true
molar for the amount of individual variety presented by the
greater or less development of the well-known talon compli-
cation, and for its being usually more complex in the lower

‘1 From téocapes and Ad¢gos, four-ridged.

as distinct from M. Humboldtii (Trilo-
This difference of three and four ridges

phodon). In this species the last ridge

was, so far as Iam aware, first pointed
out as a distinctive character between
two European species, namely Mas-
todon angustidens and M. Arvernensis,
by Von Meyer, as far back as 1834, but
without being extended to the three
intermediate molars. The name J.
Arvernensis was applied by him to
the Eppelsheim species, M. longirostris
of Kaup. (Die Fossilen Zahne und
Knochen yon Georgensgmiind, p. 33).
The ridge-formula in the two subgenera
is as follows :—

Milk molars True molars
14243 3+34+4
In Trilophodon 5 Ee Tea
24+3+4 4+4+45
In Tetralophodon Sree T4445

the numerals exhibiting the ridges in
each tooth, exclusive of the ‘talons.’

2 The only apparent exception which
has come under my observation occurs
in the dentition of the South American
species, to which the name of Mastodon
Andium (Tetralophodon of our arrange-
ment) has been restricted by the French
paleontologists, Laurillard and Gervais,

in most of the intermediate molars is
considerably reduced in size; and the
teeth have been, in consequence, deseribed
by Gervais (Zoologie de l’expédition dans
l'Amérique Méridionale par Le Comte de
Castelnau, p. 19) as three-ridged. The
specimens represented by him, figs. 2
and 5 of Plate y. in the ‘ Voyage de Cas-
telnau,’ the former an antepenultimate
upper true molar, and the latter a penul-
timate lower, are distinctly four-ridged,
while the last lower milk molar (fig. 4)
is apparently three-ridged, with a large
talon, My attention was directed to
the subject by M. Lartet. More speci-
mens are required for the exact determi-
nation of the point than yet exist in any
of the European museums; 7.¢., whether
in the intermediate molars of the form
called MM. Andiwm the ternary and
quaternary formule are mingled. Nine-
tenths at least of the specimens of South
American Mastodons in the British
Museum belong to the other species,
M. (Trilophodon) Humboldtii.— July
1857, H. F. (See p. 15, note 3, and
also yol, i. p. 106.—Eb.)

GENERIC DISTINCTIONS OF PROBOSCIDEA. 9

jaw. The ‘ridge-formula’ thus determines, with precision,
five out of the series of six molars developed in horizontal
succession in all the true Mastodons.

For reasons which will be explained in the sequel, it would
seem that there has existed in nature another subgeneric
eroup of Mastodon, of which only a single form is at present
known, iw which the crowns of the ‘intermediate molars’
are divided upon a quinary ridge-formula. This group in
our arrangement would be characterized, in harmony with
the others, as Pentalophodon; and it may with some confi-
dence be predicated that, when the dentition shall have been
well determined, the second milk molar will present four
ridges, and the last true molar six ridges in the upper jaw.

The Elephants, on the other hand, are distinguished from
the Mastodons by the absence of an isomerous ridge-formula
to the three intermediate molars of the upper and lower jaws ;
and by the circumstance that the ridges, instead of being
limited to three or four, range from six up to an indefinite
number in these teeth, in the different groups of species. We
have found that the numerous forms, fossil and recent, may
be conveniently arranged in three natural subgeneric groups,
founded upon the ridge-formula, in conjunction with certain
other dental characters.

In the first of these groups, corresponding with the forms
collectively designated Mastodon Elephantoides by Clift, the
ridge-formula may be said to be hypisomerous, as the difference
between the crowns of any two of the consecutive intermediate
teeth does not exceed more than one ridge, and the ciphers
range in the different species from 6 to 8. The ridges are not
more elevated than in the true Mastodons, so that, when the
teeth are sawn through longitudinally, the section yields a
succession of salient and re-entering angles, the height of
the chevron-shaped ridges not much exceeding the width of
their base. The enamel is very thick, and the coronal inter-
spaces in most of the species are filled up with an enormous
quantity of cement. To this group we have assigned the
subgeneric name of Stegodon.' It is limited to extinct forms
confined at present to the Indian Tertiaries. The Stegodons
constitute the intermediate group of the Proboscidea from
which the other species diverge through their dental cha-
racters, on the one side into the Mastodons, and on the other
into the typical Elephants.

In the second group, which includes the species allied to
the African Elephant, the ridge-formula is also hypisomerous,
as in the Stegodons, the ciphers ranging from 7 to 9 in the

1 From otéyn tectum, and ddovs dens, having reference to the gable-end form
of the section of the ridges.

10 BRITISH AND EUROPEAN FOSSIL MASTODONS.

crown ridges of the intermediate molars of the different
species. But the colliculi, instead of yielding a gable-shaped
or ‘tectiform’ section as in the Stegodons, are much more
elevated and compressed, so that when the teeth are sawn
longitudinally and vertically, the ridges present the appear-
ance of elongated wedges, with thinner plates of enamel. For
this subgeneric group, the name of Lowodon,' first indicated
by Frederick Cuvier, has been adopted. It comprises both
extinct and living species.

The last group, which is numerically the largest and most
important, including the Elephants with thin-plated molars,
as in the existing Asiatic species, is characterized by the
ridge-formula being regulated in the ‘intermediate molars,’
not by hypisomerous ciphers, but by progressive increments
(anisomerous), which may be expressed (e.g. for the Indian
Elephant) by the series 12+14+418.? These ciphers, be it
remarked, are not put forward as being rigidly exact in every
case; for the higher the numerical expression of the ridge-
formula in the species, the more liable to vary within certain
limits, dependent on the race, sex, and size of the individual,
is the number of the plates; and they do not rigidly corres-
pond throughout in the upper and lower molars, the latter
often exhibiting an excess. But it may safely be asserted
that the numbers are never transposed or reversed—+.e., the
younger tooth among the ‘intermediate molars’ never nor-
mally exhibits in the same individual a higher number than
the older; the increments may not always be symmetrical,
but they are invariably more or less progressive. For this
subgeneric group we propose the term of Huelephas.*

The following systematic Diagnosis of the genera Mastodon
and Elephas was prepared as an Appendix, but its insertion

1 From Aofbs obliguus, and 500s dens,
having reference to the rhomb-shaped
dises of the worn molars; an adapta-
tion of the term ‘ Loxodonta’ proposed
by Fred. Cuvier (‘Hist. Naturelle des
Mammiferes,’ tom, iii. Article ‘ Eléphant
d'Afrique.’ 1835).

2 The illustration in this case is taken
from the existing Indian Elephant, EF. ( Hu-
elephas) Indicus, in which the ridge-for-
mula of the whole series is nearly thus:—

Milk molars
4+8+12 14+18+24
44+8+12 14+ 18+24—27*
the numerals representing the ridges in
each tooth, exclusive of the talons. A
progressive increment runs throughout

True molars

the series; but the selected numbers
refer only to the ‘intermediate molars.’
In the species which approach nearest to
Loxodon, the numerical expression of
the ridge-formula is lower.

8 From ed bene, and édépas, having
reference to the typical Elephants most
familiarly known, In the illustrations
of the ‘Fauna Antiqua Sivalensis,’ the
term Hlasmodon was applied to this sub-
generic group; but, the designation of
Elasmodus having been preoccupied by
Sir Philip Egerton for a series of fossil
fish (Proc. Geol. Soe. vol. iv. p. 163, 1843),
to prevent confusion, the term of Ewele-
phas has been substituted for it.

* The ridge-formula for the true molars was subsequently altered to ks. is

12+16+24—27

See note to description of Hlephas antiquus, and Memoir on E. Oolwmbi.—[Ep].

GENERIC DISTINCTIONS OF PROBOSCIDEA. 11

in this place more conveniently elucidates the subject-matter
of this memoir.

Genus Mastopon (Cuv.).

: : . Lae a j
Formula Dentium deciduorum.—Primores 5 vel5; Lani-
e. 3

aru x3 Molares = 8-7. .

Formul. Dent. persist.—Primor. : vel ; ; Lan. 53 Proemol. =
3 )vel : ; Molares vert + =12-8.

Primores eburnei plerumque exserti: superiores maximi
vario modo porrecti, inferiores horizontales vel leviter
deflexi, recti, minores. Molares complicati, tritores ;
coronidis rimaé longitudinali obsolete bifide colliculi con-
cavi e tuberculis mammillaribus per paria transverse
aut alternatim dispositis, constantes : adamante crasso,
cemento in valliculis parco aut subnullo. Premolares
aut ceteris forma simpliciores minores, aut nulli.
Molares veri 3, deinceps majores, altero alterum extrusum
a tergo vicissim excipiente, demum utrinque solitarii.—
Molares 3 utrinque intermedzi (nempe deciduorum post-
remus et verorum antepenultimus penultimusque) colli-
culis isomeris aut 3, 4, aut 5 conformes.

Proboscis longissima, prehensilis. Corpus vastum artubus
elevatis insistens. Pedes 5-dactyli.

Subgenus 1. TrrnopHopon.—Dentium molarium 3, utrinque
intermediorum coronis colliculis 3.

Subgenus 2. TerraLopHopon.—Dent. molar. 3, utrinque
intermediorum coronis colliculis 4 (raro 5).

Observations.—The adult dentition varies much in the dif-
ferent species of the genus; the premolars and inferior in-
cisors being inconstant. The typical complete formula is
best shown by M. (Triloph.) angustidens of Simorre :—

Tncis. >; Can. S; Premol. a: Mol. = 12, being identical
with that of Dinotheriwm, so far as the dentition of the latter

: : aa, 0 2
has been determined—+.e., Incis.+ ; Can.—; Premol. ; Mol.

0
- =12; the only question being in regard of upper incisors,

the presence or absence of which has not yet been clearly
ascertained in Dinotheriwm. The affinity indicated by the
agreement in number is corroborated by the last milk molar
and antepenultimate true molar being three-ridged, alike in
Dinotherium and in the section of Mastodon here called Tri-
lophodon. Premolars have not been met with in M. (T'riloph.)
Ohioticus, which, counting both sides of both jaws, has eight

12 BRITISH AND EUROPEAN FOSSIL MASTODONS.

molars less in the adult state than M. (Triloph.) angustidens ;
nor have they been observed in M. (Triloph.) Humboldti.
They occur probably in M. (Triloph.) Tapiroides. Their
presence or absence has not yet been ascertained in the other
species of Trilophodon. These teeth have been observed in
situ in the upper and lower jaws of M. (Tetraloph.) longiros-
tris, and in the upper of M. (Tetraloph.) Arvernensis. They have
not yet been seen wm situ in the other species of Tetralophodon.
Inferior incisors have been discovered in M. (Triloph.) angus-
tidens, M. (Triloph.) Ohioticus, and M. (Triloph.) Tapiroides; and
also in M. (Tetraloph.) Andiwm and M. (Tetraloph.) longirostris,
in the first of which they occasionally attain a very large
size. They do not appear to occur ever in M. (Tetraloph.)
Swalensis, nor in M. (Tetraloph.) Arvernensis. Their presence
or absence in the two other species of Tetralophodon has not
yet been satisfactorily determined. The ridge-formula, as
being respectively ternary in Trilophodon, and quaternary in
Tetralophodon, is very constant, the only doubtful case being
presented by the form or forms named Mastodon Andium by
the French palzontologists. Cement, although quantita-
tively inconspicuous in most of the species of both subgenera,
is present in considerable abundance in the valleys of the
crowns of M. (Tetralophodon) Perimensis and in M. (Triloph.)
Humboldt. In the former it fills up the bottom of the in-
terstices between the mammille. The transverse or alternate
direction of the mammille of the ridges, and the open or
interrupted nature of the valleys connected therewith, are
not equally defined in all the species, intermediate stages
being met with. But the ridges are invariably transverse
and the valleys open in M. (Triloph.) Borsoni, Ohioticus, and
Tapiroides, and in M. (Tetraloph.) latidens; while the mam-
mille are constantly more or less alternate, and the valleys
interrupted, among the Trilophodons in M. (Triloph.) angusti-
dens, Humboldtu, and Pandionis; and among the Tetralo-
phodons in M. (Tetraloph.) Sivalensis and Arvernensis. The
most complex crowns are presented in the Trilophodons by
M. (Triloph.) Pandionis (an Indian fossil species recently dis-
covered, and as yet undescribed) and M. (Triloph.) Hum-
boldtw, and among the Tetralophodons by M. (Tetraloph.)
Swalensis and Arvernensis. The upper adult molars in several
of the species—e.g., M. (Triloph.) angustidens and M. (Tetra-
loph.) Andiwn—were invested with a longitudinal belt of
enamel, disposed more or less spirally, and reaching the apex.
The lower incisors, according to Lartet, are constantly devoid
of any such belt. In M. (Triloph.) angustidens inferior incisors
would appear to have been common to males and females,
and not to have been a mark merely of sexual difference.

GENERIC DISTINCTIONS OF PROBOSCIDEA. 13

Mastodon Sivalensis, although with five-ridged ‘ intermediate
molars,’ is provisionally included under Tetralophodon.

Genus Evepas (Linn.).

Form. Dent. decid.—Primor. 2 ; Lan. = ; Mol. ae

: A 1
Form. Dent. persist.—Primor. ~ ; Lan. ae Premol. 5 vel

He
2 ; Mol. = =11-7.

Primores eburnei plerumque exserti, sursum et antrorsum
adscendentes. Molares aut complicati aut lamellosi,
tritores ; coronidis longitudinaliter integree colliculi con-
vexi e tuberculis mammillaribus, aut laminis cunei-
formibus vel compressis digitatis transversis, constantes :
adamante illis crasso, his attenuato, cemento in valliculis
copioso. Premolares rarissime utrinque 2 (szepius nulli),
ceteris forma simpliciores, minores. Molares veri 3,
deinceps majores, altero alterum extrusum a tergo
vicissim excipiente, demum utrinque solitarii.mMolares
3 utrinque intermedii (nempe deciduorum postremus et
verorum antepenultimus penultimusque) colliculis supra
5 (6-18), aut hypisomeris, aut anisomeris.

Proboscis longissima, prehensilis. Corpus vastum artubus
elevatis insistens. Pedes 5-dactyli.

Subgen. 1. SrEecopon.—Dentium molarium 3 utrinque
intermediorum coronis complicata colliculis
hypisomeris (e.g. 7+7+8), mammillatis,
tectiformibus. Premolares nondum observati.

Subgen. 2. Loxopon.—Dent. molar. 3 utrinque inter-
medior. coronis lamellosa colliculis hypisomeris
(e. g. 7+7+8), cuneiformibus. Preemolares
raro utrinque 2.

Subgen. 3. HuELEPHAS.—Dent. molar. 3 utrinque inter-
medior. coronis lamellosa colliculis deinceps
numero auctis, anisomeris (e. g. 12+14+18),
attenuatis, compressis. Przmolares nulli.

Observations—The adult dentition of the Elephants,
although typically more aberrant, is more constant than
that of the Mastodons. Inferior incisors are wanting in all
the species, fossil and recent, at present known; and pre-
molars have as yet only been met with in a single form,
E. (Loxodon) planifrons, 'The common formula is, Incis.

1 0 0 3 Saghia ]
> 3 Can. 5 3 Premol. | ; Mol. > =7: but in this exceptional
case, the premolars are as numerous as in any species of

SYNOPTICAL TABLE OF THE SPECIES

Genera, Subgenera, and Species

Spec. )
( iL M.(Trilophodon) Borsoni(Z. Hays). . |
(a.)Colliculi acuti vallicule- } 2. M. (Triloph.) Tapiroides (Cuv.) .
as z que transversi . 3. M. (Triloph.) Ohioticus (Blumb.)
marl
°
gs 4, M. (Triloph.) angustidens (Cwv.)
°
B & | (.) Colliculi obtusialternatim | 5. M. (Triloph.) Pyrenaicus (Lart. MSS.)
| S mammillati, vyallicule ) 6, M. (Triloph.) Humboldtii (Cuv.)
= interrupt
= 7. M. (Triloph.) Pandionis ? (Fale.)
4
ZI C . M. (Tetralophodon) longirostris (Kaup)
¥ . (Tetralophodon) longirostris (Kaup
S eel Cee aliens {3 (Tetraloph.) —latidens (Clift)
: eae ae es M. (Tetraloph.?) | Andium (Cuv.)?
& | (d.) Colliculi obtusi alternati . :
: 8 (2) "mamma vale ; 11. M. (Tetraloph.) Perimensis (Fale.g Caut.)
Se interrupt
fe} A . .
3 | Arvernensis (Croizet
ae es - M. (Tetraloph.) {" "Und Jobert). 4
H | (e.) Colliculi numero 5, ob- . M. (Tetraloph.) Sivalensis (Wale. § Caut.)
tusi alternatim mammil-
lati, vallicule interruptee
. _ | (a) Colliculi circiter 6, caemen- spies
dy | to in yalliculis parco 1. E. (Stegodon) Cliftii (Fale. § Caut.) .
[4 $ 2. EB. (Stegod.) bombifrons (Fale.§-Caut.)
22 (b.) Colliculi 7-8, exxmento in 3. E, (Stegod.) Ganesa? (Fale. § Caut.)
nn
yalliculis copiosissimo .
4, E. (Stegod.) insignis (Fale. § Caut.)
c 5. E, (Loxodon) planifrons (Fale. § Caut.
: (c.) Colliculi grosso digitati,
Loz adamante crasso .
alae . 6. E. (Loxod.) meridionalis (Vesti)
< | &2) (d.) Colliculi medio angulatim - 7. E. (Loxod.) priscus (Goldf.)*.
mH 13 a dilatati, machzridibus {
fe a per detritionem rhom- 8. E. (Loxod.) Africanus (Blumb.)
boideis
a
Colliculi sub ti ada- :
r (2) Sara dts ‘subremoti ada: 9. E. (Euelephas) Hysudricus (Fale.§ Caut.)|
(f.) Colliculi approximati, me- (10. E. (Eueleph.) antiquus (Fale.) . !
dio leviter dilatati, ma- | 11. E. (Eueleph.) Namadicus (Fale. §-Caut, )
cheridibus undulatis .
7 (12. E. (Eueleph.) Columbi (Failc.)
3 3 ..
d eB (g.) Colliculi approximati, ma- 13. E. (Eueleph.) Indicus (Linn.)
a0 8 “iF :
2s ae evi is ya ioe wT 14s BL (Hencloph:) Armeniaeus (Fale.)
ne atis .

(h.) Colliculi confertissimi, ada-
mante valde attenuato,
macheridibus yvix undu- 15. E. (Eueleph.)
latis

1 The so-called Mastodon Australis of Owen Dr. Falconer subsequently (as indeed he had done in 1846; see }
vol. i. p. 106) referred to M. Andium instead of to M. Humboldtii. (See notes to memoir on Elephas Columbi. )—{Ep. 1]
2 Extract of letter from Dr. Falconer to M. Lartet, December 18, 1856 :—‘ I have discovered a new species of |
Trilophodon from India, Triloph. Pandionis. It is not from the Sewalik Hills. It is allied to Afasiodon |
angustidens and M. Humboldtii’ (See also vol. i. p. 124.)—[ED.] |

primigenius (Blwmb.)

| Geological Age

| Pliocene .
| Upper Miocene
Post-Pliocene .

Upper Miocene
| Post-Pliocene ?

Pliocene?
Upper Miocene
Miocene .
Pliocene ?
Miocene .
Pliocene .

Miocene .

Miocene .

Miocene .
Miocene .

Miocene and

Pliocene

Miocene .

Pliocene .
Pliocene .
Existing .
Miocene .

Pliocene .
Pliocene .

Post-Pliocene ?

Existing .

Upper Miocene j

OF MASTODON AND ELEPHANT.

Country

France ; Piedmont
France ; Switzerland
North America

(France; Germany; Switzer-
land wee ;
France

i

Post-Pliocene .

South America

Southern India .
Germany: Eppelsheim .
Southern India; Ava

South America

Southern and Western India .
England; France; Italy
India: Sewalik Hills

Southern India; Ava

India: Sewahk Hills
India: Sewalik Hills

Sewalik Hills and Central
India - " A

|

India: Sewalik Hills

England; France; Italy
England ; Lombardy

Africa .

India: Sewalik Hills

England ; France; Italy

Central India

Mexico; Georgia; Alabama.

India . r

Armenia: Erzeroom

Europe, Asia, and
America

{

North {

Remarks

Syn. Mastodon Buffonis (Pomel)

Syn. M. Turicensis (Schinz, juxta Von Meyer)
Syn. M. maximus (Cuy.); M. giganteus (Auc-
torum). (Pl. I. fig. 2, and Pl. II. fig. 2.)
Syn. M. Simorrense (Lart.); M. Cuviert (Po-

mel). (PI.II. fig. 3, and Pl. III. figs. 3 and 4)

An Syn. M. Australis of Owen (?); of reputed
Australian origin !!

Imperfectly known, but very distinct as a
species; is the only Indian Trilophodon
(Pl. XXXIV. figs. 6 & 7 of vol i.)

(Pl. IL. fig. 8, and Plate III. figs. 1 and 2)

(Pl. II. fig. 9)

Large inferior incisors, one or two. The sub-
genus doubtful. (Pl. I. fig. 3, Pl. I. fig. 4,
and Plate VIII. of vol. 1.)

Hitherto found only in Perim Island, (PI.I. fig.
4, Pl. IL. fig. 5, and Pl. IX. of vol. i.figs. 3-6)

Syn. M. brevirostre (Gervais). (Pl. I. fig. 7,
and Plate IV.)

The only known species indicating a Pentalo-
phodon-type. (Pl. I. fig. 5, Pl. II. fig. 6,
Plate IX. of vol. i, figs. 1 and 2, and Pl. X.
of vol. i.)

Syn. M. latidens (Clift. pro parte). (Pl. II.
fig. 10, and Pl. V. figs. 1 & 2)
(Pl. I. fig. 6, and Plate II. fig. 11)

. |Distinetness as a species doubtful. (Pl. I. fig. 7,

and Pl. II. fig. 12)

Found in both the Valley of the Nerbudda
(Pliocene) and Sewalik. Hills (Miocene).
(PL L.figs. 8, 8 a,& 8b, PL II. figs. 13 & 13 a,
and Pl. V. figs. 3 & 4)

The only Elephant in which premolars have
been met with. (Pl. I. fig. 9, Pl. II. fig. 14,
and Pl. VI. fig. 2)

(Pl. I. fig. 11, Pl. IL. fig. 16, and Pl. VIII.)

Imperfectly known. Fossil remains rare.
Galva)

(Pl. I. fig. 10, Pl. II. fig. 15, and Pl. VI. fig. 1)

(P11. figs. 12 & 12a, and Pl. II. figs. 17 & 17 a)

(Pl. IX.)

Restricted to the Pliocene Fauna of the Valley
of the Nerbudda, Central India. (Pl. I. fig. 13,
and Pl. II. fig. 18)

An Syn. Z&, Jacksoni ? (Silim. Journ. 1838, vol.
xxxly. p. 863). (Pl. X. figs. 1 & 2)

Syn. E£. Swmatranus (Temminck). (Pl. I.
figs. 14, 14a, & 146, PII. figs. 19, 19 a, &
19 4, and Pl. V1. fig. 3)

In the Brit. Mus. Coll. Discovered between
Erzeroom and Moosh in 1856. The molar
plates closely approximated, and the ena-
mel edges very undulated (Pl. X. fig. 3)

(Plate I. fig. 15, Pl. II. fig. 20, and Pl. VI.
fig. 4.)

3 Dr. Falconer was afterwards inclined to think that Mastodon Andium would have to be replaced in the sub-
genus Z’rilophodon, in which he had placed it in 1846. (See vol. i. p. 99, and memoir on Elephas Columbi.)—[ED.]
* Loxodon priscus was afterwards regarded by Dr. Falconer as a form of Zlephas antiquus. E. Melitensis, Falc.
(Plates XI. and XII.), however, or the pigmy Fossil Elephant of Malta, was subsequently added by him to the
subgenus Lo.rodon.—[ED.]

16 BRITISH AND EUROPEAN FOSSIL MASTODONS.

Mastodon, the formula being, Incis. — ; Can. =: Premol.

ar Mol. = =11. It exceeds the rest of the species by 8

molars in both jaws, as M. (Triloph.) angustidens exceeds
M. (Triloph.) Ohioticus. A longitudinal belt of enamel has
not yet been observed on the tusk of any Elephant. The
molars are presented under two forms: in the subgenus
Stegodon as ‘ Dentes complicati,’ resembling those of Masto-
don in the folded form of their crown-eminences, and as
‘Dentes lamellosi’ in Loxodon and Euelephas. The con-
vexity of the crown-ridges, and the absence of the longi-
tudinal, mesial, bipartient cleft, so characteristic of the true
Mastodons, are very constant in the Elephants, the only
exception, limited to the latter character, being indistinctly
seen in an H. (Stegod.) Cliftti. The passage from the Ste-
godons into the Loxodons is effected through H. (Steg.) in-
signis and E. (Lozxod.) planifrons, and from the Loxodons
into Huelephas through EH. (Lox.) meridionalis and EH. (Euel.)
Hysudricus. The anisomerous ridge-formula in Huelephas is
not numerically the same in all the species, being in some
higher, in others lower; but they all agree in exhibiting
progressive increments. The amount of undulation pre-
sented by the worn edges of the enamel plates furnishes a
good means of distinguishing the nearly allied fossil species
in Euelephas.

The distinctive and specific characters of Mastodon and
Hlephant.—A safe criterion by which to test the soundness of
any proposed arrangement in Natural History is, that the
technical characters, however abridged, should be ex-
ponents, so to speak, of the natural and serial affinities, and
in nowise at variance with them. If this test be applied to
the ridge formula, as a consistent basis for the arrangement
of the Mastodons and Elephants, it will, we believe, not be
found wanting: thus the Mastodons ranged under Trilo-
phodon and Tetralophodon include all the Elephantoid species
which have the crowns of the molars comparatively simple
and uniformly divided into two subequal divisions by a longi-
tudinal line or cleft; the ridges limited in number, each with
fewer mammillary eminences, and invariably more or less
concave across; the enamel thick, and in conical or com-
pressed points ; and the valleys between the ridges deep and
empty, or with but a sparing quantity of cement. The
Elephants, on the other hand, as restricted by the ridge-
formula and ranged under Stegodon, Loxodon, and Euelephas,
include all the Proboscidean species which have the crowns
of the molars more complex, and usually wanting in a longi-

DESCRIPTION OF PLATE I.

Representations in outline of crania of Proboscidea: an-
terior views, drawn to scale, one-twentieth of the natural

size. The figures in this Plate have been reproduced from

the original drawings by Mr. Ford in Plates XLII. and
XLII. of the Fauna Antiqua Sivalensis (see vol. i. p. 476),
in which the crania are drawn one-eighth of the natural size.
(See page 14.)

Fig. 1. Dinothertum giganteum. After Kaup.

Fig. 2. Mastodon Ohioticus. After figure in Americ. Phil. Trans.
1838, vol. viii., PJ. III.

Fig. 3. Mastodon Andium. After specimen in British Museum.

Fig. 4. Mastodon Perimensis. After specimen figured in Plate
XXXVIII. of the Fauna Antiqua Sivalensis.

Fig. 5. Mastodon Sivalensis. After specimen figured in Plate XXXII.
of the F. A. S.

Fig. 6. Elephas bombifrons. After specimen figured in Plate XXVII.
of the F. A. §.

Fig. 7. Elephas Ganesa, After Col. Baker’s cranium figured in Plate
XXI. of the F. A. S.

Fig. 8. Elephas insignis. Described by Dr. Falconer as ‘the most
singular and grotesque of all the Indian Elephants ;’ old; after
cranium figured in Plate XV. of F. A. S.

Fig. 8 a. Elephas insignis. After cranium figured in Plate XVIL, fig. 1,
of F. A. S.

Fig. 8 b. Elephas insignis. Very young cranium after that figured in
TPA Ed. VIL, tie. 3.

Fig. 9. Hlephas planifrons. After cranium figured in P]. TX., F. A. §.

Fig. 10. Hlephas Africanus. Existing species.

Fig. 11. Hlephas meridionalis. After Nesti’s drawing of cranium from
the Val d’Arno.

Fig. 12. Elephas Hysudricus. After cranium figured in Plate IV. of
Pras.

Fig. 12 a. Elephas Hysudricus. Cranium of young animal after that
figured in Plate VI., fig. 1, of F. A. S.

Fig. 18. Elephas Namadicus. After that figured in Plate XII, A of
B.eAgae:

Fig. 14. Hlephas Indicus. Existing species; var. Dauntela.

Fig. 14a. Elephas Indicus. Existing species ; var. Mukna.

Fig. 14 6. Elephas Indicus. Existing species; young.

Fig. 15. Elephas primigenius. After Fischer.

VOL. I. a

Vol. Il Plate 1

“4
- ‘G.HFora del. J. Dinkel bth. y WiWest imp
1

DESCRIPTION OF PLATE II.

Representations in outline of crania of Proboscidea : profile
views, drawn to scale, one-twentieth of the natural size. The
figures in this Plate have been reproduced from the original
drawings by Mr. Ford in Plates XLIV. and XLV. of the
Fauna Antiqua Sivalensis (see vol. i. p. 476), in which
the crania are drawn one-eighth of the natural size. (See
page 14.)

Fig. 1. Dinotherium giganteum. After Kaup.

Fig. 2. Mastodon Ohioticus. After figure in American Phil. Trans.
1838, vol. vili., Plate III.

Fig. 3. Mastodon angustidens. After figure in De Blainville’s Ostéo-
graphie. Plate III.

Fig. 4. Mastodon Andium. Brit. Mus. specimen.

Fig. 5 Mastodon Perimensis. After specimen figured in Plate
XXXIX. of F. A. S.

Fig. 6. Mastodon Sivalensis. After specimen figured in Plate XXXIILI.,
fig. 4, of F. A. S.

Fig. 7. Mastodon Arvernensis. After Nesti.
Fig. 8. Mastodon longirostris. After Kaup.

Fig. 9. Mastodon latidens. After specimens figured in Plate XXXL,
fig. 4, and Plate XXX., fig. 6 a, of F. A. S.

Fig. 10. Hlephas Cliftii. After Clift’s specimen.

Fig. 11. Hlephas bombifrons. After specimen figured in Plate XXVIII,
fig. 1, of F. A. 8.

Fig. 12. Hlephas Ganesa. After specimen figured in Plate XXIL, fig. 1,
of F. A. 8.

Fig. 18. Elephas insignis. After specimen figured in Plate XVIL,
fig. 2, of F. A. S.

Fig. 13a. Elephas insignis. After specimen figured in Plate XVIL..
fie. 4? of F. A. S.

Fig. 14. Elephas planifrons. After specimen figured in Plate X., fig. 1,
of F. A. 8.

Fig. 15. Hlephas Africanus. Existing species.

Fig. 16. Hlephas meridionalis. After Nesti.

Fig. 17. Elephas Hysudricus. After specimen figured in Plate V., fig. 1,
of F. A. 8.

Fig. 17 a. Elephas Hysudricus, young. After specimen figured in Plate
VI, fig. 2, of F. A. 8.

Fig. 18. Elephas Namadicus. After specimen figured in Plate XII. B,
fig. 1, of Hl. A. 8:

Fig. 19. Elephas Indicus. Existing species; var. Dauntela.

Fig. 19 a. Elephas Indicus. Existing species; var. Mukna.

Fig. 19 b. Elephas Indicus. Existing species; young individual.
Fig. 20. Elephas primigenius. After Fischer.

VOL. II.

Vol... Piate 2.

GH Fordael J. Dinkellith

Crania & Lower Jaws of Proboscidea Profile views ——

DISTINCTIVE CHARACTERS OF MASTODON AND ELEPHANT. 17

tudinal line of division; the ridges more numerous and less
definite, each being composed of a greater number of mam-
millary or digital points, which are most elevated in the
middle, rendering the ridges convex across, instead of con-
cave; the processes of enamel thinner, higher, and more
divided; and the deep narrow valleys between them entirely
filled up with cement. The limitations of the two genera
agree pretty well with the views generally entertained by
paleontologists regarding them; with the exception, that
the group comprising the collective Mastodon Elephantoides
of Clift, and by some called ‘ Transitional Mastodons,’ | is here
regarded as more properly belonging to the Elephants.

A Synoptical Table? is annexed of the species of Mastodon
and Hlephas, ranged under subgenera, after the manner here
indicated. The species were first determined or adopted
after a careful examination of all the original materials
accessible, in the foreign collections already referred to
(p. 4), or in various museums in the United Kingdom.
They were then arranged serially, according to their relative
affinities, as indicated by the molar teeth; the common cha-
racters were next analyzed, to furnish a key for breaking up
the mass of species into groups representing genera and sub-
genera ; and the Synoptical Table shows the result. It is put
forward as exhibiting a fair representation of the subject, so
far as the materials and state of knowledge at the present time
admit, but with no pretension to being either unexception-
able or complete. The progress of investigation, by the
discovery either of new forms, or of more abundant materials
of the species which are now the most imperfectly deter-
mined, will in all probability modify more or less, or break
down, any generic or subgeneric limitations that may be at
present devised. For the daily experience of every depart-
ment of Mammalian Paleontology tends to show that, while
the characters of species are persistent over wide areas and
through long periods of time, genera are nothing more than
ideal or conventional centres, around which groups of species
are arranged, subject to incessant modifications through the
discovery of new forms. It would be foreign to the main
object of the present communication, and beyond the limits
within which it is necessarily restricted, to discuss in detail
the grounds on which the arrangement is founded. As this
will be done more fully elsewhere, I shall content myself
here with stating them in a general way, and with indicating
where the assailable points are. Although the Mastodon of

Owen’s ‘Odontography,’ p. 624. (See | in outline the skulls of the several

vol. i. p. 68.—Eb.) species of Mastodon and Elephant.—
* In Plates I. and II. also are figured | [Ep.]

VOL. II. C

18 BRITISH AND EUROPEAN FOSSIL MASTODONS

North America and the Mammoth are so widely different in
the form of their molar teeth that they must be ranked
under distinct genera, the intermediate gradations are so
complete as to establish a passage from the one into the
other. Failing the characters of premolars and inferior in-
cisors, previously relied upon as distinctive of the Mastodons,
and abundant cement as distinctive of the Elephants, the
constancy of the ridge-formula in being isomerous (whether
ternary, quaternary, or quinary) in the intermediate molars,
appeared to furnish a sufficient technical demarcation be-
tween Mastodon and Hlephas, and to subdivide the former
satisfactorily into the natural subgeneric groups of T'rilo-
phodon and Tetralophodon. It remains to be seen whether
there is any intermediate species in which the characters of
these two groups are blended.

Mastodon Sivalensis is regarded as having five ridges to
the ‘intermediate molars,’ instead of four; but this remark-
able character being restricted at present to a single species,
it was deemed inexpedient to form a systematic section for
it alone, and it is ranged at the end of the Tetralophodons.

Although a mesial, bipartient, longitudinal cleft along
the summit of the crown is very common in the molars of
most of the species of Mastodon, and usually absent in the
Elephants, there is one species of the former, M. (Triloph.)
Borsoni, in which the cleft is so obsolete, that Isaac Hays!
founded the specific character upon the supposed absence of
this cleft. But the cleft, although but slightly pronounced,
is distinctly present in unworn germ-teeth of this form; and
it is even visible in the original molar described by Abbé
Borson, upon which Dr. Hays relied for its absence.

The plurality of the species in the first subgeneric group
of Elephas, namely Stegodon, are sufficiently distinguished
from the Mastodons by the higher numerical expression of
the crown-formula, in showing 7 or 8 ridges instead of 3 or
4; by the great quantity of laminated cement which fills the
transverse valleys; by the ridges being convex, as in the
typical Elephants ; by the greater number of points to each
ridge; and by the absence of a mesial dividing furrow.
But in one of the species, H. (Stegodon) Cliftu, there is an
obsolete indication of this furrow; and its affinity to the
Mastodon is further evinced by the low or senary expression
of the ridge-formula. This species constitutes a frontier
form, through which the passage between the two genera is
effected; but the details of the other dental characters show
that it is most nearly allied to the Stegodons, and the cha-
racters of the subgeneric group were constructed to admit of

1 Transactions of the American Philosophical Society, ser. 2, vol. iv. p. 334.

DISTINCTIVE CHARACTERS OF MASTODON AND ELEPHANT. 19

its reception among them. ‘Two of the Loxodons, namely, EH.
(Low.) planifrons and H. (Low.) Africanus, have a ridge-
formula which is identical or nearly so with that of Stegodon
imsignis ; but the separation of the group is indicated by the
great increase of vertical height in the colliculi, and by the
layers of enamel assuming the character of plates, instead of
the mastoid eminences of Stegodon. HH. (Lox.) meridionalis
has a higher number of plates in the ‘intermediate molars’
than those two species, and constitutes a frontier form, lead-
ing towards the next group, Huelephas. But the ridge-
formula in this form would appear to be hypisomerous, and
the aggregate characters indicate its position among the
Loxodons. The majority of the species in the group Huele-
phas are well marked by the progressive increments and high
numerical expression of the crown-ridges of the intermediate
molars, by the great vertical height of the colliculi, and the
attenuated plates of enamel. One species among them, E.
(Hueleph.) Hysudricus, constitutes a frontier form leading
towards EH. (Lox.) meridionalis. More ample details respect-
ing the Elephants will be given in the second part of this
memoir, when treating of the European fossil species.

To revert specially to the Mastodons, Trilophodon and Te-
tralophodon (including under the latter the exceptional five-
ridged Mastodon Siwvalensis), as regards the number of forms
at present known, are of nearly equal value, the former in
our view comprising 7, and the latter 6, well-marked species ;
and they are each divisible into two parallel subordinate
eroups, the exact appreciation of the characters of which is
of much service in the determination of the European fossil
species. In the one series, the ridges are broad, transverse,
more or less compressed into an edge, with the valleys open
throughout and uninterrupted by subordinate tubercles:
these are well represented in Trilophodon by M. (Triloph.)
Ohioticus, and in Tetralophodon by M. (Tetraloph.) latidens. In
the other series, the ridges are composed of blunt conical
points, which are fewer in number, more elevated, and flanked
in front and behind by one or more subordinate outlying
tubercles, which disturb the transverse direction of the
ridges and block up the valleys, interrupting their continuity
across. This series is represented in Tvilophodon by the
Miocene European species, M. (Triloph.) angustidens, and in
Tetralophodon by the Pliocene species, M. (Tetraloph.) Arvernen-
sis of the Crag. (See Plates ITI. & IV.) The species with
transverse compressed ridges, in both subgenera, may be
compared with Dinotheriwm as regards their molar crowns,

_and the other series with Hippopotamus.
c2

20 BRITISH AND EUROPEAN FOSSIL MASTODONS.

The European fossil species of Mastodon at present known
are the following,! all of which are of Miocene age, with
the exception of M. (Trilophoden) Borsoni and M. (Tetralo-
phodon) Arvernensis, which are Pliocene :—

M. (Triloph.) Borsoni, Isaac Hays (Pliocene).

M. (Triloph.) Tapiroides, Cuvier.

M. (Triloph.) angustidens, Cuvier, pro parte.

M. (Tetraloph.) longirostris, Kawp.

M. (Tetraloph.) Arvernensis, Croizet & Jobert (Pliocene).

The British fossil Mastodon, and its comparison with M.
angustidens, M. Arvernensis, and M. longirostris.—The remains
of only one species of Mastodon have hitherto been discovered
in the British Isles, in what is called the Older Pliocene
‘Red Crag,’ at Felixstow and Sutton in Suffolk, and in the
Newer Pliocene, ‘ Fluvio-marine,’ or ‘Mammaliferous Crag,’
in various localities near Norwich and in Suffolk. I shall now
endeavour to ascertain what this species is; and as I con-
sider that the question is one of considerable importance, as
a turning-point upon which the independent character of the
British Pliocene fauna hangs—that is to say, whether it is
distinct, or merely a long-lived offset from the Miocene—I
shall not hesitate to enter at length upon the details caleu-
lated to throw light upon the subject.

Professor Owen is the only English palzontologist who has
undertaken to identify and describe in connexion all the
Mastodon remains of the Crag, which he has done very fully
in his valuable work ‘ On the British Fossil Mammalia,’ pub-
lished in 1846. He there designates the species Mastodon
angustidens or Mastodonte a dents étroites of Cuvier; and
gives as synonyms, in his opinion, M. Arvernensis of Croizet
and Jobert, and M. longirostris of Kaup. He heads the
chapter with a woodcut of the upper and lower jaws of the
Eppelsheim WM. longirostris, after Kaup, under the name of
Mastodon angustidens ; and in his description of the dentition
of M. angustidens, in the ‘Odontography,’? he draws his
details of the various teeth indifferently from the three nomi-
nal species above mentioned—namely, M. angustidens, M.
longirostris, and M. Arvernensis. In his memoir on the Crag

1M. Aymard has added largely to | giganteus, Aym. But the specific dis-

the nomenclature of the Proboscidea by
creating a new genus, and new species
for the remains found in the Velay and
Auvergne—viz. Anancus macroplus, as
a generic form distinct from Mastodon ;
and Mastodon Vellavus and M. Vialettii,
regarded by Pomel as synonyms of M.
Borsoni; also a fossil Elephant, £#.

tinction of these nominal species is
exceedingly doubtful (vide Bulletin de
la Société Géologique; and Congrés
Scientifique de France, 1855, p. 276).
The species referred to in a preceding
page as haying been made out by M.
Lartet has not yet been published.
2 Op. cit. p. 619 et seq.

MASTODON ANGUSTIDENS. 21

Mammalia, contained in the 47th number of the ‘Quart.
Journ. Geol. Soe.,’ published in August of the present year
(1857), he reiterates the opinion that Mastodon angustidens
and Mastodon longirostris are synonyms of the English Crag
species. Any opinion emanating from so distinguished a
paleontologist as Professor Owen, and repeated by him after
mature study, at various intervals, between 1843 and 1856,
must necessarily carry great weight with it. The first point,
therefore, to determine is, what is the species to which Cuvier’s
name of M. angustidens is legitimately applicable ?

Mastodon angustidens.—The fluviatile or lacustrine Mo-
lasse of the basin of the Sub-Pyrenees has from a very remote
time been worked, at Simorre, by mines for what was called
the ‘ Turquoise de nowvelle Roche,’ this substance being the
ivory of Mastodon-tusks chiefly, highly injected with a metallic
infiltration, so as to simulate the natural mineral Turquoise. !
The excavations brought to light the numerous Miocene re-
mains found in this rich depét, and among others the molars
of Mastodon. These were vaguely referred to by the old
naturalists under the name of the ‘Animal de Simorre.’ ?
Some of them found their way, in the progress of time, to
the Museum of Natural History in Paris, about the middle
of the last century ; and Daubenton described them under the
title of ‘ petrified teeth having relations to those of Hippopo-
tamus,’ to which indeed, in some important respects, they bear
a very striking analogy. Cuvier, having established his
‘erand Mastodonte’ of North America, next directed his
attention to the Huropean remains of the genus, the first of
which he published under the title of ‘ Mastodonte a dents
étroites,’ or M. angustidens. It has been proved upon the
clearest evidence, by various palzontologists, and admitted,
among others, by his devoted friend and follower Laurillard,?
that Cuvier has included more than one species under this
nominal designation of M. angustidens. Itis requisite, there-
fore, to ascertain precisely what were the original types which
suggested a name of such palpable signification to a shrewd
and philosophical observer like Cuvier. On referring to his
original memoir, it will be found that Cuvier commences, *
as his first illustration, with a description of one of the
Simorre molars previously described by Daubenton. The
second piece is the Dax specimen from near Sort, Départe-
ment des Landes, and obtained from a fluvio-marine Molasse

3 Dictionnaire Universel d’Histoire
Naturelle, tom. viii. pp. 29, 30.

4 Annales du Muséum, tom. viii.
p. 412.

1 Reaumur, ‘Mém. de Acad. des
Sciences, 1715, p. 174; and Lartet,
* Quelques Apercus Géologiques dans le
Département du Gers,’ p. 19.

2 Id. op. cit. p. 24.

22 BRITISH AND EUROPEAN FOSSIL MASTODONS.

formation, probably of the same age as the Simorre lacustrine
beds. The third specimen is a South American fragment,
brought to Europe by Humboldt, which has no connexion
with the European species: on this head all later palzeonto-
logists who have investigated the subject, without exception
(exclusive of mere compilers), are agreed; among others,
Laurillard,' who identifies it with M. Andiwm, as restricted
by him. The fourth specimen which Cuvier quotes is another
Simorre fossil. The sixth, a very important and character-
istic specimen, is from the same locality. Now, all these
Simorre specimens, with the exception of the third—which is
a premolar, and therefore a normal exception—are charac-
terized by having their crowns divided into three principal
ridges. ‘It is therefore,’ as we have elsewhere? stated, ‘ 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.’ (See Plate
ITI. figs. 3 & 4.)

Since the time of Cuvier, Simorre and Sansan have become
classical paleontological ground, through the important dis-
coveries made by M. Lartet of the first-announced fossil
monkey in Hurope, of Macrotheriwm, Anisodon, &. Among
others, a vast quantity of Mastodon remains have been met
with, including the whole dentition, from the young sucking-
calf up to the adult and old animals. A superb skeleton was
disinterred by Laurillard at Seissan, so complete in every
respect, that it has been set up in the Paris Museum, along-
side of the skeletons of the existing Indian and African
Elephants. Two points, which have been invariably exhibited
by all these teeth, are of special importance in their bearing
upon the present question: the first is, that the intermediate
molars are constantly three-ridged, or, in other words, belong
to the Trilophodon type—no Tetralophodon molars having ever,
within the knowledge of M. Lartet, been discovered, either
at Simorre, Sansan, or Lombez; the second is, that they
entirely agree with the original Simorre types described by
Cuvier, upon which his M. angustidens is founded, and that
they are absolutely the narrowest of all known Mastodon
molars. Another remarkable character of the species is this—
that, in harmony with the narrow teeth, the horizontal ramus
of the lower jaw is more compressed, and higher in relation
to the width, than in any other known Mastodon. This is
well shown in the Paris skeleton, and in numerous lower

* Dictionnaire Universel d'Histoire ? Fauna Antiqua Sivalensis, par. i.
Naturelle, tom. viii. p. 29, (1846), p. 57. (See vol. i. p. 90.—Ep.)

DESCRIPTION OF PLATE ITI.

Mastopon (TETRALOPHODON) LONGIROSTRIS AND MastTopon

(TRILOPHODON) ANGUSTIDENS.

The figures in this Plate have been copied from those by
Mr. Ford in Plate XI., vol. xiii., of the Quart. Journ. Geol.

Soe.,

with which the memoir on Mastodon was originally

illustrated. They are rather less than one-fourth (two-
ninths) of the natural size.

Fig. 1.

Fig. 4.

Mastodon ( Tetralophodon) longirostris, Kaup, from Eppelsheim :
plan-view of the penultimate true molar from the left side of the
upper jaw. 4a, anterior talon; ¢, posterior talon ; J, c, d, e, the
four principal ridges which .compose the crowns of the ‘ inter-
mediate molars’ in the Tetralophodons. An irregular longi-
tudinal cleft along the middle divides the crown into an inner

and outer division. (See page 24.)

. The same tooth seen in profile. Froma cast in the Geological

Society’s collection.

. Mastodon (Trilophodon) angustidens, from the Dép. Gers, in the

Sub-Pyrenees: plan-view of the penultimate true molar from the
left side of the upper jaw, showing the worn discs of the three
principal ridges which compose the crowns of the ‘ intermediate
molars’ in the Trilophodons. a, anterior talon; t, posterior
talon; b, c, d, the three ridges. The longitudinal cleft is par-
tially worn out. (See page 22.)

The same tooth, seen in profile. From a specimen in the
collection of M. Lartet, For. Mem. G. §., Seissan, Gers,

VOL. 1.

Vol.IL. Plate 3.

GHForddal J‘Dinkellith, - “W.West imi

1-2. Mastodon (Tstralophodon) longirostris.. 3-4..M. (Tilophodin) amgustidens.

MASTODON LONGIROSTRIS AND M. ARVERNENSIS. 238

jaws contained in the paleontological gallery. M. Lartet
possesses, in his rich collection at Seissan, several lower jaws
exhibiting the same character. <A nearly entire skeleton of
this species was discovered, in the latter part of 1855, in the
sandstone-quarry of Veltheim, near Winterthur, in Canton
Zurich ; this I was enabled to examine minutely through
the kindness of M. Ziegler-Hrnst of Winterthur. It is the
largest specimen of the species that I have anywhere seen.
The lower jaw, although in fragments, is nearly complete,
and shows the extreme compression of the horizontal ramus,
and its great depth. I found, by measurement with the
callipers, that this compression was even greater than is seen
in Dinotheriwm, while the lower jaws of most of the known
Mastodons and Hlephants yield more or less of a circular
section.! This tenuity of form is carried on throughout the
skeleton in the Mastodon of Simorre.

From these remarks it would appear sufficiently evident
that, whether we are guided by priority of description and
reference to the original specimens, or by the obvious signi-
fication of the term, the title of Mastodon angustidens is legi-
timately applicable to the Trilophodon of Simorre, and to no
other species; for it is, par excellence, the ‘Mastodonte a
dents étroites’ of Cuvier. The species, thus limited, has
nowhere been met with in the fossil state in England.

Mastodon Arvernensis and M. longirostris.—Cuvier, as
already stated, included under the name of M. angustidens
other remains which do not belong to it. Upon this head
nearly all the French paleontologists are agreed, although
at variance as to the details. Of the specimens figured in
the four plates devoted to ‘Divers Mastodontes’ in the ‘ Os-
semens Fossiles,’ all those from South America, amounting
to 10 in number, are by common consent referred to one or
two species peculiar to that country. Seven are referable to
the Mastodon of Simorre with narrow molars; one to M.
Tapiroides; five are doubtful, either from inexact know-
ledge as to their origin, or from their undecided characters ;
and all the rest, being eleven in the aggregate, are from Italy,
with the exception of one specimen from Trevoux in France.
It is curious to observe the different views that have been
taken of them. De Blainville? limits the South American
remains to a single species, while Laurillard and Gervais
range them under two well-defined forms. De Blainville and
Owen agree with Cuvier in referring the so-called narrow-
toothed remains from Simorre, Italy, Auvergne, and Eppel-
sheim also, to a single species. Laurillard, devoted as he was

1 See Appendix, p. 73.—[Ep.] ? Ostéographie: Des Eléphants.

24 BRITISH AND EUROPEAN FOSSIL MASTODONS.

to the traditions of his great leader, was compelled by the
evidence to admit two species—namely, M. anqustidens, under
which title he included the Italian, Auvergne, and part of the
Eppelsheim remains; and M. longirostris, under which he
ranged both the principal part of Kaup’s Eppelsheim species,
and the whole of the Simorre remains.! Misled by the undue
importance which he attached to the presence of mandibular
incisors common to the two forms, he sunk the characters
presented by the molars, and confounded ternary-ridged and
quaternary-ridged forms under the same name, although it is
distinctly evident that he was aware that two of the European
species severally possessed 3 and 4 ridges to their inter-
mediate molars, and that the ternary formula was common to
the Mastodons of North America and of Simorre. In 1828,
four years before the demise of Cuvier, Croizet and Jobert?
proposed the name of M. Arvernensis for the Auvergne
remains, as distinct from M. angustidens; and soon afterwards
Dr. Kaup? published his magnificent series of the Eppel-
sheim form as equally distinct, under the designation of M.
longirostris, which has been regarded by Hermann von Meyer
to be identical with M. Arvernensis.4 Lartet> had accurately
determined the milk and permanent dentition (so far as the
true molars are concerned) of the Simorre form as far back as
1847. He assigned three ridges to the last milk molar and
to the antepenultimate and penultimate true molars in both
jaws; and in his ‘ Notice,’ ® published in 1851, he proposes
to distinguish it by the name of Mastodon Simorrense, retain-
ing the designation of Mastodon angustidens for the Italian
and Auvergne remains, characterized by four ridges in the
penultimate true molar, instead of three.’ Lartet at the same
time’ proposed the name of Mastodon Gaujaci for a sup-
posed small form from the same Miocene deposit at Lombez.
Laurillard considered it as furnishing a tonfirmation of the
conjectural species named Mastodon minutus by Cuvier.®
Gervais followed Laurillard in considering the Simorre M.
(Triloph.) angustidens and M. (Tetraloph.) longirostris as be-
longing to the same species, M. longirostris; but adopted
for the Auvergne form the name of M. Arvernensis; and went
a step beyond his predecessors in proposing a new name for
the Mastodon remains found in the arenaceous deposits near

1 Dictionnaire Universel d'Histoire | p. 113.
Naturelle, tom. viii. p. 29. 5 Dictionnaire Universelled’ Histoire,
? Recherches sur les Ossemens | Naturelle, tom. viii. p. 29.
Fossiles du Département du Puy-de- ® Notice sur la Colline de Sansan,
Dome, p. 133. p. 24.
° Ossemens Fossiles de Darmstadt, 7 Op. cit. p. 27.
8 Dietionnaire Universel d’ Histoire

pt. iv.
* Nova Acta Acad. Nat. Cur. vol. xvii. | Naturelle, tom. viii. p. 31.

MASTODON LONGIROSTRIS AND M. ARVERNENSIS. 25

Montpellier, which he identifies with the Mastodon of the
Astesan and the Val d’Arno under the name of M. breviros-
tre! Pomel, in his memoir of 1848, proposes a new name for
the Simorre Trilophodon, namely M. Cuviert, and he retains
that of M. angustidens for the Auvergne and Italian forms,
admitting their distinctness from the M. longirostris of Eppel-
sheim.? In his ‘Catalogue Méthodique’ of 1854, he adopts
the name of M. Arvernensis for the Auvergne and Montpellier
form, to which he assigns the additional foreign localities of
the Val d’Arno, Piedmont, and the Crag in England; but in
a remark on the next page, he reiterates the view expressed in
his previous memoir, that he has retained the name of M.
angustidens for the species of Italy.? Nesti,‘ in his des-
cription of the Tuscan remains, adopts the name of M.
angustidens (Mastodonte a denti stretti) in the loose compre-
hensive sense in which it was used by Cuvier; while Eugenio
Sismonda, aware of the various and contradictory opinions
upon the point, guardedly described the fine skeleton found
at Dusino, in Piedmont, under the title of ‘Osteographia di
un Mastodonte angustidente.’°® My friend and collaborateur,
Colonel Sir Proby Cautley, in 1836, figured and described
some teeth of the Indian species, to which we subsequently
restricted the name of M. (Tetralophodon) Sivalensis, as iden-
tical with the ‘ Mastodonte a dents étroites’ of Cuvier; and he
expressed at the same time the opinion, that the Italian form
(which he had more particularly in view) would, with the
Sewalik one, constitute a subgenus of the Angustidens type,
in contradistinction to the type of Clift’s M. latidens.®

These, so far as I am aware, are the leading opinions which
have been put forward by original writers on this much-
disputed question. Those which have been expressed by the
compilers of systematic works on Paleontology, however use-
ful, are of little weight in the discussion, as they express more
the balance of the authorities numerically, than opinions
formed upon independent examination of the subject by
themselves. The specific name Mastodon angustidens is even
struck out of the list of Huropean species, except as a
synonym, in the last edition of Bronn’s ‘ Lethea,’ and
replaced by the terms M. Arvernensis, M. longirostris, and M.
Cuviert.’ Paleontologists would confer a great boon on
Geology, if they could be brought to agree in applying this

1 Annales des Sciences Naturelles, | tom. xii. pp. 17-34.

3me série, tom. v. p. 268. 5 Mem. del Reale Acead. di Torino
2 Bull. de la Soc. Géologique, (1848), (1851, pp. 175-235.

tom. v. p. 257. § Journ. of the Asiat. Soe. of Bengal,
3 Catal. méthod. et descript. pp. 74, | vol. v. p. 294. (See vol. i. p. 126.—Ep.)

75. 7 Lethea Geognostica, 3rd edit. vol.

‘Nuovo Giorn. di Letterat. (Pisa), ' ii. pp. 827-832 (1856).

26 BRITISH AND EUROPEAN FOSSIL MASTODONS.

name (M. angustidens) to the Simorre form, for which it was
devised by Cuvier.

The views which we entertain were partially disclosed in
the first part of the letter-press of the ‘Fauna Antiqua
Sivalensis,’ and fully elucidated in the four plates of outline
heads (from Plate XLII. to Plate XLV. of Part 5 of the Illus-
trations),! 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 distinct species, namely :

M. (Triloph.) angustidens.
M. (Triloph.) Andium.

M. (Tetraloph.) longirostris.
M. (Tetraloph.) Arvernensis.

The only change which subsequent investigation on fresh
materials has led us to make, is to transfer M. Andiwm from
the subgenus Trilophodon into that of Tetralophodon, for
reasons which it is not necessary to detail on the present
oceasion.2 Of these forms, the only one which I believe has
been met with in the fossil state in England is M. (Tetralo-
phodon) Arvernensis (Plate IV.) ; and I shall now proceed to
the consideration of the evidence in support of this conclusion.

British specimens of Mastodon.—Remains of two out of the
three species of Mastodon with which we are chiefly concerned
now-—viz., M. (Trilophodon) angustidens, M. (Tetralophodon)
longirostris, and M. (Tetralophodon) Arvernensis—have been
discovered on the Continent, in the localities where they
prevail, in such a perfect condition, that very little remains
to be desired in regard to their entire osteology. 'The skele-
ton of M. (Trilophodon) angustidens from Seissan, set up in
the Gallery of Comparative Anatomy in Paris, is so complete
in every_respect, from the cranium down to the digital pha-
langes, that it may be compared, bone for bone, throughout
the frame, with the skeletons of the African and Indian
Elephants which adjoin it. Of M. (Tetraloph.) Arvernensis,
a nearly entire skeleton was disclosed by a railway excavation
at Dusino, near Asti in Piedmont, and is now deposited in
the Turin Museum. It is deficient only in the cranial por-
tion of the head, right hind-leg, part of the scapula and
pelvis, and some of the bones of the carpus and tarsus. The
upper and lower jaws, with the tusks entire to their tips, are
preserved ; and Prof. Sismonda was only deterred by the
brittle condition of the bones from attempting to reconstruct
the whole. A skeleton of the same species, nearly as perfect,
which I have examined, was discovered in the lower Val

1 See Plates I. & I. of this volume, 2 See Note 8 to Synoptical Table.—
and page 476 of vol. i.—[Ep.] [ Ep. |

ee ep a a

DESCRIPTION OF PLATE IV. -

MasrTovon (TETRALOPHODON) ARVERNENSIS FROM THE CRAG.

The figures in this Plate have been copied from those by
Mr. Ford in Plate XII., vol. xiii., of the Quart. Journ. Geol.
Soc., with which the memoir on Mastodon was originally
illustrated. They are rather less than one-fourth (two-
ninths) of the natural size.

[The letters to the figures refer to the same parts as in fig. 1, PJ. HI. ]
Fig. 1. Mastodon (Tetralophodon) Arvernensis, from Ramsey, near

Fig

Harwich: plan-view of the germ of the penultimate true molar
from the right side of the upper jaw. (See page 33.)

. The same tooth, seen in profile. A large flanking mammilla is

seen to occupy the middle of each valley. This specimen is in
the collection of the Rev. J. R Marsden, Great Oakley, Essex.

. Mastodon (Tetralophodon) Arvernensis, from Suffolk: plan-

view of the germ of the last true molar from the left side of the
lower jaw. (See page 31.) 0b, ¢, d, e, f, the five ridges
composing the crown, the mammille of which are disposed

alternately.

. 4, The same tooth, seen in profile.

[Figs. 3 and 4 are drawn from a cast in the Geological Society’s
Museum ; the ‘ bourrelet’ being partly restored from a Crag
molar of similar age, also in the Society’s collection. ]

7

VOL. Il.

a

Vol. IL. Plate 4:.

GE-Ford dal. J.Dinkel lith. W.West imp.

Mastodon (Tetralophodon) Arvernensis, from the Grag.

MASTODON ARVERNENSIS. 27

d’Arno in a marine stratum, along with the skeleton of a
Whale. It is now laid out in the Museum at Florence,
together with numerous other bones of the same species.
From the Miocene sands of Eppelsheim, Kaup disinterred
the upper and lower jaws, with an immense quantity of
molars, showing the entire dental series, milk and adult,
besides various other portions of the skeleton, of M. (Tetralo-
phodon) longirostris. The materials are therefore so abundant
now, that it is in a measure easy to institute a comparison,
more or less rigorous, between the three species.

But as regards the English remains of Mastodon, it is quite
the reverse. Only solitary teeth detached from the jaws, or
part of a mutilated young cranium, have hitherto been de-
scribed, and the teeth in most cases are mutilated. The
beautiful vignette which heads the chapter upon Mastodon
angustidens in the ‘ British Fossil Mammalia’ would convey a
very exaggerated notion of the English remains as they are
ordinarily met with, but that the author takes care to apprize
his readers that it is derived from Kaup’s figure of the
Hppelsheim species. No good specimen of the lower jaw, so
far as Iam aware, has yet been found in Britain; nor have
any of the large bones of the extremities been identified,
although it is more than probable that such do exist in the
numerous collections which have been formed in Norfolk and
Suffolk. The pieces are usually more or less mutilated; and
it is clear that the bones have been broken up before the
fragments were deposited in the strata where they are now
found. Nothing approaching the remains of a perfect skele-
ton has been seen in any one locality, with the exception of
the notable case recorded by the Rev. J. Layton, in which the
entire skeleton of a Mastodon is stated to have been found
lying on its side, stretched out between the chalk and gravel,
at Horstead near Norwich, on a bed of marl. The bones in
this instance were heedlessly broken up by the workmen, or
dispersed before any steps could be taken for their pre-
servation.!

The molars or other fragments occur scattered and
detached. Prof. Owen mentions a well-preserved atlas of
(apparently) Mastodon angustidens as being preserved in the
Ipswich Museum.? Mastodon molars have been found both
in the Red Crag:of Suffolk and in the Fluvio-marine Crag of
Norfolk and Suffolk; in the former at Sutton and Felixstow—
in the latter at Postwick, Whitlingham, Thorpe, Horstead,
and Bramerton near Norwich, and at Easton near Southwold.
Mr. Charlesworth, in reference to their supposed raity.

' Fairholme’s ‘ Geology of the Serip- ? Quart. Journ. Geol. Soe. vol. xii.
tures,’ p. 281. p. 228.

28 BRITISH AND EUROPEAN FOSSIL MASTODONS.

mentions that, within the twelve months preceding September
1851, upwards of a dozen of Mastodon molars had been dis-
covered, in washing the Crag to get out the phosphatic
nodules.!_ Prof. Owen notices their occurrence in the Crag-
pits of Suffolk.2 I am not aware that they have yet been
discovered in the Fluvio-marine Crag of Bridlington in
Yorkshire, nor in any of the freshwater deposits below the
Drift, where remains of Elephant and Hippopotamus are
more or less abundant.

It is no part of the object of this communication to describe
the numerous remains of Mastodon from the Crag, that are
to be met with in different English collections. All that is
intended is to determine what the species really is, and only
such characteristic specimens will be referred to as exist
either in original or as casts, in public museums, or as have
been so accurately figured and described in works of authority
as to be susceptible of satisfactory identification.

First, as regards the Molars.—The most perfect specimen
that has yet been discovered, is the famous Whitlingham
tooth, which forms the frontispiece of Mr. W. Smith’s ‘ Strata
Identified,’ and of which (reversed) a beautiful woodcut is
given in fig. 97 of the ‘ British Fossil Mammalia.’ It is also
very carefully represented, unreversed, both as regards the
plan and profile views of the crown, in the ‘ Fauna Antiqua
Sivalensis’ (Plate XXXVI. figs. 8 and 8 a).? It is the last true
molar (upper jaw, right side), being 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. Hach ridge consists of about two pairs of thick, high,
conical mammille, with very thick enamel. Deep clefts or
valleys intervene between these ridges; but the valleys,
instead of being transverse, are interrupted in the middle by
one or more large accessory conical mammille, which are
interposed between the ridges, and alternate with the outer
and inner divisions. These mammille are usually connected
with the inner division of each ridge in the upper jaw, and
with the outer division in the lower. They are much thicker
than in the other species of Mastodon which possess them, and
have a large conical core of ivory. The consequence of this
complex composition of the crown is that, when the ridges
have been worn down by continued grinding, they present a
ereat number of distinct alternate trefoil discs, surrounded by
a ring or belt of enamel, instead of the single or double trans-
verse disc exhibited by the Mastodon of Eppelsheim, M.
(Tetralophodon) longirostris, or by the M. (Trilophodon)

1 Warren, ‘On Mastodon giganteus, 2nd edit. p. 204. 2 Op. eit. p. 228.
3 See vol. i. p. 468.—[ Eb. ]

MASTODON ARVERNENSIS. 29

angustidens of Simorre. In the Eppelsheim species, of which
T have carefully examined all the molars contained in the rich
collection at Darmstadt, there is a considerable range of
variety, as regards the accessory tubercles which flank the
main ridges. In some of the last molars, the main ridges are
perfectly free from any outlying or flanking mammille; they
are regularly transverse, and the valleys between are equally
transverse, and open throughout. In others, the ridges bear
numerous small warty tubercles of enamel, which jut into the
valleys and distort them. But the transverse continuity of
the valleys is never blocked up by the large conical mastoid
mamumille, which in the molars of M. (Tetralophodon) Arver-
nensis invariably alternate with the main ridges, and reduce
the valleys to disconnected gorges, occupying the outer and
inner sides of the crown. The accessory tubercles in the
Eppelsheim species are so unimportant, that their only effect,
after advanced wear, is to expand the diameter of the disc,
or communicate to it something of a trefoil pattern. The
discs are always transverse, while in the English Crag Mas-
todon they are invariably more or less alternate. In illustra-
tion of this very obvious differential character in M.
(Tetralophodon) longirostris, the beautiful series of figures given
by Kaup in the ‘Ossemens Fossiles de Darmstadt,’ from
Plates XVI. to X XI. inclusive, may be quoted; but I would
more especially refer to figs. 4 and 5 of Pl. XVI., figs 1, 3, 4,
and 9 of Pl. XVIIL., and figs, 2, 6, and 7 of Pl. XXI.

We have endeavoured to exhibit these differences in a
well-marked and obvious manner, by contrasted figures
(drawn with the greatest care and fidelity by Mr. George
Ford) of the same tooth in three species, placed side by side,
in Plate XXXVI. of the ‘ Fauna Antiqua Sivalensis.’ Figs. 6
and 6 a! represent in plan and profile the last molar (upper jaw,
left side) of M. (Tetralophodon) Sivalensis, an Indian fossil spe-
cies, which is the most nearly allied to the English Crag Mas-
todon, so far as the alternate disposition of the crown-mam-
mille is concerned, but differing in the ridge-formula. Figs. 8
and 9 represent two specimens of the same tooth of M. (Tetra-
lophodon) Arvernensis, the one being Mr. William Smith’s
Whitlingham fossil, and the other (fig. 9) Captain Alexander’s
specimen, dredged up between Southwold and Haston, of
which there is a cast in the Geological Society’s Museum.
Figs. 12 and 13 represent two specimens, of different races
or sexes, of the same tooth of M. (Tetralophodon) longirostris,
from Eppelsheim. The Sewalik molar (fig. 6) exhibits six
ridges and a hind ‘talon;’ the Crag and Eppelsheim molars

1 See vol. i. Plate ix. figs. 1 and 2.—[ Ep.]

30 BRITISH AND EUROPEAN FOSSIL MASTODONS.

show only five ridges and the ‘talon.’ In the Eppelsheim
teeth (figs. 12 and 13), the crown is broad; the mammille are
thicker in proportion to their height; the ridges are less
elevated, and consist of a greater number of coronal points
(there being often as many as six or seven to each ridge) ; the
outer and inner lines of points converge less towards the apex
of the crown as they rise upwards; and the valleys are either
entirely open and transverse, or interrupted only by an insig-
nificant amount of warty tubercles. In the Crag fossils (figs.
8 and 9), the crown is narrower in proportion ; the mammille
form more attenuated cones, and are more elevated; the
ridges consist of fewer coronal points, which, instead of run-
ning across in a wide crest, appear, so to speak, as if they had
been squeezed together, and their transversality disturbed ;
the outer and inner lines of points, especially the latter, con-
verge rapidly as they ascend, rendering the apex of the crown
much narrower than the base; the outer and inner divisions
of the crown are more or less alternate, and the vallicular
mammille that flank and alternate with them are large
conical points, which yield discs of wear approaching in size
to those of the principal points; the valleys are completely
obstructed by these mammille, and reduced to a gorge on
either side of them. When the teeth are viewed in profile
(such as fig. 8, on the above-mentioned plate, compared with
fig. 13), the difference is very marked, the latter yielding a
series of salient and re-entering angles, corresponding with
the prominent points and valleys, which the former does not,
the re-entering angles being intercepted by a dark shade,
which represents the accessory mammille. If the eye is next
directed to figs. 6 and 6 a, the differences are stlll more
marked, M. (Tetralophodon) Swalensis exhibiting a greater
amount of alternation of the crown-mammille, and more
complexity of pattern, than is even seen in the English Crag
Mastodon. To summarize the distinctive characters of the
two HKuropean species, it may be stated, that M. (Tet.)
Arvernensis (Pl. IV.), with M. (Tet.) Sivalensis, resembles the
Hippopotamoid type, and that M. (Tet.) longirostris (Plate ITI.
fig. 1), with the Indian species of M. (Tet.) latidens, the Dino-
therian type, in so far as the form of the crowns of the molar
teeth are concerned.'

1 Tn opposition to this view, Kaup has
lately endeavoured to show that his own
species, Mastodon longirostris, is merely
the adult form of M. Arvernensis (Beit-
rage, genus Mastodon, 1857). Herr Ed.
Suess, however, of the Imperial Museum
of Mineralogy of Vienna, thus writes, in
a letter to Mr. Rupert Jones, respecting
Dr. Falconer’s view:—‘I agree wholly

with Dr. Falconer in distinguishing’ the
Italian Tetralophodon from the Mast.
longirostris of Eppelsheim, and believe
this latter one to be quite as good a
species, although the author (Kaup) him-
self has thought good to cancel this
species and to unite both the Tetralopho-
dons of Europe into one species in his
last book.’—[Eb. |

MASTODON ARVERNENSIS. 31

Of the last true molar of the lower jaw, no good entire spe-
cimen, so far as I am aware, has yet been published as having
been yielded by the English Crag. But in the Museum of
the Geological Society, there is a cast of a very fine specimen
of this tooth from the left side of the lower jaw, which,
according to the label on the cast, was found in the Crag of
Suffolk (see Pl. IV. figs. 3, 4). It is a nearly unworn germ,
measuring about 84 inches long, by 3 inches in width in
front, without fangs, and the anterior ridge alone being
slightly touched by wear. It is composed of five ridges and a
talon of two mammille. The anterior ridge shows two pairs
of mammille; the next four ridges present only two large
conical mammillee each, which converge rapidly towards the
summit of the crown, and are disposed in an alternate man-
ner. One or more large accessory mammille are interposed
between the ridges, blocking up valleys in the manner
described as characteristic of the species, and the ridges are
inclined with a slope, which increases successively backwards.
The talon appears to have been composed of a pair of points,
one of which is mutilated on the inner side, and a small
portion of the back end of the tooth is wanting. The slope
of the posterior ridges is so pronounced as to approach nearly
to the character of ‘imbrication.’ In this respect the spe-
cimen closely resembles the Val d’Arno molar figured by
Cuvier (Divers Mastodontes, Pl. IV. fig. 7), which he describes
as the last of the upper jaw;' but it would seem, from the
form and contour, to be an entire germ of the last inferior
molar, and, in our opinion, of the same species as the Crag
Mastodon, namely, M. (Tetraloph.) Arvernensis.

A fragment composing the posterior half of the last
inferior true molar has been noticed and figured by Mr. 8S.
Woodward.? It is composed of seven prominent conical
mammille, disposed in three ridges, which contract very
much behind, and terminate in an odd _ talon-tubercle.
These tubercles form two lines, an outer and an inner, and
they are placed in regular alternation with each other. A
corresponding fragment, of which there is a cast in the
Geological Society’s Museum, is represented by Pl. XXXVIL
fies. 9 & 9a of the ‘Fauna Antiqua Sivalensis.? The mammilla,
in this case also, form two alternate rows, each ridge being
composed of a pair of points.®

The finest detached specimen of the Crag species that I
have anywhere seen is a last lower molar, left side, found
below the citadel of Montpellier, and which has been figured

1 Oss. Foss. 4to edit. tom. i. p. 258.

2 Mag. of Nat. Hist. (1836), vol. ix. p. 152, fig. 22.
3 See vol, i. p. 470.—[Ep. ]

82 BRITISH AND EUROPEAN FOSSIL MASTODONS.

and described by Gervais.! Casts of this piece are to be met
with in many of the principal European Museums. It is of
large size, being about 94 inches long by 3} inches wide at its
anterior end, and it consists of five principal ridges and adouble
talon. The five anterior ridges are well worn, and exhibit
the characteristic alternation of the discs ina very prominent
manner, nothing approaching which has ever been seen in an
Eppelsheim specimen. M. Gervais refers this tooth to his
Mastodon brevirostre, as distinguished from Mastodon Arver-
nensis ; but the grounds upon which he has attempted this
separation are wholly insufficient.

The same peculiar characters in the alternate disposition
of the mammille of the crown are finely exhibited in the last
lower molar of the nearly allied Indian M. (Tetraloph.) Siwa-
lensis, as represented in Pl. XXXVII. figs 8 & 8a of the
‘Fauna Antiqua.’? The specimens resemble each other so
closely, that my colleague Sir Proby Cautley,* in his earliest
description, considered them to be identical species.

Tf, on the other hand, we examine the equivalent teeth in
the lower jaw of M. (Tetralophodon) longirostris of Eppelsheim,
the same differences as occur in the upper molars are con-
stantly presented. The ridges are transverse, and the coronal
eminences ina greater number than a pair of mammille,
which latter, on the outer and inner sides, are opposed, and
not alternate. Figs. 1, 2, & 3 of Pl. XTX. of Kaup’s ‘ Osse-
mens Fossiles’ furnish excellent illustrations.

When these molars, upper and lower, are ground down by
well-advanced use, the alternation of the discs in the one
species, and their transversality in the other, become still
more conspicuous. The former character is exhibited in a
very marked manner by figs. 1, 3, & 6 of Pl. IV. of Cuvier’s
‘Divers Mastodontes,’ above referred to. All the specimens
are of Italian origin, being either from the Val d’Arno or
from the plains of Lombardy and Liguria. Cuvier remarks
upon one of them (fig. 6), that it is ‘remarquable par des fes-
tons plus nombreux que dans les autres.’* They are all refer-
able to M. (Tetralophodon) Arvernensis. The alternate dis-
position of the discs of wear is also seen well in the speci-
mens of M. (Tetraloph.) Arvernensis discovered in the Astesan
near Dusino, and figured and described by Sismonda ;°
while the transverse discs of the HEppelsheim species are
more or less apparent throughout Kaup’s Illustrations, and

1 Annales des Sciences Naturelles, 4 Oss. Fossiles, 4to edit. tom. i.
3me sér. tom. y. p. 268, and Paléontol. | p. 259.
Frangaise, p. 38, tab. 11. fig. 7. 5 Osteograph. di un Mastodonte an-
2 See vol. i. p. 470.—[Ep. | gustidente, tab. 1, figs. 2 and 3 (Mem.

$ Journal Asiatic Society of Bengal, | R. Accad. Se. Torino, sér. 2, vol. xii.)
vol. v. p.294. (See vol. i. p. 126.—Ep.)

Sah 2

MASTODON ARVERNENSIS. 33

more especially in figs. 4 & 5 of Pl. XVL., figs. 1, 2, 3, & 5 of
Pl. XIX., and figs. 2 & 6 of Pl. XXI. of his work above quoted.
Of the other true molars, 7.e. the antepenultimate and
penultimate, various specimens, more or less perfect, have been
yielded by the Crag. Several existed in Mr. Robert Fitch’s
interesting collection at Norwich when I examined it in 1846,
and probably a considerable addition has been made to it
since. Two of these are figured in the ‘ British Fossil Mam-
malia’ (pp. 280, 281). Fig. 98 is described by Professor
’ Owen as the penultimate upper. The anterior portion is
broken off ; what remains of the crown shows four ridges and
atalon. But for the position assigned to it by so able and
practised a paleontologist, the figure would convey the im-
pression of its being a lower instead of an upper molar, from
‘the narrowness of the crown in comparison with the width,
and from the form of the hind talon. Fig. 99 represents a
corresponding penultimate lower molar, also from Mr. Fitch’s
collection. Both teeth—the one of which has the crown re-
presented in plan, the other in profile—show in a strongly
marked manner thecharacteristic alternation of the mammille,
which is never seen in the corresponding molars of the Eppel-
sheim species. Moreover, the mammille are more elevated,
their conical isolation more defined, and the enamel layer
thicker than in M. (Tetraloph.) longirostris. There is a pecu-
liar wavy and finely grooved rugosity of surface, which is seen
on the enamel near the basis of the crown and ‘bourrelet’
where it exists in the molars of the Crag Mastodon (see PI.
IV. fig. 4), and of the nearly allied Indian species M. (Tetra-
lophodon) Sivalensis. It may be compared to the appearance
yielded by a bound book when the edges of the leaves slightly
overlap, and they are bent in a flexuous curve. This peculiar
rugosity is not nearly so conspicuous in the Eppelsheim
species, nor in the M. (Trilophodon) angustidens of Simorre.
The finest English specimen of one of the ‘ intermediate
molars’ of the Crag Mastodon that has come under my obser-
vation is a germ of the penultimate true molar (upper jaw,
right side), lately discovered by the Rev. Mr. Marsden in the
bed of coprolitic or phosphatic nodules in the parish of Ramsey
in Hssex, about three miles west of Harwich, and kindly
lent to me for description. Itis represented (about two-ninths
of the natural size) by figs. 1 & 2, Pl. IV. of the accompany-
ing illustrations. It consists of the shell of the crown quite
entire, the nucleus of the ivory core not having become fully
calcified, nor any of the fangs developed. The crown pre-
sents four intact ridges, with a front and a back talon. The
mammille of the outer and inner lines are very high, and
converge as they ascend, so that the apex of the crown is
VOL. II. D

34 BRITISH AND EUROPEAN FOSSIL MASTODONS.

much narrower than the base. Two large outlying mammille
are interposed between the first and second ridges, one
between the second and third, and one between the third and
fourth. A large tubercle, flanking the inner division of the
first ridge, forms the commencement of the anterior talon.
The posterior talon consists of a line of about six low tuber-
cles. The intermediate flanking mammille, as is usual in
the species, interrupt the transverse continuity of the valleys,
which are reduced to an outer and inner gorge. It is mani-
fest that, if the crown were ground down by wear, the dis-
position of the tubercles is such that a series of trefoil discs,
more or less alternate, would be the result. The dimensions
of this specimen are—length 4°9 inches, width of crown in
front 2-6 inches, width at the last ridge 2°9 inches.

Premolars.—That vertically successional teeth replace one
or more of the milk molars in M. (Tetraloph.) Arvernensis
has been proved by the original specimens from Auvergne,
upon which the species was founded by Croizet and Jobert.
Fig. 7 of Pl. XI. of their work! exhibits an upper jaw of a
very young animal, containing the antepenultimate and penul-
timate milk molars in situ, the former consisting of two pairs
of points, disposed in two ridges, the latter showing three
ridges. Behind the second tooth there is introduced, in the
figure above referred to, a germ-fragment consisting of two
ridges (marked A), as if of the third milk molar; but Abbé
Croizet states, in the descriptive details, that the fragment
was found detached, and that for various reasons, which are
detailed, he considers it to be incorrectly placed in the figure.
In the original specimen, which I had an opportunity of
examining at Paris, the remains of part of the alveolus of a
vertical premolar were distinctly visible above the penulti-
mate milk molar; and M. Laurillard informed me that he
had seen the germ of this premolar, the tooth ‘ A’ above re-
ferred to being the one in question, 7.e. the penultimate pre-
molar. The last premolar, which is the vertical successor of
the last milk molar, has not as yet been observed im situ, so
far as I have had the means of ascertaining.

No premolars of the Crag Mastodon, from English locali-
ties, have either been figured or described in the ‘ British
Fossil Mammalia,’ or elsewhere, up to the present time.? This
is of little moment, in so far as the mere identification of the
species is concerned. Premolar specimens may probably be
found either in Mr. Fitch’s or in some other of the Norfolk
collections.

1 Oss. Fossiles du Départ. du Puy-de- | Fitch, as cited in the ‘British Fossil

Dome, pp.134, 135. Mammalia,’ p. 290, is not a premolar of
2 The specimen referred to by Mr. ! the Crag Mastodon.

MASTODON ARVERNENSIS. 35

Milk molars.—Of the milk series it is not necessary on the
present occasion to enter on many details. I will refer only
to one or two characteristic specimens. The most perfect
and instructive yet met with was discovered in the Crag at
Postwick by Mr. Wigham, to whom I was indebted for the
means of comparing it carefully with a corresponding speci-
men of M. (Tetraloph.) longirostris from Eppelsheim, belonging
to the Harl of Enniskillen. It consists of the left upper jaw
of a calf Crag Mastodon, with the last milk molar beautifully
preserved in situ, and the remains of the empty alveolus of
the penultimate milk molar in front of it. The tooth is
stated, in Sir Charles Lyell’s memoir on the ‘ Relative Ages
of the Tertiary Deposits of Norfolk,’! to be the ‘ second true
molar.” But itis really the last milk molar. He adds:—
‘This fragment was sufficiently perfect to enable Mr. Owen,
to whom I submitted it, to refer it to Mastodon longirostris, a
species also found at Eppelsheim.’ The crown measures 3
inches long by 1°8 inch broad, and is composed of four ridges,
with a front and hind talon, and a well-pronounced basal
‘bourrelet.? The three anterior ridges are more or less worn,
especially along the inner division; the last ridge is nearly
intact. Two views of this tooth, drawn with the utmost care
by an artist of well-known power and fidelity, Mr. George Ford,
are shown in the ‘ Fauna Antiqua Sivalensis,’ Pl. XXXVI.
figs. 7&7a.? The ridges are seen to be connected by one or
two stout conical mammille, which occupy the middle of the
valleys, interrupting their transverse continuity, and alter-
nating with the divisions of the main ridges, in the manner
characteristic of the older or true molars previously described.
If these figures are compared with Pl. XL. figs. 6 & 6 a? of the
same work, by the same artist, which represent Lord Ennis-
killen’s very beautiful specimen of the young calf Mastodon
from Eppelsheim, the distinctive characters of the two species
will be found to be carried on throughout. The Eppelsheim
specimen is a little younger than the Crag fragment ; it shows
the series of three milk molars in situ. The third milk molar
is nearly intact; the four ridges of which it is composed 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 nowise intercept their transverse continuity.
The back talon forms a low transverse free ridgelet, as in M.
(Tetralophodon) latidens of India ; while in M. (Tetralophodon)
Arvernensis the talon tubercles are huddled together and
accrete to the last ridge.

' Mag. of Nat. Hist. (1839), p. 318. 8 See vol. i. p. 472.—[Ep., }
® See vol. i. p. 467.—[Ep.]
D2

36 BRITISH AND EUROPEAN FOSSIL MASTODONS.

Sir Charles Lyell, in his ‘Manual,’! gives a figure of Mr.
Wigham’s tooth of the natural size, in which a very notable
character of the young molars is well brought out. The
enamel of the mammillz is seen to be furrowed vertically by
numerous deep parallel grooves, presenting the appearance
of a reeded column or of a number of cords pressed close
together, and disposed around a thick central axis. The
shell of enamel shows as if it were composed of distinct
narrow pieces glued together. The same character is at-
tempted to be represented by b & ¢ of fig. 7, Pl. XXXVI. of the
‘Fauna Antiqua Sivalensis,’ also of the natural size. It does
not occur in the corresponding young molars of M. (Tetralopho-
don) longirostris. The enamel-surface in these is superficially
wrinkled and furrowed with numerous irregularities, without
however exhibiting the symmetrical fluting observable in the
Crag Mastodon. So conspicuous is this character, that I
believe that the young teeth of the two species could be
distinguished from each other by portions of their enamel
alone, occurring mixed in a collection. I would refer to a
figure given by Kaup of the dentition of a young Mastodon
longirostris in Pl. XVI. fig. 1 of his ‘Ossemens Fossiles,’ for
the comparison. It is of the natural size, and the last milk
molar may be contrasted with the corresponding tooth of the
Crag species figured by Sir Charles Lyell. It was this
character in the young teeth which chiefly led Croizet and
Jobert,? in 1828, to propose Mastodon Arvernensis as a distinct
species. They met with specimens in Auvergne, mostly of
very young animals, of both the upper and lower jaws, in
which the last milk molar was unworn ; and they were struck
with the remarkable complexity of the crown-ridges, com-
posed of a great number of small wart-like cones, separated
by the decurrent vertical grooves, which we have referred to.
But the indicated character was not accepted by Cuvier as
of sufficient importance to distinguish the species from his
too comprehensive Mastodon angustidens.

Another fine specimen of the last upper milk molar, from
Mr. Fitch’s collection, is figured in the ‘British Fossil
Mammalia’ (fig. 100, p. 284). Like Mr. Wigham’s specimen,
the crown is composed of four ridges and a talon, with the
same complexity of pattern, alternation of the flanking
tubercles, and interruption of the valleys. Itis only necessary
to cite it here, as proving the constancy of the characters of
the Crag specimens. Prof. Owen describes this tooth as the
‘fourth upper molar ;’ while he assigns a different position
and value to Mr. Wigham’s specimen, considering the latter

1 Manual of Element. Geol. 5th edit. 2 Croizet et Jobert, Oss. Foss. du Dé-
(1855), p. 166. fig. 133. part. du Puy-de-Déme (1828), p. 1338.

MASTODON ARVERNENSIS. 37

to be the ‘second true molar.’ In our view they are both
last milk molars, which would be the equivalent of what
Prof. Owen designates as ‘the fourth in the order of size,
and the third in the order of position, counting backwards in
the upper jaw, before any of the teeth are shed.”!

There is some intricacy in the terms expressive of the
numerical values which Prof. Owen assigns to the different
molars of Mastodon in his descriptions, both in the ‘ British
Fossil Mammalia’ and in the ‘Odontography.’ This, I
believe, has arisen from the peculiar views there advanced
as to the order of succession of the premolar teeth in this
genus; andas it is a point of systematic and paleontological
importance in reference to the disputed affinities between
Mastodon and Dinotheriwm, I think it desirable to make a few
remarks on the subject. In both the works here cited,” it is
affirmed that Mastodon is distinguished from Elephant, in a
well-marked and unequivocal manner, by two dental charac-
ters: the first is the presence of tusks in the lower jaw; the
second ‘is the displacement of the first and second molars ’
(meaning milk molars) ‘in the vertical direction by a tooth
of simpler form than the second, a true dent de remplacement,
developed above the deciduous teeth in the upper, and below
them in the under jaw.’ Prof. Owen, in his remarks upon
Mr. Fitch’s specimen of the last milk molar (fig. 100), goes
on to say, ‘In Dr. Kaup’s figure the tooth in question’ (ve.
the third) ‘is associated with the first and second molars of
the Mastodon angustidens, which are much worn, and are
true deciduous teeth—the only ones, in fact, which strictly
correspond with the deciduous teeth of ordinary Pachy-
derms.’? In this view, when the antepenultimate and
penultimate milk molars are shed, and the penultimate pre-
molar has made its appearance, he designates the latter as
the ‘ third molar tooth ;’ and the last milk molar, which is
behind it in position, but anterior in appearance, he calls the
‘fourth molar tooth,’ although fully aware that there were
good grounds for regarding it ‘as the last of the theoretically
deciduous series, although it has no vertical successor.’
But this conclusion as to the absence of a vertical successor
to the tooth in question was premature. I detected both the
penultimate and last premolars im situ in the jaws of H.
(Lowodon) planifrons, a Sewalik fossil Elephant, upwards of
twelve years ago. The evidence is published in the ‘ Fauna
Antiqua Sivalensis.’4 M. Lartet has found the same two
premolars repeatedly in the upper and lower jaws of M.

' Op. cit. p. 284. 4 Op. cit. p. 31, Pl. vi. figs. 4-6; Pl. xn.
2 Brit. Foss. Mamm. p. 274; and Odon- | figs. 8-11. (See vol. i. pp. 68, 427,
tography, p. 615. and 433.—Eb.)

S Brit. Foss, Mamm. p. 284.

38 BRITISH AND EUROPEAN FOSSIL MASTODONS.

(Trilophodon) angustidens.'. In a manuscript note appended
to the work here quoted, with which I have been favoured
by M. Lartet, he adds, ‘J’ai pu depuis lors vérifier plusieurs
fois le remplacement effectif de la 2™° et de la 3™° molaire de
lait, tant a la machoire supérieure qu’a l’inférieure comme
cela a lieu dans le Dinotherium; la premiére n’est jamais
remplacée.’ In a beautiful specimen of the lower jaw of a
young M. (Trilophodon) angustidens, belonging to M. Ziegler-
Ernst of Winterthur, I found both the penultimate and last
premolars present—the former protruded, the latter in germ.
When a single premolar is developed, it is the successor to
the last milk molar, and not to the penultimate, as stated by
Prof. Owen in the passages above referred to.2 I have
seen detached specimens of this last premolar, both of the
upper and lower jaws of M. (Tetralophodon) longirostris of
Eppelsheim, in the Museum at Darmstadt. The order of
succession here indicated is alone consistent with what occurs
in other Pachyderms, where, when suppression in either of
the milk or premolar series takes place, it is constantly the
anterior or feebly developed and rudimentary teeth that are
suppressed. In them we never find the last premolar sup-
pressed while the penultimate is developed, but the reverse.
The penultimate premolar replaces only the corresponding
milk molar: the antepenultimate milk tooth is never re-
placed in Mastodon or Elephas so far as observation has yet
shown. The molar dentition of the permanent or second
set (ze. the premolars and true molars) in M. (Triloph.)
angustidens and Dinotherium is numerically identical, con-
sisting of two premolars and three true molars; each of
these having also two well-developed mandibular tusks: and
the close affinity thus indicated by the number of teeth is
further borne out by the correspondence of a ternary-ridged
formula in two of their ‘intermediate molars,’ and by other

’ Lartet, ‘Notice sur la Colline de, and the antecedent smaller grinders, as
Sansan’(1851), p. 25. the homologues of the milk molars of

* In his subsequent memoir on the | other Diphyodonts, which milk molars
molar teeth of Phacocherus, &c. (Phil. | have no successors in the Elephants. In
Trans. 1850, p. 496), Professor Owen | certain Mastodons, however, which are
takes a different view of the premolar | the earliest known forms of the Probos-
teeth in Mastodon from that set forth in | cidean family, the last milk molar was
the ‘ British Fossil Mammalia’ and the | displaced by a vertical successor or pre-
‘ Odontography.’ It is expressed thus :— | molar.’ The tooth, which, in his earliest
‘The existing species of the gigantic | descriptions, was considered as the suc-
Proboscidean family, viz. the Asiatic and | cessor of the antepenultimate and pe-
African Elephants, are totally devoid of | nultimate (first and second) milk molars,
incisors in the lower jaw, and all their | is here regarded as the successor of the
grinding teeth succeed each other hori- | last milk molar. But the presence of
zontally; so that it is only by a more | both the penultimate and last premolars
than proportional increase of size that | in Mastodon angustidens is not recog-

the antepenultimate grinder is recogni- | nized.
zable as the first of the true molar series,

ub),

a?

MASTODON ARVERNENSIS. 39
osteological characters,' which leave little room for doubt
that they were both Proboscidean genera, Dinotherium having
close affinities to the Tapirs, as Cuvier sagaciously inferred
in his earliest memoir on ‘ Les Tapirs Gigantesques.’ *

Of the antepenultimate and penultimate milk molars no
specimens from the Crag have been as yet figured or described ;
but the characters presented by these teeth in M. (Tetraloph.)
Arvernensis are well known, both in the upper and lower jaws,
through the specimens discovered by Croizet and Jobert,
Bravard, and others in Auvergne or the Velay. The ante-
penultimate presents two ridges, and the penultimate three
ridges, with the usual talon complications. They are readily
distinguishable—the upper from the lower—when met with
detached, from the circumstance that the milk molars of the
lower jaw are narrower and more compressed, the antepenul-
timate being reduced to a single cusp. Figures of these teeth
are given by Croizet and Jobert in the work already referred
to.3 .

The ridge-formula in the molar teeth of the Crag Mastodon,
including milk and true molars, but exclusive of premolars,
as inferred from the various data detailed in the previous

pages, is—

Milk molars. True molars.

—_——_—.
24344 | 44445
24+3+4 44445

The assigned numbers have not been verified in every instance
upon Crag remains; but they are all founded on an examina-
tion of specimens, of which some were of foreign origin, when
materials were not available from the Crag.*

}In M. Ziegler-Ernst’s Winterthur
specimen of the young lower jaw above
referred to, five molar teeth are present,
viz. the penultimate and last premolars
—the former extruded, the latter em-
bedded; the last milk molar far advanced
in wear, and immediately over the last
premolar; and the antepenultimate and
penultimate true molars. both in germ,
but the former partially emerged and
in incipient use.

2 Annales du Muséum, tom. iil. p. 182.

3 Oss. Foss. du Puy-de-Déme, Pl. i.
figs. 1-8 and PI. ii. fig. 7. ;

4 In the description of the various
teeth throughout this memoir, the terms
antepenultimate, penultimate, and Jast
have been used, instead of the numeral
expressions of /irst, second, and third,
when designating the position either of
the milk or of the permanent molars.
This would seem indispensable when

symbols are not employed, to avoid con-
fusion in the designation of the milk
molars, since the typical first or the most
anterior of the milk molar series, which
is present in many other pachydermatous
genera, is constantly suppressed in the
Mastodons, and generally also in the
Elephants. When, therefore, the terms
Jirst, second, and third milk molars are
applied to Mastodon, they are not the
equivalents of the same numerals applied
to Rhinoceros or Hippopotamus, in which
all the four milk molars are developed ;
whereas the terms antepenultimate, pe-
nultimate, and last in every case repre-
sent homologous teeth in the milk molars
of all the ungulate genera. This is the
more necessary, as the theoretical jirst
or pre-antepenultimate milk molar is
occasionally met with in the African
Elephant. De Blainyille (Ostéographie :
Des Eléphants, tab. ix. p. 81) has given

40 BRITISH AND EUROPEAN FOSSIL MASTODONS.

Lower jaw.—The characters furnished by the lower jaw are of
great significance in distinguishing the nearly allied species of
Mastodon, more especially in what relates to the form of the
symphysis, and the presence or absence of mandibular incisors.
The differences between the lower jaws of M. (Trilophodon)
angustidens, M. (Tetraloph.) longirostris, and M. (Tetraloph.)
Arvernensis are so pronounced that they would have been
sufficient to discriminate the species, supposing the mciar teeth
were unknown to us. As above stated, no good specimen, so
far as I am aware, has hitherto been discovered of the lower
jaw of the Crag species, M. (Tetr.) Arvernensis, in England ;
but several have been met with in the Pliocene strata of
Italy and France; while abundant remains of the lower jaw
of M. (Tetraloph.) longirostris have been disinterred from the
Hppelsheim sands by Dr. Kaup, and of M. (Trilophodon)
angustidens by MM. Lartet and Laurillard from the Falunian
deposits of the Sub-Pyrenees.

First in regard to M. (Triloph.) angustidens. The lower
jaw of this species is at once distinguished by the great elon-
gation, downward direction, and slender form of the symphy-
sial portion, which contains the sockets of the two inferior in-
cisors (Plate IT. fig.3).!_ The ascending ramus is of moderate
height, corresponding in that respect with M. (Trilophodon)
Ohioticus (fig. 2), and approaching that of Dinotheriwm gigan-
tewm (fig. 1). The horizontal ramus is very high in front, in a
line with the mentary foramen, and low behind; the anterior
portion is compressed ; and the lower margin stretches some
way infront of the mentary foramen, in a straight line; it is
then bent a little downwards, and continued forwards in nearly
the same straight line; the under surface of the elongated
portion forming an obtuse angle with the corresponding
surface of the horizontal ramus. The elongation of the
symphysial beak is enormous, far exceeding that of M. (Tetra-

an illustration of its presence in the | numerical category, is exhibited in the

latter species, on one side of the lower
jaw, regarding it as a ‘supernumerary’
tooth ; and a corresponding occurrence
in the lower jaw of the same species is
represented in the ‘Fauna Antiqua
Sivalensis’ (Pl, xiv. fig. 4a). Itis usually
restricted to one side, and I regard it
as not very uncommon, As the true
molars never exceed, nor are below,
three in number in the Pachydermata
and Ruminantia, the same terms may be
conveniently used in describing them.
The inconvenience of designating the
molars in Mastodon and Elephas by
successive numbers ranging from 1 to
6 or 7, which inelude both milk and true
molars without distinction in the same

descriptions of the Elephants given
throughout by De Blainville in his
‘Ostéographie;’ and more recently in the

| otherwise excellent descriptions by Ger-

vais of the dentition of Mastodon Andium
in the ‘Expédition de Castelnau.’ The
penultimate and last milk molars are
there figured and described as the ‘third’
and ‘fourth’ molars, involving a confusion
of the ridge-formula, which is seen to be
of so much importance in the subgeneric
distinctions. (Recherches sur les Mam-
miféres Fossiles de l’Amérique Méridio-
nale, 1855, pp. 20-22, Pl. v. figs. 1-5.)
_} De Blainyille, ‘ Ostéographie : Des
Elépkants,’ Pl. xiv.

MASTODON ARVERNENSIS. 41

lophodon) longirostris, or even of Dinotheriwm; the length °
from the mentary foramen forwards being more than double
that of the horizontal ramus, measured from the same point
backwards to the base of the anterior margin of the coronoid
process. A constant character of the species is the presence
in both sexes of two long, closely adpressed, and straight, or
but slightly curved, incisors. This fact has been established
by M. Lartet upon a very large number of specimens. !

These lower incisors differ notably in form from the upper.
In the adult specimens they are nearly of uniform diameter
from the base to the point, which is bevelled on the upper
side by wear, so as to yield a flat or spathulate surface. In
section they are pyriform, with frequently a longitudinal
channel on the upper and inner side. The section closely
approaches that of the inferior canines of the fossil Hippopo-
tamus named Hexaprotodon Sivalensis. The tusks or upper
incisors of this species are nearly circular in section, and
taper gradually to a conical point. They are invested along
their length on the inner side by a broad band of enamel,
which runs in an obsolete spire, so as to be presented on the ~
upper surface near the tip. M. Lartet has never observed
any indication of a belt of enamel on the lower incisors. In
the superb complete skeleton which was disinterred by Lauril-
lard at Seissan, the extruded portion of these lower incisors
measures 203 inches, that of the upper tusks being 41 inches.
The characters above indicated are constant, wherever the
lower jaw of this species has been discovered—whether in the
Faluns of the Orleannais and Touraine, in the lacustrine
deposits of Gascony and Languedoc, or in the Miocene Molasse
of Switzerland, where I found them confirmed by the
examination of two very fine specimens, young and old, found
in the neighbourhood of Winterthur. They are well shown by
the representations given in De Blainville’s ‘Ostéographie: des
Eléphants,’ Pl. XIV. The molar teeth in all these specimens
have constantly presented the normal marks of the Trilopho-
don division—namely, three ridges to the last milk molar, and
to the antepenultimate and penultimate true molars.

In M. (Tetralophodon) longirostris the ascending ramus is
considerably more elevated than in M. (Triloph.) angustidens,
approaching more the character which is seen in the
Elephants proper; the horizontal ramus is less compressed
and more circular in section; instead of presenting the
ereatest height in a line with the commencement of the
alveolar border, or mentary foramen, it is contracted there
in consequence of the lower margin rising upwards to slope

1 Lartet, ‘ Notice sur la Colline de Sansan,’ p. 24.

42 BRITISH AND EUROPEAN FOSSIL MASTODONS.

off into the base of the symphysial beak (Plate II. fig. 8).!
This beak is very massive and comparatively short, not ex-
ceeding the length of the horizontal ramus, from the mentary
foramen to the anterior margin of the ascending ramus.
Instead of being, as it were, a deflected continuation of the
inferior border of the jaw, as is seen in M. (T'rilophodon)
angustidens, the beak in the Eppelsheim species is thrown
off in a plane nearly parallel with the inferior border, but
separated from it and raised above it by astep. It is deflected
slightly downwards; but, instead of forming a long slender
apophysis as in the other species, it shows a thick mass tra-
versed by a broad gutter. The greater extent of the beak is
made up of the alveoli of two mandibular incisors, as in M.
(Trilophodon) angustidens. These teeth have not yet been
found in situ. Kaup has figured three specimens? which he
conjecturally considers to be lower incisors. The greatest
diameter of the largest he states to be 2°75 inches. The
molars of these Eppelsheim jaws have constantly exhibited
the Tetralophodon character of four ridges to the crowns of
- the intermediate teeth ; the ridges being transverse, with the
valleys nearly uninterrupted.

In the Pliocene M. (Tetraloph.) Arvernensis, the lower jaw
differs widely from that of the other two species. The as-
cending ramus is well elevated above the grinding plane of
the teeth, as in M. (Tetraloph.) longirostris. The horizontal
ramus is very massive, without compression, and yields a
section which is nearly circular, as in that species. But the
symphysis, instead of being elongated into a process composed
of the alveoli of two mandibular incisors, terminates suddenly
in a short beak, as in the Elephants and other Proboscidean
species that are destitute of inferior tusks. This beak does
not project much more beyond the anterior rounded surface
of the jaw than in the African Elephant, or in M. (Tvrilophodon)
Humboldtii, also a species without mandibular incisors ; but
it differs from them and all other known species in the
diastemal ridges expanding at the point, so as to form a short,
blunt, dilated spout. This character is well shown by the
Val d’Arno specimen delineated by Cuvier in the ‘ Ossemens
Fossiles,’ tom. i. tab. ix. figs. 5 & 6, after Nesti. Itis one of
the pieces upon which Nesti founded his Elephas meridionalis ;
but which, although the molars are wanting, Cuvier saga-
ciously inferred, from the general form, to belong to a
Mastodon. I was enabled, by the obliging permission of
Professor Gaspero Mazzi, to examine the specimen minutely,
and to compare it with the numerous lower jaws of H. (Loxodon)

1 Kaup, Oss. Foss. de Darmst. tab. 2 Ossemens Fossiles de Darmstadt,
Sap ep sites 1d tab. iil. figs. 1, 2, 3.

MASTODON ARVERNENSIS. 43

meridionalis and of M. (Tetraloph.) Arvernensis contained in
the Natural History Museum at Florence, and was satisfied
that it belonged to the latter species, as Cuvier had inferred
from the drawing. The same Museum contains the greater
part of a skeleton of Mastodon, found in a marine deposit of
the lower Val d’Arno above Leghorn. The lower jaw of this
specimen presents the same character of a short symphysial
beak without incisors. The same is exhibited by the lower
jaw of the Dusino skeleton from the Astesan, described by
Prof. Hugenio Sismonda.! They all agree in the common
characters, so far as these are shown, of a Tetralophodon
formula to the crown ridges of the three molars here called
intermediate ; of alternate mammille to the ridges with
blocked-up valleys; and of a short obtuse beak with no
incisors.

Sismonda describes and figures the lower jaw of the Dusino
specimen as being without tusks, or remains of their sockets.
But, predisposed to believe that they must have been present
at some period of the animal’s existence, from their occur-
rence in other Mastodons, he conjectures that those tusks had
fallen out early, and that the alveoli had been obliterated by
filling up; and he has given a representation of a very muti-
lated fragment of a Proboscidean symphysis of the lower jaw
as exhibiting the alveoli of two mandibular incisors.2_ I was
enabled, by the obliging kindness of Signor Bartolomeo
Gastaldi of Turin, to examine the specimen in question, which
is very much rolled, and in a different mineral condition from
the fossils of the Dusino Mastodon bed, and found that the sup-
posed incisive alveoli were only the anterior terminations of the
dentary canals, which are of large size in all the Proboscidea.
The form impressed me with the conviction that it was more
probably the symphysis of an Elephant than of a Mastodon.
This case, therefore, gives no support to the belief that M.
(Tetralophodon) Arvernensis had lower incisors.

Professor Owen, in his ‘ British Fossil Mammalia,’ gives a
very beautiful representation (p. 291, fig. 101) of a fragment
of a tusk discovered by Mr. Fitch in the Mammaliferous Crag-
pits near Norwich. He describes it as a portion of the lower
tusk of the Mastodon angustidens. The specimen is about 15
inches long, with a greatest diameter of 3 inches. It is of
a straight, compressed, conical form. The fragment is
crushed, and it is manifest that the outer layers of the ivory
are detached, and that the original tusk was of a larger
diameter than the specimen now exhibits. The marked
conical form and great size are irreconcileable with this

* Osteograph. di un Mastod. angustidente, tab. i. fig. 1.
2 Op. cit. tab. 1. fig. 7.

44 BRITISH AND EUROPEAN FOSSIL MASTODONS.

fragment, being referable to an inferior incisor of the Simorre
M. (Triloph.) angustidens of Cuvier; andit would seem to me
that they are equally irreconcileable with its being considered
as alower tusk of M. (Tetralophodon) longirostris, for the sym-
physial beak required for the implantation of a tusk of such
magnitude would be enormous, and is unknown among any
of the species of Mastodon. Professor Owen describes the
specimen as being traversed from end to end by a sub-central
canal. The same character has been observed in the upper
tusks of other fossil Proboscidea, and is nowise characteristic
of a lower incisor. I consider that the specimen in question
is not a fragment of a lower, but of an upper tusk near the
point; and it differs in no important respect from the un-
doubted upper tusks of M. (Tetralophodow) Arvernensis seen
in the Museums of Florence and Turin, which are either
slightly curved or twisted in a gentle spiral direction, as
represented in the figure given by Sismonda’ of the Dusino
skeleton.

In corroboration of this view, it may be stated that the
Tndian fossil species which we have named M. (T'etralophodon)
Sivalensis is in some respects more nearly allied to the Crag
species than the latter is to either M. (Trilophodon) angusti-
dens or M. (Tetralophodon) longirostris. It shows the same
alternate character of the mammille of the ridges of the
‘intermediate molars,’ and it appears to have been equally
destitute of inferior incisors. JI have examined a large
number of lower jaws of this species, of all ages, from the
sucking calf up to the adult animal, specially with a view to
the detection of these teeth, and never observed the slightest
indication of their presence in any specimen, whether in the
Indian fossil collection of the British Museum, at the India
House, or in the rich series belonging to the Asiatic Society
of Calcutta.

This completes what I have to bring forward in the shape
of descriptive and comparative details, in order to indicate
the most prominent diagnostic characters derivable from the
teeth and jaws of the Crag Mastodon. I believe that the
differences of the three species included by Cuvier under the
name of Mastodon angustidens will be found to be carried out
through all the principal bones of the skeleton. It would
be wholly out of place to enter upon such osteographical par-
ticulars on the present occasion; but a good idea of the
general character of the skeleton in each may be attained by a
reference to two well-known standards of comparison, namely,
the existing Indian Elephant and the Mastodon of North

! Op. cit. tab. i. figs. 4 and 6.

GEOLOGICAL AGE OF MASTODONS. 45

America, M. (Trilophodon) Ohioticus. Cuvier found that the
latter differed from the Elephant in having a more elongated
carcass sustained upon shorter, thicker, and more robust legs.'
The Crag M. (Tetralophodon) Arvernensis appears to have
had a heavy carcass, with legs still shorter in proportion,
approaching more the character of the Hippopotamus, and to
have been without lower tusks. The Eppelsheim Miocene
species, M. (Tetralophodon) longirostris, would appear to have
resembled the Crag species in its general proportions; but
the necessary detailed comparison has not yet been sufficiently
carried out; it is distinguished at once by the possession of
inferior tusks. On the other hand, the Miocene M. (Trilopho-
don) angustidens differed remarkably from both in presenting
a comparatively slender build throughout; so that it stood
higher in proportion, and with longer limbs, than either the
Indian or African Elephants. This is well exhibited by the
mounted skeleton in the Paris Museum.

Geological Age of the Mastodons (M. angustidens, M. longiros-
tris, and M. Arvernensis).—I shall now consider the geological
age and associated faunas of the formations in which these

pecies severally occur.

M. (Trilophodon) angustidens is a characteristic species of
the Miocene Falunian beds throughout Europe. It has been
met with in immense abundance in the lacustrine deposits of
Gascony and Languedoc ; in the Faluns of Touraine and the
Orleannais; in the Miocene Molasse of Switzerland, more
especially in the lignites of Ellg, Kepfnach and Buchberg,
and in the sandstone in the neighbourhood of Winterthur;
in the Georgensgmiind Miocene in Germany ; and in the lig-
nite of Gandino in the Val Seriana of Lombardy. The mam-
malian genera and species with which it was associated are
very constant, although, for obvious reasons, they have not
been found equally or uniformly distributed all over the area.
In the French Falunian deposits there occur M. (Trilophodon)
Tapiroides, a species first conjecturally named by Cuvier, but
subsequently made out well by MM. Pomel, Lartet, and other
French paleontologists; Dinotheriwm gigantewm, or the smaller
variety, as I consider it, called D. Cuvieri, Chalicotherium
Goldfussi, Anchitherium Aurelianense, Aceratherium incisiveum
(Rhinoceros tetradactylus, Lartet), Aceratherium Groldfussi
(Rhinoc. brachypus, Lartet), Rhinoceros Sansaniensis, Lophio-
cherus Blainvillii,Macrotherium giganteum, Dicrocerus and Dor-
catheriwm, &¢.; besides various Carnivorous forms, large and
small, with remains of Chelonian genera, together with
scanty indications of Crocodile.’

1 Oss. Foss. 4to edit. tom. i. p. 249.
2 Lartet, ‘ Notice sur la Colline de Sansan,’

46 BRITISH AND EUROPEAN FOSSIL MASTODONS.

In the Upper freshwater Molasse, of Switzerland, M. (Tii-
lophodon) angustidens occurs along with M. (Triloph.) Tapir-
oides (which has been named Mastodon Twricensis, as a dis-
tinct species, by Schinz and Von Meyer), Aceratheriwm incisi-
vum, Acerather. Goldfussi, Dinotheriwm gigantewm, Lophioche-
rus Blainvillii, a species of Tapir, Paleomeryx and other
Ruminants, &c., with several species of Chelonians.

M. (Tetralophodon) longirostris occurs abundantly in the
sands of Eppelsheim, associated with Dinotheriwm gigantewm,
Chalicotherium Goldfussi, Rhinoceros Schleiermacheri, Acera-
therium incisiwum, Acerath. Goldfussi, Macrotherium gigantewn,
Hippotherium gracile, and species of Dorcathervwm, Machairo-
dus, Amphicyon, &e.

The agreement in so many remarkable generic and specific
mammalian forms leaves little room for doubt that the
Eppelsheim sands and the lacustrine deposits of the Garonne
and other parts of France are of the same Miocene age. But
there are some notable peculiarities in the Eppelsheim fauna.
No well-marked specimen, so far as I am aware, of M. (Tetra-
lophodon) longirostris, as here defined, has hitherto been met
with beyond the limited area of the Eppelsheim sands, and
probably the valley of the Danube; nor has either M. (Trilo-
phodon) angustidens or M. (Triloph.) Tapiroides—which usually
go together——-been discovered within it. It is very improbable
that the range of this species should have been confined to a
small district in the Valley of the Rhine ; but the fact is un-
doubted, that it occurs there in great abundance, and has
either not yet been found, or is very rare, elsewhere. The
only exception out of Germany, with which I am acquainted,
is a specimen of unknown origin in the Museum of the
Faculty of Sciences at Toulouse, which, on the indication of
M. Lartet, I was enabled to examine by the kindness of M.
Leymerie. It consists of an upper right maxillary containing
the penultimate and last molars in situ. They present all the
characters of M. (Tetralophodon) longirostris, as distinguished
from M. (Tetraloph.) angustidens. M. Leymerie informed me
that the specimen is supposed to have been found either in
Gascony or Languedoc, but that there was no record of the
exact locality. It is not improbable that another exception
is formed by the specimens mentioned by Cuvier (Oss. Foss.
additions, 4to edit. tom. ili. p. 318) as having been discovered
by M. Lourteau at Sairac in the Sub-Pyrenees. Two of the
molars are described as having the Tetralophodon character
of four ridges ; but, no figures having been given, the details
are not sufficiently precise or exact to admit of any decided
opinion upon the subject.

A satisfactory geological limitation of the Hppelsheim

GEOLOGICAL AGE OF MASTODONS. 47
deposit and its organic contents is attended with some diffi-
culty. The loose incoherent sand of which it is composed is
spread out horizontally like the Loss, and the margin thins
out to spread over a portion of the ‘Lower Miocene’
Mayence Basin ; so that where the beds are in contact the
fossil remains of the two are liable to be confounded. But in
all its leading features the Mammalian Fauna of Eppelsheim
resembles that of the Falunian deposits of France and
Switzerland.

I shall now consider the relations of the Pliocene fauna in
which the Crag Mastodon occurs. M. (Tetraloph.) Arvernensis,
as here defined, had a wide range of habitat in Europe,
embracing Italy, France, and England. The principal locali-
ties in which it has been found are—in Italy,’ the Val d’Arno
(in great abundance), associated with the Elephant called EH.
meridionalis by Nesti (Lo«od. meridionalis), Rhinoceros lepto-
rhinus,? Hippopotamus major, with species of Tapirus, Sus,
Equus, Ursus, Hyena, Felis, Machairodus, &c.; in the marine
‘Panchina inferiore,’ of the Lower Val d’Arno, an entire
Mastodon skeleton was found along with those of extinct
Whales ;* in Piedmont and Lombardy, in various localities in
the Sub-Apennine strata along the Valley of the Po, but more
especially in the Astesan, Romagnano, and Duchy of Piacenza,
along with the M. (Triloph.) Borsoni (M. Buffonis of Pomel),
a well-marked ternary-ridged species, first brought to notice
by Abbé Borson, and the extinct Elephants H. (Loxod.) meri-
dionalis, EZ. (Loxod.) priscus, and EH. (Huelephas) antiquus, and
Rhinoc. leptorhinus, Hippopotamus major, &c., which occur in
some places in fluviatile deposits along with species of Helix,
Paludina, and Clausilia, and in others in marine deposits
along with sea-shells; in France, in various parts of the
southern Departments, in Pliocene strata, such as the marine
sands of Montpellier and its vicinity, the Valley of the Rhone
near Lyons and Trevaux, the Vivarais, Velay, Auvergne, &c.

Great diversity of opinion holds among the French pale-

1 «The Florence Museum is not so rich

the tusk nearly as in the existing Indian
in Mastodon as in Hlephas meridionalis.

Elephant. The tip is very much flat-

But what I saw, viz. six lower jaws, were
allof Mast. Arvernensis. The characters
were yery well marked. The beaks
were all short, nearly horizontal, and
neyer inclined downwards or elongated ;
and they never showed any trace, in young
or old, of inferior tusks. They have in
the Museum one superb tusk, found
along withaskeleton, about 9 feet long by
about 43 inches in diameter, the section
nearly round, and the curved direction of

tened (not conical), but as I could not
see it close enough, I could not deter-
mine whether this was owing to a layer
of enamel. The Mastodon skeleton was
found along with a small Balena and
covered over with oyster-shells Letter
to M. Lartet, Florence, July 17, 1856.—
[Ep.]

* Rhinoceros Etruscus, Fale-—[Ep.]

S$ Savi e Meneghini sulla Geolog.
Stratigraph. della Tuscana, p. 508.

48 BRITISH AND EUROPEAN FOSSIL MASTODONS.

ontologists as to the association of the mammalian species
among which M. (Tetraloph.) Arvernensis occurs in French
deposits. I shall refer briefly, on the present occasion, to the
disputed cases at Montpellier or its vicinity, and in Auvergne.
De Christol! has described the marine sands of Montpellier
and the gravel-beds of the contiguous basin of Pézénas as of
the same age. From the latter he procured remains of
Elephant which he ascribed to the Hleph. meridionalis of
Nesti, Hippopotamus major, two species of Hquus, one of Bos,
and two of Cervus. Gervais, on the other hand, insists that
the gravels of Pézénas are of the age of the Diluvian fauna
(Pleistocene), the sands of Montpellier being Pliocene. To
the former? he attributes Hlephas primigenius, Hippopotamus
major, two species of Equus, Bos priscus, and Cervus martialis ;
and to the latter * Mastodon brevirostre (Tetraloph. Arvernensis),
Rhinoceros megarhinus, Tapirus minor, with species of Sus,
Cervus, Ursus, Machairodus, Halitherium, Hoplocetus, &c. M.
Gervais does not admit Elephant remains in the Pliocene
fauna of Montpellier; but there are two circumstances which
diminish the authority of this opinion upon the subject—the
first being, that he refers all the fossil elephants found in the
South of France to the Mammoth, EL. primigenius of the Dilu-
vian fauna, of which he considers H. meridionalis to be a va-
riety; the second, that he does not admit that any species of
fossil Elephant have been discovered anywhere in Pliocene
strata in Europe. He considers that in the instances asserted
by Croizet, Christol, Marcel de Serres, and others, Mastodon
bones have been mistaken for those of Elephant. But, put-
ting aside the disputed French cases, it will be seen in the
sequel that there are undoubted instances of the occurrence
of remains of Mastodon and Elephant in the same strata in
the Sub-Apennine beds of Italy and in the Crag of Norfolk.
In Auvergne and the Velay, the lacustrine and regenerated
alluvial strata of all ages, from the Miocene up to the Post-
Pliocene, have undergone such complicated disturbances from
successive volcanic eruptions, that great difficulty has been
experienced in separating the members of the various faunas,
more especially of the subdivisions of the Pliocene and later
period. The utmost diversity of opinion holds among the
paleontologists who have paid most attention to the later
types of the fossil Mammalia of Auvergne, regarding the
groups of species which were coexistent at different times.
Without going into details, I may observe that Bravard has
endeavoured to make out three distinct faunas after the
1 Annales des Scien. Natur. 2 sér. tom. | Pl. xxi.

iy. p. 193. 3 Op, cit. tom. ii. deseript. Pl. xxx.
? Paléontol. Franc. tom. ii. descript. 4 Gervais, op. cit. tom. 1. p. 36.

GEOLOGICAL AGE OF MAST. ARVERNENSIS. 49

Miocene lacustrine beds of the Limagne: 1st, a Mastozoic, or
Pliocene fauna, characterized by-the presence of species of
Mastodon and the absence of Elephants, Horse, and Hippo-
potamus ; 2nd, an Hlephantine fauna, comprising these genera ;
and 3rd, a Diluvian fauna,in which Hlephants and Rhinoceros,
&c. are wanting.! Pomel, on the other hand, in his last de-
tailed memoir, has attempted to distinguish after the Miocene
lacustrine deposits of the Limagne; Ist, a Pliocene fauna,
characterized by two species of Mastodon, a Rhinoceros, Sus,
Tapir, and twelve or fourteen species of Cervus, but no Hle-
phants ; 2nd, an alluvial fauna, which he divides into two
distinct series of different ages: the one more ancient, com-
prising EHlephas meridionalis, Rhinoceros leptorhinus and
Rhinoc. Aymardi, Hippopotamus major, Tapirus elegans, Ursus
speleus, Bos priscus, Megantereon latidens, and two species of
Deer, &c.; the other, more modern, consisting of Hleph.
primigenius, H. priscus, Rhinoceros tichorhinus, Hyena spelea,
Cervus Guettardi, &c.? But there are grave objections to both
these arrangements, inasmuch as the association of the
species does not correspond with what holds elsewhere in the
Pliocene and Post-Pliocene deposits of Italy, England, and
Germany, which are free from the volcanic intrusions that
have overwhelmed and confused the deposits of Auvergne.
It suffices for my purpose on the present occasion, to state
that, where M. (Tetralophodon) Arvernensis occurs in Auvergne
and the Velay, the same species are met with in different
localities as are found together in the same Pliocene stratum
in the plains of Piedmont and Lombardy, namely, M. (T'rilo-
phodon) Borsom, Rhinoceros leptorhinus,? Hippopotamus major,
and the Hlephants called H. (Loxodon) meridionalis and FE.
(Loxodon) priscus (?), with species of Tapirus, Sus, Cervus, &c.
The numerical agreement of the Auvergne fossil species with
those which occur in the richer fauna of the Val d’Arno is
still more considerable. But it is, at the same time, to be
remarked, that at the late Meeting of the ‘ Congrés Scienti-
fique’ of France, held at Puy in Sept. 1855, MM. Croizet,
Aymard, and Pichot* were agreed that the Mastodon re-
mains in the Velay and Auvergne were of an older age than
the beds containing Elephant remains.

Mastodon of the Crag.—I shall now pass under review the
circumstances under which M. (Tetralophodon) Arvernensis
occurs in British strata.

1 Bravard, cited by Pomel, ‘Bullet. de | et Deseript., &c. 1854, pp. 172-184.

Ja Soc. Géol. de France,’ 2e sér. tom. iii. % Written in 1856.—/ Ep. ]

p. 178 et seq. : ‘ Congrés Scientifique de France, 1835,
2 Pomel, loc. cit. and Catalog. Méthod. | tom. i. p. 326.

VOL. II. E

50 BRITISH AND EUROPEAN FOSSIL MASTODONS.

First, in the ‘ Fluvio-marine ’ or ‘Norwich Crag.’ Un-
doubted remains of this species have been discovered in this
deposit: at Whitlingham by Mr. William Smith ; at Horstead,
by Messrs. Layton, 8S. Woodward, and Gunn; “at Postwick,
Thorpe, and Norwich, by Messrs. Fitch and Wigham ; at
Bramerton, by Mr. 8. Woodward and Capt. Alexander ; and
in Suffolk, at Easton and Sizewell Gap, by Capt. Alexander.
The entire skeleton, of which so circumstantial an account
has been given by the Rev. Mr. Layton, is stated to have
been found on the surface of a bed of marl, ‘between the
chalk and gravel,’ at Horstead, without indicating the precise
relation of the bed to the Crag and the superincumbent
blue clay or submerged forest- bed. I have examined the
most of these specimens, either in original or as casts, at
the Museums in Norwich and London, and found them all
referable to the species, as here limited.

Various statements have been made by different writers
regarding the fossil Mammalia associated in the Fluvio-
marine Crag, with the Mastodon or without it. Mr. William
Smith’s celebrated Whitlingham specimen is said to have
been found along with the horns of Deer and Crag-shells.!
Mr. R. C. Taylor? mentions that the Crag of Bramerton has
yielded ‘the Mastodon, the Elephant, the Gigantic Elk, and
the Enormous-horned Bison.? Mr. Charlesworth?’ states,
that bones of Elephant and other herbivorous animals are
more frequently associated with shells in the Mammaliferous
than in the Red Crag, but he does not mention what the
species are. Sir Charles Lyell,‘ in his memoir on the
‘Relative Ages of the Norfolk and Suffolk Crag,’ states that
the Fluvio-marine Crag, near Southwold, has yielded the
remains of the Elephant, Rhinoceros, Horse, and Deer,
mixed with marine, terrestrial, and freshwater shells; and
that in the inland pits near the same place he found mamma-
lian remains associated with the Cyrena trigonula of Grays
and elsewhere. He mentions, that ‘the horns of Stags,
bones and teeth of Horse, Pig, Elephant, and other quadru-
peds,’ associated with Mastodon, had been obtained at Post-
wick, Thorpe, Bramerton, and other localities near Norwich.
The tusk of an Elephant was obtained at Bramerton, covered
with Serpule, showing that it had lain for some time at the
bottom of the sea of the Norwich Crag.*

Professor Owen in the conspectus of genera and species
contained inthe introduction to his ‘British Fossil Mammalia,’

1 Taylor, Geology of East Norfolk, | p. 89.

1827, p. 14. * Geol. Proc. vol. iii. p. 127; and -
2 Loe. cit. Mag. Nat. Hist. new ser. vol. iii, p. 316.
$ Phil. Mag. 8rd ser. yol. vii. 1835, 5 Op. cit. p. 128,

GEOLOGICAL AGE OF MAST. ARVERNENSIS. 51

enumerates in the list of the fossils of the Pliocene Fluvio-
marine Crag the following genera and species, viz. Mastodon
angustidens, Hlephas primigenius, Rhinoceros tichorhinus,
Hquus fossilis, Cervus elaphus, Arvicola, and Lutra. But, influ-
enced probably by the opinion at which he had arrived, that
the Crag Mastodon was identical with the M. angustidens of
Cuvier and M. longirostris of Kaup, he adds in a note, that all
the other species, except Mastodon, were probably derived
from the overlying blue clay.’ The contemporaneous asso-
ciation of these species is unquestionably in the highest
degree improbable, as it would include a Miocene Mastodon
along with a Post-Pliocene Elephant and Rhinoceros, and
the existing Red Deer, in the same fauna. But it admits of
no doubt that species of the genera above enumerated have
been found in the Fluvio-marine Crag, and it is of import-
ance to ascertain what these species really are. I carefully
examined the Hlephant molars from the Crag, blue clay, or
submerged forest-bed, contained in the different collections
at Norwich, and arrived at the conclusion that none of them
belonged to H. primigenius, the Mammoth of Siberia, properly
so called; but to two distinct species, the one, H. (Loxodon)
meridionalis, which occurs in vast abundance in the Val
d’Arno; and the other, H. (Huelephas) antiquus, which is
found in the plains of the Astesan, in Piedmont, in various
- other parts of Europe, and in the so-called ‘ Newer Pliocene’
freshwater deposits and caves of England. The evidence
upon which these species are founded will be considered in
the sequel. The occurrence of the Siberian Rhinoceros
tichorlinus (Rhin. antiquitatis of Blumenbach) in the Crag
would seem exceedingly improbable; for, elsewhere, it has
invariably been met with in company with the Mammoth, in
the northern fauna of the Glacial Drift period, and nowhere
as yet, upon undoubted evidence, in Pliocene formations.
Professor Owen (Brit. Foss. Mamm. p. 381) states, that ‘Mr.
Fitch of Norwich possesses specimens of upper and lower
molar teeth of the Rh. leptorhinus from the freshwater (lignite)
beds on the Norfolk coast near Cromer, which demonstrate
the occurrence of this species in the same deposit with the
Rh. tichorhinus” The contemporaneous association of the
two species in these beds would seem as improbable as the
occurrence of Rh. tichorhinus in the Crag, and the explanation
may be sought for in an adventitious mixing of the speci- .
mens.” The evidence adduced in support of the existing

1 Op. eit. p. xlvi. with the shells of the Crag in Norfolk,

* Mr. Charlesworth, in remarking that | adds that, ‘in that county the formation

the bones of Elephants and other quad- | in many places exhibits such irregulari-

rupeds are more frequently associated | ties, and is sometimes so mingled with
E 2

52 BRITISH AND EUROPEAN FOSSIL MASTODONS.

common Otter (Lutra vulgaris!) and Red Deer (Cervus elaphus)
having also been found in the same deposit, would require to
be very conclusive before the facts alleged could be received
as well established. For no fewer than eight species of
Cervus, belonging to the subgenera Rusa and Strongyloceros,
with round antlers, have been described by the French
paleontologists as occurring in the Velay and Auvergne,
besides eleven other species in Pliocene or Post-Plocene
strata.2 Several species with round-antlered horns have
also been obtained from the Val d’Arno, which would seem
to be identical with Auvergne forms (making liberal allowance
for doubles emplois in the specific names), and it is much more
probable, from the agreement in the other associated mammals
that the Crag species belonged to one of these than to the
existing Cervus elaphus.’ Hippopotamus major and Rhinoceros
leptorhinus,' if not hitherto obtained from the Fluvio-marine
Crag, occur in abundance either in the blue clay or in the
ancient forest or lignite-bed, which immediately overlies the
Crag in the sections along the Norfolk Coast; and evidence
will be adduced in the sequel, that these beds are of the same
Pliocene age, in so far as is shown by the paramount proof
of identity of mammalian fauna. Taking together the
ascertained fossil Mammalia of these two beds, they agree
very closely with the Pliocene fauna of the Sub-Apennines,
viz. M. (Tetralophodon) Arvernensis, E. (Loxodon) meridionalis,
Ei. (Euelephas) antiquus, Rhinoceros leptorhinus, Hippopotamus
major, large Bovide, and large Deer with round-antlered
horns. Among the Proboscidean forms the principal excep-
tion isthe absence of the Mastodon here called M. (Trilophodon)
Borsoni from the Crag and blue clay. This species, which
occurs both in the Astesan and Auvergne and other parts of
France, is so nearly allied to the Mastodon of North America
that the first discovered Huropean specimens were regarded

immense accumulations of sand and, pliocéne. Tels sont le Cerf, la Loutre,

gravel, that it becomes almost impossible
to distinguish the specific Crag deposit
from the accompanying diluyial strata.’
—Phil. Mag. 8rd ser. vol. vil. p. 89.

1 Owen, Brit. Foss. Mamm. p. 121.

2 Pomel, Catal. Méthod. et Descript.
p. 103,

3 Gervais has expressed doubts re-

specting the veritable association of these
living with extinct forms :—

‘Tl est également a supposer, que les
nouvelles recherches des géologues
d’Angleterre démontreront aux paléon-
tologistes de ce pays que certains ani-
maux reconnus par M. Owen comme
étant d’espéces actuelles n’ont pas appar-
tenu, comme ils le supposent, a l’époque

et le Sanglier ordinaires. Le Rhinoceros
tichorhinus, que nous considérons comme
caractéristique du _ pléistocéne, nous
parait aussi devoir étre rayé de la liste
des animaux pliocenes. On pourrait sup-
poser qu'il s’est glissé quelque ereur dans
la détermination des piéces osseuses, re-
gardées comme telles, mais cette déter-
mination est garantie par la citation que
M. Owen fait de cette espéce dans sa liste
chronologique des Mammiferes fossiles en
Angleterre, et 11 est plus probable que
cest sur age du terrain lui-méme que
Yon s'est trompé.’—Gervais, Paléontol.
Francaise, tom. 1. p. 180.
4 See note 3, p. 49.— [Ep.]

ee SRA a te Se A ae eee eT ee

1

SlLkcen

GEOLOGICAL AGE OF MAST. ARVERNENSIS. 53

by Cuvier! as belonging to that species; but its specific
distinctness has been clearly established by the French
palzontologists, and its occurrence in the Crag or overlying
beds may yet be expected, if it has not been heretofore over-
looked by collectors. The species would seem to be exceed-
ingly rare in Italy, since tooth specimens referable to it are
either unique or nearly so in the public collections there.

Next, as regards the ‘Red Crag’ of Suffolk. Mammalian
remains were formerly so rare in the ‘Red Crag’ that their
abundance in the Norwich Crag was seized upon by Mr.
Charlesworth as furnishing a significant designation for the
latter under the name of ‘Mammaliferous Crag.’ But lat-
terly the excavations for phosphatic nodules have led to the
discovery of these remains in abundance. Among others,
molars of M. (Tetralophodon) Arvernensis have been obtained
in very considerable numbers. By the liberal kindness of
Professor Henslow, I have been enabled to examine at leisure
those which are contained in the Ipswich Museum, presented
to that institution by Mr. George Ransome. They were
found in the Red Crag pits. Some of these remains are now
on the table before the Society. One, a very characteristic
specimen, consists of the greater part of the last true molar,
upper jaw, left side. It presents all the distinctive marks of
M. (Tetralophodon) Arvernensis, namely, the discs of the worn
tubercles decidedly alternate, and the valleys blocked up by
large outlying tubercles. These ‘Red Crag’ molars differ in
no respect specifically from those found in the Fluvio-marine
Crag. They are highly impregnated with ferruginous infil-
tration, and present a vitreous polish, very much like that of
the Mastodon molars from Perim Island on the western coast
of India. They are mutilated by fracture, but do not present
the appearance of having been rolled. The fractured edges
of the enamel are sharp; and the only indications of abrasion
which the teeth present are the natural results of wear, from
long service as grinders. This is a point of some importance,
as indicative that they were not washed into the Red Crag
out of some older Miocene deposit.

Mr. Charlesworth, in his memoir on the ‘ Crag of Suffolk,’
&c., after enumerating the genera of fossil fish that prevailed
in the ocean of the Red Crag, adds—‘ It is here also that we
first meet with the higher orders of the animal kingdom.
The teeth of the Mastodon, Elephant, Hippopotamus, and
other Mammalia are deposited with the Mollusca of this
period, and in addition to them I may mention the bones of
Birds, which I have recently obtained from several localities.”?

1 Oss. Foss. 4th edit. tom. iii. p. 375. * Phil. Mag. 3rd. ser. vol. viii. p. 535.

54 BRITISH AND EUROPEAN FOSSIL MASTODONS.

Professor Owen has on three occasions described the fossil
mammalia of the Red Crag: first, in 1840;! next, in his
‘British Fossil Mammalia’ in 1846; and latterly, as a
Supplement, in No. 47 of the ‘Quarterly Journal of the Geolo-
gical Society.’? In neither has he included two of the genera
cited by Mr. Charlesworth, viz. Hlephas and Hippopotamus,
both being of great significance as diagnostic of the age of
Huropean tertiary strata. No specimen of a tooth of Hippo-
potamus from a Red Crag locality, so far as I am aware, has
hitherto been figured or described ; and the occurrence of this
genus in the deposit cannot at present be regarded as an
established fact ; but several molars of fossil Hlephas, present-
ing the characteristic mineral condition of the Mammalian
remains of the Red Crag, have long been deposited in public
and private collections, bearing labels as being from Red Crag
localities in Suffolk. One specimen, in particular, in the
Museum of Practical Geology, is marked as being from Felix-
stow, and other reputed instances of the same kind will be
noticed in the sequel.

In the ‘Conspectus’ contained in the ‘British Fossil
Mammalia,’ Prof. Owen enumerates, as Mammalia of the
‘Miocene Red Crag,’ remains of Ursus, Meles, Felis pardoides,
Sus, and Cervus. But he adds in a note, that ‘ the nature of
the stratum renders the actual age of these fossils doubtful.’
To the enumeration of the five Hocene species of Cetacea in
the same conspectus, he appends a note, ‘ that most of them
occur in the Miocene Crag, but there is little doubt that they
were washed out of the underlying Hocene clay.’ The
‘ Cetotolites’ in question were discovered by Professor Hen-
slow in the Red Crag at Felixstow,’ which has yieldedabundant
Mammalian remains of herbivorous quadrupeds. In his late
paper, Professor Owen gives an account, more or less detailed,
of the remains of twelve species of Mammalia (exclusive of
Cetacea) from the Red Crag, belonging to the genera Rhin-
oceros, Tapirus, Sus, Equus, and probably Hipparion, Mastodon,
Cervus (of the subgenera Dicranoceros and Megaceros), Felis
(two species), Canis, and Ursus. He sums up the following
conclusion, which, from its importance, I quote i eatenso :—

‘From the foregoing details it will be seen that the
researches now applied during fifteen years to the Mammalian
fossils of the Red Crag of Suffolk have led to the very inter-
esting result, that the majority of them are identical, or

closely correspond, with Miocene forms of Mammalia, and

especially with those from the Eppelsheim locality, described

? Ann. and Mag. of Nat. Hist. vol. iv. | % Quart. Journ. Geol. Soe. vol. i.
p. 186. p. 87

2 Op. cit. 1856, vol. xii. p. 217.

GEOLOGICAL AGE OF MAST. ARVERNENSIS. 55

by Prof. Kaup. In Suffolk, as in Darmstadt, we find the
Mastodon longirostris, Rhinoceros Schleiermachert, Tapirus
priscus, Sus paleocherus, and Cervus dicranoceros, associated
together in the same formation; and, with these Miocene
forms of extinct Mammalia in the Red Crag, we have likewise
a Cetacean which most closely resembles a Miocene species of
that order, previously recognized in the Crag or Molasse of
the Continent. At the same time there are, as e.g. in the
Megaceros, specimens of newer Pliocene or Pleistocene forms
of Mammalia mingled with the older tertiary species; whilst,
on the other hand, Hocene forms of fish, as, e.g. Hdaphodon,
with Myliobatide and Eocene Crustacea, have been obtained
from the Red Crag pits.

‘As, however, several of the Mammalia which occur in
Miocene formations are also found in the older Pliocene
deposits in parts of France, it would be rash, perhaps, to pro-
nounce positively on the Miocene age of any of the above-cited
Crag fossils; but it is certain that the majority of those
Mammalian fossils, and by far the greatest proportion of indi-
vidual specimens, belong to an older tertiary period than the
Mammalia of the newer Pliocene drifts, gravels, brick-earths,
and bone-caves.’ (Loc. cit. p. 229.)

In this view, regarded in the most restricted sense, a very
mixed origin and complex character are attributed to the
Mammalian fossils of the Red Crag, and it would seem to be
open to several objections, some of which I shall now state.
Professor Owen, having satisfied himself that the Mastodon
of the Crag was identical with the Miocene species of Eppel-
sheim, was naturally predisposed, where the evidence was at
all ambiguous or indecisive, to regard the remains of the
other fossil Mammalia with a leaning towards a Miocene
origin. First, as regards the Rhinoceros; the HKuropean fossil
species of this genus, including Aceratheriwm, are at present
involved in such a maze of confused synonymy that no two
living paleontologists are agreed about the number, or upon
the names which ought to be applied to them. In conse-
quence, it is exceedingly difficult to arrive at any satisfactory
conclusion where a fossil Rhinoceros older than the Siberian
species forms an element of the discussion. In the ‘ British
Fossil Mammalia,’! Professor Owen adopts the opinion of
Christol, that Rhin. Schleiermacheri and Rhin. megarhinus are
synonyms of the same species, the former having been founded
by Kaup upon Miocene remains discovered at Eppelsheim, the
latter by Christol upon Pliocene remains from Montpellier.
From his late memoir it would appear that he now considers

1 Op. cit. p. 370.

56 BRITISH AND EUROPEAN FOSSIL MASTODONS.

them distinct, and he leans doubtingly towards the opinion
that the Crag molars of this genus, upper and lower, belong
to the Miocene Rhin. Schleiermacheri, rather than to Rhin.
megarhinus. But without going into details, it may be stated
that these teeth present no characters, so far as they have
been described, inconsistent with their being referred to the
so-called Rhin. megarhinus of the South of France and Italy.
The premolars possess the basal ‘ bourrelet’ which Christol
pointed out as one of the distinguishing marks of his Rhin.
megarhinus; it occurs, as stated by Professor Owen, in the
same teeth of Rhin. Schleiermacheri, and it is met with also in
the premolars of the Rhin. leptorhinus of Cuvier. Further, it
would seem to be clearly established now, that Cuvier was
quite correct in characterizing his Rhin. leptorhinus as
destitute of a nasal bony septum, and that Christol was misled
by the deceptive appearance of a drawing in assigning this
peculiarity to the original Italian specimen, and confounding
it with Rhin. tichorhinus.' There are also the strongest
grounds for believing that the Rhin. megarhinus of the
Pliocene sands of Montpellier is specifically identical with
Rhin. leptorhinus of Cuvier. The Red Crag specimens, figured
and described by Professor Owen, are undoubtedly very like
the corresponding teeth of Rhin. Schleiermacheri; but it seems
to me that the materials are not sufficient to establish a
satisfactory paleontological identification, and that it is at
present an open question whether they belong to Rhin.
leptorhinus of Cuvier, or to Rhin. Schleiermachert of Kaup.
The same remark applies to the Tapir of the Red Crag, which
Prof. Owen refers, on the evidence of a single upper and sin-
gle lower molar, to the Miocene Tapirus priseus of Kaup.
Pliocene species of Tapir have been met with both in Italy
and France, one of which has been named 7. Arvernensis
(Croizet and Jobert), and the other 7. elegans? (Pomel) ; and
a supposed third species, 7’. minor of Marcel de Serres, has
been yielded by the marine sands of Montpellier. The
adduced evidence would seem hardly sufficient to establish
that the Crag molars do not belong to either of these.

' Cornalia, in Duvernoy’s ‘ Nouvelles

Etudes sur les Rhinoc. Fossiles’ (Archiv. |

du Muséum, tom. vii. p. 99). He de-

scribes the original specimen, which is |

depositedin the Natural History Museum
at Milan, as perfectly free from any trace
of a bony septum, whether along the
median line of the nasals or upon the
floor of the nasal cavity. Christol, not
haying had access to the specimen, mis-
interpreted a shaded portion of a drawing

of it as a representation of the septum. |

| Dr. Cornalia’s remarks confirm, in every
essential respect, the previous description
by Cuvier.

2 Pomel, Catal. Méthod. et Descript.
| p. 84.

3 Gervais, Paléontol. Francaise, tom.
ii. p. 4, Pl. v. figs. 4 & 5. Gervais
doubts, with De Blainville, whether the
materials are sufficient at present to
prove that these Pliocene nominal species
really differ from the Tapirus priscus of
Eppelsheim. .

En a

GEOLOGICAL AGE OF MAST. ARVERNENSIS. 57

So also in regard to the Crag Suide referred by Professor
Owen to the Eppelsheim species, Sus paleocherus and Sus
antiquus of Kaup. The Crag specimens upon which the
identification is founded are limited in each case to a single
detached upper molar. The tooth referred by Prof. Owen to
Sus paleocherus assuredly bears a very close resemblance to
the figure of that of the Eppelsheim species with which he
compares it; but the evidence, it must be admitted, is too
limited to bear out a satisfactory specific identification; for
aught that is shown to the contrary, except a slight difference
of size, both of the Crag teeth may belong to the same
species. An extinct species of Sus, S. Arvernensis of Croizet,
has been found in the Pliocene strata of Auvergne; another
supposed species, S. provincialis of Gervais, in the marine
Pliocene sands of Montpellier; and species, as yet undeter-
mined, of the same genus, occur in the Pliocene deposits of
Italy. Is it certain that the ‘Red Crag’ molars of Sus differ
from all these ?

The Hquus of the Red Crag is stated by Professor Owen to
resemble in the molar teeth his Hquus plicidens of the Oreston
Cavern, reconcileable with a Pliocene origin. The evidence
respecting the teeth of the form considered by Prof. Owen
‘as probably of the subgenus Hipparion’ has not been ad-
duced. This subgenus had hitherto been regarded as strictly
confined to Miocene strata, but Gervais! has attempted to
distinguish several species from the marl beds of Curcuron in
the Vaucluse, the age of which, whether Miocene or Pliocene,
he alleges, still remains to be determined.

As regards the two Cervine Ruminants from the Red Crae
the determination of the form which Prof. Owen refers to
Cervus dicranocerus of Eppelsheim rests upon two shed antlers
and two detached molars. The horns undoubtedly closely
resemble those figured by Kaup of that species; but, as
Prof. Owen states, a species presenting the rare character of
a similar bifurcate form of antler, and named Cervus australis
by Marcel de Serres,” has been discovered in the Pliocene
marine sands of Montpellier ; and it has not been shown that
the Crag form differs specifically from it. The identification
which is most at variance with the conclusions hitherto ac-
cepted is that of the shed antler, said to be from a Crag pit
at Felixstow, which Professor Owen (in the reference to the
figure) describes as the ‘base of the antler of the Megaceros
Hibernicus;’ inferred to occur in a formation where the
majority of the Mammalian species are regarded as Miocene.

' Paléontol. Francaise, tom. i. p. 177.
* Gervais, Paléontol. Francaise, tom. i. p. 85. Pl. vii. figs. 1-8.

58 BRITISH AND EUROPEAN FOSSIL MASTODONS.

Any determination emanating from so distinguished a pale-
ontologist as Professor Owen must be entertained with the
respect which his great authority carries with it. But the
specimen in question, although (like most of the fossils of the
‘Red Crag’) highly impregnated with iron, and of corres-
ponding gravity, is encrusted with fresh patches of Lepralia
Peachii. Prof. Busk, to whom I am indebted for this identi-
fication, after a careful examination of the original, informs
me that the pearly appearance and transparency of the walls
of the cells indicate the modern origin of this marine Bryozoon.
Other species of the same genus are found in abundance upon
the fossil shells of the Red Crag, but they are invariably
more or less tinged with an ochreous colour, and the walls of
the cells are opaque. Instead, therefore, of having been
found in a Crag pit (the statement under which the specimen
came before Professor Owen), it would seem most probable
that it was dredged out of the present sea, from some locality
off the coasts of Suffolk or Essex. Teeth and bones of
Elephants and of other herbivorous mammalia, highly im-
pregnated with iron, and encrusted with marine Bryozoa, are
brought up by the dredge, or found upon the beach, at
intervals all along the coast from Mundesley to Harwich. A
large number of molars of Hlephas (Loxod.) meridionalis, pre-
senting a highly vitreous polish, heavy, and dark-coloured,
exist in Mr. Fitch’s collection at Norwich; and analogous
remains are to be met with in various collections in Suffolk
and Essex; yet it is not a little remarkable, considering the
numerous descriptions of the coast section which have been
made by different English geologists, that the particular
beds from which these remains have been derived have not
yet been determined with precision. No authentic case has
as yet been made out of remains of the Irish Elk in strata of
an older date than the period of the Mammoth, Siberian
Rhinoceros, and Ursus Speleus of the Glacial fauna; and the
palzontological evidence would require to be very conclusive
before the range of this species could be extended so as to
include the Pliocenes of the Sub-Apennine period.

As regards the Carnivora of the ‘Red Crag’ enumerated
by Prof. Owen, the evidence, so far as it has been published,
is of a very limited nature, being confined to detached teeth,
and is adequate for little more than the identification of the
respective genera. No Miocene species of Ursus has yet been
met with in Europe. The tooth from a Red Crag pit at
Newbourn, which Professor Owen guardedly describes as
‘somewhat smaller than the corresponding tooth of the
Ursus speleus,’ would correspond in size with that of the
Pliocene Ursus Arvernensis, found abundantly in Italy and

GEOLOGICAL AGE OF MAST. ARVERNENSIS. 59

Auvergne. Professor Owen admits that the carnassial teeth
specimens, from Newbourn and Woodridge, of his Felis par-
doides, do not differ in size from the Pliocene Felis pardinensis
of Croizet and Jobert, found in Auvergne, and it remains to
be shown that the former is specifically different from the
latter form. The remarkable sectorial tooth from the Red
Crag, which, according to Professor Owen, closely resembles
one of the ancient Carnivora called Hyenodon and Pterodon,
and which he suspects to be an indication of an extinct oscu-
lant genus, linking on the true Felines to the Hyana or
Musteline family, has not been generically determined;
and it may have been washed in from strata of the Hocene
age.!

If, on the other hand, a paleontologist, having satisfied
himself that the Red Crag Mastodon is an undoubted Pliocene
form, and finding the same species in the Fluvio-marine
Crag, were to infer that they were both of the same geologi-
cal age, and if he were then to take a group of some of the
well-established species as a starting point, he would expe-
rience little difficulty in reconciling many of the more doubtful
Mammalian species with a consistent Pliocene association.
The species would run in the following order :—The Probos-
cidea, M. (Tetralophodon) Arvernensis, EH. (Loxodon) meridion-
alis, and H. (Huelephas) antiquus; the Pachydermata, Rhino-
ceros leptorhinus or Rhinoc. ?, Tapirus Arvernensis, and
Hquus plicidens; the Carnivora, Felis pardinensis, Ursus
Arvernensis, and probably a Pliocene species of Canis. With
such an harmonious agreement in the great leading forms, he
would naturally look to Pliocene forms for comparison when
he met with scanty and indecisive remains of such a widely
distributed and extensive genus as Cervus, unless the charac-
ters were so pronounced as to be decisive of species of an
earlier age.

This is the manner in which I have been led to regard the
fossil Mammalia of the Red and Fluvio-marine Crag; and it
has appeared to me that (where remains obviously of an
anterior epoch have not been adventitiously intermixed) they
agree generally, so far as the species have been well deter-
mined, with the great Pliocene fauna of Italy, as exhibited
along the valleys of the Po and of the Arno. But it must at
the same time be freely admitted, that the materials upon
which the determination of many of the species of the Red
Crag Mammalia at present rests are so scanty and indecisive,

? Quarterly Journal of the Geol. Soc, vol. xii. ps 237, fig. 20.

60 BRITISH AND EUROPEAN FOSSIL MASTODONS.

that the identification, either way, whether as Miocene or
Pliocene forms, must be regarded as little more than approxi-
mative.

There are other considerations which corroborate the Plio-
cene view of the Mammalian fauna of the Crag. The debate-
able species referred by Prof. Owen to a Miocene origin all
belong to genera that are common to the Miocene and Plio-
cene periods, such as Mastodon, Rhinoceros, Tapirus, Sus,
Cervus, and Felis. But of the more remarkable types which
are limited to the upper Miocene deposits, and which abound
in them all over Europe, such as Dinotherium, Chalicotherium,
Aceratherium, Anchitheriwum, Amphicyon, &c., not a single
remain has ever been cited as having been found in the Crag
deposits. The question naturally arises, how does it happen,
if the majority of the Red Crag Mammalia are Miocene, that
there has been this selective admixture of species of long-
termed ‘ Miocene’ genera in the Crag, and why the exclusion
of the strictly characteristic genera ?

Another view may be taken, that, as the Red Crag contains
Fish and Crustacean remains which have been inferred to
have been washed out of denuded Hocene deposits, so the
Pliocene sea-bottom of the Red Crag may have had Miocene
mammalian remains washed into it, thus causing an extrane-
ous admixture among the Pliocene mammalian fossils. But
it may be asked in reply, where are the Falunian deposits, in
proximity with the Crag in England, from which such a
washing-in could have taken place? And, if they were
transported from a distance, they ought to show marks of
abrasion from rolling, which, so far as my observation goes,
are not seen in a great many of the Red Crag Mammalia to
which a Miocene origin has been attributed. Many flattened
pieces of bone, exhibiting a high vitreous polish, and bearing
palpable marks of having been long rolled in the sea among
shingle, have unquestionably been met with in the Crag;
but it does not necessarily follow that they were all washed
out of an older deposit. It is intelligible that the effect
may have been produced by attrition caused by the waves of
the Crag-sea upon bones of animals of the same geological
period.

It now remains to consider how far the Cetacean fossils of
the Crag are in accordance with the inferred Pliocene cha-
racter of the Land Mammalia. Professor Owen has described
‘ Cetotolites’ of five species of Balenide from the Red Crag.
He states (Brit. Foss. Mam., p. 527), that they ‘appear to
have been dislodged from a subjacent Eocene deposit ;’ and
the same opinion is repeated in the note appended to the

. GEOLOGICAL AGE OF MAST. ARVERNENSIS. 61

‘Conspectus’ of British fossil species which I have already
cited. They are there arranged under the head of Eocene,
and excluded from the Miocene fossils. Cetacean remains
have been met with in abundance in the Pliocene deposits of
Italy, under circumstances which leave no doubt that they
are of the same age as the land quadrupeds found associated
with them. I have already mentioned the case examined by
myself, where the skeleton of M. (Tetraloph.) Arvernensis,
covered with marine incrustations, was found in the ‘ Panchina
inferiore’ of the lower Val d’Arno near Leghorn, associated
with the entire skeleton of a whale referred by the Italian
naturalists to Physeter, and with Dolphin remains. A still
more remarkable and conclusive instance is furnished by the
rich and well-known deposit of Pliocene Mammalia investi-
gated by Cortesi in the Sub-Apennine deposits near Piacenza.
Monte Pulgnasco is stated to attain an elevation of about
1,700 feet above the level of the Adriatic,’ and near it there
are lower elevations, Monte Zago and Della Torazza. The
upper beds in all three alike, to a great depth, consist of red-
dish calcareous sands full of marine shells; and below these
there are beds of blue clay (‘Marna cerulea’), also loaded
with similar shells, both being of the Sub-Apennine Pliocene
age. Cortesi discovered in the blue clay, at different points,
nearly entire skeletons of extinct whales, referred by Cuvier
to the Rorquals (Balenoptera Cortesii and Balenoptera Cu-
vier *), and of Dolphins allied to Phocena Orea, but differing
in the form of the cranium (Phocena Cortesii,? and other
species unnamed). Near the summit of Monte Pulenasco, in
the overlying stratified sands, the greater part of a skeleton
of the Val d’Arno Elephant, EH. (Loxod.) meridionalis, was
discovered ; and upon Monte Zago the original skull, together
with many other bones, of the individual Rhinoceros upon
which Cuvier founded his Rhinoc. leptorhinus as distinct from
R. tichorhinus. The Rhinoceros skeleton was found in the
sandy strata, but resting immediately upon the blue clay, and
with upwards of 200 feet of strata above it. I was enabled, by
the kind permission of Dr. Emilio Cornalia, to examine the
fine collection of these Monte Pulgnasco remains deposited in
the Natural History Museum at Milan, including, among
others, the palate specimen of the Elephant described by
Cortesi,‘ which I found to be identical with EH. (Loaodon)
meridionalis of the Val d’Arno and fluvio-marine Crag.

Here are two cases of the association of Pliocene Cetacea
with terrestrial Mammals, under circumstances where extra-

1 Cortesi, Saggio Geolog. 1819, p. 72. 8 Op. cit. tom.iy. p. 634.
_ ? Diction. Univers. d'Histoire Natur. * Cortesi, op. cit. p. 68, pl. vi, figs.
tom. ii. p. 443, 1 ee,

¢2 BRITISH AND EUROPEAN FOSSIL MASTODONS.

neous admixture is inadmissible. Cetacean remains were
long ago described by Cuvier from the Crag of Antwerp.’
Lyell found in the same formation numerous specimens of
bones said to be of Balenoptera and Ziphius, which bore no
marks of rolling as if washed out of older beds; and he
inferred that the animals to which they belonged once co-
existed in the same sea with the associated Crag Mollusca.’
He considers the strata to be Older Pliocene, equivalents of
the Red Crag and Coralline Crag.

Professor Owen, in his late memoir, enumerates some
additions to the Cetacean remains from the Red Crag described
in the ‘ British Fossil Mammalia.’ Among these are portions
of an upper jaw very closely resembling the Dioplodon Becanu
of Gervais (Ziphius of Cuvier), and ‘ water-worn teeth corres-
ponding in size and form’ with those of the Hoplocetus
crassidens, an obscure and as yet imperfectly determined form
provisionally so named by Gervais,’ from the Miocene Faluns
of La Dréme. Another supposed species of the genus, named
Hoplocetus curvidens by the same paleontologist, is founded
upon specimens procured from the Pliocene sands of Mont-
pellier. The Crag ‘ Cetotolites’ (i.e. the same species) have
nowhere as yet been described as occurring in Eocene beds
in England; and the whole bearing of the evidence would
seem to indicate that at least a considerable part, if not the
whole, of the ‘Red Crag’ Cetacea are of the same age as
the associated terrestrial Herbivora.

[Since the preceding pages were in type, I have had an
opportunity of examining specimens in some of the principal
collections in Essex, Suffolk, and Norfolk, which throw light
upon some of the points discussed above. In the Town Hall
of Colchester there is a fine specimen, comprising both
maxillary bones of a young Elephas (Eueleph.) antiquus, and
presenting the last milk molar (right side) in place. The
matrix is very ferruginous, and the bones and tooth are of a
dark-chocolate colour, with a vitreous polish. It was dredged
up from off the ‘West Rocks’ on the Essex coast; and it
resembles in its mineral condition the large Cervine horn
reputed to be from a Crag pit at Felixstow, and referred by
Professor Owen to Megaceros (see above, p. 57).

In the rich and valuable collection of Red Crag fossils
belonging to William Whincopp, Esq., of Woodridge, there

1 Oss. Foss. tom. y. p. 352. yais throws out a suggestion, that his
? Quart. Journ. Geol. Soe, vol. viii. | Hoplocetus may have a connexion with
p. 281; and Manual of Geology, 5th | the Balenodon of Professor Owen, but
edit. p. 174. does not enter into a detailed compari-
% Paléont. Frang. tom. i. p. 161. Ger- | son,

GEOLOGICAL AGE OF MAST. ARVERNENSIS. 63

are two upper and three lower molars of a species of Hippo-
therium from the Red Crag pits at Sutton. They beara close
resemblance to the Miocene H. gracile, Kaup, from Eppelsheim.
The same collection contains several molars, upper and lower,
of the genus Rhinoceros, one of which (an upper antepenulti-
mate milk molar) agrees, in most of the characters, with an
original specimen of a corresponding tooth of Rhinoceros
Schlevermacheri from Eppelsheim, with which it was compared.
Mr. Whincopp also possesses an upper maxillary bone con-
taining a series of the molar teeth of Hyracotheriwm leporinum;
also detached molars apparently of the smaller species, Hyrac.
ewniculus, both said to have been procured from the Red Crag
at Felixstow. Besides these, Mr. Whincopp possesses: Ist,
several perfect Cetacean teeth, resembling those referred to
Hoplocetus by Gervais ; 2ndly, two remarkable molar teeth of
a form which has not hitherto been described as a British
fossil; and 3rdly, numerous remains of Red Crag Del-
phinide.

In the rich collection of Edward Acton, Hsq., of Grundis-
burgh, there are specimens referable to both species of
Hyracotherium, and reputed to be from Red Crag localities in
Suffolk, besides molars of Tapirus and Rhinoceros. Myr.
Acton also possesses a singularly perfect antepenultimate
true molar from the lower jaw of M. (Tetralophodon) Arver-
nensis, Showing the peculiar characters of the species strongly
marked.!

In neither of these collections did I observe any specimen
referable to M. (Tetraloph.) longirostris of Eppelsheim, nor to
the peculiar Mammalian genera of the Upper Miocene period,
enumerated in a preceding paragraph as being usually
associated with that species (p. 59). It is manifest that the
Hyracotherian remains must have been derived from broken-
up HKocene deposits ; and the teeth of Hippotheriwm indicate
a similar inference of Miocene remains being mixed up with
Pliocene forms in the reconstructed materials of the Red Crag
deposit.—H.F., Oct. 20th, 1857.]

Conclusion.—On a review of the various facts and con-
siderations discussed in the preceding pages, it seems clear
that the Mammalian fauna of the Fluvio-marine Crag is of a
Pliocene age. The undoubted association of M. (Tetraloph.)
Arvernensis and of H. (Loxodon) meridionalis in this deposit
admits of no other inference. The mixed contents of the
Red Crag, including Mammalian remains of different strata

1 In Mr. Whincopp’s collection there | molar, from the Red Crag, closely re-
is a very beautiful specimen of an intact | sembling the specimens figured by
germ of an antepenultimate upper milk | Croizet and Jobert.

64 BRITISH AND EUROPEAN FOSSIL MASTODONS.

from the Eocene period upwards, are inferred to have been
deposited in the reconstructed strata also within the Pliocene
period, since M. (Tetraloph.) Arvernensis, which occurs so
abundantly in the Red Crag, has not been met with anywhere
on the Continent of Europe except in deposits of a Pliocene
age. The Red Crag sea appears to have breached a previously
established and populated Pliocene land, and to have buried
the bones referable to various epochs in the same sea-
bottom.

In the preceding remarks I have purposely excluded any
reference to the Shell-evidence, and confined the comparison
strictly to the Mammalian Fauna. The Mollusca have
unquestionably been wielded as a most powerful exponent of
geological chronology, and of the successive physical changes
which have taken place on the surface of the earth. But it
will hardly be denied that the evidence presented by Mamma-
lianremains, when obtained in sufficient variety and abundance,
is of greater significance as a test of contemporaneous forma-
tion in geology, or the reverse :—Ist, Because Mammalian
genera and species are everywhere shown to be of more
limited duration in time than the Mollusca; and, 2ndly,
because from the vastly greater complexity of their relative
functions, they are much more susceptible of being affected
by the altered climatal conditions which are necessarily
involved in every great physical change, and which conduce
most to the extinction of species.

The conclusions to which the comparison has led are :—

1. That the Mastodon remains which have been met with
in the ‘ Fluvio-marine Crag’ and ‘Red Crag’ belong to a
Pliocene form, Mastodon (Tetralophodon) Arvernensis.

2. That the Mammalian Fauna of the Fluvio-marine Crag
bears all the characters of a Pliocene age, and is identical
with the Sub-Apennine Pliocene Fauna of Italy.

3. That the Red and Fluvio-marine Crags, tested by their
Mammalian Fauna, must be considered as beds of the same
geological age.

MASTODON TAPIROIDES IN RUSSIA. | 65

APPENDIX TO MEMOIR ON MASTODON.

I.—Note on J. F. Branpt’s Memoir ON THE SKELETON OF A MAstopon
DISCOVERED NEAR NikoLaJEW (NIcoLAIerF), IN SouTuern Russia.
By Dr. Fatconer anp Mr. T. Rupert Jonss.!

Early in 1860 the Imperial Academy of Sciences at St. Petersburg
received a notice, with drawings and photograph, of the remains of a
large Elephantine animal found in the South of Russia, twelve werst
from Nikolajew, to which attention was first called by the army-sur-
geon, M. Wassilijew. From an examination of the photograph, and
from information (from Admiral Butakow) as to the shape of the
lower jaw, M. Brandt suggested that the remains may have belonged
to. Mastodon angustidens. ‘'The portions of the skeletons of Masto-
dons hitherto found, so far as I know,’ says M. Brandt, ‘in the
Middle and Upper Tertiaries of the various countries of Europe, such
as Germany, and here and there in Russia, have been only isolated
parts, principally molars, and more rarely fragments of the lower jaw.
The Museum of the Academy possesses the half of a lower jaw, fur-
nished with two molars, dug up in the Chersonese Government, near
the town of Ananjew. Nordmann and Hichwald have described some
molars of Mastodon likewise found in Southern Russia.’ M. Brandt
recommended the acquisition of the Nikolajew specimen for the
Academy.

In June 1860 M. Brandt sent from Nikolajew to the Academy a
report of the proceedings of the expedition to that place, entrusted
by the Academy to his management. After giving an account of
the collections inspected at Moscow, Charkow, and elsewhere, he
describes the arrival of himself and scientific companions at Nikolajew
on the 3lst May, the welcome they received from Admiral von
Glasenap, and the cordial cooperation of that gentleman and others in
the examination of the bones and in the search for other remains.

The skeleton of the Mastodon had been found in a ravine (formed
by spring-floods) about a werst distant from the village of Waskress-
ensk (or Gorochowo), and disappearing on the Ingul, at the place
where this river (an affluent of the Bug) makesa bend. The ravine
bears at first, from its head, from S. to N., then it takes a NW. direc-
tion. In the upper part of the ravine the rocky strata are denuded,
and subsequently they disappear with the change of direction, and allu-
vial soil only is seen at the entrance of the gully.

As early as 1854, after a very rainy season, several large bones had
been found here; subsequently the nearly perfect skeleton of the
Mastodon was found near the upper part of the ravine, at a depth of 3
‘sajen’ and 2 ‘arschin;’ the arrangement of the strata being, in
descending order, as follows :—

1. Black humus; 9 inches (English).

Reprinted from the ‘Quarterly Journal of the Geological Society’ for May,
1862.

VOL. ITI. EF

66 BRITISH AND EUROPEAN FOSSIL MASTODONS.

2. A thin calcareous layer, compact, made up of shell-fragments, 6
inches thick, passing into—

3. A soft grey and white limestone, of oolitic structure, with casts of
shells; 5 inches thick.

4. Soft yellowish-grey sand, here and there brownish-red, with oxide
of iron, harder beneath, without fossils; 8 inches.

5. Harder sandstone, alternating with beds coloured with oxide of
iron, and traversed by layers of clay of various thicknesses, passing
downwards into sandy clay with siliceous concretions, but no fossils; 7
feet (English).

In this bed was found the Mastodon; and not far off, in the same
stratum, was found a layer of a kind of brown coal, an inch thick.
Under this layer a stratum of limestone was observed only a few feet
thick ; it contained a Cardium. Of all these beds the bottom clay and
limestone are the only two which are constant. The bones of the
Mastodon skeleton that have been saved consist chiefly of the tusks and
molars of the upper and lower jaws, the lower jaw, an almost perfect
shoulder-blade, nearly all the ribs, a great number of cervical and dorsal
vertebra, and the tolerably perfect bones of the fore foot.

The bones were ina very fragile condition, and their extrication from
the firm, moist, loamy earth required great caution. Careful drawings
were made of the relative positions of the bones on the spot; and the
fragments were carefully numbered, so that it is hoped they will serve
to construct, in the Museum of St. Petersburg, a tolerable skeleton, that
in its completeness will be one among the best of the preserved speci-
mens of the ancient Mastodons. The bones have already reached St.
Petersburg, and have been placed in their proper collocation by the
Conservator Radde.

In November, 1860, a supplemental notice, illustrated by drawings, of
these remains, was read before the Academy by M. Brandt. The
drawings are represented by a large lithographic plate in the ‘ Bul-
letin,’ and are described at pp. 507-509. All the bones appear more
or less displaced, some only slightly; the skull was crushed, and its
bones nearly all destroyed by the action of the weather. The back
upper molars lay apart from each other. The almost straight tusks,
6 feet 8 inches long, and thickest at the base, were but slightly dis-
placed, although their alveoli had been destroyed, and they themselves
broken into many pieces. The tusks of the well-preserved lower jaw
were in their natural position, in sockets in a short characteristic
symphysial process. The imperfect cervical vertebrae were partly
displaced, and, like most of the anterior dorsal vertebra, were
more or less broken or decayed. Only a few of the middle and
posterior dorsal vertebree were tolerably preserved; indeed, but a
small proportion of them were found in their natural position. The
number of the ribs remaining nearly perfect indicated, as a general
rule, that all those which lay obliquely were, for the most part, in a
tolerably good state of preservation. 'The majority of these appeared
more or less dislocated, with the exception of the posterior ribs of the
left side, which were only slightly displaced. ‘The greater part of
the left shoulder-blade was preserved. The right humerus, greatly

displaced from its natural position, and lying close upon the vertebral —

column, is more entire than the left, which is, in connexion with the

Ape $\-

MASTODON TAPIROIDES IN RUSSIA. 67

bones of the fore arm, crushed outwards. The figures, however,
represent only a part, although certainly the chief portion, of the
original depot of the bones of the Woskressensk skeleton,—to wit,
those which M. Brandt and his colleagues had been able to observe
in their natural position. Before their arrival, several detached
bones or fragments were found, lying scattered close to the excava-
tion of the principal remains, and belonging chiefly to the extremities ;
these fragments were separately preserved, and presented to the
Commission on their arrival. Moreover, the lower end of the right
scapula had been sent to Odessa, to the Governor-General Count
Strogonow, from whom they subsequently received it.

The bones in question are evidently a part of the imperfect remains
of the bones of the extremities, which, as stated in the preceding re-
port, had been discovered a few years ago. They lay in a superficial
stratum of earth; so that the figured part of the remains, such as the
lower portion of the head and the greater part of the trunk, parti-
cularly the anterior and middle portions, lay at a lower level, and
were covered by a somewhat deeper layer of soil. From this dispo-
sition of the remains, it is intelligible how the displacements of the
bones and the destruction of the skull took place.

The close study of the remains places it beyond doubt that they
belong to an Elephantine form ; and further, from the mammillated
crowns of the molars as well as the lower jaw, that they are of a
Mastodon. From the drawings which were in the first instance sent
here, says M. Brandt, I was disposed to ascribe them to Mastodon an-
gustidens, Cuv., ep. Mast. angustidens, Owen (Brit. Foss. Mamm.
p- 271), Blainville (Ostéogr. Gravigrades)= Mastodon arvernensis,
Croizet et Jobert (Ossem. Foss. du Puy de Déme)= Mastodon longi-
rostris, Kaup (Ossem. Foss. de Darmstadt, p. 65)=Mastodon Cuvieri
Pomel (Bullet. géolog. 1848, p. 257). A closer but in nowise satis-
factory study of the involved and tangled synonymy of the Masto-
dons led me, however, to abandon the earlier opinion formed from
the drawings, in consequence of the different form of the crowns of
the molars, as also the exceedingly short, straight symphysial pro-
cess of the lower jaw. Aastodon angustidens, Cuv. (magna e.p.),
Owen (=Mastodon, longirostris, Kaup), possesses a very prolonged
and deflected symphysial process, half as long as the entire length of
the lower jaw, with moderately stout tusks, while the crowns of the
molars are characterized by the circumstance of constantly presenting
in the unworn state a small and accessory outlying tubercle, interposed
between each pair of the strongly compressed principal tubercles on
their broader surfaces.

In contrast to the characters just indicated of Mastodon angusti-
dens, which would be considered identical with I. longirostiis, it
may be said that in the Nikolajew remains the symphysial process,
together with the straight tusks, does not attain a quarter the length of
the lower jaw. The broad surfaces of the crown tubercles of the
molars are but slightly folded, and have no accessory tubercles between
them. The upper and elongated tusks are quite straight.

With reference to their form resembling that of the Tapir, the
molars of our skeleton agree best with those of the Mastodon Tapi-
roides, Cuy., figured by De Blainville (Ostéog. Gravigrades, PI. XVII.).

F 2

63 BRITISH AND EUROPEAN FOSSIL MASTODONS.

The Nikolajew skeleton may therefore be referred, on the best
erounds, at any rate provisionally, to Mastodon Tapiroides. The
remains in question thus determined, since they cannot well be re-
ferred to Mastodon longirostris, would appear to possess a positive
scientific value, and are calculated to establish on more definite grounds
a species hitherto accepted only from the characteristic form of the
molars. At the same time they demonstrate that, at least in Europe
and Russia, another species of the genus J/astodon existed, besides
Mastodon longirostris.

The significant fact referred to in the preceding report is worthy
of attention, viz. that a few steps from the site of the Mastodon remains,
and in one and the same deposit, there was found a layer, about an
inch thick, of a rusty, incompact wood, approaching the condition of
lignite. The origin of it can only be explained thus, that the place -
where the remains were found bore forests during the period of exist-
ence of the Mastodon, whilst at the present time its surface presents
bare tracts of steppes or prairie-land. From what we know of the
habits of the existing Elephants, it may also be reasonably inferred
that the wood in question constitutes a part of the remains of arbo-
real forms, the young twigs and leaves of which furnished at least a
part of the food of the Mastodons. We may lay the greater stress on
this view, as the remains of our Mastodon, which were tolerably con-
nected with each other, or at any rate not very far separated, belonged
to an individual that died at no very great distance from the place
where they were found. [H. Poe "FoRade]

Notre.—The Nicolaieff Mastodon, as above indicated by Professor
Brandt, appears to belong to M/. Tapiroides; but, as De Blainville,
to whose figures the author refers, confounded two distinct species
under this name, viz. Jf. Borsoni and M. Tapiroides, the former a
Pliocene form, and the latter from the Middle Miocene deposits of
France and Switzerland, it is important to add, that the Nicolaieff
skeleton belongs, so far as a determination can be rested on the
figures, to the M. Tapiroides proper of the French paleontologists,
being the M. Turicensis of Schinz, from the lignite beds of Keepfnach.
See Schinz, Schweitz. Denkschr. vol. vii. p. 58, Pl. I. fig. 1; De
Blainville, Ostéographie, Gen. Eléph. Pl. XVII. sup. 5 & 6¢, infer. 1 &
62; Lartet, Bulletin Soc. Géol. de France, vol. xvi. p. 486, Pl. XV.
fig. 8.—H. F.

IJ.—Extracts rrom Dr. Fatconrer’s Note-Booxs.

A. Mastopon ANGUSTIDENS AND M. Borsont.
Zurich Museum, August 29, 1856.

Returned here from Basle this morning, to examine the Probosci-
deans fossils from Keepfnach, &e.

Examined : 1st. The tooth figured by Schinz, Tab. I. fig. 6, being
the penultimate upper right of Mast. angustidens found in Keepfnach. —
It is of avery broad oblong shape, with three ridges and intermediate
mammille interrupting the continuity of the valleys. The large
mammillz are very blunt and converging, as in Mast. angustidens.
The enamel is very smooth; the posterior talon descends outwards
from the last inner tubercle, and is crenulated with a number of —

MASTODON ANGUSTIDENS. 69

minute mammille, as in Schinz’s figures. It is from the lignite of
Koepfnach, and is black.

Length of crown, 4°4 in. Width of crown, Ist. ridge, 2°7 in. Width of crown,
2nd ridge, 3-1 in. Width of crown, last ridge, 2°9 in. Height of crown (about)
1:7 in.

The internal basal bourrelet is very strongly developed.

2nd. Examined the original of the lower jaw, young specimen from
Buchberg, which is exceedingly interesting from showing the germ of
a vertical dent de remplacement in front of the worn 8rd milk molar.
It is of the right side, showing the horizontal ramus nearly entire,
(except at the posterior angle), and the ascending ramus. The leafy
expansion is exhibited, but the condyle and coronoid are broken off.
Two exserted teeth are seen in the horizontal ramus, and one behind ;
an included germ is disclosed by a lateral breach of the alveolar cavity
on the inner side.

The posterior protruded molar is the third milk molar, of which the
greater part of the crown is broken off, showing only the anterior ridge,
nearly entire, and the posterior talon intact. The anterior ridge is
only slightly touched by wear. It shows a very marked anterior
dise of pressure, and the accessory mammille behind interrupting the
valley.

Length of crown, 2°8 in. Width of crown at 1stvidge, 1-4 in.

Of the penultimate milk molar a small portion remains posteriorly,
the great mass of it having been extruded by the germ of the cor-
responding vertical successional premolar, which is seen considerably
protruding, but much below the level of the 3rd milk molar, and
quite intact and in germ, indicating its nature; only the last ridge
remains, showing complex converging minute mammillx, very different
from the transverse character of the ridges of Mast. Tapiroides. The
cast of the shell remains, indicating a length, including fang, of about
2 inches, by a width of about 1:2 inches.

Behind the 3rd milk molar is seen the shell of the Ist or antepen-
ultimate true molar, showing three ridges and a heel, but not sufficiently
disclosed for description, except that the inner tubercle of each ridge
is blunt and rounded (the W/. angustidens character), and not trenchant.

The anterior part of the jaw is broken off, giving no hint as to the
form of the symphysis.

8rd. Examined also a superb undescribed specimen of the penultimate
and last molars of the lower jaw left side, of the same species, adhering
together, but retaining no portion of the jaw. The anterior tooth is
well worn on the three ridges, but the anterior ridge only has its points
worn down into one common disc. The discs are simple in their out-
line, and very much like those of figs. 9 and 9 a of M. angustidens in
Plate XL. of the ‘Fauna Antiqua Sival.,’ but less touched by wear.
The greatest difference is in the talon (posterior) which has one large
point with numerous crenulated points forming a small transverse ridge.
The intermediate mammille are exactly as in fig. 9 a of the ‘ Fauna
Antiq. Sival.’; there is no basal ridge. The tooth in form of crown
is less cucumber-shaped than usual.

Length of crown, 4°8 in. With at lst ridge, 2°35 in, Width at 2nd ridge, 2°76 in.
Width at 3rd ridge, 3° in.

The lateral outline of the crown, scen from above, forms salient pro-
jections, and re-entering hollows causing constrictions.

70 BRITISH AND EUROPEAN FOSSIL MASTODONS.

Constriction at base between 1st and 2nd ridges, 2°3 in. Constriction between
2nd and 38rd ridges, 2°5 in.

The last true molar is in germ, and shows only the two anterior
ridges, the posterior half being broken off. The points are very high
and convergent, the outermost on either side the largest. The bridge
connecting the 1st and 2nd ridges, across the valley, consists of numerous
minute mammille, chiefly connected with the posterior surface of the
Ist ridge.

4th. The tooth, fig. 7 of Schinz’s Plate I., shows nearly a square
crown, with four discs of wear, indicating four points. Itis ef the right
side, and appears to be the 2nd milk molar upper right, as it bears a
well marked impression of disc of pressure, both in front and behind,
showing that it had acted on a tooth in front, and had also been pushed
from behind. It has a well marked basal bourrelet.

Length of crown (about) 1°5in. Width of crown (about) 1°46 in.

Although worn there is very little difference of level between the
enamel and ivory of the disc depressions.

Schinz’s figure is tolerably good. The tooth is described by him
as being from Elgg; but this is denied by himself on inquiry, and
also denied by Mons. Escher de la Linth. It is stated to be from
Keepfnach.

5th. Examined also a beautiful little detached germ-tooth, correspond-
ing very closely with Schinz’s fig. 8, with a complex oval converging
crown. It is intact, and as it bears no disc of pressure either in front
or behind, although the fangs are well developed, I regard it as the
upper vertical successional premolar, which pushed out the second
above. But it is very much smaller than the ascertained premolar of
the lower jaw, and may therefore be the first milk molar upper jaw.
The fangs are double and well developed. It is from Keepfnach.

Length of crown, 1°3 in. Width of crown about 1° in.

I have not been able to find the specimen fig. 3 of Schinz, described
by him as an ‘ Elgg Mastodon.’

6th. Examined also two superb fragments of the extremity of the tusk
of a Mastodon from Keepfnach.

One of these measures 18 inches in length. It is remarkable in two
respects: Ist, that the upper concave side is gradually compressed into
atrenchant edge, the transverse section being pyriform; 2nd, that, when
the outer layer is removed, the surface below shows the grooves of broad
channelling, noticed by Cuvier, and figured by Schinz. The grooving
is especially shown on a flattened surface, near the point on one side.

Length about 18° in. Depth at thick end, 4:2 in. Width, 2°8 in.

The upper edge is so sharpened off as to be like a keel to the concave
side. I have seen nothing like it in any other Mastodon. The ivory is
as black as the lignite, and so light and altered that it seldom shows
the guilloché, or engine-turning. I could not make out that there is
enamel, but it may be there. Some other fragments nearer the base
are oval in section, and not pyriform.

Vertical diameter, 4:7in. Transverse diameter, 3°7 in.

1 Extract from letter to M. Lartet, dated ‘Jn the hurried note which I sent you
Sept. 30, 1856 :— from Toulouse I forgot to mention that

ae re

MASTODON BORSONI. 71

In these fragments there are broad shallow channels.

7th. Mastodon Borsoni. Second species of Swiss Molasse Mastodon.

A superb, conglomerated, lignite specimen, containing three teeth in
situ, but dislocated, and judged to be of the upper jaw, right side. I
will describe the teeth from behind forwards, in the order of their com-
pleteness.

These molars are at once distinguished, all of them, from the Mastodon
angustidens specimens, by the ridges being transverse, elevated, and
trenchant, without intermediate mammille interrupting the valleys, and
by their general similitude to Mast. Ohioticus.

The last tooth is in germ, and partly embedded on the inner side in
the alveolus, concealing the bourrelet. It is the penultimate or second
true molar, showing three transverse ridges intact. A mesial line of
longitudinal bipartition is seen. The points on the inner side are the
largest, and the inner division of the two front ridges consists chiefly
of this point. The posterior mesial surface of the front ridge sends
down a vertical keel as in the specimen of M. Borsoni at Turin. The
anterior talon is a transverse basal bourrelet, connected with the
front inner point by a vertical ridge. The bourrelet is continued
around the base on the inner side, but is partly concealed. The pos-
terior talon is an insignificant basal bourrelet, very little developed.
The three ridges are nearly of the same height and width, the crown
being oblong.

Length of the tooth, 3°5 in.

Width, at 1st ridge, 2°2 in. Width, 2nd ridge,
2°45 in,

Width, 3rd ridge, 2°45 in. Height of crown (shell), 1°75 in.

It agrees very nearly with the cast-measurement taken at Geneva, and
is evidently the same tooth.

The next tooth in front is the 1st true molar (for reasons connected
with the next to be described). The posterior ridge and the inner half
of the middle ridge are broken off, giving no measurements. The
anterior ridge is entire and very slightly affected at the edge by wear,
trenchant as in M/. Ohioticus, with a very pronounced basal bourrelet,
which is continued in front transversely in a basal slightly developed
talon. There is no interruption of the valley.

Length of crown about 3:1 in. Width of Ist ridge, including bourrelet, 1-9 in.
Width of 2nd ridge, including bourrelet, 2°3in. Width of 8rd ridge, about 2:4 in.

The cast-specimen at Paris shows that this cannot be the third milk
molar.
Immediately in front of the first true molar, but dislocated to the

inside of it, is an entire germ of a

Mons. L. showed me several fragments of
a well-preserved tusk of compact tex-
ture and very hard, very different from
anything I saw with you at Seissan.
The section is round or oval, but has no
tendency to be pyriform. The original
surface is well preserved with a kind of
cortical outer layer, in some parts a good
deal sil/oné transversely with fine wavy
lines and no appearance of enamel (pro-
bably thealveolar portion). It was found,
if I remember rightly, where Mastodon

square tooth, with two transverse

Tapiroides and Dinotherium are found.
The fragments do not fit together, but the
tusk seemed to be concave at the point
and a good deal curved. This specimen
has made me feel uncomfortable about
my inferences respecting the M. Tupi-
roides origin of the tusk specimens I saw
at Zurich. There is not the least ap-
pearance when the cortical layer is
removed, of the sé//ons longitudinales.
Can it be the incisor of Dinotherium ?’—
[ Ep. |

72 BRITISH AND EUROPEAN FOSSIL MASTODONS.

trenchant ridges, a basal inner bourrelet, and a front and back talon
developed like the bourrelet. Each ridge consists of about six points.
This is evidently the vertical premolar succeeding the last milk molar,
as it shows no disc of pressure either in front or behind.

Length of crown including talons, 1°8 in. Width of crown at front ridge,
1:3in. Width at second ridge, 1:5 in.

The tooth is partly embedded; the width of the trenchant edges is
nearly as great as of the base.

In front of this there is an obscure appearance of another minutely
cusped tooth, probably another premolar ! in which case it would be the
successor of the second milk molar. But the mass is too conglomerated
to make this certain. There had evidently been a tusk along the side
which has left its grooved impression.

8th. Another beautiful small conglomerated fragment consists of :
1st, a portion of grooved tusk distinctly showing guilloché; 2nd, a square
two-ridged tooth exactly the counterpart of the last, and also in germ ;
and 3rd, a small two-fanged oval crowned complex germ tooth, exactly
like that described as being the vertical premolar of Mast. angustidens ;
but dislocated and at right angles to the other tooth.

Length of two-ridged tooth, 1°8in. Width at front ridge, 1:4 in. Width at
back ridge, 1°55 in.

In every respect, in size, in form of ridges (which are not parallel
in either), it is exactly like the other, and probably of the opposite
side.

The small tooth in the first specimen would therefore seem not to be
of Mast. angustidens, but of 11. Borsoni.

The grooved tusks would also seem to be of this species.

All these fragments are marked on the labels as being from Elgg, but
this is denied by Mons. Escher.

9th. Examined a beautiful germ-specimen of the penultimate of lower
jaw, left side, with a disc of pressure in front, but quite unworn. The
ridges very trenchant; a vertical keel to the inner tubercle of front
ridge; the ridges sloping a little backwards and outwards. No inter-
ruption to valleys; outer and inner points most elevated ; middle points
concave; posterior talon transverse, consisting of small crenulated points;
anterior talon very little developed ; no basal bourrelet ; and no constric-
tion as in Mast. angustidens.

Length of crown, 411 in. Width at 1st ridge, 2°25 in. Width at 2nd ridge,
2°55in. Width at 3rd ridge, 2°75 in.

I infer this to be the penultimate of Mastodon Borsoni, but the
position is not absolutely determined.

In all these specimens, a little connecting keel descends from the
anterior and posterior sides into the valleys without in the least in-
terrupting them; from the inner points (outer Lartet) of the upper
teeth, and outer points (inner) of the lower; being a very reduced
condition of the bridge seen in Kaup’s Mastodon longirostris; it is
very minutely crenulated, forming a keel. There is also a marginal
crest to the outer of upper, and inner of lower, but less marginal in the
upper.

10th. Another specimen, containing the third milk molar (anterior
ridge broken off) of the lower jaw, presents the same characters. It has

Ly

a

MASTODON ANGUSTIDENS. 73

also the first true molar behind it, the crown being entirely broken off,
but shown to be in germ. A portion of a grooved tusk is attached. If
of the lower jaw, it would be of left side. It corresponds very much
in size with the upper third molar above described in the principal
conglomerated specimen ; the vertical ridge is very well marked.

Width of 1st ridge, about 2°2 in. Length of crown, about 3°3 in.

The ridges are trenchant, and made up of about six points; valleys
perfectly open; the posterior talon has a crenulated bourrelet, as in
the small second milk upper of Mastodon latidens, but much less
developed. The two posterior ridges are very little touched by wear.

B. Mastopon ANnGusTIDENs.
Winterthur, August 30, 1856.—Museum of Mons. Johann Ziegler-Ernst.

Got a note from Professor Heer, and visited M. Ziegler, who
showed me a superb specimen of the lower jaw, both sides nearly
complete (but broken up, the left side especially, into several fragments,
which fit together, and also to a large block of sandstone matrix), of a
half grown or younger individual of Mastodon angustidens. It exhibits
the milk-dentition, also one of the vertical premolars, both sides, and a
true molar embedded in the alveolus; also a considerable portion of
the prolonged beak, which shows the transverse section of two very
compressed lower incisors.

The following teeth are seen in succession :

A.—An oval shaped bicuspid tooth with an entire crown, but chiefly
concealed in matrix, so that the top of it cannot be seen, only the side.
It resembles in form very much the small cuspid tooth seen in the
Museum at Zurich. By the section it is seen to have two compressed
fangs. The larger cusp is in front, the smaller behind; but both are
partly concealed by matrix.

The crown measures 1:2 inches in length. The posterior edge of the
anterior cusp is vertically lobed by three indentations. The crown
would appear to be quite entire and unworn.

B.—A tooth of which the crown is entirely concealed in matrix,
but the side of the basal enamel is disclosed, and also the broken ivory
nucleus. This tooth presents two large and separated fangs: the
posterior being much the largest. Judging from the length, one would
almost suppose that it had three ridges; but this is unfortunately
concealed by matrix, nor is it visible what amount of wear it had
undergone; it is assuredly a milk molar, from there being a vertical
premolar below it.

Length of crown enamel, 2°4 in.

C.—Immediately below this tooth and pushing it up is seen the germ
of a vertical premolar of large size (so to speak), exactly as in Lartet’s
specimens in the Paris Museum.

Length of this tooth so far as disclosed, 1-4 in. Height of anterior cusp, 1°1 in.

D.—Immediately behind the tooth B is placed another molar (with
three ridges), of which on the left side the top of the crown is seen, but
the two anterior ridges are broken off. The posterior ridge and the
talon are quite entire. The last ridge has the tips of the tubercles just

74 BRITISH AND EUROPEAN FOSSIL MASTODONS.

abraded into little round discs, showing that it had been in wear, and
that the anterior ridges must have been worn. The talon is of two
points, exactly as in Plate XL. fig. 7, of the ‘ Fauna Antiqua Sivalensis.’
The tubercles of the last ridge are 4 or 5; the outer and inner are very
blunt. The intermediate mammillz are not shown.

Extreme length right lower jaw from beak to posterior edge of ascending ramus,
23: in. Length of beak from broken tip of incisives to posterior curve of symphysis,
6-4 in. Height of jaw below the Ist small molar, 3:15 in. Width of jaw below
(about) 1-8 in. Length of crown, 2°75 in. Width at base of last ridge, 1°75 in.

This tooth I infer to be the third milk molar evidently of MW. angus-
tidens, but there is certainly no premolar below it. It must be added,
however, that the fangs press down into the dentary canal leaving no room.
Taking into account the great length of the crown of the tooth B, which
could have three ridges, is it possible that that tooth is the third or last
milk molar, and this one the first truemolar? This matter must be well
weighed. (See antea, p. 39, note 1.)

The vertical height of the corresponding tooth is well seen on the right
side entire, and of three ridges; but the top of the crown is not shown.
The length is estimated to be about 2°8 in. to 2°85 in.

E.—Behind the tooth D, on the right side, but embedded in the
posterior part of the alveolar cavity in the angle of the ascending ramus,
is seen the enamel-shell of a germ which shows three ridges. The side
of it only is seen, and imperfectly, without the talons. The visible part
of the three ridges measures in length about three inches; it is therefore
a small tooth, and probably the first true molar. On the left side a
transverse section view is had of one of the ridges, giving a width of only
1-7 in. to the crown. This ridge is seen to consist only of four distinct
obtuse tubercles in two pairs; the outermost is the largest, with the well-
marked form of M/Z. angustidens.

The left incisive gives a vertical section of about 1:1 in.

It is very much compressed with a sort of reniform groove to the
inner side. The surface is very much channelled superficially.

The anterior part of the lower jaw is very narrow and compressed, as
in the Paris specimen. It was discovered last year (1855) in a sand-
stone quarry about a mile from Winterthur, at a place called Veltheim
Quarry, embedded in a fine grained greyish sandstone, nearly hori-
zontal, being the Upper Molasse. The matrix is exactly like the tunnel
specimen at Zurich.

Found fragment also of what I believe to be the upper tusks, super-
ficially channelled :

Transverse ? diameter, 1:0 in. Vertical? or smaller, 0°6 in.

C. Mastopon From AMERICA.
Extract of Letter from Dr. Falconer to M. Lartet, September 12, 1856.

‘At Genoa I saw a cast of a large lower jaw of a Mastodon from
Mexico, with an enormous bec abruptly deflected downwards and con-
taining one very large lower incisor. The beak is much thicker than in
M. ( Trilophodon) angustidens and larger than in AT. ( Tetralophodon) lon-
girostris. You know that every one (Laurillard, Gervais, &c.) has
insisted on the absence of the lower incisors from both of the South

Pah i.

>

*

a

z
3
i
s
7
Bic
=

MASTODON ANGUSTIDENS. 75

American species. The outline of the jaw resembles very much the
figure in Alcide D’Orbigny’s Voyage, described by Laurillard as J.
Andium. The specimen is unpublished material, and I was therefore
only allowed to examine it very cursorily. The Genoese paleontolo-
gists had provisionally named it Rhynchotherium, from the enormous
development of the beak, approaching Dinotherium.’

76 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

Il. ON THE SPECIES OF MASTODON AND ELEPHANT
OCCURRING IN THE FOSSIL STATE IN GREAT
BRITAIN.

PART Il. ELEPHANT.!

J. INTRODUCTION—II. SUBGENERA OF ELEPHAS—III. CHARACTERS OF
STEGODONS :—1. ELEPHAS (STEG.) CLIFTII—2Z. ELEPHAS (STEG.) INSIGNIS
IV. PENTALOPHODON—V. CHARACTERS OF LOXODONS:—1. AFRICAN
ELEPHANT—2. ELEPHAS (LOX.) PLANIFRONS—3. ELEPHAS (LOX.) PRISCUS
—4, ELEPHAS (LOX.) MERIDIONALIS—A. TUSCAN SPECIMENS—B. BRITISH
SPECIMENS—VI. CHARACTERS OF EUELEPHAS :—1. INDIAN ELEPHANT—
2. ELEPHAS (EUEL.) PRIMIGENIUS—3. ELEPHAS (EUEL.) ANTIQUUS—
VII. GEOLOGICAL AGE OF ELEPHANTS.

In the remarks introductory to the preceding part of this
essay I adverted to the importance, for sound reasoning in
geology, that every Mammal found in the fossil state should
be determined specifically with precision, and I endeavoured
to illustrate the point by the entanglement and confusion of
the Faunas of the Miocene and Pliocene periods, which had
arisen from so many distinct forms of different ages having
been ranged by Cuvier and later palzeontologists under the

common name of Mastodon angustidens.

1 This memoir was communicated to
the Geological Society of London on
June 3, 1857, and an abstract of it was
published in the ‘ Quart. Jour. Geol.
Soc.,’ vol. xiv. p. 81. The publication
of the paper, in extenso, was postponed
in order that the author might incorpo-
rate the results of further investigations,
and, unfortunately, this had not been
effected at the time of his death in Janu-
ary 1865. In August 1865, the memoir
was published in a very imperfect form
in the ‘Quarterly Journal of the Geo-
logical Society,’ the entire description of
Elephas antiquus, &c., portion of that
Elephas primigenius, together with all the
general conclusions, being wanting. The
description of E. antiguus and a por-

tion of that of EZ. primigenius appear
never to have been written; but an
attempt has now been made to remedy
these omissions by extracts from Dr.
Fs Note-Books. The manuscript of
the concluding portions of the paper,
which treats of the bearing of the fossil
elephants, regarded as distinct species
upon the classification of the newer ter-
tiary strata, was discovered among Dr.
F.’s papers, subsequently to the appear-
ance of the first part of the paper in
August 1865, and is now for the first
time published. The illustrations have
been obtained from the ‘ Fauna Antiqua
Sivalensis,’ and from the sourees speci-
fied in each case.—[Eb. ]

INTRODUCTION. 77

The observation applies with still greater force to the case
of Elephas primigenius, to which a scope in space and time,
taken together, has been assigned, without a parallel, I
believe, within the whole range of the Mammalia, fossil or
recent. D’Archiac, in his excellent ‘Histoire des Progrés,’
so late as 1848, gives a brief summary of the localities in
which the remains of the ‘Mammoth (H. primigenius) have
been said to occur, namely, from the British Isles across the
whole of the temperate zone of Europe and of Asia, and
along all the coasts and islands of the Icy Sea, as far as the
frozen cliffs of the east coast of Behring’s Strait; in Hsch-
scholtz Bay; in Russian America as high as 66° of N. lat.;
over most of the United States of North America ; in the great
valley of the Mississippi; and along the coasts of the Gulf
of Mexico.’! Struck with the extent of this vast area, in-
cluding all the emerged lands between the parallels of 40°
and 75° N. lat., he puts a query whether the EHlephantine
remains met with by Humboldt on the plateau of Quito and
at Cumanacoa, in Columbia, did not also belong to the same
species.2. De Blainville, going a step beyond most other
palzontologists, doubtingly referred the fossil remains of
Elephants found so abundantly in tropical India to the same
species,® thus assigning at least half of the habitable globe
for the pasture ground of the Mammoth.

The duration allotted to the same species is equally re-
markable. Discovered fresh, either in the frozen cliffs or in
ice-blocks at the mouth of the Lena, it has been traced,
through its osseous remains, in the superficial gravel-beds
over nearly the whole of northern and the greater part of
central Europe. Here it has consistently been found in
company with the Siberian Rhinoceros (R. antiquitatis,
Blum.), the Musk-ox, and the Reindeer. The same specific
form has been carried down into the so-called ‘ Pleistocene’
clay, loam, and mud deposits which are so massively deve-
loped on the Norfolk and Suffolk coast, in company with R.
leptorhinus, Hippopotamus major, and other extinct forms;
thence through the submerged forest and lignite bed of
Happisburgh and Mundesley into the Crag in company with
Mastodon (Tetralophodon) Arvernensis; and abroad into the
‘Older Pliocene’ beds of the Sub-Apennines, and of Monte

1 Bronn enumerates the following lo- | scholtz Bay ; on the east side of North
ealities: Spain, Apulia, and Sicily ; the | America, in Ohio, Kentucky, and South
Islet of Gozo near Malta, Athens, and | Carolina, including the parallels be-
Odessa; the whole of Europe except | tween 40° and 75°N. lat. (Lethaea Geog-
Seandinavia ; from the Caucasus, through | nostica, Band iii. p, 819.)
the whole of Siberia, north to the Polar 2 Op cit. tom. i. p. 378.

Sea, and Kamtschatka; on the north-| * Ostéographie: ‘Des Eléphants,
west coast of America, as far as Esch- | p. 222,

78 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

Mario, Monte Verbo, and other localities in the south of
Italy. The measure of time involved in the thus implied
duration of the species is best appreciated by considering
some of the changes that appear to have taken place in
Europe during the interval. The Alps, the Pyrenees, and
the Apennines have all undergone a considerable amount of
elevation. When the earliest Elephants were roaming over
the emerged land of Italy, a wide and open sea communica-
tion would seem to have existed between the Mediterranean
and the Atlantic Ocean, admitting of a common province for
the Mollusca of the shores of the Crag-sea and of Italy,
and a common resort for the Whales and Dolphins which
abounded at that period in European waters. Portions of
the Pliocene sea-bottom of the Sub-Apennines, consisting of
stratified beds full of marine shells, and containing nearly
entire skeletons of Elephants and Rhinoceros, have been
thrown up into hills, which, after a long series of ages of
degradation, still maintain an elevation of 1,700 feet above
the level of the adjoining sea. Yet, if we are to accept the
confidently expressed opinion of Cuvier, long after his early
inferences had been questioned, the same form of Mammoth
lived through all these mighty changes, and it is only yes-
terday as it were, in relation to the human epoch, that its
last remnant was exterminated and frozen up in the perennial
ice-cliffs of the Arctic Circle.

It will hardly be denied by any one who attempts to
reconcile the English and Continental classifications, that
the arrangement of the newer Tertiary and Glacial deposits
in successive chronological order is at present in a very
unsatisfactory state, probably more so than that of any part
of the older Tertiary series; and it appears to me that
nothing has contributed more to retard the progress of this
section of geology in Britain than the generally accepted
belief in the specific unity of the Mammoth, wherever fossil
remains of Elephants were discovered in Huropean strata.
The percentage of extinct Mollusca, so valuable a guide in
the identification of the middle Tertiaries, becomes in the
newer Tertiaries an evanescent quantity—at every step more
elusive as we ascend upwards; and if the geologist tried to
extract some help from the associated Mammalian remains,
he was at once perplexed by the ubiquitous presence of the
Mammoth. The very name of Hlephas primigenius was sug-
gestive of ‘transported gravel,’ ‘diluvial action,’ ‘glacial
drift,’ or some other explanation suggested by the image of
the Woolly Mammoth, frozen in, flesh and bone, at the
mouth of the Lena; so that every stratum in which Ele-
phant bones were met with was regarded in some degree

INTRODUCTION. 79
under the influence of a foregone conclusion. Numerous in-
stances might be cited of the force of this bias upon the
views of some of the ablest writers on the geology of the
later Tertiary deposits.

The object of the present communication is to show that
several European fossil species, belonging to two distinct
subgenera, have been generally confounded under the name
of Hlephas primigenius, that these species are susceptible of
being discriminated, not on mere trivial or uncertain, but
upon broad and well-founded distinctions, and that their
range in time is consonant with what is known of other well-
determined species of Mammalia, namely, that they have
been restricted within definite eras. In order to give any
weight to the specific distinctions among the fossil Hlephants
which I shall endeavour to point out, it will be necessary to
explain the grounds upon which they are founded in greater
detail than is set forth in the remarks introductory to the
preceding part of this essay, when treating specially of the
Mastodons; and, at the risk of being chargeable in some
measure with repetition, I must solicit the indulgence of the
Society on the subject.

The specific name of Hlephas primigenius, adopted from
the eminent German naturalist, Blumenbach, was applied
by Cuvier to all the fossil Hlephantine remains occurring
in Kurope, Northern Asia, and America, up to the date of
his last edition of the ‘Ossemens Fossiles.’ De Blainville,
swayed by his adherence to the dogma of a single and simul-
taneous creation of living beings, subject to incessant extine-
tions, but never repeated, in admitting Elephas primigenius,
extended its area for the reception of the living Indian Hle-
phant, as he held the opinion that there were not sufficient
grounds for regarding them as specifigally distinct... Owen
adopted Cuvier’s limitation of the Mammoth; but, struck
with the wide differences presented by molars from various
British strata, he endeavoured to account for them on the
hypothesis of a gradation between thick and thin plated
varieties.? Gervais,’ while fully admitting the a priori im-

1 «En sorte que le résultat définitif
auquel on est conduit par une logique
rigoureuse, c’est que dans |’état actuel de
nos collections du moins au Muséum de
Paris, il est encore a peu prés impossi-

ble de démontrer que 1’Eléphant fossile, |

dont on trouve tant de débris dans la
terre, différe spécifiquement de lElé-
phant de l’Inde encore vivant aujour-
Whui,— De Blainville, ‘ Ostéographie :
Des Eléphants,’ p. 222,

2 <Tf these varities’ (7.e. thick- and
thin-plated) ‘actually belonged to dis-
tinct species of Mammoth, those species
must have merged into one another, so
far as the character of the grinding
teeth is concerned, to a degree to which
the two existing species of Elephant, the
Indian and African, when compared to-
gether, offer no analogy.’

% Paléontologie Frangaise (1848-52),
p. 30,

80 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

probability that the same species of Elephant ranged from
the Pliocene up, through the Pleistocene, to the Post-Pliocene
period, adheres to the specific unity of Elephas primigemius ;
and he endeavours to escape from the difficulty by assuming
that the so-called Pliocene remains of Elephants have been
wrongly determined, and ought to be referred to the genus
Mastodon. To avoid cumbering the present communication
by a tedious citation of other authorities, I may refer to the
two latest compilations on paleontology, respectively by
Bronn and Pictet, for the existing state of knowledge and
opinion upon the subject. Bronn, after an exhaustive expo-
sition of the literature on fossil Elephants, sums up by
stating that the number of fossil species, exclusive of two or
three Indian forms and of H. priscus (upon which he does
not venture to decide), is limited to a single, or, at the
utmost, two fossil species; and he ranges all the European
forms, with the exception of H. priscus, under the synonymy
of EH. primigenius.’ Pictet doubts the veritable fossil nature
of the specimens upon which EH. priscus was founded; the
other nominal species he considers as not established on
sufficient grounds, and he would continue them all, inclusive
of H. meridionalis, under the common designation of EH. pri-
migenius. He questions the occurrence of Elephant remains
in the Pliocene period, leaning to the opinion of Gervais,
that the asserted instances should be referred to the genus
Mastodon.”

The restriction of the European fossil Elephants to a single
Species was first called in question by Nesti, as far back as
1808, upon fossil remains discovered in the Val d’Arno, for
which he proposed two new designations. Nesti was in
possession of the most ample materials for the establish-
ment of one of these,, H. meridionalis; but, unfortunately for
science, he described the lower jaw of Mastodon (Tetralo-
phodon) Arvernensis as that of an Elephant, and abandoned
the characters furnished by the molar teeth as untrustworthy
and uncertain ; and his Hlephas meridionalis and EH. minutus
succumbed to a criticism by Cuvier. The former was revived
by Croizet and Jobert in 1828, for remains found in the
Velay* under the name of Hléphant de Malbattw; it has been

1 «Die Anzahl der fossilen Arten mag
sich, ausser 2 bis 8 Ostindischen am
Fusse des Himalayah gefunden, und ab-

(1856) edit. 3, Band iii. p. 814.
2 Pictet, Paléontologie, 1853, tom. 1.

gesehen von E. priscus, uber den wir
nicht entscheiden wollen, auf eine bis
héchstens zwei beschranken, womit auch
die americanische dickplattige Form, £.
Americanus, Leidy, tibereinzustimmen
scheint.’ -— Bronn, Lethea geognost.

3 Annali del Museo di Firenze, tom. i.
‘Di aleune ossa fossili de’ Mammiferi
che s’ incontrano nel Val d@’ Arno.’

4 Oss. Foss. du Puy-de-Déme, pp.123—
132.

ee es ree

|

SUBGENERA OF ELEPHAS. 81

admitted by Christol! and Pomel? for others from Auvergne
and Montpellier; and by Morren, in his account of the
Elephant remains occurring in the fossil state in Belgium.?
In 1847 it was applied, in the ‘Fauna Antiqua Sivalensis,’
to remains from the Norwich Crag and lignite bed.

Goldfuss, in 1821, proposed the name of Hlephas priscus
for some supposed fossil molar teeth, bearing a strong resem-
blance to the molars of the existing African Elephant. Cuvier
disputed their authenticity as real fossils; and it is not a
little curious that Goldfuss would appear in this case to have
founded a veritable species upon spurious materials. I de-
tected in the British Museum molars of indubitable fossil
origin from the brick-earth deposit of Gray’s Thurrock, in
the valley of the Thames, presenting characters closely re-
sembling Goldfuss’s species, and figures of them were pub-
lished* under the name of H. priscus in 1847. Pomel applies
the name to some fossil molars described by Laizer in
Auvergne.

Fischer de Waldheim, Eichwald, and Morren together
have proposed eight nominal species as distinct from H. pri-
migenius ; but these were based, for the most part, on such
obviously trivial characters that discredit was reflected on
the species which had a better foundation.

In 1847 I proposed the name of H. antiquus for molars
which are met with in vast abundance in certain of the
newer Tertiary beds in England, and in corresponding de-
posits on the Continent, more especially in Italy; but no
descriptions having accompanied the published figures, the
species has hardly been noticed, and nowhere admitted, by
other paleontologists.

Il..—Tue SusGENERA OF ELEPHAS.

In the first part of this essay (page 11) it was attempted to
be shown that the species of Mastodon, with the single ex-
ception of M. Siwvalensis, are susceptible of being arranged in
two natural groups, Trilophodon and Tetralophodon, according
to a definite and isomerous numerical expression of the crown-
ridges of the three ‘intermediate molars’ of both jaws, and
that this formula implies the ridge-characters of the other
molar teeth.

In the Elephants, the divisions of the crowns of any one

1 Ann. des Sci. Nat. 1835, 2™¢ sér. | d’Eléphans trouvés en Belgique, 18384,
Zool. tom. iv. p. 197. p. 13.

2 Catal. Méthod, et Descript. 1854, p. 4 Fauna Antiqua Sivalensis, Pl. xiv.
74. figs. 6 & 7. (See vol. i. p. 441.—Eb.)
3 Mémoire sur les Ossemens fossiles
VOL. II. G

82 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

of the ‘intermediate molars’ are never less than six; and in
the species, fossil and recent, that are furthest removed from
Mastodon in affinity, they range as high as 16 or 18 in the
penultimate true molar, or third of the ‘ intermediate’ series.
‘They are not isomerous, as in the Mastodons, but deviate
from the numerical symmetry either by an augmentation of
one ridge to the crown of the last ‘intermediate molar,’
constituting the hypisomerous forms, or they are more nume-
rous, and augment by progressive increments corresponding
with the increase of age, including the amisomerous forms.

The Elephants with hypisomerous-ridged molars are divi-
sible into the two natural groups, Stegodon and Loxodon; the
anisomerous species form a third natural group, for which, as
already explained, the term Ewelephas is proposed.

ILI1.—CHARACTERS OF THE STEGODONS.

1. General Remarks.—The Stegodons form the nearest
approach in natural affinity to the Mastodons, and more
especially to that subdivision of the section Tetralophodon
which comprises M. (Tetraloph.) longirostris and M. (Te-
traloph.) latidens. This is evinced by the low elevation and
transverse direction of the crown-ridges, by their nearly
uniform height throughout the length of the crown, by their
thick enamel, and by the mammillary form of the ridge-
processes. A fragment of one of these teeth, denuded of its
coat of cement, and seen by a naturalist for the first time,
would at once be referred to Mastodon rather than to Elephas ;
and it was this broad resemblance which struck Clift so for-
cibly that he applied to them the designation, at the time
very appropriate, of Mastodon Elephantoides. But when the
essential characters are analyzed, the species are seen to
partake more of the nature of true Elephants :—

1st. In the greater number of the crown-ridges and of the
mammille or points that enter into the composition of each.
2nd. In the agreement of the ‘ ridge-formula’ of certain of
the species with that of the existing African Elephant and
other Loxodons. 3rd. In the convex outline of each ridge
in the transverse direction when unworn, the central mam-
mille being the most elevated; and in the absence of the

longitudinal line of division along the middle of the crown

which is so characteristic of the Mastodons on the one hand,

and so generally absent in the Hlephants on the other. 4th. f

In the enormous quantity of laminated cement that fills
up the valleys in most of the species. 5th. In the pro-

nounced are of a circle described by the molars as we trace

them forwards in the jaws, as in the Elephants, instead of — :

a

|
|

CHARACTERS OF STEGODONS. 83

the nearly horizontal line of protrusion observable in the
most typical Mastodons, such as the species of North Ame-
rica and of Simorre. 6th. In the obverse relation of the
planes of detrition of the opposed teeth during wear, the
inner side of the upper teeth, and the outer side of the lower,
continuing higher in the Stegodons, as in the typical Ele-
phants, while the converse holds in the Mastodons. 7th.
In the absence or extreme rarity of premolars in both
jaws, and of mandibular tusks, neither of which, though
occurring among certain Mastodons, have been as yet de-
tected among the Stegodons. The aggregate weight of so
many points of agreement turns the balance strongly on the
side of the Elephants.

It is deserving of remark, that all the species of the
Stegodon group at present known belong to the series indi-
cated in the preceding part of this paper, as being of the
Dinotherian or Eurycoronine! type, in that the crowns of
the molars are broad, the ridges uniformly transverse, and the
valleys open, without being in the least degree interrupted
by outlying tubercles, as is seen in the Hippopotamine or
‘Stenocoronine’ type. Sir Proby Cautley and myself have
thought we could distinguish four species of Stegodon, namely
Hi. (Steg.) Cliftii, H. (Steg.) bombifrons, EH. (Steg.) imsignis, and
H. (Steg.) Ganesa? The first, besides other distinctive marks,
is at once characterized by the broad distinction of the ante-
penultimate and penultimate true molars being six-ridged, or
hexalophodont in number, the last true molar conformably
presenting an additional ridge and ‘talon.’ The first of the
‘intermediate series,’ namely the last milk molar, has not
yet been observed entire in sitw in the jaw, but I am prepared
to expect that, when determined, it will present five or six
ridges. This species, the remains of which were discovered
by Mr. Crawfurd in Ava, constitutes the passage into the
Mastodons ; this is indicated both by the limited (¢.e. senary)
number of ridges, and by the circumstance that the crowns
of the molars exhibit a very obsolete or indistinct trace of a
longitudinal bipartient cleft, as inthe Mastodons. Further,
in the only well-preserved palate-specimen at present known,
the outer side of the upper molars is higher, and the inner
side lower and more worn, being another point of agreement
with the Mastodontoid rather than with the Elephantoid type.
Where nearly allied groups inosculate, the intermediate

1 Tt has been suggested to me that the | difference than is intended. I propose
contrasted terms of Dinotherian and | therefore to substitute for the former
Hippopotamine types may mislead, | ‘ Eurycoronine’ or broad-crowned type,
through being supposed to imply a| and for the latter ‘Stenocoronine’ or

' greater amount both of affinity and of | narrow-crowned type.

G2

84 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

forms commonly partake more or less of the character of
both. But the sum of the characters, and more especially
the identical form of the divisions of the crowns and the |
ridge-formula, connect this species more with the other Ste-
godons than with any group of Mastodon. The next two
species, namely, H. (Steg.) bombifrons and EH. (Steg.) msignis,
have from seven to eight, and occasionally even nine ridges
in their different intermediate molars; and their teeth are
exceedingly alike in character, although the species are dis-
tinguished by an excessive amount of difference in the form
of the cranium, greater even than that between the African
Elephant and the Mastodon of North America. Regarding
the specific distinctness of H. (Steg.) Ganesa lam by no means
so well assured; this species is chiefly founded on a huge
cranium in the British Museum with long tusks, presented
by Colonel Baker. I have not been able to reconcile the form
of this cranium with either that of H. (Steg.) msignis or E.
(Steg.) bombifrons ; but at the same time I must confess that
T have failed in tracing its dentition satisfactorily as a distinct
form through different ages. Three species of this group
appear to be distinct beyond question ; and I cite them chiefly,
on the present occasion, in reference to determinations in the
sequel, to show that Elephantine forms may approach very
closely in their dental characters, as occurs in other Mam-
malia, and still be distinct species.

The Stegodons, so far as is at present known, are exclu-
sively confined to Tropical Asia. It is therefore unnecessary,
on the present occasion, to describe in detail the peculiarities
of their dental characters; and I shall confine myself to the
leading points in their ‘ridge-formula,’ that place them in
connexion with the Mastodons on the one hand, and with the
Loxodons on the other.

2. Hlephas (Stegodon) Cliftii.—Of this species the youngest
milk teeth are as yet unknown. The third upper milk molar,
or first of the intermediate molars, is seen in sitw in the
specimen represented in the ‘ Fauna Antiqua Sivalensis,’ Pl.
XXX. fig. 1 b, entire on one side, but worn down to the common
base of ivory, so that the divisions of the crown have entirely
disappeared, leaving no certain data for determining the
ridge-formula of this tooth. Behind it, in the same palate
specimen from Ava (presented by Colonel Burney to the
British Museum), the three anterior ridges of the antepenul-
timate true molar are seen in situ, the posterior half being
broken off. But the detached tooth on the upper jaw is seen
entire, and beautifully preserved, in the specimen fig. 2 of
the same plate, presenting six ridges and a small hind talon.
The same tooth is represented by fig. 6 of Pl. XX XIX. of Mr.

E, (STEGODON) INSIGNIS. 85
Clift’s Memoir (Geol. Trans., vol. ii. 2nd series). It is there de-
scribed as an upper molar tooth of Mastodon Hlephantoides,
under which title Mr. Clift included specimens that are re-
ferred in our arrangement to two distinct forms.! The Ele~-
phantine affinities of this tooth are indicated by the absence
of a longitudinal line of division along the crown, and by the
great number of points (about eleven in each) that enter into
the composition of the ridges. The penultimate true molar
(or third of the intermediate series) is presented in situ on
both sides of the superb palate specimen represented by Clift
in Pl. XXXVI. of the memoir above referred to. It is proved
to be the penultimate by its large dimensions, and by the
circumstance that part of another tooth of still larger size,
and inferred to be the last, is seen behind it in the jaw. The
Same specimen is more carefully represented by figs. 3 and 3 @
of Pl. XXX. of the ‘ Fauna Antiqua Sivalensis.’ 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 last true molar
of the lower jaw is represented by fig. 5 of Pl. XXX. of the
‘Fauna Antiqua Sivalensis.? The crown consists of eight
ridges and a talon. The anterior large fang had been ab-
sorbed, but the portion of the crown sustained by it remains.
The six posterior ridges have their fang elements confluent
into a continuous plate or shell, thus maintaining the Ele-
phantine affinity indicated by the crown characters. (See
Pl. V. figs. 1 & 2.) Taking the data furnished by these teeth,
the cipher 6 is seen to prevail in the two last of the interme-
diate molars, indicating a Hexalophodont type, or 6+6+8
for the ridge-formula of the true molars.

3. Hlephas (Stegodon) insignis.—The only other form among
the Stegodons which it is necessary to notice is that for which
the name of H. (Steg.) insignis has been proposed. (See Pl. V.
figs. 3 & 4; and vol. i. Pl. IV. fig. 1.) In this species the
crown-ridges are constructed very closely upon the model of
H. (Stegodon) Clift, the principal difference consisting in the
much greater mass of laminated cement that fills up the

1 Mr. Clift, in his excellent memoir,
includes the Ava fossil Proboscideans
under two species, Mastodon latidens and
Mastodon Elephantoides. In the ‘Fauna
Antiqua Sivalensis,’ and in the synopti-
cal table appended to the preceding part
of this paper, the former name is re-
tained for the specimens of the Tetralo-
phodon type, figured by Mr. Clift, Pl.
Xxxvii. figs. 1 & 4+; Pl. xxxviii. fig, 1, and
Pl. xxxix. figs. 1, 2,& 3. Of the others, the

palate specimen, Pl, xxxvi. (Mastodon |

latidens,Clift), together with the detached
molar, Pl. xxxviii. fig. 6 (Mastodon Ele-
phantoides, Clift), are referred to E. (Ste-
godon) Cliftii; and the lower jaw speci-
men, Pl. xxxviili. fig. 2 (also MW. Hlephan-
toides, Clift), is referred to F. (Stegodon)
insignis. The specimens regarded by him
as of his M. Elephantoides being here
considered to belong more properly to
the genus Elephas, it became necessary
to resort to another specific designation,
Hence the origin of F. (Stegodon) insignis,

&6 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

valleys. In some sections as many as eleven distinct strata of
this substance may be counted.! But the ciphers yielded by
the ‘ridge-formula’ place the species in close affinity with the
Loxodons, and more particularly with the species named EH.
(Lox.) planifrons. Remains of H. (Steg.) imsignis have been
discovered in immense abundance in the Sewalik hills, and
specimens illustrative of the dentition of every age and in
every stage of wear are contained in the great Indian col-
lection of the British Museum. ‘The rigid constancy in the
number of ridges observable in the two subgenera of Mastodon
is no longer maintained. As stated in the preceding part of
this paper, the higher the numerical expression of the ‘ ridge-
formula’ in the species, the more liable is the number of
ridges to vary within certain limits dependent on the race,
sex, and size of the individual; and the molars of the lower
jaw often exhibit an excess. After examining a very large
number of specimens of all ages, the prevailing numerical
expression of the ridge-formula, exclusive of ‘talons,’ in H.
(Stegodon) insignis has appeared to me to be thus :—

Milk molars. True molars.
24+547: 7+ 8 +(10-11). °
2+5+7: 7+ (8—9)+(11—18).

I have already remarked that all the known species of
Stegodon belong to that species of the Proboscideans in which
the ridges are transverse, and the valleys open. It may be
expected, without much temerity, that other species remain
to be discovered in the fossil state, in which the mammille will
be disposed more or less alternately, with outlying tubercles
and interrupted valleys, as in the ‘ Stenocoronine’ type.

ITV.—PENTALOPHODON.

From the circumstance that so many Mastodons present
the ciphers either 3 or 4 constantly in the ridges of the inter-
mediate molars of two groups of species, and that in the next
allied group, Steqodon, Elephas (Stegodon) Cliftii in like
manner presents the cipher 6 in two of the same teeth, while
the prevailing number augments in H. (Stegodon) bombifrons
and H. Stegodon insignis, with faith in the harmony of nature —
it might have been with some confidence anticipated that
another Proboscidean type remained to be discovered in the
fossil state, intermediate between Tetralophodon and Stegodon, —
in which a quinary ridge-formula would be presented, consti- —
tuting a third subdivision of the genus Mastodon, to which
the name of Pentalophodon would be applicable.

It appears to me that the Indian fossil species M. (Tetralo-

1 Fauna Antiqua Sivalensis, Pl. vi. fig. 7.

DESCRIPTION OF PLATE Y.

Euerxas (Strcopon) Cuirrir AND HLEPHAS (STEGODON)
INSIGNIS.

Figs. 1 and 2. Elephas Cliftii. Views in plan and profile of last true
molar, lower jaw, left side, rather less than one-fourth (two-
ninths) of the natural size. Copied from drawings by Mr. Ford
in Plate XXX., figs. 5 and 5a, of the Fauna Antiqua Sivalensis.
(See page 85, and vol. i. p. 462.)

Figs. 3 and 4. Elephas insignis. Viewsin plan and profile of a frag-
ment of lower jaw containing the last true molar, rather less than
one-fourth (two-ninths) of the natural size. Copied from draw-
ings by Mr. Ford in Plate XX., figs. 8 and 8 a, of the F. A. 8.
(See page 85, and vol. i. p. 452.)

VOL. 0.

Vol. ll. Plate 5

del L_ Aldous bth. Harrison & Sons Imp*

1.2. Elephas (Stegodon) Clifti. 3.4. Elephas (Stegodon) insignis.

* ‘ Str a, ae ci .
ov 4 i A o ‘ a! Apa i
; .
sy .
*
4 a
ie
* -
’ a
s a
ay
r
~
. - %
. . 4
. . a
> A =i =
- 4

PENTALOPHODON. 87

phodon) Swalensis, figured in the ‘ Fauna Antiqua ’,! presents
the first indication in that direction. In the ‘intermediate
molars’ of this form, both upper and lower, besides the usual
anterior ‘talon’ and four large ridges, there is a fifth ridge,
somewhat reduced in size, but exactly corresponding with
the other in form, composed of several large mammillary
tubercles, separated from the next ridge by a valley, and
throwing off an outlying tubercle, which reduces the valley,
as in M. (Tetralophodon) Arvernensis, to lateral gorges. This
fifth ridge is not a mere offset from, or subordinate appendage
to, the fourth ridge after the ordinary manner of a ‘talon.’
It is supported directly by the last fang, and is separated,
both on the outer and inner sides, from the latter by the in-
tervening valley. In most of the species of Mastodon having
alternate mammille, the hind ‘talon’ in the upper molars
(and conversely in the lower) forms a crenulated ‘ bourrelet,’
which is given off from the inner posterior mammilla, de-
scending obliquely around the base of the outer division, and
generally more or less effaced by the pressure of the next
posterior molar during its progress forwards. In a fine
specimen of a penultimate upper molar of Mastodon Sivalensis,
which is now before me, the fifth ridge, although well de-
veloped and attaining the height of the fourth, bears no trace
of a ‘talon’ appended to it; while an antepenultimate lower,
which I have also before me, shows distinctly five ridges, the
last differing in no respect of complexity or development from
the others, except in being a little smaller, and it bears a
distinct crenulated adpressed ‘talon’ appendage, having the
appearance of a terminal ‘ bourrelet.’

In the preceding part, when discussing the conditions of
the ‘ridge-formula’ in Trilophodon and Tetralophodon, it was
stated that while the penultimate milk molar always presents
one ridge less than the ‘intermediate molars,’ the last true
molar presents one ridge more. Conformably, the last true
molar in M. Sivalensis presents six ridges, besides the hind
‘talon,’ thus maintaining throughout, so far as the dentition
is known, the numerical characters to be inferred from the
ridge-formula, as ascertained in Trilophodon and Tetralopho-
don.” I consider it sufficient, on the present occasion, to call
attention to this as a point of some interest and importance
in the systematic and paleontological relations of the Pro-
boscidean family, in reference to the indications they present
of an order of successive serial development, without entering
in detail upon the evidence in support of the view here taken.
That the species is a distinct form is abundantly borne out

1 Op. cit. Pl. xxxvi. figs. 1-6. (See vol. i. p. 467.—En.)
? See vol. i. Pl. vii. fig. i—Eb.

88 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

by the marked characters of the skull,! independently of the
strong dental distinctions. The ridge-formula for the true
molars in Mastodon Sivalensis is inferred to be

5+5+6

5+5+(6—7)’
and when the dentition is fully made out, it is anticipated
that the complete ridge-formula will be nearly thus :—

Milk molars. True molars.
24445 5+5+ 6
24445. 5+5+(6—7).

V.—CHARACTERS OF THE LOXODONS.

1. General Remarks.—The existing type of this group is the
African Elephant, which Fred. Cuvier, in 1835, proposed to
erect into a distinct genus under the name of Loxodonta,
having reference to the rhomb-shaped discs of wear of the
molar teeth. He held the opinion that, in its general form,
in the structure of its grinders, in the form of the head, and
in that of some of the external parts of the organs of sense,
the African differs as much from the Indian Elephant as the
Dog from the Hyzena, the Paca from the Agouti, the Lagomys
from the Hare, and the Hog from the Phacochere.* Besides
the African Elephant, the group Lowvodon comprises three
fossil species, of which one is Indian, H. (Loxod.) planifrons,
from the Sewalik hills, and two European, namely, H. (Loxzod.)
priscus and EH. (Loxod.) meridionalis.2 The essential charac-
ters by which the molar teeth of the Loxodons differ from
those of the Stegodons is that the ridges or colliculi, while
closely corresponding in regard of number, are considerably
more elevated and compressed. This is best seen when they
are sawn up longitudinally and vertically. The section in the
Stegodons exhibits a series of chevron-shaped ridges, of which
the height does not much exceed the base, with thick enamel
and assimilating closely in form to the true Mastodons; *
while in the Loxodons* it presents a succession of elongated
wedge-shaped processes, with thinner enamel, constituting
an intermediate stage between the former and the nearly
parallel thin-plated ridges of the next group, Huelephas. In
the technical definition of the subgenera appended to the
preceding part this distinction is attempted to be expressed
by the terms ‘ coronis complicata’ applied to the teeth of the

1 Vide ‘Fauna Antiqua Sivalensis, | sequently discovered by Dr. Falconer,
Pl. xxxii. (See Pl. x. and p. 464 of | viz. . (Loxodon) Melitensis (Fale.)—Ep.
vol. i—ED.) 4 «Fauna -Antiqua Sivalensis,’ Ilus-

2 F. Cuvier, ‘Histoire. Naturelle des | trations, Pl. 11. figs. 6a & 6d.

Mammif.’ tom. iii., ‘Eléphant d'Afrique,’ 5 Ibid. Pl. ii. figs. 4a & 40.

8 A third European Lorodon was sub-

CHARACTERS OF LOXODONS. £9

Stegodons, and ‘ coronis lamellosa’ to those of Loxodon and
-EHuelephas.! It forms the basis of the arrangement of the
species of the Proboscidea, by De Blainville, into two groups,
i.e. ‘ Hléphants mastodontes ’ and ‘ Eléphants lamellidontes,’
the whole comprised in a single genus, Hlephas.

2. African Hlephant.—De Blainville has attempted to de-
scribe and figure in detail the dental succession, from the
first milk molar of the young calf to the last true molar of the
adult state, in H. (Loxod.) Africanus. Of some of the ‘inter-
mediate molars,’ he was not in possession of perfect specimens ;
in these cases, his determination of the ridge-formula can
only be regarded as approximative. Another point, which
materially affects the numerical estimate of the ridges as-
sioned by him to the different teeth is, that in every case he
counts the accessory ridgelets, or ‘talons,’ as ridges. His
results may be expressed thus for the number of ridges in the
different teeth :—

Milk molars. True molars.
44+7+6 7+(9—10)+ 10
44+7+ ?. ?+(8—9) +(10—12).

This determination is open to the objections that the third
milk molar has a smaller number of plates assigned to it than
the penultimate, which is very much smaller in size, and that
the penultimate upper true molar is described by De Blain-
ville as possessing the same number of ridges as the last.
This occurs in no species of Mastodon or Elephant. De
Blainville has figured an instructive specimen, which proves
that the theoretical first or pre-antepenultimate milk molar
is occasionally developed in the lower jaw of the African
Elephant.

Professor Owen has briefly described the ridge-characters
of the teeth of this species in the ‘ Odontography,’ and
assigns the following numbers to the ridges in the six suc-
cessive molars, 7.e. 4+ 7+7:7+(8—9) + (10—-12), the last of
the ciphers being attributed to the sixth (or last true) molar
of the lower jaw.. In this estimate, the ‘talon’ is apparently
reckoned in some of the cases as one of the principal ridges.
I have examined the specimens upon which De Blainville’s
descriptions were founded, and various molars of all ages in
different collections contained in museums in this country or

{

teeth of Mastodon and Elephant appear
to convey the same meaning respectively
as the ‘dens complicatus’ and ‘dens
‘Pyramidal,’ ‘Prismatisch, applied by | lamellosus’ of Illiger (vide Lliger’s
Von Meyer and Bronn to characterize | ‘Prodrom.’ p. 22, and Bronn’s Lethzea
the difference in structure betweon the | Geognost. Band ii. pp. 753 and 797),

' These terms are adopted from the
logical and accurate Illiger. The ex-
pressions ‘ Bildung’ or ‘ Entwickelung,’

90 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

abroad, and, excluding the two ‘ talons,’ the ridge-formula in
the African Elephant has appeared to me to be thus :—

Milk molars. True molars.
8+64+7 7+ 8 +10
3+64+7 7+(8—9)+11.

The amount of development in the posterior talon is subject

to considerable variation. In some cases it forms merely an-

insignificant splent appended to the last ridge; in others it
attains the proportions of a reduced ridge, and, according to
the degree of its evolution, it may be differently regarded by
different naturalists, either as a distinct ridge or as an ap-
pendage. The hypisomerous character of the ridge-formula,
in the intermediate molars of the Loxodons, is exhibited by the
succession of ciphers assigned above to the ridges in the last
milk molar and the antepenultimate and penultimate true
molars, i.e. 7+7+8. I have seen specimens in which the
penultimate true molar of both the upper and lower jaws
presented nine ridges. Cuvier states that he had never
observed a tooth of the African Elephant showing more than
ten plates. A last molar of the upper jaw, left side, procured
from Cape Coast by Mr. Samuel Turner, exhibits, in a
length of 11 inches, thirteen plates, 7.e. eleven principal
ridges, besides front and back talons.

The well-known and very constant distinctive characters
in the molars of the African Elephant consist of the rhomb-
shaped pattern yielded by the dise of the ridges after ad-
vanced wear, together with the relative narrowness of the
crowns as compared with those of the Indian Elephant.

(See Pl. VI. fig. 1, and vol. i. Pl. IV. fig. 3.) The sys- -

tematic signification of these peculiarities appears to me to
be, that this species among the Loxodons represents the
eroup of forms in Trilophodon and Tetralophodon, described
in the preceding part as having the ridges of their molars
characterized by outlying flanking tubercles and blocked-up
valleys, and as belonging to the ‘Stenocoronine’ type. In
the African Elephant, the digital processes are less divided
and more speedily confluent than in .the Mastodons; each
ridge throws out, in front and behind, a mesial angular pro-
jection, which meets or overlaps the corresponding part of
the next contiguous ridge; and the transverse continuity of
the valleys, which are filled up with cement, is interrupted
in consequence. The adjoining rhombs, in the process of
wear, are in contact by their opposed angles, and at length
become confluent in a common disc. The angular expansions
of the discs are the modified homologues of the flanking
tubercles of the Mastodons; and as the character prevails in
several forms among the latter, its presence in so pro-

la ,
Were . of. at gee ee ae

DESCRIPTION OF PLATE VI.

Exrprxas (Loxopon) Arricanus, Exuepuas (Loxopon) PLA-
NIFRONS, ELuepHas (HvELEPHAS) InDIcUS, AND HLEPHAS
(HUELEPHAS) PRIMIGENIUS.

Fig. 1.

Elephas Africanus. First true molar, lower jaw, left side, one-
half of the natural size. Reproduced from a drawing by Mr.
Ford in Plate XIII. A, fig. 8, of the Fauna Antiqua Sivalensis.
(See page 90, and vol. i. p. 440.)

. Elephas planifrons. Last true molar, lower jaw, left side, three-

eighths of the natural size. Reproduced from a drawing by
Mr. Ford in Plate XI., fig. 2, of F. A. S. (See page 92, and
vol. i. p. 430.)

. Elephas Indicus. Last true molar, lower jaw, right side, of the

existing Indian Elephant, three-eighths of the natural size.
Reproduced from a drawing by Mr. Ford in Plate XIII. A, fig. 6,
of the F. A. S. (See pages 147 & 151, and vol. i. p. 440.)

. Elephas primigenius. ast true molar, lower jaw, right side,

three-eighths of the natural size. Copied from a drawing by Mr.
Ford in Plate XIII. A, fig. 1, of the F. A. S. (See pages 147 &
159, and vol. i. p. 439.) »

VOL. Il.

3

Vol. 1. Plate 6.

da L Aldous hii: Harrison & Sons Imp

1. Elephas (Loxodon) Africanus 2 E (Loxodon) plamfrons.
3.E.(Euelephas) Indicus 4.E.(Euelephas) primigenius .

E. (LOXODON) PLANIFRONS. 91

nounced a degree in the African Elephant might have led us
a priori to expect in nature other allied species in which it
would be more or less exhibited. Premolars have not as yet
been observed among the teeth of this species.

3. H. (Loxodon) planifrons.—In order to show the constancy
of the hypisomerous character of the ridge-formula among
the Loxodons, as furnishing a reliable aid in the distinction
of certain of the Huropean fossil Elephants, it is necessary to
refer briefly to the dentition of another form in the sub-
genus, being the Indian extinct species from the Sewalik
hills, H. (Low.) planifrons, the characters of which, yielded
both by the skull and teeth, are so pronounced, and the
accessible materials in European collections so abundant, as
to place its specific distinctness wholly beyond question. In
this form the ridge-formula of the deciduous and true molars
is thus :—

Milk molars. True molars.
3+6+7 7+ 8 +10.
amaeae 7+(8—9)+(10—11).

Vertical sections of an upper and lower true molar con-
trasted with corresponding teeth of the African Elephant are
shown by figs. 5a and 5b of Pl. I. of the ‘Fauna Antiqua
Sivalensis.’ (See also vol. i. Pl. IV. fig. 2.) The ridges
are seen to be much more elongated vertically than those of
H. (Steg.) insignis (vol. i. Pl. IV. fig. 1), but to be considerably
less so than in the African Elephant (vol.i. Pl. IV. fig. 3).
Other distinctive characters from the latter species consist in
the enormous quantity of cement which fills up the valleys
and envelopes the ridges, and in the much greater thickness
of the folded plates of enamel. When the teeth are regarded
from the crown aspect, the discs of wear assimilate more in
general form to those of the existing Indian than of the
African Elephant. They form transverse bands, which are
broader, fewer in number, and wider apart than in the Indian
Elephant, sometimes with the bounding edges of enamel
nearly parallel, in other cases showing a slight angular
expansion, or throwing out a salient loop (outlying tubercle)
near the middle, as in figs. 8 and 9 of Pl. XIV. (F.A.S.), but
never exhibiting the systematic lozenge-shaped expansion so
characteristic of the African Elephant. (See Pl. VI. fig. 1.)
The enamel edge or macheris is very thick, and generally
free from plaiting. The tips of the digital processes are
thick, and yield well-marked circular discs before they
become confluent by wear. These characters are represented
throughout by the figures of Plates XI. and XII. of the
‘Fauna Antiqua Sivalensis,’ which include the principal
varieties in the form of the molar-crowns. (See also Pl. VI.
fig. 2.)

92 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

Of the different teeth in the upper jaw the antepenultimate
upper milk molar is represented by fig. 1 of Pl. XII. of the
‘Fauna Antiqua,’ and in vertical section by fig. 1 6, showing
three principal ridges, with a basal ridgelet in front, and a
hind talon. The penultimate milk molar, of which a finely
preserved unfigured specimen is now before me, presents six
principal ridges, with a distinct front and back talon. It
measures 2 inches in length by 1 inch in front, and 1 inch
behind. The third milk molar and the antepenultimate or
first true molar are present in situ on both sides in another
palate specimen. The former exhibits the discs of six worn
ridges, besides a seventh ridge behind, which is enveloped by
cement. The first true molar is in a germ state, and presents
seven intact principal ridges, together with a front and back
talon. The penultimate true molar is seen im sitw in the
upper jaw specimens, figs. 4 and 6 of Pl. XII., presenting
eight main ridges. The last true molar is seen in germ,
intact on one side and well worn on the other, in the cra-
nium-specimen, figs. 1 to 4 of Pl. X., with eleven ridges and
talons. In other cases, such as fig. 6 of Pl. XII., the last
upper molar presents only ten ridges.

Of the inferior molars, the antepenultimate and penultimate
milk teeth are seen in situ in the lower jaw fragment, fig. 10
of Pl. XIV. F.A.8., drawn to the natural size.1 The former
presents three ridges with talons, and the latter six principal
ridges besides talons. Another lower antepenultimate is
yielded by the young mandibule, figs. 7 and 7a, of Pl. XIL.,
also presenting six principal ridges. The empty alveolus of
the small antepenultimate tooth is exhibited in the same
figure at b. The last milk molar, or first of the intermediate
series, is seen in fig. 8 of the same plate to possess seven
principal ridges, with a front talon. The antepenultimate or
first true molar is represented by fig. 10, showing also seven
principal ridges. The two specimens last mentioned are
further remarkable in showing each a premolar tooth in situ.
The penultimate true molar varies in presenting eight or
nine ridges. The specimen, fig. 8 of Pl. XI. of the same
work, exhibits a penultimate, with nine ridges and a small
back talon. The last true molar is beautifully preserved on
either side in the mandibular specimen, fig. 2 of Pl. XL,
showing about eleven principal ridges.? Other specimens of
the same tooth, presenting nearly the same number of ridges,
are seen in the specimens figs. 12 and 13 of Pl. XII? The

1 At the bottom of the plate in the 3 Fig. 13 a of the same plate has no
‘Fauna Antiqua,’ the specimen is re- | connection with fig. 13. Itis misplaced
ferred to E. Hysudricus, but this is an | there, and belongs to the series of illus-
error.—See vol. i. page 442.—[ Ep. ] trations of E. (Huelephas) Hysudricus,

2 Reproduced in Pl. VI. fig. 2—[Ep.] | not to £. (Lowod,) planifrons.

a ee

E. (LOXODON) PLANIFRONS. 93

last true molars, upper and lower, are subject to a certain
amount of variation in the number of the ridges, which will be
more fully considered in the remarks upon the next subgenus
Huelephas. All the molars, both upper and lower, are rela-
tively to the length of the crown much broader in this
extinct species than in the existing African Elephant.

A very important part of the dentition of H. (Loxodon) pla-
nifrons, in relation to the systematic affinities and characters
of the Elephants, remains to be considered, namely the pre-
molars. The presence of these teeth in this species is ad-
verted to in the preceding Part (p. 6), and in the observations
which follow the technical definition of the genus Hlephas
(p. 13). The lower jaw fragment, fig. 8 of Pl. XII. F.A.S.,
cited above, displays three teeth in situ, viz. in the posterior
extremity the last milk molar, in front of it the penultimate
milk molar (b) nearly worn out, and emerging from below the
latter a small vertically succeeding premolar (c) is exhibited.
The penultimate premolar is represented of natural size (c)
in fig. 9 of the same plate. It is considerably smaller in all
its dimensions than the antepenultimate milk molar, fig. 1a,
drawn to the same scale. It is of a roundish form, and
shows no distinct indications of ridge-divisions, It was
therefore, in all probability, of but small importance func-
tionally in the economy of the species. In like manner,
figs. 10 and 10a of the same plate furnish an illustration of
the last lower premolar in situ, in front of the first or ante-
penultimate true molar. That the latter is one of the true
molars is clearly proved by its large dimensions and by the
mature form of the jaw. Fig. 116 represents the last pre-
molar (natural size) 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.

Of the upper premolars only the penultimate has been
discovered in situ. A beautiful example is seen in the cranial
fragment represented by figs. 4,5, and 6 of Pl. VI. of the
same work, on the left side of the palate, the tooth of the
right side having dropped out. Behind the premolar are the
last milk molar, well worn, and the antepenultimate true
molar in germ. The premolar, in plain view, is of a very
broad and round oval form. 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. The surface of the crown has attained

94 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

the level of the first disc of wear of the tooth immediately
behind ; it bears but slight marks of abrasion, which, however,
appear to indicate that it was opposed to a corresponding
tooth below. This penultimate premolar is represented of
the natural size by fig. 6 of Pl. VI. of the work above
referred to.

Dr. Kaup has given a very instructive illustration’ of the
beautifully preserved young lower jaw of M. (Trilophodon)
angustidens, which 1 detected (antea, pp. 88, 73) im the collec-
tion of M. Ziegler-Ernst at Winterthur. The two premolars
are seen in situ in this specimen, the penultimate emerged, the
last embedded in the jaw, below the last milk molar. The
specimens cited above place it beyond question that the
Sewalik species, H. (Loxodon) planifrons, had the premolar
series as complete numerically as either Dinotherium gigan-
teum or M. (Trilophodon) angustidens. I have already ex-
plained that paleontologists have heretofore entertained an
opinion adverse to this being found to occur in any species of
Elephant.

The above remarks may appear to be beside the professed
object of the essay, but they are essential to the proper
estimate of the characters to be adduced in the sequel, in
proof of the specific distinctness of the British fossil Lox-
odons, which will now be considered.

4, Elephas (Lowod.) priscus.—Has the African Elephant
ever been found in the fossil state in Europe? and if so,
within what geographical limits? These are questions of the
highest interest, and to which a new kind of importance
attaches, from the investigations of some of the later French
paleontologists. In 1821 Professor Goldfuss, of Bonn, pub-
lished an account, with figures, of a reputed fossil molar,
found in the collection of the Canon Mehring, of Cologne, the
precise origin of which was not well ascertained. The crown
presents seven discs of wear, with the well-marked rhombs,
shaped exactly as in the existing African Elephant. The
dimensions of the tooth are—length, 5:4 inches, by an ex-
treme width of 2:3 inches, determining it to be a true molar.
The specimen is described as being much decomposed ; the
crust of cement friable and of an ochre-yellow colour, the
ivory greyish white, and the plates of ivory and enamel
separated by fissures. In another memoir of a later date,
the same author describes other teeth, presenting similar
characters, and asserted to be derived from a diluvial deposit,
on the banks of the Riithr in Westphalia; and he affirms that
he had seen in other collections similar fossil teeth. He

1 Beitrige, Heft iii. 1857, p. 9, tab. i. figs. 1-3.

E. (LOXODON) PRISCUS. 95
inferred that the valley of the Rhine was formerly inhabited
by a species of Elephant which more nearly resembled the
existing African species than EH. primigenius does the existing
Indian. But he did not hazard an opinion whether or no it
was specifically different from the existing African, which
could only be satisfactorily established by the discovery of a
skull, and he named the species provisionally. Cuvier ques-
tioned the fossil authenticity of these specimens, and of other
instances of the same nature, which he enumerates. In the
autumn of 1847 I had an opportunity of examining the spe-
cimens above referred to, in company with Dr. Goldfuss, at
Bonn. They were much sun-cracked, resembling in this
respect grinders of the existing Asiatic Elephant as they are
presented in India after long exposure to atmospheric agen-
cies; but the fracture and texture of the ivory yielded the
glistening sericeous appearance characteristic of recent teeth,
and conveyed to my mind a corresponding impression that
the molar was probably of modern origin.

The celebrated C. EK. von Baer describes, with exemplary
caution, two reputed fossil molars from the north of Germany,
resembling exactly those of the African Elephant. One of
these he unhesitatingly regards as being of modern origin,
from the circumstance that some of the cellular membrane
lining the alveolus was still preserved upon the tooth.! The
other, discovered in the sandy foundations of the monastery
of St. Adalbert, near Dantzic, is, from the description, mani-
festly of the African Hlephant; and the tooth, from its
partially worn condition, is evidently not one that had been
naturally shed. Von Baer cites the opinion of Rathke, that
it may have been derived from a casualty in some travelling
menagerie, but he with reason doubts if an African Elephant
was ever brought to Dantzic, either during the Roman
empire or subsequently. After carefully balancing the tex-
ture and consistence of the specimen and the circumstances
under which it appears to have been found, he could arrive
at no satisfactory opinion whether it was really fossil or not,
and he leaves the point undetermined.? Itmay be remarked
on this head, that the freshness of preservation of teeth and
tusks of Hlephants discovered in Post-Pliocene deposits fur-

nishes no argument against

1 ‘Quid amplius, tunice et tele cellu-
losee alyeolum vestientis partem in dente
siccatam invenimus. Partes molles per
tot secula in nostris regionibus servari
posse eredat Judeus Apella! Nos vero
dentem non fossilem suspicamur. He
also describes a tooth certainly fossil,

their being of really fossil

but of uncertain origin, in the Museum
of St. Petersburg, referring it to 2. pris-
cus (Mém. de l’Acad. de St. Pétersbourg,
tom. i., Bulletins Scientif. p. 16).

2 «Sie status fossilis testimonia certa
non cognoscimus et non habemus quo -
catalogi assertum confutemus,’

96 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

antiquity; for ivory tusks of the Mammoth have been found
in silt in Britain, in such perfect preservation as to have
been fit for turning into chess-men.' I have examined a skull
of the Mammoth, discovered in the Lehm of the valley of the
Rhine, and now preserved in the Museum at Mannheim,
which is quite as fresh, and appears to retain as much animal
matter as crania of existing Elephants that have long been
exposed in public collections. It is in a better state of
preservation than skulls of domestic animals that have been
buried for a long time within the historical period and sub-
sequently disinterred.

The most characteristic specimen of LHleph. (Loxodon)
priscus that has yet been discovered in British deposits is
a tooth which was purchased by the late Mr. Konig, then
keeper of the Mineralogical and Paleontological Gallery, for
the British Museum, of Mr. Ball, a well-known trading
collector. It was stated to have been procured from the
brick-earth excavations at Gray’s Thurrock, in the valley
of the Thames—a locality rich in Mammalian fossils, and
first brought to notice by the able investigations of Mr.
Morris. No precise particulars as to the history of the
specimens were ascertained or put on record by Mr. Konig.
But on paying a visit to Gray’s Thurrock, in company with
the late Professor Edward Forbes and Colonel James, in
the summer of 1845, with the express object of examining
the association of extinct Mammalia in this very interesting
deposit, I was informed on the spot that the tooth in question
belonged to the skeleton of an Elephant, the greater part of
which was found spread out in one place by the workmen,
when digging for brick-earth. Most of the bones were
destroyed in the operation; but besides this molar, another,
belonging to the same animal, was retained by Mr. Meeson,
the proprietor of the brick-field. ;

The specimen (No. 39,370 of the Brit. Mus. MS. Cat.) is a
last molar, left side, of the lower jaw. The mineral charac-
ters, friability, test by the tongue, colour, dull fracture, and
general appearance, leave no doubt as to its being a veritable
fossil. Mr. Kénig, to place this important point beyond
question, permitted it to be sawn up, and the condition of
the interior was equally conclusive of its fossil nature. The
longitudinal section is represented by fig. 76 of Pl. XIV. of
the ‘Fauna Antiqua Sivalensis;? (reproduced in Pl. VII.

1 The fossil tusks of the Mammoth| 2 The dimensions of the tooth are
form an article of commerce in Siberia, | given in the description of it in the
and are largely used in the manufacture | ‘Fauna Antiqua Sivalensis.’ See vol. i.
of ornaments and statuettes, &c. p- 441.—{ Ep. ]

——

E. (LOXODON) PRISCUS. 97

figs. 1 and 2); and if it be compared with fig. 4) of
Pl. II. of the same work,' representing a vertical section of
a penultimate lower molar of the existing African Elephant,
it will be seen that there is the closest general resem-
blance between the two, 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. If
the comparison be extended to the sections of the teeth of
the Mammoth and of the existing Indian Elephant, figs. 1
and 2a, Pl. I., of the same series, the difference from them
is equally apparent. The specimen consists of the part of
the tooth extending from the sinus between the first and
second fangs to the last ridge. The anterior portion sup-
ported by the first fang, and which in the African Elephant
consists of the front talon and the two foremost ridges, is
wanting. The fragment exhibits the discs of eight worn ridges
finely preserved. The three anterior discs are worn low;
the next four are successively less and less abraded ; the last
ridge shows only the tips of two digitations, with a consider-
able interval between them. There is no distinct hind talon.
The discs of wear present an unmistakable resemblance to
those of the existing African Elephant, in breadth, lozenge-
shaped outline, and mesial expansion; but when examined
in detail, there are obvious points of distinction. In the
living species the lozenges are more strictly rhomb-shaped;
the salient edge of enamel is distinctly crimped; the lateral
terminations of the rhombs are flattened; and the mesial
angles of the contiguous discs are either more approximated
or overlap each other laterally. In EH. (Loxodon) priscus, the
dises are rounded at their lateral terminations, and broader.
Although the mesial expansion is quite as great as in the
African Elephant, it is less sudden, and in the general outline
there is a tendency to a reniform or obsolete crescentic shape,
the anterior enamel boundary of each disc being somewhat
concave, and the posterior convex. The horns of the crescents
are bent abruptly forwards. This is best seen in the fourth,
fifth, and sixth discs ; the first three, being more worn, show
this peculiarity less distinctively. Another obvious character
is that the enamel-plates are thicker, and present a less
degree of crimping than in the African Elephant.

When viewed laterally, the resemblance to the existing
species is as marked as in the crown aspect. The ridges are
alike broad in both, and the fangs are similarly disposed,
those which support the five posterior ridges being confluent
into a common shell. (Compare figs. 7aand5a of Pl. XIV.,

[ i ao of the penultimate upper molar is shown in PI, iy. fig. 3 of vol. i.
/ =) up.

VOL. II. H

98 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

‘Fauna Antiqua Sivalensis.’) The vertical height of the
seventh ridge, although but slightly worn, does not exceed
24 inches, while the greatest width of the crown is 3 inches.
The flexuous bend of the enamel-plates vertically, at the
posterior end as seen in the section, is not a distinctive
peculiarity, since it is met with in inferior molars both of the
Indian Elephant and of H. (Hueleph.) antiquus.

The principal dimensions of this specimen are—

Extreme length of crown-surface, 8° in. Width of crown-surface at first ridge,
2:35 in. Width of crown-surface at the fourth ridge, 2°8 in. Width of crown-
surface at the seventh ridge, 1'8 in. Height of the seventh ridge, 2°5 in. Width
of second disc at mesial expansion, 0°95 in.

There are eight discs of wear to a length of 8 inches,
being an average of one ridge to the inch, a proportion
corresponding closely with that presented by the oldest teeth
of the African Elephant. The front fang, in the last lower
molar of the latter species, generally supports two principal
ridges besides the anterior talon. It is inferred, therefore,
that the corresponding fossil tooth of H. (Loxedon) priscus,
when entire, was composed of ten or eleven ridges, and that
it was about 11 inches long.

Another specimen (No. 18,966 of the British Museum Col-
lection), also reputed to have been procured from the brick-
earth deposits of the valley of the Thames, is represented by
fig. 6 of Pl. XIV. of the work above cited. It is a fragment
mutilated at both ends, showing 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 Grays Thurrock, described above.
There is the same mesial angular expansion, and a still
greater tendency to the discs assuming a crescentic form.
The mutilated condition of this specimen renders its identi-
fication somewhat doubtful; but it is inferred to belong to
E. (Loxod.) priscus, and to be a penultimate molar of the
lower jaw, left side. The dimensions are—

Length, 5° in. Width of the crown behind, 3-in. Height of the crown behind,
2°8 in.

Besides the five entire ridges, the fractures pass through
the middle of a dise at either end; so that the specimen may
be considered to possess, six ridges in a length of 5 inches,
being an average width of -83 to each, near the summit, where
but little worn.

The only other British specimen referable to this species,
that has come under my observation, is a fragment of a lower
jaw, with which I have lately become acquainted, in the rich
and valuable collection of Mammalian remains from the
Norfolk coast, between Cromer and Lowestoft, formed by the

E. (LOXODON) PRISCUS. 99

Rey. John Gunn, of Irstead, who has liberally placed this
specimen, with many others, at my disposal for description.}
It is a rolled and mutilated fragment, comprising the sym-
physis and anterior part of both rami, the beak-apophysis
being entirely rubbed off. A single molar is present on the
left side ; none on the right, which is very mutilated.

The tooth, which is inferred to be the penultimate true
molar, presents the crown nearly entire and well worn. Its
length is determined by the anterior and posterior fangs,
which are exposed. The front talon, together with a portion
of the first principal ridge, supported upon the anterior fang,
are partially broken. The crown exhibits the discs of the
seven posterior ridges and part of the first, indicating in all
eight main ridges. There is no posterior talon, the last ridge
descending continuously, for insertion upon the fang. The
crown is very narrow in front, and expands gradually as far
as the sixth ridge. The discs of wear are broad, with a
mesial angular expansion as in the African Elephant; but at
the same time they exhibit a very pronounced crescentic
outline, the horns or lateral terminations being much more
bent forwards than in the specimen from Grays Thurrock.
The general contour of the anterior enamel-plate of each
disc is markedly concave, and the posterior one convex. The
mesial expansion of the third disc, measured between the
outer surfaces of the enamel, is exactly 3ths of aninch. The
discs are uniform in shape, from the first to the last, the
difference between them depending solely upon the greater
or less amount of wear. The projecting edge of enamel is
irregularly crimped, and to a much more obvious degree than
in the Grays Thurrock specimen. In this respect it ap-
proaches more nearly the character of H. (Huelephas) antiquus,
to be described in the sequel. (Pl. VII. figs. 3 & 4.)

The principal dimensions of this fragment are—

Vertical height of the ramus, measured from the lower margin to the summit of
the first ridge of molar, 9:1 in. Length of the molar crown (part wanting in
front), 6°7 in. Width of crown at the second dise, 1:9in. Width of crown at the
fourth disc, 2:4 in. Width of crown at the sixth disc, 26in. Height of the

seventh ridge, about 3:0in. Mesial expansion of the second dise of wear, 0°75 in.
Mesial expansion of the seyenth dise of wear, 0°7 in.

In front of the molar there is a well-marked triangular
cicatrix, indicating the remains of a nearly filled up fang-
cavity, at the anterior angle of which a small portion of an
ivory stump is visible. The plane of the cicatrix slopes
_ suddenly downwards upon the diasteme, and the line of the
interior margin does not follow the direction of the alveolar

‘ ' Mr. Gunn kindly forwarded this specimen to me to be figured by Mr. Dinkel.
See Plate vii. figs. 3 & 4.—[Ep.]

H 2

100 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

margin of the molar in situ. The length, measured from
the anterior angle of the fang-scar to the back of the tooth,
is 8:1 inches. A question arises, whether this alveolar scar
belongs to a distinct younger tooth that had been shed; or
does it represent the remains of an anterior portion of the
tooth now seen in situ, which was supported by a fang
anterior to that described above as being the large front fang ?
In the latter case, at least three other ridges would have to
be added to the crown, making eleven in all, and the tooth
would then present the character of the last lower of the
African Elephant instead of the penultimate. There is no
positive character to decide the question either way ; but L
am led to consider that the tooth has its full proportions in
what now remains, from the circumstance that the crown
narrows so. much at the first ridge, z.e. to less than two
inches, while it is three inches wide behind. The fang-scar
in this view is regarded as indicating the position of a shed
antepenultimate.

The jaw is so rolled and mutilated, that it affords but few
distinctive characters for description. The most striking
point is the very great proportional height of the ramus,
which in a line with the anterior termination of the molar
is upwards of 9 inches. This proves that the jaw is that of
an adult animal, and that the molar is certainly not of a
younger age than.a penultimate. The inner side of the
ramus is flat, and the jaw appears to have been very high
and compressed in front. What remains of the symphysis
indicates that the gutter was broad, and that the rami
diverged considerably. Two mentary foramina are present
on the left side, with an interval of about 3 inches between
them, and close to the edge of the diasteme.'

This valuable specimen was found on the Palling beach,
near Happisburgh, where fossil molars of elephants are so
abundant. There is no certain information from what bed
it was derived, whether from the ‘ Elephant bed’ of Mr.
Gunn, below the ‘submerged forest-bed,’ or from the ‘lami-
nated blue clay’ above it. But he is satisfied that it was
derived from a deposit below the ‘ dark-mud Boulder-clay.’
The same uncertainty applies to the greater part of the
Mammalian remains found along the beach from Happis-
burgh to Mundesley. They have rarely or ever been observed
in the cliffs in situ; and in the present instance there is no

matrix upon the specimen to aid in arriving at an opinion

' Tn this respect the jaw in question | are always placed a considerable distance
would seem to differ from E. Africanus, | from the edge.—Note by G. Busk, F.R.S.
in which species the mentary foramina | in Quart. Journ. Geol. Soe.

bis 5 el

DESCRIPTION OF PLATE VII.
ELEPHAS (LOxoDON) PRISCUS.

Fig. 1. Crown view of last lower molar, left side, one-third of the
natural size. The drawing is a copy of one by Mr. Ford in
Plate XIV., fig. 7, of the Fauna Antiqua Sivalensis. The spe-
cimen is from Grays Thurrock, and is now in the British Museum.

(Cat. No. 39,370.) It is fully described at page 96.

Fig. 2. Vertical longitydinal section of same tooth, one-half of the
natural size. Copied from Plate XIV., fig. 7 6, of the F. A. S.

Figs. 3 and 4. Views in plan and profile of the penultimate true molar,
lower jaw, left side, one-half of the natural size. These draw-
ings have been executed by Mr. Dinkel from the original
specimen in the collection of the Rev. John Gunn, which Mr.
Gunn forwarded to London for the purpose, subsequently to Dr.
Falconer’s death. The specimen is fully described at page 99.

VOL. Il.

oe

hs

ee eee ee eee ee et ae

Wee, 7

Vol. TL.

priscus.

:
a

8
SS

a
a
=
Fa

ae

E. (LOXODON) PRISCUS. 101

upon this point. The specimen is entirely free from ferru-
ginous impregnation. The ivory is white, and adheres freely
to the tongue.

Authentic remains, referable to this obscure form, are so
rare in Kuropean collections, that it is of importance to make
known any specimen calculated to throw light upon it. By
the liberal and obliging permission of Dr. Emilio Cornalia, IL
was enabled to examine minutely a very fine fossil molar,
preserved in the Natural History Museum of Milan, which I
refer to H. (Loxodon) priscus. This specimen is a last molar
of the lower jaw, left side, nearly entire, the only deficiency
being in the anterior talon and part of the first ridge borne
by the Jarge anterior fang. The crown exhibits twelve
principal ridges and a posterior talon, the ten anterior of
which are worn down into transverse discs, while the last
three are but slightly abraded. All the discs of wear present
a broad rhomboidal expansion in the middle, as in the African
Elephant, but modified by a crescentic tendency as above
described in the fossil molars from Grays Thurrock. The
first dise is fractured vertically, and confluent at either side
with the second, which is also nearly confluent, from advanced
wear, with the third. The fourth disc is very broad (antero-
posterior diameter), and exactly corresponds in form with
the first disc of the Grays Thurrock specimen, the mesial
expansion being ‘75 of an inch. The outer termination of this
disc is bent forwards somewhat like the fourth in that
specimen, but more abruptly pronounced. The ivory surface
is deeply excavated, so that the enamel edge projects in high
relief above it. The fifth and sixth discs are of a similar
form; but, being less worn, they are less expanded. Their
cornua are bent forwards on the inner side, with the cres-
centic character seen in the fourth, fifth, and sixth discs of
the Gray’s Thurrock specimen. The anterior enamel-plate
of the sixth disc projects very much (to the extent of seven-
tenths of an inch) above the contiguous stratum of cement,
while the included ivory surface is but slightly depressed.
The seventh, eighth, and ninth discs present a corresponding
form, getting narrower successively in consequence of being
less worn, but each showing more or less of a mesial angular
expansion. The tenth ridge is but slightly worn, and the
disc is barely continuous across. The eleventh and twelfth
ridges show each two distinct discs. The posterior talon
shows the tips of two denticles or digitations, like the last
ridge of the Grays Thurrock specimen. To the posterior
surface of the last there is appended a single thick digitation,
which projects backwards in a salient gibbosity, being the
converse of what is seen in the re-entering sinus, on the
posterior surface of that specimen.

102 BRITISH AND EUROPEAN FOSSIL ELEPHANTS,

The grinding-surface is very concave from back to front, a
chord stretched from the front to the last ridge being fully
1:4 inch above the level of the sixth and seventh dises. It
is also a good deal contorted, the anterior inner side sloping
‘backwards and outwards, while the posterior outer angle
slopes forwards and inwards, corresponding precisely in this
respect with the specimen of the last lower molar of HF.
(Huelephas) Hysudricus, represented by fig. 13 a, Pl. XII. of
the ‘ Fauna Antiqua Sivalensis.’ ;

The enamel plates are very thick; and their outer edges
present an appearance of crimping, caused by the deep
vertical grooving of the outer surface, namely, that in con-
tact with the stratum of cement; but they are not plaited.

The principal dimensions are as follows :—

Length of crown, about 12° in. Width of crown at fourth ridge, 3:1 in. Width
of crown at sixth ridge, 3°35 in. Width of crown at eighth ridge, 3-4 in. Width
of crown at tenth ridge, 3-1 in. Width of crown at twelfth ridge, 2°6 in. Width
of posterior talon, 1°65 in. Height of crown at fourth ridge (much worn), 1°6 in.
Height of crown at sixth ridge, outer side, 3:1 in. Height of crown at ninth ridge,
outer side, 45 in. Height of crown at twelfth ridge, outer side, 5-2 in. Mesial
expansion of fourth dise, 0°75 in.

The crown includes twelve ridges in a length of about 12
inches, being an average of one inch to each. The specimen
was compared with an exact drawing, of the natural size, of
the Grays Thurrock molar; and the two agreed in the closest
manner, making allowance for their different stages of
wear. This very important specimen bears a record of having
been discovered by Count Gazzola, in a calcareous deposit
upon Monte Serbaro, in the valley of Pantena, about eight
miles from Verona, along with the remains of other herbi-
vorous quadrupeds. Inits mineral condition and appearance
it presents undoubted evidence of being a true fossil.

This completes what I have to adduce in proof of EH.
(Loxodon) priscus being a distinct species ;' and it must be
freely admitted that, considering the area explored and the
number of museums examined, both in Britain and abroad,
the evidence, although strong in kind, is, in the form of
authentic materials, quantitatively very limited. It may be
asked, if this be a well-founded species, how does it happen
that determinable remains of it are everywhere so rare? To
which it is replied that the Pliocene Mastodon (Triloph.)
Borsoni, respecting which there is now no question among
those Mammalian paleontologists who have -studied the
remains attributable to it, is in the same predicament, and

1 Subsequently, in 1859, Dr. F. found | Monte Verde, having all the characters
in the University of Rome, and des- | of the Milan specimen. The crown was
scribed in his note-book, a magnificent | 12in. long and had twelve plates.—[ Ep. ]
last lower molar of E. priscus from

E. (LOXODON) PRISCUS. 103

almost equally rare. A single molar of that species was dis-
covered by Abbé Borson in 1820, in the Pliocene deposits
of the Astesan, since which date, up to 1856, not a single
additional specimen had been acquired for the collections of
Florence, Pisa, Turin, Milan, or Pavia, although the ossiferous
Pliocene strata of the Astesan had subsequently been largely
laid open by railway-cuttings. The detailed proofs have been
obtained from deposits in Auvergne and elsewhere in France.
Tn 1845, I was unable to reconcile the characters yielded by
the Grays Thurrock specimen with those of any recognized
species of fossil elephant, except H. priscus; and after an
interval of twelve years, with a large addition of experience
in the investigation of the subject, and with more materials,
my conviction of its being distinct is as strong as that in
favour of any other species in the genus. In Mammalian
paleontology, when the evidence furnished by the teeth can
be crucially tested, by means of the varied characters of the
cranium and of the bones of the extremities, a safe and
satisfactory conclusion as to the distinctness or otherwise of
the species can generally be attained. But when a few teeth
only are available, the area of the evidence becomes very
limited, and there is a constant unperceived tendency in the
observer either to magnify the value of the differential
remarks, or to underrate them, as the case may be, according
to his inclination, from some extraneous influence, to make
the species distinct or merely a variety of some other form.
In this case I have tried to guard myself against a bias either
way, and the evidence has appeared to me to be conclusive of
the distinctness of the species. Although so little is known
of the details of the different teeth, the ridge-formula is
inferred to be 7:7 + 8+ 11 in the last milk molars and
three true molars, as in the African Elephant. The known
limits to the dimensions of the molars in the Elephants,
coupled with the average great antero-posterior extent of the
ridges, in this form, namely, one inch to each, necessarily
involves a limited number of the latter. The distinction
from the African species is founded upon the characters that
the lozenges are regularly rhomboidal in the one, and some-
what crescentic, with the angular expansions more apart, in
the other. Both species belong to the ‘ Stenocoronine’ type
of Loxodon. The distinction from the fossil form to be next
described, EH. (Lowodon) meridionalis, is borne out by well-
marked characters, the crowns of the molars in the latter
being constantly very broad, the digitations thick and distinct,
and the discs of wear free from mesial rhomboidal expansion.
If E. (Lowod.) priseus could be reconciled with any other
fossil species, it would be with H. (Huelephas) antiquus, in

104 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

which the dises of the worn ridges exhibit a certain amount
of expansion. But in this species the ridges are very
numerous, elevated, and attenuated, their number, as in the
existing Indian Elephant, ranging as high in the last lower
molar as from twenty-four to twenty-seven, while in H.
priscus they do not exceed twelve or thirteen. These points
will be brought out more in detail in the sequel, when treat-
ing of these species.’

Hlephas (Loxodon) priscus occurs in Italy in the Sub-A pen-
nine Pliocene strata of the Romagnano; in England, in the
fluviatile deposits of the valley of the Thames, and in unde-
termined strata on the coast of Norfolk, but believed to be
below the ‘ Boulder-clay.’ I have not observed it among the
exposed speciméns in the public collections in Paris, nor in
any museum in France that I have visited. The name is
enumerated by Pomel in his ‘ Catalogue of the Fossil Remains
of the Loire and the Alliére,’ as having been found in the
Plain of Saliéve, Collection of Laizer. He describes it briefly
as Elephas priscus (Goldf.) : ‘Espéce ayant les lames de ses
molairesdisposées comme dans] Hléphant d’Afrique.’ Whether
this means the ancient race of the existing species attributed
to the valley of the Rhine, or the distinct fossil form, with
crescentic discs of wear, I am unable to determine.

5. H. (Loxodon) meridionalis, Nesti.—Of this species, the
materials in Huropean collections, more especially in Italy
and England, are fortunately so abundant and perfect as to
place its specific distinctness beyond question. But this
conclusion has been so long opposed by the highest paleeon-
tological authority, namely, by Cuvier, De Blainville, and
Owen, and the geological inferences involved in it are of such
importance that I consider no apology necessary for entering
fully upon the evidence bearing on the subject.

The ‘Val d’Arno Superiore’ has, from remote ages, been
celebrated for the vast abundance of fossil remains found
there. Huge bones and teeth of Elephants were especially
numerous. A large collection of these was formed by Tar-
gioni Toretti, which ultimately found its way into the Grand
Ducal Museum at Florence; and numerous additions were
made by Nesti, who, in 1808, soon after the publication of
Cuvier’s ‘Memoir of the Mammoth’ (Annales du Muséum,
tom. vill.), examined the Tuscan Elephantine remains, and
was so satisfied of their difference from those of the Mammoth,
that he proposed for them two specific designations, namely,
Filephas meridionalis and E. minutus: Influenced by the fact

' In his memoir on £. Columbi, written | the opinion that Loxodon priscus was
five or six years subsequently to that on | a form of ZL. antiguus.—t|Ep.]
Mastodon and Elephant,Dr. F', expressed

ee

E. (LOXODON) MERIDIONALIS. 105

that Cuvier had laid so much stress upon the peculiar form
of the lower jaw, and guttered beak of the symphysis, as dis-
tinctive marks of H. primigenius, Nesti (not a professed ana-
tomist) was naturally led to direct his attention, in the first
instance, chiefly to the same parts in the Val d’Arno remains.
Unluckily the specimen that presented the most pronounced
beak had lost its molar teeth ; Nesti assumed it to be of an
Elephant. But this selected ‘ piéce justificative ’ for his Hle-
phas meridionalis was proved by Cuvier to be the lower jaw
of Mastodon Arvernensis,' and EH. minutus to be merely a young
Elephant.

After a long interval, during which Cuvier had visited the
Tuscan collections, Nesti brought out another memoir upon
the subject, in which, upon greatly extended observations on
specimens of all ages, from the foetus upwards, including
crania, lower jaws, molars, tusks, and bones of the extremities,
he upheld the soundness of his first inference in regard to
the distinctness of H. meridionalis, while he admits tacitly
the force of Cuvier’s criticism upon his second species, H.
minutus. The memoir is accompanied by figures of the cra-
nium, lower jaws, and molars, but so imperfectly executed,
that they proved of little service either in establishing his
case or in guiding other paleeontologists to a satisfactory con-
clusion. Another circumstance, which materially damaged
the authority of Nesti upon a question of such difficulty and
importance, is that he states that, after examining a vast num-
ber of molars of all ages, he had found them to vary so much—
some having thick plates, others thin, and the same tooth
presenting such different patterns, according to its age and
degree of wear—that he had abandoned the characters
yielded by the molar teeth as worthless (!) for any reliable
marks of specific distinction. In the teeth themselves he
had discovered no sensible differences from the characters
figured and described of those of HE. primigenius. This sin-
eular conclusion is, in some measure, explained by the fact
that hardly a specimen of a molar of the true Mammoth ex-
ists in the Florentine Museum for comparison. It is, per-
haps, still more remarkable that the experienced eye of Cuvier
should have glanced over the multifarious evidence supplied
by the Tuscan collections, without being convinced that FH.
meridionalis was a well-founded species, considering the rapi-
dity with which he seized, and the logical precision with

1 Note from Dr. F.’s Note-Book.— Flo- | It is clearly a Mastodon. Another lower
rence, May 20, 1859. Again examined | jaw near it has the spout even a little
this lower jaw. It is of an animal not | longer, but not like M. longirostris or M.
quite adult, and probably contained the | angustidens.’—| Ep. ]
antepenultimate and penultimate teeth.

108 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

which he characterized, the distinctive marks of the Mam-
moth from the existing Indian Elephant.

Failing the teeth, Nesti drew his specific distinctions from
the form of the cranium and lower jaw. Ample evidence is
afforded by them for establishing EH. meridionalis as an inde-
pendent form.

It would be both tedious and beside the scope of this
essay to detail the various opinions that have been expressed
by different paleontologists respecting EH. meridionalis. They
will be found embodied in systematic works upon the science ;
and, on the present occasion, I shall confine myself to such
as have had most influence, either in throwing light upon
the characters of the species or in discrediting it.

Cuvier rested the specific distinction of the molars of the
Mammoth upon three characters, namely, the great width of
the crown, the attenuation of the plates, and the absence or
small amount of crimping in the edges of the enamel. He
admits that he had observed some notable exceptions as re-
gards the two last. The first example that he cites is the
‘dent de Porentrui,’ from the valley of the Rhine, above Stras-
burg, which is remarkable for a great amount of plaiting in
the enamel-plates. This specimen, however, is not fossil, but
belongs to the existing Elephant.'! The only other exceptions
cited are three Italian specimens from Romagnano, Monte
Verde, and the Val d’Arno, in allof which the plates are very
thick, and which in reality belong to EH. meridionalis. No
exceptional illustration is adduced from Siberia, or any other
northern locality, where the true Mammoth prevails. It is
implied that they constantly present attenuated and un-
crimped plates. Cuvier therefore supposed the two last cha-
racters to be inconstant, and adhered to the great width of
the crown, which, however, is common to the Mammoth and
to EH. meridionalis. It is obvious that the prepossession in his
mind in favour of a single European fossil species of Ele-
phant, which is manifest throughout the ‘ Ossemens Fossiles,’
had unconsciously led the great anatomist to undervalue the
very characters which he was the first to inculcate.

The Abbé Croizet, to whom paleontology is indebted for
so much valuable research on the fossil fauna of Velay, was
the first who had the courage to question the decision of
Cuvier against EH. meridionalis. In his work upon Puy-de-
Dome, he has figured and described a fragment of an upper
(?) molar (lower left of Croizet and Jobert) discovered at
Malbattu. It is a good deal mutilated, and the figure is not

* “It is a lower left, and beyond all | out to me by M. Lartet.’—WNote by Dr.
question not fossil, but of the existing | /.—[Ep.]
Indian Elephant. This was first pointed

E, (LOXODON) MERIDIONALIS. 107

so exact as to be conclusive ; but in the form of the discs of
wear, in the thickness of the enamel-plates, and in the slight
degree of crimping along the edges, it differs alike from H.
(Huelephas) primigenius and H. (Huelephas) antiquus, and
corresponds with Italian specimens of H. meridionalis. In
Plate X. fig. 1 of the same work, he gives a representation of
a fossil molar discovered by Lecoq at Clermont, which exhi-
bits similar characters. He refers to Nesti’s researches, and
sums up by inferring that, as there are two living Hlephants,
so there were two fossil species—the one with attenuated
plates, being the Mammoth of Siberia, the other with thick
plates, as seen in specimens from Porentrui, Romagnano,
Monte Verde, Laufen (in Germany), and the Val d’Arno. He
corsidered the facts sufficient, but assigned no other name to
the second species than that of ‘ Eléphant de Malbattu,’ and
awaited the results of further discovery for confirmation of
the inference.

Professor Owen has entered very fully into the question of
distinct species, in the part devoted to Hlephas of his Report
to the British Association for 1843, and subsequently repro-
duced in his separate work on the ‘ British Fossil Mammalia.’
The result at which he arrived, after examining a vast num-
ber of specimens, was that there had been only one species of
fossil Elephant in Britain, namely, H. primigenius; and while
fully recognizing the marked differences presented by mo-
lars from different localities and different deposits, he had
found so many intermediate gradations, that he was unable to
draw a well-defined line between the thick-plated and
thin-plated varieties. The consideration of the grounds upon
which this opinion was founded will fall more properly into
the discussion on the fossil species of Hwelephas. In regard
to H. meridionalis, he alleges that the variety of molar (i.e.
thick-plated) on which this proposed species is founded occurs
not only in England, but in Siberia and as far north as
Eschscholtz Bay ;' and, in proof, he appeals to the specimens
described by Buckland in the appendix to the ‘ Voyage of the
Blossom.’ Professor Owen refers to this thick-plated variety
of the Mammoth certain British molars, which will be noticed
in the sequel, as belonging to H. (Loxod.) meridionalis. I
may remark that the conclusions to which I have been led on
all the points involved in the question of distinct species or
varieties in the European fossil Elephants are widely dif-

1 T got yesterday from Mr. Christy the | meridionalis, although the plates are a
molar of the Elephant from the Ural | little more approximated than usual. If
mountains. It is certainly neither of} really from the Ural mountains, it is of
E. primigenius nor of E. antiquus, and | great importance—Lctter to M. Lartct,
so far as the evidence goes it is of Z| Dec. 31, 1863.—[Ep. |

108 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

ferent from those set forth in the ‘ Report to the British
Association ’ and in the ‘ British Fossil Mammalia.’

Early in 1844, my attention was directed to the European
fossil Hlephants as subjects of comparison with the Indian
fossil species from the Sewalik hills. I had satisfied myself,
upon the indisputable evidence of entire crania and well-pro-
nounced dental distinctions, that, exclusive of the Stegodons,
there were three Indian fossil species of Hlephas, two from
the Miocene Sewalik deposits, namely, H. (Huelephas) Hysu-
dricus and HE. (Loxodon) planifrons, and one from the Pliocene
beds of the Nerbudda, HL. (Huelephas) Namadicus, which were
as distinct as the two existing species are from each other.
On comparing them with British specimens, I found that
there was one series among the latter which resembled the
molars of EH. (Lowxod.) planifrons, and that they were chiefly
derived from the ‘ Norwich Crag’ or its vicinity; while an-
other series, found in vast abundance on the ‘ Oyster-bed ”
and in other localities along the Norfolk coast and elsewhere
in England, differed constantly from characteristic specimens
of the Mammoth of the superficial glacial deposits, and were
closely allied to H. (Huelephas) Namadicus from the Nerbudda.
I was in this manner convinced that there were two British
fossil species, besides EH. primigenius and HE. (Loxodon) priscus.
The prevailing opinion, at that time, among the best geolo-
gical authorities in England, was that the Crag deposits were
either of a Miocene or very old Pliocene age. On referring
to the description and figures given by Nesti, Croizet, and
Jobert, and by Cuvier, of molars attributed to H. meridionalis,
I found that they were so indecisive, either from their re-
duced scale or their imperfect execution, that it was impos-
sible to identify the British specimens satisfactorily by them ;
and in the metropolitan collections I could discover no good
series of Val d’Arno specimens to assist me. In consequence,
I came to the conclusion, but hastily as it proved, that the
fossil species from the Norfolk coast and fluviatile beds of the
Thames Valley was the same as the extinct Elephant of the
Val d’Arno; and the figures illustrative of it in the ‘ Fauna
Antiqua Sivalensis’® were published under the name of FH.
meridionalis, while those from the ‘Crag’ and superjacent
‘ Hlephant-beds’ were designated EH. antiquus, under the im-
pression that it was the oldest of European Elephants then
known. But, on paying a visit afterwards to the Oxford
Museum, I found Val d’Arno specimens in Dr. Buckland’s
collection which satisfied me that [had made a mistake, and
that the ‘Crag’ molars were identical with those of EH. meri-
dionalis. It was too late to correct the error in the published
plates ; and it appeared to me that less confusion would arise

E. (LOXODON) MERIDIONALIS. 109

from my continuing, in the subsequent plates, the nomencla-
ture which I had adopted in the earlier ones, than if altered
names were partially introduced, as I intended to give a full
correction of the whole in the letter-press. I regret to find
that the delay in the publication of this correction has led to
a good deal of misconception and to misgiving as to the
validity of the species both at home and abroad. I beg leave
to explain now, that all the plates bearing the name of F.
meridionalis in the ‘ Fauna Antiqua Sivalensis,’ including the
outline-figures of crania in Plate XLII., belong to H. anti-
quus, while those that bear the latter name belong to H.
(Loxodon) meridionalis.! In the descriptions which follow
they will be cited as such.

Before entering upon the details of the British specimens
of #. meridionalis, I think it best to communicate the results
of my examination of the Tuscan collections, as the evidence
furnished by all parts of the skeleton is more complete and
abundant in them than anywhere else.

A. Tuscan Specimens.—The Grand Ducal Museum at
Florence contains seven crania, or considerable portions of
crania, of this species. One of these, a late acquisition, is
attached to a mounted skeleton, the trunk part of which is
complete, but the extremities wanting. Another specimen
consists of a crushed cranium, with the lower jaw attached,
containing the three milk molars, more or less consolidated
both above and below, in situ. The first milk molar is free
from wear, proving that the animal must have died, if not in
the foetal state, at least very soon after its birth. Another
specimen, also of a young calf, shows both maxillaries, with
the palate and floor of the nasal cavity entire, the rest of the
cranium being wanting. The two anterior milk molars in
this specimen, and in the corresponding lower jaw, are worn
to a degree indicative of the animal having been about a year
old. There are five adult crania, indicating by the form of
the tusks both sexes. Three of those described by Nesti, of
enormous size, are still extant. In another, of a very old
animal, the tusks are beautifully perfect. Another speci-
men, limited to the incisive sheaths, also shows the tusks in
their natural position quite perfect. There are numerous
lower jaws and bones of the extremities of colossal dimen-
sions, and an abundance of detached molars of all ages and
in every stage of wear. These Elephantine molars (including
probably both H. meridionalis and EH. (Huelephas) antiquus)
were so common in the Val d’Arno, near Figlinie, that the

1 This correction must be taken with the important qualification already pointed
out in vol, i. p. 448, note 1—[Ep. ]

110 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

peasants were formerly in the habit of using them pro-
miscuously with boulders in constructing the stone-walls sur-
rounding their fields. The osteological materials available
for the determination of the Val d’Arno Elephant, which exist
in the Florentine Museum, are therefore as abundant, and
nearly as complete, as those of the Mammoth at Moscow.

a. Upper Milk Molars.—The beautiful specimen comprising
both maxillaries shows the two front or antepenultimate and
penultimate milk molars in place on both sides, the alveolar
part of the third being wanting. The antepenultimate, on the
right side, is perfectly entire in its contour, but well worn.
The general form is a broad oval, narrowest in front and
broadest in the middle. It presents three principal ridges,
with a front and back talon. The discs of wear are very
wide (antero-posteriorly), with thick enamel-plates, exactly
like fig. 4 of Plate IX. of Cuvier’s ‘Ossemens Fossiles.’ The
dimensions of this tooth are ‘95 inch in length by 75 inch in
width at the second ridge where broadest.

The penultimate upper milk molar of the same specimen
is fully formed and consolidated. It presents a broad oblong
crown, narrow in front, but wide behind, composed of six
principal ridges, with a front and back talon. The anterior
talon and three first ridges are touched by wear, the other
three being intact. The ridges are wide apart, and the discs
of wear show thick enamel-plates. The enamel-surface,
where denuded of cement, is very rugose from deep and in-
tricate grooving, as is seen in specimens from the Crag. The
tooth bears no mark of pressure behind from an impelling
last milk molar. The dimensions are: length of crown,
2°5 in. by 1-1 in. of width at first ridge, and 1-6 in. at the fifth
ridge where broadest. The height of the crown at the fifth
ridge is also 1-6 in., the tooth thus presenting at a very early
age one of the distinctive marks of the species, namely, a
proportionally broad crown, with a low elevation to the ridges.

The original of fig. 4. of Plate IX. of the ‘Ossemens Fos-
siles ’ is also a penultimate upper milk molar, of which a part
is worn away. What remains presents five ridges and a hind
talon well worn. The ridges are wide apart, with thick
enamel plates.

The last (third) upper milk molar is seen in a detached
specimen in the Florentine Museum (marked No. 98), the
cement of which is covered with dendritic crystallizations of
manganese. It is well worn, but quite entire, showing the
anterior talon and the disc of pressure against the preceding
tooth. The crown presents eight ridges, besides a front and
back talon. All of them are more or less touched by wear,
but none confluent, except the first with its adjoining talon.

E. (LOXODON) MERIDIONALIS, 111

The three anterior ones present continuous transverse discs,
with thick unplaited enamel, the outer surface of which,
where in contact with the cement, shows a crimped edge
caused by the section intercepting the superficial grooves.
The third disc exhibits a small mesial loop in front. The
fourth and fifth ridges show each three distinct transversely
oblong discs, with about three digitations to each. The sixth
and seventh show five distinct oval or roundish discs. The
apex of the eighth ridge is barely touched, the posterior
talon being enveloped by cement. The general contour of
the crown is a broad oblong. The ridges are separated by
wide open intervals, and the enamel-plates are thick. The
dimensions of the specimen are :—

Extreme length of crown, 4°6in. Width of ditto at first ridge, 2-0in. Width at
seventh ridge, 2°5in. Height of seventh ridge, barely worn only, 2°0 in.

From these dimensions, it will be seen that the length of the
crown is less than twice the width, and that the width
exceeds the height of the seventh ridge; or, in other words,
a broad crown with low ridges, wide discs, and thick enamel.

b. Upper True Molars.—The antepenultimate (or fourth of
the entire series in the order of antero-posterior succession)
is presented im situ in a mutilated cranium of a semi-adult
and probably female Elephant, which comprises both maxil-
laries with two molars in each, and the incisive bone of the
left side with the corresponding tusk. The anterior of the
molars is the antepenultimate, the crown of which is so far
advanced in wear that the anterior ridges are ground down
into a common flat disc. There are six distinct discs of as
many ridges behind, witha talon. The enamel is very thick,
with deep grooving on the exterior surface, but scarcely any
plaiting. The digital tips of the little-worn back ridges are
thick, well separated, and they yield well-defined rings by
abrasion. It is inferred that the crown possessed eight
ridges besides the talons.

A detached left antepenultimate, entire as regards the
crown, but without fangs, shows nine ridges with a front
and back talon; the first two ridges are worn, the next
intact. It agrees with the specimens above described in the
leading characters of well-separated ridges, with thick un-
plaited enamel, and a low elevation to the plates, the dimen-
sions being :—

Length of crown, 6°2in. Width of ditto in front, 2-4in. Height of the third
ridge, 3°8 in. Height of the eighth, 3:1 in.

Another detached antepenultimate shows only eight ridges
besides a front and back talon. It has the three first ridges
barely touched by wear, showing annular discs. The enamel
is very thick and rugous, the digitations are deeply divided

112 BRITISH .AND EUROPEAN FOSSIL ELEPHANTS.

and distinct, and the ridges wide apart. The length of the
crown of this specimen is 64 inches. The other dimensions
were not taken. The number of plates in the antepenulti-
mate upper molar appears to vary from eight to nine.

Of the penultimate upper (fifth in the order of succession)
molar there are numerous noble specimens in the Florentine
Gallery. One of the left side, having the enamel tinged
black, and grey cement, shows nine principal ridges and a
front and back talon. The first four ridges only are worn ;
the digitations are thick, well-separated, and distinct, the
ridges wide apart, the crown broad, and the height low. The
dimensions of this specimen are :—

Length of crown, about 9:0 in. Width of ditto at first ridge, 3-2 in. Width of

ditto at base of fifth, 4:0in. Extreme height of fifth ridge, 5:4 in.
Tn this specimen we have an illustration of the constancy of
the distinctive characters—namely, 2 broad crown and the
height of the enamel-ridges, not much exceeding the width,
being nearly in the ratio of 11: 8.

Another detached penultimate upper molar, having the
first five ridges worn, shows ten ridges and a talon. The
digitations in this case are so distinct that the discs of each
of the first three ridges present three subordinate discs. The
dimensions of this specimen are :— :

Length of crown, 8°75 in. Width of crown at first ridge, 35 in. Height of
crown at sixth ridge, unworn, 5:2 in.

This tooth exhibits all the distinctive characters noted of the
other teeth. The number of plates in the penultimate upper
molar of EH. meridionalis appears to range from nine to ten.

Of the third or last upper true molar (sixth in the order of
succession) there are numerous specimens in the Florentine
Museum, some of them im situ in entire crania, others de-

tached. They are distinctly shown in good preservation in-

three huge male skulls, with enormous tusks, and in one
female (?) head with smaller tusks; but in each of these
cases the most anterior ridges are, from extreme age, worn
out; and I prefer drawing an illustration from a perfect
detached specimen for the exact determination of the ridge-
formula. Among the most instructive of these are a pair
belonging to opposite sides, and so much alike that they
were probably of the same individual. The molar of the
right side (No. 9,261 of the old Cat. Florent. Mus.) shows
thirteen ridges, besides talons. The discs of the first two
ridges and talon are nearly confluent into one common wide
surface ; but the presence of the large anterior fang proves
that no part of the crown is lost in front. The eight
succeeding ridges are more or less abraded, the three last

vate be.

i at

E, (LOXODON) MERIDIONALIS. 113

being intact. In consequence of the plane of advanced wear
intercepting the ridges obliquely, the enamel-plates appear
to slope a great deal where they emerge from the cement,
and the edges project much above it. The ninth ridge,
which is but slightly abraded, exhibits eight or nine distinct
thick digitations. The crown contracts a good deal towards
the hind talon, which is enveloped by a thick mass of cement.
The principal dimensions are—

Extreme length of crown, about 11:0in. Width of crown at the fourth ridge,

43 in. Width of ditto at base of ninth ridge, 4:0 in. Height of enamel-plate at.
tenth ridge, 4°5 in,
In this specimen, all the characters noticed as distinctive of
the anterior true molars are strongly marked. There are
thirteen main ridges in a length of 11 inches, being an
average of about 0:85 inch to each; and, taking the talons
into account as distinct ridges, there would still be an ave-
rage of about 0°75 inch to each ridge. The relatively low
elevation of the ridges and the very great width of the
crown are also remarkable.

The last molars present in the crania above referred to
differ in no respect from the one just described, more than
is necessarily dependent on their more advanced state of
detrition. The lower down they are ground the wider is the
expansion of each disc, and the more approximated are the
enamel-plates of the contiguous ridges. In all of them the
enamel-plates are thick, deeply channelled on the outer
surface, but hardly ever plaited, the inner edge being even
or disposed in easy flexures.

A very fine illustration of the characters of the palate and

_ two last true molars on either side is presented by the Monte

Pulgnasco specimen discovered by Cortesi, and figured by
him in the ‘Saggi Geologici,’ tab. 1, fig. 1. It was found at
no great distance from the classic cranium of Monte Zago,
upon which Cuvier founded his Rhinoceros leptorhinus, as an
extinct species devoid of any bony partition between the
nostrils. Both specimens are now preserved in the Natural
History Museum of Milan, and, by the permission of Dr.
Hmilio Cornalia, I had an opportunity of examining them
minutely. The precise identification of both is of consider-
able importance in the general argument of the Mammalian
Fauna of the Pliocene period in Europe. The skull of the
Rhinoceros is exactly as Cuvier in the first instance, and
Dr. Cornalia subsequently described it, i.e. without a trace
of an external nasal septum. The mutilated cranium of the
Hlephant is a superb fragment, comprising the maxillaries

and palate, with the penultimate and last true molars of LH.

meridionalis. The penultimate is nearly worn out, the discs
VOL. II. I

114 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

of the last four ridges being confluent in the middle, but
separated laterally by a ‘ Dedalian’ channel-macheris, show-
ing the thick unplaited lamin characteristic of the species.
The last molar presents from twelve to thirteen principal
ridges, with front and back talons. Five of these ridges are
worn, the rest are intact and enveloped by cement. The
crown is very broad; the thick digitations have their apices
worn off into circular discs, exactly as in the Val d’Arno
specimens, and the ridges are low relatively to the width of
the crown. The opposite lines of teeth converge in front.
The figure of this specimen, given by Cortesi, is very imper-
fect in execution, and inexact. Fig. 2 of the same plate, a
supposed representation of the lower jaw, is made up of two
fragments of opposite sides joined by their anterior ends,
and therefore highly deceptive. The tusks of this cranium
are of enormous dimensions, and yield an oval section with
diameters of 94 by 74 inches. Other bones of the same
skeleton are preserved in the Milan Collection, one of them
being a sacrum of immense size.

c. Lower Milk Molars.—The antepenultimate and penulti-
mate milk molars, beautifully preserved, are present in a fine
specimen of a young lower jaw of the same age as the
fragment comprising the corresponding upper teeth. The
two fragments are considered by the authorities of the
Museum to be upper and lower of the same individual, and
they agree exactly in their mineral condition and appear-
ance. On the right side the antepenultimate is wanting ;
on the left it exhibits a well-worn crown, composed of three
principal ridges with front and back talons. It is much
smaller and more compressed in front than the upper tooth,
and in the general form it is somewhat cusp-shaped, like the
corresponding tooth of the Sewalik H. (Loxod.) planifrons.

The penultimate (or second) inferior milk molar presents
Six principal ridges, besides a front and back talon; the
three anterior ones more or less worn, the next intact.
Making allowance for the difference of upper and lower, the
tooth is exactly like the corresponding penultimate above.
The plates are thick, and the ridges wide apart, the vallicular
intervals being but imperfectly covered with cement. On
the right side there are about six loose unconsolidated plates
of the third milk molar in the alveolar cavity; on the left
side only the empty alveolus. The principal dimensions of
the specimen are :—

United length of the two milk molars, 3-0 in. Length of the first, 0°7 in. Length

of the second, 2°4in. Width of ditto at first ridge, 0°38 in. Interval between the
anterior edges of the two milk molars, 1°7 in.

The last milk molar is beautifully preserved in an older

E. (LOXODON) MERIDIONALIS. 115

jaw, although still young, comprising the right ramus, with
the remains of the second milk molar in front, and the
empty alveolus of the antepenultimate true molar behind.
The age of the interposed third milk molar is therefore very
pointedly indicated. The crown presents eight principal
ridges, with front and back talons. The four anterior ridges
alone are affected by wear—the first showing two distinct,
curved and reniform discs, with the convexity in front; the
second, three continuous but separate discs; the third, four ;
the fourth barely abraded, but exhibiting the rings of five
digitations. The tooth, in its longitudinal direction, has the
usual curve, being concave on the outer side. The ridges
are wide apart, and enwrapped by an enormous layer of
cement, very much as in the young teeth of H. (Lowxod.) pla-
nifrons. In its general form, the crown differs notably from
the second milk molar, in presenting a nearly uniform width
from front to rear. There is no. indication of plication in
the enamel of the plates, nor any outlying mesial loops. The
specimen is hard and heavy, resembling in its mineral con-
dition the hard Sewalik fossils when a little weathered. The
dimensions of the tooth are :—

Length of crown, 4°6in. Width of ditto at first ridge, 1:6 in. Width of ditto
at fifth ridge, 1°8 in.

Another very fine detached specimen of the last milk
molar, lower left, with the first four ridges worn, shows also
eight principal ridges, besides talons. Posteriorly it is de-
nuded of cement; the ridges are wide apart, and the enamel
very deeply grooved, as in Crag specimens, the grooving
being strongly marked below, and disappearing towards the
apices of the ridges. Hence a corresponding difference in
the edging of the enamel-plates, according to the stage of

detrition. This specimen is very ferruginous, and might
pass for a ‘Crag’ fossil. The dimensions are :—

Length of crown, 4'7 in. Width at first ridge, 1-5in. Width at fifth ridge, 1°85
in. Height of enamel-plate at fifth ridge, 2°3 in.

Another smaller-sized detached specimen of the same
tooth shows only seven principal ridges, with front and back
talons, proving that there may be a difference of one ridge,
more or less, in the ridge-formula of the last lower milk

‘molar. Taking the data furnished by the young jaws, upper
and lower, and using « as a symbol for the talons, the ‘ ridge-

“formula’ of the milk-series in EH. (Loxod.) meridionalis is
proved to be thus :—

Milk molars.
U3x + LOL + LBL,

U3x + vbr + 28x
This specimen is also very ferruginous and ‘ Crag ’-looking.

12

116 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

d. Lower True Molars.—The antepenultimate true molar is
represented by a finely preserved detached tooth of the right
side, having the crown and fangs complete. It presents
eight ridges, with the usual talons. The large undivided
anterior fang supports two ridges, and the anterior end bears
the pit of pressure against the preceding tooth, proving the
crown to be entire. The first four ridges are worn low,
exhibiting thick enamel-plates, more or less grooved or chan-
nelled exteriorly, and thus presenting a spurious appearance
of crimping, but unplaited, on the inner border. The disc
is but slightly expanded in the middle, and without angu-
larity. The last three ridges have only the tips of the
digitations abraded, showing very distinct rings of thick
enamel, free from grooving or plaiting. The ridges are all
well in relief from the cement, and wide apart. The fangs
supporting the last four ridges are confluent into a common
shell. The principal dimensions are :—

Length of crown, 5°5 in. Width of ditto at second ridge, 2°25 in. Width of
ditto at seventh ridge, 2°6 in. Height of crown at seventh ridge, 2°3 in.

This tooth, though of larger dimensions and with a much
greater relative width than the last milk molar, retains the
same number of plates.

The same tooth is presented i situ, in a mutilated left
ramus of the lower jaw, containing the last milk molar worn
low, and the antepenultimate true molar nearly intact. Like
the specimen described, the crown is composed of eight
principal ridges, with talons. The anterior talon and three
front ridges are a little worn, the others entire. The spe-
cimen is red and ferruginous, like molars from the ‘ Crag.’
The dimensions are :—

Length of crown, 6'4in. Width of ditto at first ridge, 2-4 in. Height of fifth
ridge, 4:0 in.

The penultimate or second true molar (fifth in antero-
posterior succession, and third of the ‘ intermediate’ molars)
of the lower jaw, is well shown by a detached tooth of the
left side, the crown and fangs of which are complete. It
_ presents nine principal ridges, followed by a smaller tenth,
and a talon-splent behind, so that it is open to regard it as
having nine main ridges, with a front and complicated rear
talon, or as having ten with a small talon. All the ridges
are more or less worn. The tooth, in all its leading cha-
racters, so closely resembles the others already described,
except in the implied condition of larger size, that it is
unnecessary to describe the crown in detail. The wide
separation of the ridges, ample width of the discs of wear,
thickness of the enamel-plates, and their freedom from pli-

E. (LOXODON) MERIDIONALIS. 117

cation, are exactly as in the other teeth. In the central
discs there is a tendency to an annular expansion or loop,
which is directed backwards. In this specimen, also, the
two first ridges are supported on a single fang, and the
posterior ridge es on a Shell of confluent fangs. The dimen-
sions are :—

Length of crown, 7'8in. Width of ditto in front, 2-2in. Width of ditto atthe
seventh ridge, 3°3 in.

The same tooth is presented in numerous fragments of the
lower jaw, but, in most instances, more or less mutilated or
worn out, so as to be less adapted for a distinctive descrip-
tion in reference to the ridge-formula.

Of the third or last lower true molar detached specimens
are numerous in the Florentine Museum, besides a quantity
of lower jaws containing it i situ.

One mutilated mandibular fragment of the left ramus
contains the entire tooth, but not of the largest size, and
probably of a female. The crown presents thirteen ridges,
with a talon in front and a small talon behind. Of these, the
anterior ten ridges are more or less worn—the first three into
a continuous disc; the fourth, fifth, and sixth show a tendency
to annular expansion (an outlying denticle) in the middie,
the loop of enamel being invariably appended to the posterior
plate of enamel. The seventh, eighth, and ninth ridges have
each about six distinct roundish discs, indicating the same
number of massive digitations, with very thick enamel. The
tooth contracts very much behind. In all respects the discs
of wear agree with those of the other teeth already described.
The dimensions are :—

Extreme length of the crown, 10:25in. Width of ditto at second ridge, 2°7 in.
Width at third ridge where widest, 3°3in. Width at eighth ridge, 3:0 in.

The summits of.the worn plates of Soaey in this specimen,
are remarkable for projecting about + “ths of an inch above
the cement; and, as usual (for the reason already given), the
salient macherides recline towards the talon.

A very remarkable detached last lower molar, right side, in
which the grinding-surface is somewhat contorted (as de-
scribed of H. priscus, antea, p. 102), presents the unusual
number of fifteen ridges to the crown, with a complicated
talon. The large anterior fang and part of the first ridge are
broken off. The tooth is singular in this respect, that al-
though eleven of the ridges are worn, and the anterior ones
low down, none of the discs are confluent across, the crown
being traversed longitudinally by a deep fissure filled with
cement, dividing it into two unequal portions, two-thirds be-
longing to the inner and one-third to the outer. Throughout

118 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

the crown there is a great tendency to distinctness in the
digital processes. The cement in this specimen is in a great
measure denuded. The plates of enamel project high above
the level of the cement. The enamel is very black and highly
rugous at the sides of the crown, with transverse wavy grooves.
The dimensions are :—

Extreme length, 13:0 in. Width in front,about40in. Greatest width near the
middle, 4:3 in. Greatest height of the plates behind, only 4:9 in.

From the abnormal characters of this molar, it cannot be
safely taken for a guide as to the: ridge-formula. The dis-
tortion of the crown may account for the unusual number
of ridges. I have seen a still more remarkable case of similar
malformation in a British fossil molar of H. (Huelephas) an-
tiquus.

cee weathered or decomposed lower last molar of the
right side is entire in front, but deficient in the posterior
talon. The crown presents twelve ridges and a front talon.
. The enamel in this specimen is very thick and rugous. The
dimensions are :—

Extreme length of crown, about 11°5 in. Width of the second ridge (enamel
surface), 8°5 in, Width of the sixth, 3:7in. Height of the eighth plate, 5:0 in.

The number of ridges in the last lower molar is inferred to
vary from thirteen to fifteen. A similar but still greater va-
riation is known to occur in all the species of Huelephas, and
it in no respects throws uncertainty upon the constancy of the
ridge-formula in the other teeth.

e. Premolars.—From the close affinity of the molar teeth
in H. (Loxodon) planifrons and E. (Loxodon) meridionalis I in-
stituted a close search in the latter for the premolar teeth,
which are soremarkably developed in the former ; but I could
detect no indication of their presence. (See antea, p. 93.)

f. Ridge-formula.—Taking the data yielded by the pre-
ceding descriptions, and using # as a symbol for the talons,
the ridge-formula of the true molars in H, (Lowodon) meri-
dionalis appears to be thus :—

x8a + x(8-9)e+ 2132
w8x + 2(8-9)a + (13-15)2’

and for the whole series, milk and permanent, rejecting
talons :—

Milk molars, True molars.
34+6+8. 8+ (8-9) +13
3+64+8 8 + (8-9) +13-15°

If this be compared with the ridge-formula of EH. (Loxodon)
Africanus and H. (Loxodon) planifrons (antea, pp. 90 & 91),

E. (LOXODON) MERIDIONALIS. 119

it will at once be perceived that they agree in the hypisome-
rous character of the intermediate molars, here indicated as
distinctive of the group from Huelephas, the obvious difference
being that, besides a greater number of plates in the last true
molars, upper and lower, the cipher 8 prevails in H. meri-
dionalis, and the cipher 7 in the others. In order to show
how essentially distinct the Italian fossil species is in its
molars from H. (Huelephas) primigenius, I may anticipate the
results to be found in the sequel, so far as to contrast the
ridge-formula of the true Mammoth,! viz. :—

Milk molars. ; True molars.
4+84+12, 12+(16-18) + 24
44+8+12° 12 + (16-18) + 24-27.

For the manner in which this difference operates in modifying
the form and relative proportions of the alternate layers of
ivory, enamel, and cement, I may refer to the longitudinal and
vertical sections of the molars, represented in fig. 1 of Pl. L.,
and fig. 5, Pl. I. of the ‘ Fauna Antiqua Sivalensis,’ the former
being of the Mammoth, the latter of H. (Loxodon) planifrons,
in which the section closely resembles that of H. (Loxodon)
meridionalis. (Reproduced in Pl. V. fig. 3 and Pl. IV. fig. 2
of vol. 1.)

2. Characters of the Tusks.—In some of the crania the tusks
are preserved entire; and the specimens are sufficiently
_ abundant to furnish a correct idea of their form and direction.
In one cranial fragment, comprising the united incisive bones,
they are finely preserved in their natural position. In this
case, the extruded portions diverge for some little distance in
a straight line; they are then directed outwards, and curve
gradually upwards and inwards, so that the points are closely
approximated. When this incisive fragment is placed erect,
the included area (between the tusks) gives a truncate, ovate,
or lyrate outline, with the point towards the tips. Viewed
sidewise, they appear to be produced forwards and upwards
in a very gentle curve. On the whole, they do not differ
much in this instance from varieties seen in the existing In-
dian Elephant. In the majority of cases they diverge, and
are produced forwards and upwards in an easy curve, with the
points directed outwards, very much as in the African Ele-
phant, or in the skeleton of Mastodon Ohioticus in the British

1 You will observe that this formula | planifrons and FE. Hysudricus, but is
is very different from that of the exist- | nearest the latter, which it also closely
ing Asiatic Elephant, and of the Hweleph. | resembles in the form of the ecranium.—
antiquus (the true one) of Chartres and | Letter from Dr. F. to M. Lartet, Florence,
Grays in Essex. The HE. meridionalis | July 17, 1856. Sce also p.176, note.—
formula is intermediate between Z. | [Ep.]

120 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

Museum, figured by Owen.! They never present the pro-
nounced arc, sometimes amounting to three-fourths of a
circle, which is seen in large tusks of the Mammoth, nor the
double or spiral curve so characteristic of the latter species.
When attached to the cranium, they are often found in the ma-
trix, lying flat, and curved horizontally outwards like a sickle,
in the Theristocaulodon-fashion so grotesquely represented by
Koch in his fanciful restorations of the North American
Mastodon. This I believe to be an accident,? after the de-
composition of the soft parts, from torsion of the tusks within
their alveoli, in consequence of the excessive weight of the
extruded portions. It occurs only in the largest specimens,
and the tusks have been restored in this position in some of
the crania in the Florentine Museum. In one enormous skull,
a late acquisition, there is but a single tusk, on the right
side. On the left, the alveolus is in a great measure filled
up, but not withered, which would indicate that the left tusk
had been lost late in life. The borders of the incisive sheaths,
in this case, diverge widely apart and suddenly. In the
Indian Elephant, the tusks are sometimes broken with prodi-
gious violence in combats between savage males, and the
fracture may take place either within or outside the alveolar
sheaths. My colleague, Sir Proby Cautley, has witnessed
an accident of the kind in an Hlephant-fight at Kotah, in
Central India.

The tusks attain an enormous size, commensurate with the
colossal stature and bulk of this species. In a huge male
cranium, having the zygomatic arches entire, they measure
outside the incisive sheaths 24 inches in girth. A detached
fragment of another tusk measures about 254 inches; the
section is nearly circular. A polished frustum of another
yields upwards of 27 inches in girth, being an average
diameter of 9 inches. The section varies between round and
elliptical. In a finely preserved cranium, in which the tusks
are entire, they measure 6 feet 9 inches, including the alveolar
portion, with a diameter of 5 inches. Cuvier gives the di-
mensions of only a single Tuscan tusk, namely, 6 feet 8 inches
long. A specimen of fossil tusk from Rome, presented to the
Paris Gallery by the Duc de la Rochefoucault and M. Des-
marets, measures fully 28 inches in girth (vide Cuv. Oss. Foss.
tom. 1. p. 178). It is probably of H. meridionalis.

In varieties of one of the living species, the tusks are known
to vary so considerably in their contour and in direction that
no absolute distinctive characters can safely be founded upon

' British Fossil Mammalia, p. 298, | Warren’s ‘ Mastodon giganteus,’ Boston,
fig. 102, 185

? See the frontispiece and p. 88 of

ee ee ee eae

E. (LOXODON) MERIDIONALIS. 121
them. All that can be said of the Val d’Arno specimens is,
that they are invariably without the double or spiral curva-
ture and circular arc, with recurved points, which are so
generally observable in the tusks of the Mammoth, and that
they most resemble those of the African Elephant.

h. Oraniwm.—The characters yielded by the ridge-formula
are so pronounced, and so distinctive of H. meridionalis from
EH. primigenius, that the inquiry is immediately suggested,
‘ Are they borne out by a corresponding amount of difference
in the form of the cranium?’ The reply is in the affirmative ;
but I cannot pretend to establish this part of the case with
the precision and metrical proofs which I have endeavoured
to adduce in regard to the teeth. This duty should devolve
upon some of the anatomists or paleontologists of Italy. The
time and means of access at the disposal of a mere traveller
are unequal to the satisfactory accomplishment of a laborious
task of this nature. But it is to be hoped that the deside-
_ratum will not continue long unfulfilled. In the remarks
which follow, I shall combine the results of my own observa-
tions with those of Nesti, which I was enabled to verify at
Florence, and with the avowal that they are to be considered
more as a contribution ‘pour servir’ than as an exact or
complete description of the subject.!

The following materials, in relation to the cranium, exist
in the Museum at Florence :— .

1. A very young cranium with the lower jaw attached,
containing the earliest milk-teeth unworn. It is complete,
_ but crushed. Nesti mentions that he had seen another feetal
cranium in the possession of Count Bardi.

of the Palazzo Pitti.

1 The Grand Dukes of Tuscany have
long evinced the enlightened spirit of pa-
_ tronage of science and the liberal arts
which was bequeathed to them by the
illustrious Medici. But art-worship, and
reverence of the relics of Galileo, have
cast some branches of inquiry into the
cold shade of disregard. The Grand
Ducal Museum at Florence contains a
collection of Mammalian remains from
the Pliocene deposits of the Val d’Arno,
unrivalled in Europe both for their
abundance and for the perfect condition
in which they are preserved. Elsewhere
paleontologists are compelled to grope
their way by the faint light of mutilated
specimens; there the fossil remains of
the same forms are presented entire. A
good monograph, liberally illustrated,
upon the fossil Mammalia of the Val
dArno would reflect as bright a lustre
on the Italian diadem as do the chefs-
@euvre of the Tribune or the Galleries

The patronage of
the Court has been for centuries be-
stowed upon the wax-models of the Mu-
seum, but withheld from the magnificent
fossil remains that are laid out under
the same roof. Except a few and inade-
quate memoirs by Nesti, nothing worthy
of the subject has been brought out in
Italy upon these Tuscan collections dur-
ing the last half century; and it is not
overstating the fact to say that the pro-
gress of research on the extinct faunas
of the Upper Tertiary formation in Eu-
rope has been retarded a quarter of a
century in consequence. Had these col-
lections been yielded either by Siberia
or by the northern part of the valley of
the Po, the general results would have
been familiar knowledge long ago. At
present, a journey to Florence is the
only means of becoming acquainted
with them,

122 BRITISH AND EUROPEAN FOSSIL ELEPHANTS. .

2. Another very young cranial fragment, comprising both
maxillaries, palate, milk molars on either side, and the lower
jaw detached, in fine preservation.

3. The cranium C of Nesti’s description,! and represented
reversed in figs. 1 and 2 of his plate, copied in outline in
fies. 19 of Pls. XLII. and XLIV. of the ‘ Fauna Antiqua Si-
valensis,’ under the misnomer of EH. antiquus. It is nearly
perfect in the frontal and occipital regions, condyles, maxil-
laries, and molars, but imperfect in the facial portion, the
border of the nasal opening being broken, together with the
terminal portion of the incisive alveoli and the zygomatic
arches. Since Nesti’s figures were taken, this specimen has
suffered considerable damage, the upper lamina of the right
incisive alveolus having disappeared, together with the salient
tip of the nasals and the lateral margin including the left
orbit. The last molar is present on either side, far advanced
in wear. (See Pl. I. fig. 11, and Pl. II. fig. 16.)

4, The cranium A of Nesti’s references, fig. 3, comprising
the palatine, maxillary, and temporal regions, the inferior
part of the occiput, and the zygomatic arches, the only de-
ficiency being in the facial region. The specimen, which is
highly ferruginous, has now joined on to it the entire incisive
sheaths (not represented in Nesti’s figure) and two enormous
tusks, which are spread out horizontally in the Theristocau-
lodon-manner above noticed. Nesti, in his memoir, cites the
tusks of this specimen as yielding a diameter of 0°26", or
10:2 inches. The last molar, much worn, is present on either
side.

5. Fragments of a cranium of colossal dimensions, com-
prising, besides unjoined pieces, the maxillaries, palate, and
the last molar on either side, with the incisive bones entire,
and of enormous size. They form a plane, at the distal end,
of fully a métre in width (Nesti), or 395 inches. The incisive
alveoli diverge at their extremity, and contract very con-
siderably upwards. This is cranium B of Nesti’s references,
unfigured. ~The tusks in this specimen, according to Nesti,
are only 0°19 , or 74 inches in diameter, and the molar is
11 inches long.

6. A skull, with very old molars, and the entire incisive
sheaths, together with the tusks finely preserved in their
natural position.

7. A cranium, mutilated as regards the incisive bones and
zygomatic arches, with large tusks and much-worn molars ;
very white in its mineral condition.

1 Lettere sopra aleune ossa fossili del | nuova specie dell’ Elefante (EZ. meridion-
Val d Arno non per anco descritte sulla | alis) fossile del Val d’Arno, Pisa, 1825-6.

E. (LOXODON) MERIDIONALIS. 123

8. A cranium nearly entire, attached to the mounted trunk,
with the incisive sheaths very long, perfect, parallel, and
containing moderate-sized tusks. The upper and lower jaws
of this specimen were fixed in apposition, concealing the
crowns of the molars; and I am unable to say, with confi-
dence, that it belongs to H. (Loxod.) meridionalis.

9. A fragment, comprising the incisive alveoli, with the
perfect tusks in their natural position and of moderate di-
mensions.

Viewed from the front aspect, the head is more depressed,
and wider behind the temporal fosse, and the length of brow
from the vertex to the tip of the nasals is markedly less in EH.
meridionalis than in H.primigenius. In the latter, the frontal
region between the margins of the temporal ridges is broad ;
in the former it is much narrower, being encroached upon by
the temporal fosse. The bounding ridges sweep round by a
bold curve into the post-orbitary processes in H. meridionalis,
somewhat in the manner represented in the cranium of EH.
(Stegod.) bombifrons (Fauna Antiqua Sivalensis, Pl. XLITI.
fig. 13), in which the fronto-parietal region is much con-
stricted, while in HE. primigenius they pass into the post-
orbitary processes by a gentle sigmoid flexure. (See Pl. I.
figs. 6 and 15.)

In the Indian Elephant the posterior border of the vertex
is deeply emarginated by a re-entering sinus, corresponding
with the upper termination of the occipital fossa; in H. mert-
dionalis the line is transverse, the fossa being overarched by
a produced fold of the vertex (vide Nesti, op. cit., and Pl. I.
fig. 11).

The posterior orbitary process is very pointed and hooked;
the lachrymal tubercle is also pointed, while in E. primigenius
it is thick and prominent.

The nasals are salient, and terminate in an obtuse point;
they show no tendency to becoming lunately bifid as in the
African Elephant (PI. I. fig. 10).

The nasal aperture is situated considerably nearer the
vertex in H, meridionalis than in E. primigenius ; the bounding
margin presents a reniform outline with the cornua directed
forwards, as in the latter and in H. (Ewelephas) Hysudricus
(PL. I. fig 12).

In EH. primigenius the incisive alveoli are very much elon-
gated and parallel. The general plane of their upper surface
meets the plane of the frontal at a slight angle, from the al-
veoli being a little inflected towards the molars. This involves
a corresponding modificationin the symphysis of the mandible,
the diasteme descending nearly vertically, to terminate in a
short pointed beak. An equally remarkable elongation of

124 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

the incisive alveoli is presented by Colonel Baker’s huge cra-
nium in the British Museum, of the form named E. (Stegodon)
Ganesa in the ‘Fauna Antiqua Sivalensis.’? If the outline-
profile (Pl. IL. fig. 12) of this species bé compared with
that of the Mammoth (fig. 20), it will be seen that the plane
of the incisives in the former is continuous with that of the
frontal, with a tendency to obliquity forwards. The alveoli
are parallel in this form, as in the Mammoth.

In E. meridionalis the incisive alveoli are also much elon-
gated ; but, instead of being parallel, in all the large crania
they diverge from the sub-orbitary foramina on to their ex-
tremity, where the divergence becomes sudden and as marked
as in the African Elephant.

In the huge cranium No. 5 of the enumeration above, the
width of the incisive bones at their distal end reaches the enor-
mous spread of 394 inches (Nesti). The inter-alveolar fossa,
deep below the nasal aperture, soon becomes shallow and dis-
appears entirely near the extremity of the bones, where an
osseous plateau is interposed between the alveoli. This di-
vergence of the incisive sheaths is seen in the Florentine
specimen, represented by Cuvier in fig. 2 of Pl. IX. of the
Elephants, in the ‘Ossemens Fossiles.? In the Mammoth
they are parallel and approximated, with an interposed fossa,
throughout.

The only exception to the character here indicated that I
observed in the Museum of Florence is presented by the
cranium No. 9 of the above list, in which the tusks are com-
paratively small, indicating a female, and the specific identity
of which was not well determined. In it the incisive sheaths
are long, and, if not parallel, they are but slightly divergent,
although more dilated than in the Mammoth.

a. Lateral aspect.—When the head is rested on the plane
of the molars, and regarded sidewise, the following points
are observable :—

1. The short extent and concave arc of the surface between
the vertex and the point of the nasal bones. In LE. primi-
genius the brow is also concave; but the.curve is, gentle and
distributed over a long surface, whereas in EH. meridionalis it
is shorter and more pronounced. The concavity is much
ereater than is represented in fig. 16 of Pl. I1., copied from
Nesti’s side-view of cranium No. 3 of the list, and it is still
more pronounced in cranium No. 4, in which it approaches
the concave are presented by EH. (Huelephas) Hysudricus (fig.
17 of Pl. IL.) The upper occipital plane, as defined by
the outline of the occipital bosses, meets the frontal plane
nearly at a right angle, while the lower occipital plane joins
on with the former at an open angle, somewhat resembling

E. (LOXODON) MERIDIONALIS. 125

the profile-view of the skull of H. Africanus (fig. 15, Plate
ET).

2. According to Nesti, the plane of the zygomatic arch is
inclined to that of the molars at an angle of about 35°, while
the two planes are nearly parallel in H. primigenius. In H.
meridionalis they are also more elongated.

3. The antero-posterior extent of the temporal fossa, in
relation to its vertical height, increases progressively from H.
primgenius through H. Indicus to H. Africanus, being round
in the latter and oval in the Indian Elephant. In #. meri-
dionalis the temporal fossa has a large antero-posterior ex-
panse. According to Nesti, the proportions of length to
height are in the Indian Elephant as 37 : 44, while in #.
meridionalis they are as 16:17. The difference is still
greater when the latter is compared with H. primigenius.

4, Corresponding with these proportions, the distance
from the auditory meatus to the nasal border is greater, and
from the same point to the vertex less, in EH. meridionalis
than in the Mammoth.

5. The incisive alveoli form elongated massive cylinders
corresponding with the huge diameters of the tusks, but in-
stead of forming an angle with the frontal plane, as in JL.
primigenius, they are produced in the same plane, or with a
little outward obliquity, in H. meridionalis.

B. Occipital aspect.—The occipital face is chiefly remark-
able for two enormous bosses stretching from a little way
above the condyles up to the vertex, and leaving between
them a long and deep depression for the attachment of the
ligamentum nuche and muscles of the neck. These bosses
are continued on either side into the protuberant arches of
the parietals, that bound the temporal fosse towards the
vertex. Nesti describes them as ‘ grassi tetreedri,’? with
parallel faces where separated by the fossa, and as pointed
towards the condyles. He regarded the spacious deep fossa
as a distinctive mark from H. primigenius. But Breyne, in
his excellent description of Messrs. Schmidt’s cranium of the
Mammoth,' expressly states that there is ‘a peculiar and
very remarkable sinus of the occipital bone, deeper than an
ostrich’s egg, serving in all appearance for the insertion of
the muscles of the neck.’ These occipital bosses are distinctly
represented by two convex lines in Breyne’s profile figure, one
of which is omitted in the copy reproduced by Cuvier.? Their
development varies in the Elephants, according to the age,
sex, and size of the tusks in the individual. In some of the
Species, such as H. Namadicus and H. Hysudricus, the fossa

1 Phil. Trans. vol. xliv. for 1737-88, p. 183.
? Oss. Foss. tom. 1. Hléphans, Pl, ii. fig. 1.

126 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

terminates upwards in a deep concave notch of the vertex.
In EL. meridionalis, and also ina less degree in H. primigenius,
it is overarched by a produced lamina of the vertex. I am
unable to give any details as to the extent of the sphenoid
alze in the Italian form.

y. Basal aspect.—One of the distinctive characters of the
Mammoth, upon which Cuvier laid much stress, is the paral-
lelism of the molars in the upper jaw. In EH. meridionalis,
young and old, they invariably converge, more or less, in
front. In young specimens this convergence is very pro-
nounced ; in the worn-out molars of very old crania it is less
obvious. It is distinctively shown in the palate-specimen,
fic. 1 of Plate VI. of Cortesi’s cranium, from Monte Pul-
enasco.

The materials for comparative description of the crania of
the Elephants have been largely increased since the time of
Cuvier, and chiefly with the skulls of Indian fossil species.
The points here indicated clearly show that the cranium of
Ei. meridionalis differs more from that of the Mammoth than
does the latter from the existing Indian Elephant. The
Italian form, in this respect, resembles most the cranium of
H. Hysudricus from the Sewalik hills, and is intermediate
between it and that of the African Elephant, although widely
different from both.

i. Lower Jaw.—Much importance was attached by Cuvier
to the form of the mandible as distinctive of the Mammoth ;
and to that of H. meridionalis by Nesti. I have already ad-
verted to the error committed by the latter (pp. 42, 81, and
105) in taking the lower jaw of M. Arvernensis as the type
of his EH. meridionalis. He adhered to this opinion to the
last, notwithstanding the correction by Cuvier. The demon-
stration is so manifest that it would be unnecessary to discuss
the point again, but that De Blainville has reproduced Nesti’s
figure in the ‘Ostéographie,’ with the designation of H.
meridionalis, thus sanctioning it in some measure with his
authority.

In Mastodon Arvernensis the horizontal ramus anteriorly
bulges out with great convexity, and the symphysial beak is
projected forwards with very little inclination of the diaste-
mal ridges, and not as a continuation of the lower margin of
the ramus, which is rounded off and curved upwards to join
the beak. The latter is raised considerably above the level
of the lower margin, which is convex in the antero-posterior
direction. The beak forms a short, blunt, dilated spout, with
raised diastemal margins. On the contrary, in all the known
Elephants of the groups Loxodon and Huelephas, the beak of
the symphysis is a prolongation of the inferior margin, into

E. (LOXODON) MERIDIONALIS. 127

which the diastemal ridges descend with great obliquity ; and
it is attenuated towards the apex to terminate in an obtuse
point (vide Faun. Antiq. Sival., Pl. XITI. B. figs. 1-8). The
original of Nesti’s figure yields all these distinctive marks of
Mastodon in avery pronounced manner, and it is demon-
strable that the beak is incompatible with the ascertained
direction of the incisive bones and tusks of the upper jaw in
H. meridionalis.

Of the numerous rami of the lower jaw, young and old, of
this species in the Florentine Museum, the most perfect is an
entire mandible attached to the cranium No. 8 of the above
enumeration. ‘There are other specimens of a much larger
size. On the comparison of several, the following characters
were yielded :—

1. The teeth of the opposite sides converge in front, in-
stead of being nearly parallel, or but little inclined, as in H.
primigenius.

Much stress was laid upon this character by Cuvier in his
description of the Mammoth; but it is assuredly neither ab-
solute nor constant. In proof of this I may refer to figs. 1,
2, and 3 of Pl. XIII. A. ‘ Fauna Antiq. Sival.,’ or to fig. 1 of
Pl. I. of Fossil Remains in the ‘ Voyage of the Blossom,’ in
all of which the opposite lines of molars are more or less
convergent.

2. The length of the alveolar margin, from the anterior
edge of the ascending ramus to the commencement of the
diasteme, and the entire length of the horizontal ramus, both
absolutely and relatively to the breadth of the ascending
ramus, are greater in H. meridionalis than in EB. primigenius.

3. In the Mammoth the rami meet in front by a very ob-
tuse and rounded curve, from which a short, deflected, and
contracted beak is suddenly given off; in H. meridionalis they
unite by the curve of a flattened ellipse, and the symphysial
beak is given off by a broader base and less suddenly.

This obtuse and rounded outline in the Mammoth was
much insisted upon by Cuvier. It is constant and very dis-
tinctive of the species. The figures above cited may be re-
ferred to.

4. In H. primigenius the horizontal ramus attains a great
elevation in front, from which the diastemal ridges descend
nearly vertically, or with an abrupt inclination, into the short
beak: in H. meridionalis the ramus is longer, and proportion-
ally less elevated in front, and the diastemal margins slope
gradually into the symphysial beak from a broader base ; the
apophysis is produced more in front, and is larger in all its
dimensions than in H. primigenius. The symphysis is in
- consequence longer in H. meridionalis. In the perfect man-

128 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

dible of No. 8, the distance from the posterior surface of the
symphysis to the apex of the beak-apophysis measures 6}
inches.

5. Viewed sidewise, when the lower jaw of the last speci-
men is placed so as to rest on the posterior part of the ramus
and on the symphysis (exclusive of the beak), the inferior
margin presents a well-marked concave arc, and the beak is
produced forwards and downwards, for a considerable extent
below the plane upon which the symphysis rests. It attenu-
ates to a fine emarginate point. This concavity of the lower
border and the gradual slope of the diastemal ridges into the
beak are well seen in the young lower jaw which yielded the
description of the earliest milk molars. The latter character
is also finely exhibited by a superb British specimen from
the Elephant-bed at Happisburgh, in the Rev. John Gunn’s
rich collection at Irstead ; and in the Val d’Arno specimen in
Dr. Buckland’s collection, represented by figs. 10 and 10 a
of Pl. XIV. B. of the ‘Fauna Antiqua,’ in which, although
mutilated, the long symphysis and gradual inclination of the
diasteme are well marked. There is no good published figure
of the lower jaw of this species which can be referred to for
a visual appreciation of these differences. But an approxi-
mate idea may be had by comparing the outline of figs. 1, 2,
and 3 of Pl. XIII. A. and XIII. B. of the ‘ Fauna Antiqua
Sivalensis,’ representing different ages of H. primigenius,
with that of fig. 7, representing the lower jaw of H. Hysu-
dricus, which is allied in form to H. meridionalis ; or fig. 4
of Pl. V. in the ‘Ossemens Fossiles’ of the Mammoth, with
fio. 8 of Pl. TX., a Romagnano specimen of EH. meridionalis,
A very characteristic representation of the lower jaw of an
old Mammoth, by Scharf, is given in Buckland’s Appendix
to the ‘Voyage of the Blossom,’ fig. 1 of Pl. I. above
referred to. ,

k. Summary of the Characters.—On a review of the cha-
racters detailed in the preceding descriptions, it follows that
in all the points connected with the form of the cranium,
teeth, and lower jaw, upon which the great French anatomist
rested his distinctions among the Elephants, recent or fossil,
H. (Loxod.) meridionatis differs essentially from the Mammoth,
strictly so called. They have only two characters in common,
namely—l1st, the great width of the crowns of the molars ;
2nd, the long alveoli of the tusks. But in the former species
the height of the molar crowns is low; the ridges are cunei-
form in their vertical section and limited in number, with
thick enamel; and the incisive alveoli are divergent, with
simply curved tusks; in the latter the height of the molar
crowns is excessive, the ridges very numerous, attenuated,

E. (LOXODON) MERIDIONALIS. 129

and closely packed together, with thin unplaited enamel; and
the incisive alveoli are parallel, approximated and inflected ;
the tusks are spirally recurved. It will be seen in the sequel
that, so far from being nearly allied forms, there are several
species interposed between them.

It is no part of the design of this essay to describe the
osteography of the species more than may be subservient to
their ready discrimination when found fossil. I shall there-
fore reserve any remarks upon the peculiarities of the bones
of the trunk and extremities in the Italian form for the illus-
tration of British specimens. Bones of colossal dimensions
abound in the Museum at Florence; and Cuvier inferred from
remains in the Paris Museum that the fossil Elephant of
Monte Serbaro, here referred to H. meridionalis, attained a
height of at least fifteen feet.

B. British Specimens.—The copious details already given
regarding the dentition of this species relieve me from the
necessity of minutely describing a great variety of the
British specimens. Having the certainty, from such cumu-
lative evidence abroad, of the distinctness of the species,
it will suffice to show where the same form occurs in Eng-
land, in what. strata, under what circumstances, with what
associates, and where it is wanting. I shall refer only to
such characteristic instances as place the specific identity
of the fossils beyond question, and as are accessible for
comparison.

The finest British collection of the remains of this species
~ with which I am acquainted has been gradually accumulated
during the last thirty years by the Rev. J. Gunn, of Irstead,
from sections along the Norfolk coast. The vast abundance
in which Elephants’ teeth occur upon the ‘ Oyster-bed’ of
Happisburgh and Mundesley has been long known.’ Mr.
Gunn, favourably situated to benefit by such opportunities,
has taken advantage of his position to the full measure. The
interest and value of his collection are only equalled by the
liberality with which he makes it available for the ends of
science. I need only say in illustration that he has placed
all the specimens in his possession at my disposal, for this
essay, even to be sawn up for sections, if necessary, or for any
other use to which they could be turned. Besides a great
number of detached molars, Mr. Gunn possesses huge bones
of the extremities, an enormous pelvis, and lower jaws,

1 Periodic storms, during winter, scour | times, tears up masses of the ‘sub-
the beach and undermine the cliffs, | marine forest’ and of the ‘Elephant-
causing slips. When the detritus is | bed,’ in the latter of which the Ele-
washed away, Mammalian remains are | phantine remains occur best preserved
left in abundance upon the shore. The | and in the greatest abundance.
scouring action of the storm-wayes, at

VOL. II. K

130 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

which are only second in preservation to the Val d’Arno
specimens.

In the Norwich Museum there is also a fine series of Hle-
phant molars from the ‘ Crag’ and various points of the coast
section, including both EH. meridionalis and HE. (Huelephas)
antiquus. Therichness of the late Miss Anna Gurney’s col-
lection in Elephant remains is well known; and some very
fine specimens from the ‘Crag’ are in the possession of Mr.
Robert Fitch, of Norwich. With a single exception, up to
the present time I have not seen a fragment referable to H.
meridionalis that has not been derived either from Norfolk or
Suffolk.

a. Molars.—In the following descriptions of the.teeth I do
not consider it necessary to follow the strict order hitherto
observed of upper and lower, milk and true molars, according
to their respective succession. I shall take the most charac-
teristic specimens first.

The finest detached molar of this species that has come
under my observation is a specimen which was discovered in
the ‘ Mammaliferous Crag’ on the Thorpe road, near Norwich,
by Mr. Prestwich. The authority of so eminent and accu-
rate a geologist is a sufficient guarantee for the locality and
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 EH. antiquus. It is re-
presented, one-third of the natural size, by figs. 18 and 18 a
of Pl. XIV. B., under the misnomer already explained, of Hle- ~
phas antiquus, in the ‘ Fauna Antiqua Sivalensis.’ It is the
last true molar, lower jaw, right side, showing eleven prin-
cipal 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 trace-
able. The cement over the surface generally has been de-
composed 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 digi-
tations 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 ele-
venth, four digitations, the innermost of which is fractured.
The definition of the base of the crown behind is a little
damaged, but nothing is wanting.

E. (LOXODON) MERIDIONALIS. 131

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°8 in, Extreme height of ridges, where
the crown is broadest, 4°8 in. Width of ninth ridge, 35 in, Height of ninth
ridge, 4°6 in.
From these dimensions it is apparent that, in a length of 111 ©
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 E.
(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 EH. meridionalis are
so massive and distinct that they have occasionally been
figured and described as being of Mastodon. The specimen
now in the Norwich Museum, composed of two ridges, from
the Crag of Bramerton, described by Woodward,! is of this
nature. The enamel is very thick. I have in no case at-
tempted to express this in figures, as the plates are so ragged
and unequal that any linear measurement would be decep-
tive; but it is very obvious to the eye; and when the teeth
are sawn up and polished, their distinctness is strongly
marked. The surface of the enamel in this specimen is ex-
cessively rugous from transverse, wavy, parallel wrinkles, as
in the Italian specimens. (See Plate VIII. fig. 1.)

A Val d’Arno lower molar of the same age, from Dr. Buck-
land’s collection in the Oxford Museum, is represented, crown
side, by figs. 17 and 17 a of the same plate. The dimensions
of this specimen are :—

Length of crown, 10 in. Width of crown, 3:4 in. Height of crown, 5 in.

It presents eleven principal ridges, with front and back
talons. The English and Italian specimens agree so entirely
in their general aspect and relative proportions, that it suf-
fices to compare the figures to be convinced that they belong
to the same species, the only difference being that the latter
has the ridges divided into a greater number of digital ter-
minations—a circumstance of trivial importance, and liable
to much variation. (See Plate VIII. figs. 2 & 3.)

Tf, on the other hand, the last lower molar of EF. primi-
genius be compared with the ‘ Crag’ specimen, it will be found
to comprise, in a length of 13 inches, from twenty-four to
twenty-seven closely packed ridges, with all the dental mate-
rials attenuated, the enamel especially thin; so that when

* Mag. of Nat. History, 1836, vol. ix. p. 154, figs. 2, 3, a and 8.
K 2

132 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

sawn up vertically, the section presents an appearance closely
resembling the teeth of a comb. (See Pl. V. fig. 3 of vol. i.)

The Crag molar from the Thorpe road is so conclusive,
that, had no other specimen been met with, it would of itself
_ have sufficed to establish the existence of H. meridionalis in
the fossil state in England.

A superb right ramus of the lower jaw, in the Gunn col-
lection, dug out of the Elephant-bed between Mundesley and
Bacton, presents the penultimate and last true molars im
situ, the former half worn out and exhibiting four partly
confluent discs of wear; the latter having the first five ridges
and talon worn, the rest covered with cement, and partly
embedded in the angle of the jaw. It comprises about
thirteen ridges, exclusive of talons. The posterior ridges
are not distinctly shown, in consequence of the coat of
cement. In the penultimate, the discs of the last two ridges
are confluent by a narrow isthmus of ivory, and they exhibit
a mesial angular expansion, resembling very much that of
FE. (Loxod.) priseus. But this is simply an accident of age,
from the very low stage to which the wear of the crown has
been carried, close to the common base of ivory.

Of the last molar, the anterior talon is very broad at the
outer side, and contracts inwards. ‘The first four ridges
exhibit wide discs, bounded by an irregularly flexuous plate
of very thick enamel. The fifth ridge shows the apices of
about six very thick and distinct digitations. Between the
fourth and fifth ridges, but appended to the posterior margin
of the former, there is a single outlying mesial digitation.
The crown of this tooth is distinguished by its massive
character and width.

The dimensions are :—

Length of remains of penultimate, 3°6 in. Width of crown of penultimate,
approximatively, 3°2 in. Length of the crown of last molar, 100 in, Greatest
width of the crown at the fifth ridge, 3-9 in.

The last tooth is markedly curved in its antero-posterior
direction, the inner side being convex, the outer concave.

Another important specimen in the Gunn collection is a
detached fragment comprising the posterior half of the last —
lower molar, left side, showing seven ridges and the posterior
talon. It is inferred that from four to six anterior ridges -
with the front talon are wanting. ‘The first three ridges are
slightly worn, presenting distinct annular discs, surrounded
by a margin of thick enamel. Each of these ridges presents
about five digitations converging upwards. The specimen,
through its deficiency in front, is well adapted for showing
one of the most distinctive characters of the species, namely,
the low height of the ridges relatively to the breadth of the

r y Soden Mba
¥ Z pers +S
A

Fig. 1.

DESCRIPTION OF PLATE VIII.
ELepHas (LoxopoN) MERIDIONALIS.

Last true molar, lower jaw, right side, one-third of the natural
size. The drawing is a copy of one by Mr. Ford in Plate XIV. B,
fiz. 18, of the Fauna Antiqua Sivalensis. The specimen from
the ‘Mammaliferous Crag’ is in the Norwich Museum, and is

fully described at page 130.

Figs. 2 and 3. Views in plan and profile of a lower molar from the Val

VOL. Il.

d’Arno, in Dr. Bucklands collection, in the Oxford Museum.
The drawings are one-third of the natural size, and are copies of
those by Mr. Ford in the F. A. 8., Plate XIV.B, figs. 17 and
17a. (See page 131.)

. Crown of a last upper molar, left side, of a very old animal, in

an advanced stage of wear. The specimen is in the Norwich
Museum, and is described at page 137. The drawing, one-
third of the natural size, is copied from one by Mr. Ford in the
F. A. S., Plate XIV. B, fig. 14 a.

Blox isces2

Boe

Vol. 1. Plate

Harrison & Sons inp’

©) GiFord dal Aldous ith

"
oad
*

Flephas (Loxodon) meridionalis.

Sass

E. (LOXODON) MERIDIONALIS. 133

crown. The fracture in front passes vertically through a
ridge, exhibiting the angle of reflexion of the enamel-plate.
The extreme height of the fourth ridge of the fragment is
4-5 inches, while the extreme width of the crown is 41
inches. The height of the crown is thus seen to exceed
the width by barely half an inch. The proportions in L.
(Hueleph.) antiquus and EH. (Hueleph.) primigenius, as will be
seen in the sequel, are very different. A longitudinal section
of this specimen has been made, which exhibits very perfectly
the relative proportions of the ivory, enamel, and cement,
together with the cuneiform character of the ivory-core of
each ridge. It is highly desirable that it should be figured
of the natural size, for the guidance of English collectors in
discriminating teeth of this species.

The principal dimensions are :—

Length of the fragment, 7°5 in. Width of crown at the section, 3°8 in. Height

of front plate of enamel, 3°9 in. Width of crown at third ridge, 4:1 in. Height
of enamel-plate of fourth ridge, 4°5 in.

This specimen bears such a close resemblance to the cor-
responding tooth of the Indian fossil species, EH. (Lozod.)
planifrons, that I question if these teeth of the two forms, in
the same mineral condition, could be distinguished if found
mixed in a collection.

The Irstead collection (Gunn’s) contains numerous other
molar teeth or fragments of H. meridionalis, from Bacton,
Mundesley, Horsea, and Happisbureh, which have not been
figured ; and I, therefore, do not think it necessary to describe
them on the present occasion: some of them, it is to be
hoped, may appear shortly elsewhere.

The other illustrations of the species, to be noticed in the
sequel, are chiefly from specimens in the Norwich Museum,
which were liberally transmitted to London for identification
by the managers of that excellent institution, and are figured
in the ‘ Fauna Antiqua Sivalensis,’ Pl. XIV. B. The citations
which follow all refer to that plate, in which the figures are
drawn to one-third of the natural size.

Figs. 1 and 1 @ represent the plan and side-view of the
penultimate or second upper milk-molar of H. meridionalis.
It is a germ-specimen, 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 H. (Lowodon) 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,
14in. Height of crown at fifth ridge, 1:55 in.

134 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

The corresponding tooth of H. (Hueleph.) antiquus and of E.
primigenius yields normally eight transverse plates. The
precise origin of the specimen is not recorded; but it is sup-
posed to have belonged to Mr. Samuel Woodward, and to
have been derived from the Norfolk coast.

The specimen fig. 2 and 2a is another example of the
same tooth, a penultimate upper milk molar, right side, dis-
covered in the Norwich Crag at Easton, Suffolk, by Captain
Alexander. It presents six ridges, well advanced in wear.

The dimensions are :—

Length, 2-4 in. Width in front, 1:0in. Width behind, 1°6 in.

Figs. 3 and 3a represent another well-worn penultimate
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 4.4 represent the last milk molar of the lower
jaw, left side. The crown is worn, and comprises eight
ridges. The ends and sides of the crown are partly injured.
In mineral condition it is black and heavy, but free from
patches of marine incrustation. It is supposed by Mr.
Samuel Woodward to have been procured from the coast
(Norwich Museum).

The dimensions are :—

Length of crown, 3°9 in. Width of crown in front, 1-4 in. Width of crown at

sixth ridge, 2°0 in. Height of crown at seventh ridge, 2°1 in.
The ridge-formula in these specimens yields the same
ciphers as were 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 represent a finely preserved entire specimen
of the antepenultimate or first true molar, lower jaw, left

1Tn this and the following descrip-
tions the term ‘ marine incrustation’
means recent patches of existing Polyzoa,
two of which have been determined by
Mr. Busk to be species of Lepralia, or of
other allied forms. Their presence de-
termines the fossils to have been dredged
out of the modern sea-bottom. This is
a point of some importance in the pre-
sent case, since the Mammalian con-

tents of the ‘clay-beds’ have been so
heedlessly regarded in the geological de-
scriptions of the Norfolk coast, that there
is hardly on record a single instance
of a Mammal remain precisely referred
to any one distinct stratum above the
‘Elephant-bed’ of Gunn, although the
fossils, in many instances, bear palpable
indications of the matrix in which they
were embedded.

ihe

E. (LOXODON) MERIDIONALIS. 135

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 channeled ex-
ternally 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 are well
apart.

The dimensions are :—

Length of crown, 5-3 in. 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).

Another left lower antepenultimate true molar of a larger
individual, and more advanced in wear, is represented by
figs. 6 and 6 a. The crown presents a front talon and eight
ridges, all of them worn; the discs are wide and open, and
the vallicular interspaces are also wide ; the enamel-edges are
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 H.
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).

Figs. 7 and 7 a represent a fragment, comprising the an-
terior 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 channeling
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°3 in. 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

=

133 BRITISH AND EUROPEAN FOSSIL ELEPHANTS. ©

shown. This also belonged to Mr. 8. Woodward, and is now
in the Norwich Museum. It has all the mineral appearance
of the Mundesley and Happisburgh beds.
«Figs. 8 and 8 a represent the anterior portion of a lower
“yight molar, comprising the remains of six well-worn ridges.
It is cited to show the angular flexures that are sometimes
seen when 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.

Figs. 9 and 9q@ represent the posterior two-thirds of the
crown of a lower molar of the right side. It is inferred to
be a 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 dises are
somewhat angularly expanded, and separated by wide inter-
spaces of cement. This is best shown by the side view, fig.
9a. 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 incrustation.
It is one of Woodward’s specimens, probably from the ‘ Oyster-
bed’ (Norwich Museum).

The dimensions are :—

Length, 5°3 in. Width of crown at second ridge, 3:2 in. Width of crown at
fourth ridge, 3°1 in.

This also is a characteristic fragment of H. meridionalis.

Figs. 10 and 10a are of a specimen in Dr. Buckland’s
collection from the Val d’Arno. It is noticed 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.

Figs. 11 and 11 @ represent 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, 56 in. Width of crown in front, 2‘8in. Width of crown behind, 3-1 in.
This is also a characteristic specimen of EH. meridionalis.

Figs. 12 and 12 a represent a very notable fragment of the
posterior end of a last lower molar, comprising two discs of

E. (LOXODON) MERIDIONALIS. 137

wear and a talon. The crown is ground down low, the inter-
spaces 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°7 in. 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 spe-
cimen.

Of the upper molars, the figured specimens in Pl. XTV. B.
are less numerous; but, during the twelve years which have
elapsed since it was struck off, many specimens have been
amassed in the Norfoll collections which could furnish
complete illustrations of the upper series. I shall confine
myself to the figured specimens.

Figs. 13 and 13 a represent a mutilated fragment of a very
old molar in the collection of the British Museum (Old
Palontol. Cat. No. 7456), 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 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, 8:2 in. Width of crown, 4°3 in.
I regard it as being of H. meridionalis.

Figs. 14 and 14 a.represent 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 is
a 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
erescentic bend, the cornua being bent forwards. The plates
of enamel are very thick, and deeply channeled 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).

The dimensions are :—

Length of crown, 9°2 in. Width of crown at second remaining ridge, 3°6 in.

The antero-posterior convexity of the grinding-surface deter-
mines the tooth to be an upper molar. (See Pl. VIII. fig. 4.)

138 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

Figs. 15 and 15 a represent a very remarkable fragment 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.

The dimensions are :—

Length of the fragment, 5-4 in. Width of crown, 4-9 in.!

The same plate, XIV. B., contains a representation, fig. 16,
of an entire upper molar, comprising from sixteen to seven-
teen ridges within an extent of 11 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 E.
meridionalis.

Captain Alexander discovered in the Mammaliferous Crag
of Easton, near Southwold, a very fine specimen, of which
no figure has as yet been published, of a last upper molar,
right side, of EH. meridionalis, which I have had an oppor-
tunity of examining. The crown presented twelve principal
ridges ; the back talon was wanting. A small portion of the
tooth was broken on one side in front, but the unfractured
bend of the enamel round the opposite side proved that the
crown showed nearly its entire length. The tooth resembled
in every respect (making allowance for the difference of upper —
and lower) the specimen already described, found by Mr.
Prestwich in the Crag, near Norwich. The three first ridges
alone were touched by wear, the rest being intact. The
ridges were broad, with wide interspaces, the enamel very
thick and rugous, both from deep vertical channeling, and
from close-set, transverse, wavy wrinkles of the surface. The
digital processes were large and distinct. The ninth ridge
presented five digitations. There were no fangs. The ©
enamel-plates of the front ridges were nearly straight, and
quite free from crimping. This tooth was at once distinguish-
able from the corresponding upper molar of EH. prim-
genius or of EH. (Huelephas) antiquus, by the thickness and
low elevation of the ridges relatively to the width of the
crown.

The dimensions were :—

Length of crown, 9°6 in. Width of crown in front, 3°6 in. Height of crown at
the fourth ridge, 4-5 in. Height of crown at the penultimate ridge, 3:1 in.

The only other illustration of a molar of this species
which I shall adduce is that described and figured by

E. (LOXODON) MERIDIONALIS. 139
Parkinson,’ and reproduced in the ‘ British Fossil Mammalia,’
fig. 93, p. 239. The origin of this specimen, which is now
in the Museum of the College of Surgeons,? is not accurately
known. Parkinson states that it was purchased at the sale
of the ‘Calonnian Museum,’ by Mr. George Humphries, and
that it was said to have been found in Staffordshire. It is a last
upper molar of the left side, the crown presenting twelve
ridges and an anterior talon. The first eight ridges are worn,
the rest being enveloped by cement. The pattern of the
grinding-surface is somewhat abnormal. Interposed between
the second and third ridges there is a demi-ridge, composed
of two flattened discs, occupying only the inner half of the
interspace. The next two ridges are divided each into three
flattened annular and well-separated discs. The three last
of the exposed ridges have the apices of the digitations
barely affected by wear, but showing thick mammillary
points. Parkinson describes the tooth as differing from any
other that he had seen, the peculiarities of character being
the great thickness of the plates, the smoothness of the sides
(inner) of the line of enamel, and the appearance of the
digitated points of the plates (/.e. the interposed demi-ridge)
in the anterior part of the tooth. He adds that the width
of the plates may be taken at nearly double that of the fossil
teeth in general, and he infers that this tooth indicated a
fossil species of Elephant distinct from the Mammoth.

The dimensions are :—

Length of crown, 6°6 in. Width of crown at second ridge, 3:0 in. Greatest
width of crown at fourth ridge, 3-5 in. Length of grinding-surface in use, 5-0 in.
It will be observed that all the peculiarities which struck
Parkinson are those that are here considered characteristic
of H. meridionalis. Professor Owen has described this speci-
men carefully, and, allowing that it unquestionably offers a
great contrast to the usual form, nevertheless considers that
it exhibits the characters of the thick-plated variety of the
Mammoth simply exaggerated from the accidents of age and
attrition. The objections, founded upon teeth of the Mam-
moth, which he has raised against E. meridionalis, will be
considered with most advantage in the sequel, in the remarks

upon FL. primigenius. (See p. 148.)

___ Parkinson’s molar differs only from the ordinary character
of H. meridionalis in having the groups of digitations that
form the flattened rings more apart than usual. The inter-
calation of a demi-ridge is not uncommon in the molars of
_ fossil Elephants. This is the only ‘thick-plated’ variety

? Parkinson’s ‘ Organic Remains,’ vol. iii. p. 844, Pl. xx. fig. 6.
* Catalogue of Fossil Mammalia and Aves, p. 143, No. 599.

140 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

figured or described in the ‘ British Fossil Mammalia ;’ but
Professor Owen states that he had seen a very similar molar
of the Mammoth from the Norfolk freshwater deposits in the
collection of Mr. Fitch, of Norwich.' The authority for the
Staffordshire origin of Parkinson’s molar being unreliable,
no weight can be attributed to it as indicative of the dis-
tribution of the species over England.

b. Cranium.—No cranial fragment of H. meridionalis has
hitherto been recorded from strata in England.

ce. Lower Jaw.—A very fine lower jaw in the Irstead col-
lection has already been mentioned (antea, p. 132). It con-
sists of a right ramus, showing the whole of the body as far
as the middle of the symphysis, and the contour of the
posterior margin as high as the neck of the -condyle; the
coronoid apophysis and leafy expansion of the ala are broken
off. The greater part of the diasteme is present.

The following are the principal dimensions :—

Extreme length from the posterior margin of the ascending ramus to the broken
edge of the symphysis, 27°5in. Length of alveolar border from the anterior mar-
gin of the ascending ramus to the diasteme, 9°5in. Breadth of ascending ramus
in a line with alveolar border, 12:0 in. Height of alveolar border at outer edge
of ascending ramus, 5°7 in. Height of alveolar border in front near the diasteme,
7-7in. Length of diasteme and symphysis remaining, 6:5 in. Vertical height of
ascending ramus to neck of condyle, 12°25 in. Transverse diameter at bulge of
ramus below the coronoid apophysis, 7°2 in. Length of crown occupied by the two
molars, 14:0 in. Length of grinding-surface in use, 7°5 in. Number of plates in
use, 11.

The peculiarities distinctive of this specimen from the
lower jaw of the Mammoth are:—1, the comparatively low
elevation of the anterior end of the ramus, both absolutely
and relatively to the height at the coronoid margin; in the
Mammoth the jaw attains, in old specimens, as much as 103
to 11 inches in vertical height ; in the Irstead specimen it is
but 74 inches: 2, the long and gradual slope of the diasteme
into the beak; in the Mammoth it descends with a pitch
deviating but slightly from the vertical: 3, the long sym-
physis: 4, the greater length of the horizontal ramus in
relation to the width of the ascending ramus: 5, the less
sudden curve in the contour of the posterior angle and margin
of the ramus. The Irstead specimen differs appreciably also
from the lower jaw of EH. (Euelephas) antiquus in points
which will be noticed in the comparison of that species in
the sequel.

The Norwich Museum contains a very fine lower jaw of HE.
meridionalis, comprising both horizontal rami, and, on the
right side, part of the ascending ramus, the leaf of the ala
being broken off. The diastemal ridges are perfect, and a part

1 Catalogue of Fossil Mammalia and Ayes, p. 240.

E. (LOXODON) MERIDIONALIS. 141

of the symphysis is present; but the beak has been made up
artificially anduncouthly with plaster, and painted to simulate
the natural fossil. The last true molar is present on either
side, much worn, the anterior portion having been ground
away. There are ten discs of wear, presenting the usual
character of the species, the enamel-plates very thick and
uncrimped. The tips of the posterior ridges form well-
separated rings, and the digitations are seen to be massive.
The diastemal ridges incline with an easy slope; the outer
surface of the jaw bulges out a good deal; the height of the
ramus in front, as in the Irstead specimen, does not much
exceed the height behind under the coronoid process. This
valuable specimen was discovered in the cliff, near Mundesley,
in 1852, and presented by R. Barclay, Esq., to the Norwich
Museum. It is not stated out of what stratum it came, /.e.
whether from the ‘ Hlephant-bed,’ properly so called, or from
the ‘Laminated blue clay’ above it. It is much to be
desired that figures of these two instructive specimens should
be published. Some of the dimensions of the Norwich jaw
are as follows :—

Length of crown of left molar (last), 8-1in. Width of crown at second remain-
ing ridge, 3°0 in. Width of crown at sixth remaining ridge, 2:9in. Length of
crown occupied by six ridges, being an average of 0°77 in. to each, 4°6 in.

d. Bones of the Trunk and Hxtremities.—My remarks upon
the other bones of the skeleton will be very limited, for
several reasons. In the lacustrine and clay-deposits of the
Norfolk coast, and upon the ‘Oyster-bed’ of Happisburgh
and Mundesley, the bones and teeth of at least two of the
fossil Elephants, namely, H. (Loxodon) meridionalis and E.
(Huelephas) antiquus, occur intermixed in vast abundance.
In consequence of the prevalent belief that they were all of ©
one species, namely, the Mammoth, little attention has been
paid to the discrimination of the precise beds and divisions
of the section out of which they come, and whether from
above or below the ‘ Boulder clay.’ In no instance have the
bones of an entire skeleton been found together, and there
are no well-determined standard examples for comparison.
The identification of the species to which the bones belonged
can therefore at present be little more than approximative.
Tt will suffice to mention the principal pieces that have come
under my observation from localities in which H. meridionalis
prevails.

In Mr. Gunn’s collection at Irstead, there is an entire left
“os innominatum ’ of enormous dimensions.

[ The left side of the pelvis is nearly complete, and has the greater
part of the sacrum attached to it in the relative natural position. The
ilium is perfectly entire, from the tuberosity all along the crest back to

142 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

the sacrum. The acetabulum is complete, together with the greater
part of the pubes and ischium. The pubic portion, forming the in-
terior border of the foramen ovale, is wanting. Mr. Gunn’s specimen
is the left, which is shaded in the annexed figures. Fig. 1 represents
the pelvis, looking into the cavity; and fig. 2, the front view when
placed erect upon the pubes.

The dimensions of the specimen are as follows :—

Extreme length of ilium from the tuberosity (a) back to the sacral suture (4),
measured straight with a line and not along the curve, 48:0 in. Extreme length
of ilium along the curve from a to 4, 63°0in. Distance from @ with a cord to e,
summit of the symphysis, 34:0 in. Vertical height of symphysis pubis from c tod,
15°0 in. Width of sinus behind the acetabulum, back to sacral margin of ilium,
14-5 in. Depth from iliac crest to bottom of same sinus, 20-0 in. Distance by a
cord from ¢ top of pubes to 4 posterior end of ilium, 34-0 in. Distance from c to e,
or posterior lower spine of ilium, 20°0 in. Girth of constriction above the aceta-
bulum, 28:0 in. Girth of ischium at middle of foramen ovale, 120 in. Vertical
diameter of foramen ovale (/), 9°0 in. Transverse diameter of foramen ovale,
6-0 in. Vertical diameter of cotyloid articulating cup, 8°5in. Transverse diameter
of cotyloid articulating cup, 9°0 in. Thickness of iliac crest, 3-0 in. From upper
margin of acetabulum to summit of iliac tuberosity, 15:0 in. Length of sacrum,
rump portion, 14-0 in.

o pee

a

E. (LOXODON) MERIDIONALIS. 143

Tn consequence of the ischial portion being broken across near the
rim of the acetabulum, the exact dimensions of the cavity are not de-
terminable, but the greatest diameter is about 9 in.

In Mr. Gunn’s collection there is also the left half of another pelvis,
in three fragments, of corresponding dimensions. One fragment com-
prises the tuberosity and a portion of the ilium; and another, the in-
ferior portion of the ischium, broken off along the middle of the
foramen ovale. The outline of the acetabulum is better defined in this
specimen than in the former. |!

In the Florentine Museum there is an enormous scapula,
which has been figured by Nesti (op. cit. fig. 6), in the finest
state of preservation ; it yielded the following dimensions :—

Entire length from the coracoid process to the posterior angle, measured along
the spine, 4 ft. Transverse diameter across the spine, 3 ft. Greatest diameter
of articulating surface, 11 in.

The largest perfect humerus inthe same collection measured :—

Length 3 ft. 11 in. Transverse diameter of inferior articulating head, 1 ft. 1 in.
Girth of diameter of inferior articulating head, 2 ft. 8 in.

These dimensions are greatly surpassed by a huge humerus
in the Norwich Museum, presented by Miss Anna Gurney.
It is stated in the ‘ British Fossil Mammalia’ that it was
found in the ‘ Cliff composed of interblended blue clay and
red gravel, near the village of Bacton, in Norfolk ;’ and the
following dimensions are attributed to it :—

Entire length, 4 ft. 5 in. Circumference at the middle, 2 ft. 2°6 in.

ference at proximal end, 3 ft. 5 in. Breadth of distal end, 1 ft. 2 in.
mit of condyloid ridge to end of the outer condyle, 1 ft. 7 in.

To what species this stupendous humerus belonged has not
been exactly determined.

[In the collection of the Rev. John Gunn, at Irstead, is an enormous
left humerus of Elephas meridionalis, obtained from the Mundesley
Cliffs, uear the Paston Hill, in 1861. It is even larger than that pre-
sented to the Norwich Museum by Miss Gurney. When first found
it was entire, but in process of removal the distal and proximal ends
were broken from the shaft, so that it now consists of three fragments.
The outer surface presents a rusty colour from mineral impregnation,
and is wonderfully perfect. The dimensions exceed anything to be
seen in the collections of Europe, with the exception of the Val d’Arno
specimens of the same species in the Museum at Florence. The cha-
racters of the species are well shown by this boue, which is short and
massive as compared with the Indian elephant, or EL. primigenius. In
this respect it takes after the Tetralophodon Mastodons.

Cireumference at proximal end, 3 ft. 9 in. Circumference at distal end, 3 ft. 5 in.
Circumference at middle, 2 ft. 5 in.]?

Cireum-
From sum-

I am

1 The paragraphs in small type, within
brackets, have been put together from
entries in Dr. F.’s Note-Books, dated
June 10, 1862. The figures do not repre-
sent Mr. Gunn’s specimen, but were
copied by Dr. F. from drawings in the
*Ossemens Fossiles’ of Cuvier, to make

the measurements intelligible.

informed by Mr. Gunn that the three

fragments comprising the second spe-

cimen have been put together and depo-

sited in the Norwich Museum.—[Ep.]
? See note, p. 144.—[ Eb. ]

144 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

The largest entire femur in the collection at Florence was
4 feet 6 inches in length. The largest mentioned in the
‘British Fossil Mammalia,’ p. 254, attributed to a Mam-
moth, is stated to have been 4 feet 1 inch long.

[In Mr. Gunn’s collection there is a colossal specimen of a right
femur, which shows the shaft entire, but is mutilated at either end.
The articular head is wanting, together with the trochanter major. It
is from the forest-bed south of Bacton. Its dimensions are as follows:

Length from upper extremity where broken to lower ditto, 47 in. Girth of
lower extremity where broken, 36 in. Girth of shaft in middle, 20in. Breadth
of shaft below the base of broken trochanter, 14 in.

In the Norwich Museum are two other femora of the same size, but
incomplete—one from Happisburgh, the other from Miss Anna Gurney’s
collection. The latter is cracked and shivered like the large Norwich
Museum specimen, and resembles it so much in colour and general ap-
pearance, that it may have been of same individual. Its exact origin
is not known.

Girth of shaft, middle, 1 ft. 9 in.]!

The colossal scapula of Florence is matched by a pelvis in
the same collection, which was found entire in the Val
d’Arno ; it yielded the following dimensions :—

Expanse between tuberosities of ilium, 6 ft. 9°0in. Height of pubes at sym-
physis, 1 ft. 95 in. Transverse diameter of pelvic arch, 1 ft. 85 in. Antero-
posterior diameter of acetabulum, 7-5 in. Transverse diameter of acetabulum,
8°5 in.

VI. CHARACTERS OF HUELEPHAS.

1. General Remarks.—This group, regarded in a structural
and systematic view, is the most aberrant from the ordinary
Pachydermatous type of all the divisions of the Proboscidea,
and it is that of which the species are the most difficult to
discriminate. It is represented in the living state by the
Indian Elephant, and in the fossil state by five if not six
species at present known. The obvious manner in which
they differ from the Loxodons is, that the crown-divisions in
the molars are more numerous, elevated, and attenuated.
When the numerical values of the ridges in the successive
teeth are regarded as a series, it is manifest that they go on
augmenting by progressive increments, constituting the basis
of the technical term here applied to signify the character,
namely, an anisomerous ridge-formula, as distinguished
from the isomerous formula in the Mastodons, and the hypi-
somerous formula of the Stegodons and Loxodons.

We have seen that in three out of the four groups of the
Proboscidea already considered, each is susceptible of being

1 The paragraph within brackets has | date at which the memoir was written.
been put together from entries in Dr. F.’s | —[Ep.]
Note-Books, made subsequently to the

CHARACTERS OF EUELEPHAS. 145

divided into two subordinate series, namely, the ‘ Hurycoro-
nine,’ in which the molar crowns are broad, the ridges trans-
verse, and the valleys open ; and the ‘ Stenocoronine,’ in which
the crowns are narrow, and the valleys are obstructed by outly-
ing tubercles. These two types, under peculiar modifications,
are equally present among the forms referable to Huelephas,
and the distinctive marks upon which they are founded furnish
excellent help in determining the distinctness of the species.
They are in some respects nice in degree, but at the same
time, like all well-founded distinctions in nature, they are
very constant. In order to facilitate the determination of
the ridge-formula in the fossil forms, the characters of the
teeth in the existing species will first be considered. But it
is necessary to give some preliminary explanations of the
modifications of the dental characters in the molars of the
Kuelephants, and of the terms that are here used to express
them.

The folded crown of the molars in the groups Trilophodon,
Tetralophodon, and Stegodon is composed of three or more,
regularly or irregularly transverse, wedge-shaped cores of
ivory, arising from a common base, and covered by a shell of
enamel, which is uniformly reflected over their apices and
over the re-entering angles at their base. These divisions
are called ‘ridges’ or ‘colliculi,’ and the interstices or
valleys between them ‘vallicule:’ though usually open in
the Mastodons, the latter are in the Stegodons occupied by
an enormous mass of cement, forming reversed wedges in
relation to the ivory-cores. The layer of enamel thus alter-
nates with the ivory and cement, and, being of uniform
thickness throughout, it is the only portion of the crown
materials to which the terms ‘ plate,’ ‘lamina,’ or ‘lamella’
can with propriety be applied.

In the groups Loxodon and Huelephas these ridges go on
increasing in number, without a corresponding augmentation
of the length of the crown, so that the penultimate true
molar (or last of the intermediate series), which in the
Trilophodons has only three ridges, in the Indian Elephant
presents five times that number, or about sixteen ridges.
The law of compensation (‘ balancement’ of the French, and
‘anamorphosis’ of some German authors) comes into play
to make the necessary adjustments. The ridges are com-
pressed and close-packed, with an attenuation of the con-
stituentgivory, enamel, and cement materials ; but as there is
a limit to the lateral extension of the crown, from the
disturbance which would be thereby involved in the general
construction of the head, the ridges are attenuated and
elongated vertically, either with no increase or in an undue

VOL. Il. L

cl
146 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

proportion to the increase in the width of the crowns. But
these compressed ridges are still the homologues of the
massive divisions seen in the crowns of the molars of the
Mastodon, and, as such, it is but correct to retain the same
name for them. The obvious manner in which their elonga-
tion and compression affect the aspect of the crown is
embodied in the term ‘ coronis lamellosa,’ and the difference
of degree by the terms ‘ broad-’ and ‘ narrow-ridged,’ instead
of ‘thick-’ and ‘ thin-plated’ molars.

The mammillary divisions of the ridges in the Mastodons,
when worn, form discs, i.e. a depressed surface of ivory,
surrounded by a raised rim of enamel; and by the further
progress of wear the separate discs become confluent into
larger discs, that are either transverse or trefoil-shaped and
alternate. In Huelephas the divisions of the compressed
ridges form finger- or quill-shaped processes, which at first
are ground down into distinct ‘annular discs’; two or three
of these then become confluent into a compound oval disc;
and at length the separate oval discs run together, forming
a transverse band (‘ruban’ of Cuvier). Although it may not
be strictly logical to apply the term ‘transverse discs’ to
these narrow bands (tenize semidetrite), still they may be
regarded as very flattened ellipses; and I have found it con-
venient to use the term in this arbitrary sense in order to
maintain a uniformity of terms in designating the same object
under different modifications.

The enamel-plates furnish the most important distinctions.
1st. In regard of the thickness: in EH. primigenius they are
only half as thick as in H. meridionalis, and thinner than in
the Indian Elephant or in EH. (Huelephas) antiquus. 2nd.
Surface-characters. The inner surface, where in contact with
the ivory, is usually smooth; and the edges of the plates, in
the worn discs, are even, whether the plates are straight or
plaited. The outer surface is rugous and uneven in two
directions :—first, vertically, from parallel or divided ribs se-
parated by anastomosing channels, which are close-set and
irregular in size, and which are most marked below, disap-
pearing upon the apices of the digitations ; and secondly, trans-
versely, from parallel, wavy, contiguous, and very frequent
ruge or superficial puckering. In the vertical section these
communicate a ragged, feathered edge to the outer surface
of the plates; while the transverse section of the ribs and
channels, in the worn plates, produces a spurious appearance
of crimping, which it is important to distinguish from
plaiting or folding of the enamel upon itself. The undulated
margins caused by these alternate ribs and channels multiply
the triturating inequalities of the enamel, and they serve

CHARACTERS OF EUELEPHAS. 147

also, along with the transverse puckers, to abut the cement
firmly against the enamel-plates, and diminish its liability
to splinter during the process of trituration. This chan-
nelling is most strongly marked in the species which have
thick plates of enamel; and when the plates are denuded of
cement, the ribs between the channels simulate the appear-
ance of cords. 3rd. Flexure of the plates transversely. This
is presented under two forms: first, primary flexures, where
the plates are folded upon themselves by numerous minute
plicatures, closely applied to each other, and communicating
a continuous zigzag appearance to the worn edge of the
enamel, on both sides; this is the character to which Cuvier
applied the term ‘ festooning,’ and here called ‘ crimping’ or
‘plaiting’; second, secondary flexures, caused by the outline
of the ivory-cores upon which the enamel-plates are moulded,
and by the confluence of the discs of the separate digitations,
according to the stage of wear of the teeth.

The presence or absence of the crimping is very constant
in the different species, and very significant as a distinctive
mark. Of all the species, fossil and recent, it is most marked
in the existing Indian EHlephant,! in which the crowns of the
molars are comparatively narrow; and ordinarily it is entirely
wanting in H. primigenius, in which they are broad.? The
former belongs to the Stenocoronine type of Euelephas, the
latter to the Kurycoronine type. The effect which is brought
about in the Mastodons by the crowding of the mammille so
as to present alternate and outlying tubercles, and in the
African Elephant by the mesial rhomboidal expansion, is in
the Indian Elephant accomplished by the numerous small
plicatures of the enamel-plates. If these were unfolded, and
the plates drawn out to the extent thus gained, the molars
of the Indian species would be fully as broad, if not broader
than in the Mammoth. Both species, although differing so
importantly in these two characters of crimping and breadth
of crown, agree in oné respect—that, although presenting
more or less of secondary flexures, the discs of wear are of
nearly uniform width across; neither of them, as a general
tule, exhibits any tendency to a mesial loop or to angular
expansion; whereas in EH. (Huelephas) antiquus, which has
hitherto been so generally confounded with the Mammoth,
the molars present the threefold difference of narrow crowns,
with crimped enamel, and a certain amount of mesial rhom-
boidal expansion of the discs of wear.’ This species, in fact,
represents among the Huelephantes what the existing African
Elephant does among the Loxodons. The difference of

‘ Pl, viii. fig. 3.—[Eb.] ? Pl. viii. fig. 4-—[ Eb. ] 8 Pl. ix—[Ep.]
L 2

148 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

H. (Euelephas) antiquus from the Mammoth corresponds with
that of H. (Loxodon) Africanus from E. (Loxodon) meridionalis,
the former in each case being Stenocoronine, the latter
Hurycoronine.

Another circumstance that requires to be considered is the -
manner in which the plane of detrition modifies both the
pattern and the antero-posterior diameter of the worn discs
at different elevations. In the Mastodons, M. (Trilophodon)
Ohioticus for example, the crowns are rectangular, with only
a slight difference of height from front to back; the ridges
come successively into wear, but the plane of detrition is
nearly level in the same direction, and it makes no consider-
able angle with the vertical plane of the ridges. In the
Indian Elephant, in consequence of the large increase in the
number of ridges, the form of the crown is necessarily
modified greatly. The upper molars, instead of being rec-
taneular, are of a subtriangular and rhomboidal form, very
high in front, and falling off behind. The anterior ridges
attain in the last upper molar a height of 8 inches. In the
progress of wear, the tooth moves forward in the are of a
circle. The anterior ridges of the opposed teeth are inclined
in front, and, by their triturating action against each other,
they are worn away obliquely, and the front part of the
crown is ground down to the base before the posterior ridges
come into use. The plane of abrasion intercepts the vertical
plane of the ridges at an angle of about 60°. From this
circumstance it follows that, as the ivory-cores of the ridges,
however compressed, are wedge-shaped bodies, the discs of
wear not only necessarily become wider as they get lower,
but, from the obliquity of the plane that intercepts the
ridges, they expose, in old teeth that are used down to
the base, a broader surface than the actual width to the
ridges, measured in a straight line. From not paying due
regard to the cause, observers have been led to regard what
is in reality only an accident of advanced wear in such cases as
indicating ‘thick-plated’ varieties, and as subversive of the
specific distinction between the Mammoth and EL. meridionalis.

2. Indian Elephant.—The leading features of the dentition
of this species are so well known from the excellent descrip-
tions and figures of Corse, followed up by Cuvier, De Blain-
ville, and other comparative anatomists, and the materials
are so abundant in Kuropean collections, that I shall confine
my remarks on the present occasion chiefly to the points
which affect the determination of the ridge-formula in the
successive teeth. But it is necessary to enter with some
detail of evidence upon this part of the subject, as the results
to which I have been led differ in some important respects

E. (EUELEPHAS) INDICUS. 149

from those arrived at by previous observers, in what concerns
the ridge-characters of the intermediate molars.

a. Milk Molars.—The antepenultimate and penultimate
milk molars (m.m. 2 and m.m. 3) are seen in situ in the upper
jaw of the young cranium figured by Corse, which is now
preserved in the Museum of the India House. The antepe-
nultimate*presents four ridges, and measures but 7 inches in
length. This tooth is exceedingly rudimentary in form and
dimensions. The penultimate is composed of eight principal
ridges, with an anterior talon-ridgelet, but no posterior talon.
The eighth or last ridge is as well developed as the others,
showing eight distinct digital processes. The dimensions of
this specimen are :—length 2:4 inches, width in front ‘9 inch,
width behind 1-3 inch. The alveolus of the last milk molar,
separated from the penultimate by a partition, is present in
this specimen, but empty.

The lower jaw of the same cranium furnishes the three
milk molars in place. The antepenultimate, like the corre-
sponding upper tooth, is composed of four ridges, and measures
65 inch long. The penultimate has eight’ principal ridges,
with a small posterior talon. It is longer and narrower in
proportion than the upper ; it measures 2°55 inches in length.
Eleven germs of the ridges of the last milk molar are lying
loose in the alveolus or cavity of that tooth.

A young cranium belonging to a skeleton in the Museum
of King’s College, London, and having the lower jaw at-
tached, furnishes the next stage of dentition, namely, the
penultimate, and last milk molars, in situ. The penultimate
is much worn, the two front ridges being ground down to the
base. The crown presents eight principal ridges, with indi-
cations of an anterior talon. The discs of wear are wide, and
the enamel-border well crimped, but with no tendency to
mesial expansion. The dimensions are :—length 2-7 inches,
width of crown in front 1:2 inch, behind 1°5 inch.

The last or third milk molar, left side, has a crown com-
posed of twelve principal ridges, with a talon in front and
behind. The first ridge and unterior talon are alone worn,
the two last ridges and talon being unconsolidated and
separate. This tooth measures in length of crown 5:2 inches,
by a width in front of 1:5 inch.

In the lower jaw of the same specimen the penultimate
milk molar presents eight principal ridges in a length of
crown of 2-6inches. The last milk molar is partly embedded

in the alveolus, and the posterior portion concealed. Of the
eight emerged ridges, the four anterior are worn. The
diameter of the tusk (replaced) in this cranium is 1*1 inch.

A detached cranium, in the same Museum, furnishes the

150 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

corresponding teeth in the lower jaw, but a little older, and
with the crown fully emerged from the alveolus. The penul-
timate presents eight principal ridges, with a small talon-_
splent behind. The crown is well worn, and measures :—
length 2-4 inches, width in front ‘9 inch, width behind 1-2
inch. The last lower milk molar has a crown composed of
twelve principal ridges, with a posterior talon; the four
anterior ridges are worn, the rest being intact, and the
whole united by cement. The crown measures in length
4-5 inches, by a width in front of 1°55 inch. The cranium in
this case, although older, is of a smaller variety than that
previously described.

There are several young crania in the Museum of the Col-
lege of Surgeons yielding the same teeth. In one very im-
mature specimen (A), the antepenultimate upper is composed
of four ridges and a talon ; and the lower, of four ridges. Of
the penultimate upper and lower, each presents only seven
ridges, with front and hind talons. In another (B), which is
alittle older, the penultimate, much worn, and the last, partly
in use, are shown above and below. The penultimate upper
exhibits the remains of eight ridges; the lower is worn out.
The last upper milk molar of the same specimen and the last
lower show twelve ridges each, with a front and back talon.

Taking the data afforded by these examples and a great
many others which I have seen in different collections, the
ridge-formula of the milk molars in the Indian Elephant, ex-
clusive of talons, is ordinarily thus :—

4+84+12
448412

Tn regard to the penultimate milk molar, an exception is ad-
initted in the case of the young cranium (A), where this tooth,
both above and below, is stated to present seven ridges in
addition to front and hind talons; but the hind talons in
these cases may be regarded as last ridges. Cuvier adopted
the numbers assigned by Corse, namely, four ridges to the
tooth here designated the antepenultimate, 8 or 9 to the pe-
nultimate, and 12 or 13 to the last milk molar. But it is to
be remarked that Corse made no distinction between the
talons and the ridges proper. De Blainville, in the descrip-
tive details of these teeth, assigns to them in succession,
respectively, 4, 8, and 11 ridges to the upper, and 4, 9, and
11 to 12 ridges to the lower. Owen describes the first or ante-
penultimate as having 4 plates, the penultimate 8 or 9 plates,
and the last from 11 to 13 plates. Taking the mean of the
various numbers assigned, and making allowance for want of
precision in some of the cases in reference to the talons,

E. (EUELEPHAS) INDICUS. 151

the numbers would nearly agree with those comprised in
the above formula, which shows a progression by multiples
of 4.

b. True Molars.—The exact determination of the ridge-
formula of the true molars is embarrassed by greater diffi-
culties ; but it is a question of considerable importance, more
especially as regards the ciphers of the antepenultimate and
penultimate, in reference to the confident discrimination of
the fossil species. For if, in the living species, these teeth
should prove to be subject to any great variation in the
number of their ridges, the same might reasonably be ex-
pected to hold good in the nearly allied fossil forms, and a
reliance on the ridge-formula as a means of distinction would
not be warranted. The causes of the uncertainty are these :—
When the animal is adolescent or adult, only two at the utmost
can be present at one time, on one side of the jaw, out of the
six molar teeth developed during life ; and of these two, only
one usually is ina perfect state. If the anterior molar is in use
and complete, only a part of the posterior tooth is emerged ind
visible. If the latter is fully protruded, the greater part of the
anterior tooth will have been worn away. Itis thus impossible
ever to trace the details of the dental succession throughout,
in any one individual. Then there is a very great difference of
size between different animals of the same age. The antepe-
nultimate true molar of a large variety may be nearly as large
as the penultimate of a small one. Again, there may be a
different estimate of the number of ridges in the same tooth,
according to the manner in which different observers regard
the talons. The same last milk molar may be described by
one as having a crown composed of twelve ridges with talons,
and by another as having fourteen ridges without them.
Further, a slight amount of difference in the stage of wear
will make an upper antepenultimate present twelve distinct
ridges at one time, and only eleven when worn lower down,
in consequence of the confluence of the two anterior ridges,
exclusive of the talon, into one common disc. Cuvier, in his
remarks on the numerical determination of Corse, has ex-
pressed his belief that they are not absolute. In proof, he
cites a case observed by himself, in which the two consecutive
teeth of a lower jaw presented each fourteen ridges; while in
the corresponding upper jaw the anterior tooth had thirteen
ridges in use, and the molar in germ behind it had eighteen
ridges. With all deference to the illustrious French anato-
mist, it may fairly be asked whether in this instance the
upper and lower jaws really belonged to the same animal.
In museums it is by no means uncommon to see skulls of
Elephants fitted with mandibles that do not belong to them,

152 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

either imported thus from abroad, or having been subjected
to some accidental misplacement afterwards. A reliable in-
stance of the kind alleged, as a normal arrangement, has
never come under my observation, after the examination of a
very large number of skulls in Europe and in India. Inferring
from what is ordinarily seen in the Indian Elephant, the
teeth in the upper and lower jaws in question would be re-
garded as belonging to distinct animals of different ages.

To revert to the numerical determinations, the antepenul-
timate or first true molar is that regarding which there is the
most uncertainty. According to Corse, it consists of about
fifteen ridges. Cuvier has not specially defined the number.
De Blainville attributes to the upper antepenultimate fifteen
ridges; the lower he has not characterized. Owen describes
the tooth, in general terms, as having the crown composed of
15 or 16 plates (ridges), with a length of from 7 to 8 inches.
The result of my observation is, that although the first true
molar, in the Indian Elephant, is manifestly larger in all its
dimexsions than the last milk molar, it ordinarily repeats the
number of ridges shown by the latter. The following are
illustrations in the Museum of King’s College :—Besides the
two young crania already mentioned, there is a third, of ado-
lescent age, which contains the last milk molar and the first
true molar above and below. The third milk molar in the
upper jaw is nearly worn out; behind it the antepenultimate
true molar presents a crown composed of twelve principal
ridges, with a front and back talon. The six anterior ridges
are worn, the rest being intact. The dimensions are :—length
of crown 6°3 inches, width in front 2°1 inches, width behind 1°5
inch. In the lower jaw of the same specimen the last milk
molar is worn out; the antepenultimate true molar presents
a crown composed also of twelve principal ridges, with front
and back talons. The ten anterior ridges are worn; the discs
of wear are well crimped, and without any mesial expansion.
The dimensions are:—length of crown 6°8 inches, width in
front 2 inches ; width behind 1°5 inch. The cranium is well
marked for reference by the loss of the right tusk, the pulp-
nucleus of which had been destroyed, the third alveolus being
nearly filled up. It bears a record of having been presented
by Mr. Hammond.

In the collection of the Royal College of Surgeons there is
a young cranium, nearly of a corresponding age, in which the
same teeth are present. In the upper jaw the last milk molar
is worn down to a stump, having the indistinct remains of
about five ridges.

The antepenultimate true molar is in the middle stage of
wear. The crown presents twelve principal ridges, with front

E, (EUELEPHAS) INDICUS. 153
and back talons, making in all fourteen divisions. The five
anterior ridges and talon are worn, the rest being intact.
The dimensions are :—length of crown 6°8 inches, width of
ditto in front 2°4 inches.

In the lower jaw of the same cranium, the last milk molar
is nearly worn out; the antepenultimate true molar has a
crown composed of twelve principal ridges, with front and
back talons, the latter of which has a small splent appended
to it. The ridges may therefore be reckoned either as 12 or
13, according to the different views of observers in regard to
what ought to be considered talons. The eight anterior
ridges of the crown are in full wear. The length of the crown
is 7°8 inches.

I have now before me two very instructive detached speci-
mens, belonging to the collection of my coadjutor, Colonel
Sir Proby Cautley, and consisting of the right upper and lower
antepenultimate true molars of the same animals. They are
in the most favourable state of use for observing all the cha-
racters. ‘The upper molar has a crown composed of twelve
well-defined principal ridges, with a front and back talon.
The seven anterior ridges are in wear, presenting open trans-
verse discs with the enamel-borders strongly crimped. The
posterior talon consists of a narrow splent appended to the last
ridge. The dimensions are :—

Length of crown, 7 in. Width of ditto at second ridge, 2°5 in. Width of ditto
at eighth ridge, 2°5 in. Width of ditto at eleventh ridge, 2:-2in. Height of ditto
at seventh ridge, 6°9 in.

The corresponding tooth of the lower jaw presents a crown
also having twelve principal ridges, with a distinct front and
back talon. The nine anterior ridges are in use, the front
talon in this instance, as also in the upper tooth, being con-
fluent with the disc of the anterior ridge; the posterior talon
is a narrow splent. The discs of wear are transverse, open
and free from mesial dilatation, and the enamel-plates are well
crimped as in the upper molars. The dimensions are :—

Length of crown, 7°5 in. Width of ditto at second ridge, 2°l1in. Width of
ditto at eighth ridge, 2'4in. Height of ditto at ninth ridge, 5°6 in.

In these two specimens the character which most obviously
distinguishes the Kuelephants from the Loxodons is well
manifested, namely, the great height of the crown relatively
to the width. In the upper antepenultimate, the height of
the seventh ridge is almost equal to the length of the crown.
The dimensions of these teeth render it certain that they are
not the last milk molars.

A detached right upper antepenultimate true molar, in the
Museum of the College of Surgeons (No. 2802, Osteol. Catal.),

154 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

shows also twelve principal ridges, with front and back talons.
The dimensions of this specimen are :—

Length of crown, 6°8 in. Greatest width of crown, 2°5 in. Greatest height of

ridges, 5°5 in.
A lower jaw in the same collection (No. 2670) shows the
antepenultimate in fine preservation, presenting distinctly
twelve principal ridges, with talons. The dimensions are :—
length 6 inches, greatest width of crown 2:1 inches. As
compared with Sir Proby Cautley’s specimen, it is of small
size.

On the other hand, a perfect specimen of an upper antepe-
nultimate in the same Museum (No. 2803) shows fourteen
principal ridges, besides front and back talons. The dimen-
sions are :—

Length of crown, 64 in. Width of ditto, 2°5 in. Height at eighth ridge, 5:5 in.

Another illustration of the same kind is furnished by the
polished section of an entire upper antepenultimate, No. 2871
of the same collection. The specimen presents fourteen
principal ridges, without a posterior talon. The dimensions
are :—

Length of crown, 6°8 in. Height of ditto at the fifth ridge, 4°8 in.

In this case, if the two anterior ridges were worn somewhat
lower down, they would present but a single disc, with an
appearance of thirteen ridges to the crown. Although the
last-mentioned specimens show that the number of ridges in
the antepenultimate sometimes ranges as high as fourteen,
the other instances indicate that the prevailing cipher is 12,
or a repetition of that of the third milk molar.

The penultimate or second true molar is described by De
Blainville as beimg composed, in the upper jaw, of seventeen
ridges, and of eighteen in the lower. Owen attributes, in
general terms, to the penultimate from seventeen to twenty
ridges. Corse and Cuvier have not specially defined it. A
vertical section of an upper antepenultimate is represented in
the ‘ Fauna Antiqua Sivalensis,’ Pl. I. fig. 2 a,! composed of
seventeen ridges, with a reduced talon-splent behind, the
anterior talon being confluent with the first ridge. The di-
mensions are :—length of crown 8°5 inches, height of crown at
eighth ridge 6°2 inches. The anterior eight ridges are worn.
In the skull of a Malay Elephant in the Museum of the Royal
Asiatic Society the antepenultimate and the penultimate are
presented im situ, the former well worn, the latter in germ.
The penultimate in this case is composed of sixteen principal
ridges, with front and back talons. The typical specimen,

1 Reproduced in Pl, y, fig. 2, of vol. i—[Eb.]

E. (EUELEPHAS) INDICUS. 155

fioured and described by De Blainville,' has a crown consisting
of sixteen principal ridges, with talons. The skull, No. 2659
of the Osteol. Catal. Mus. College of Surgeons, presents the
upper penultimate on either side perfect, although partly
worn, and the empty alveoli of the germs of the last true
molar behind. The crown of the penultimate is composed of
sixteen principal ridges, with a front and back talon, of
which the eleven anterior ridges are worn. ‘The dimensions
are :—
Length of crown, 7°5 in. Width of ditto in front, 3-0 in.

Of the penultimate lower true molar, the majority of speci-
mens that I have examined have also presented sixteen prin-
cipal ridges, with talons. A fine illustration is afforded by
the left ramus of the mandible, No. 2667 of the Osteol. Catal.
Mus. Coll. of Surgeons. The inner wall of bone is removed
so as to expose the embedded crown and fangs. The penul-
timate is complete, having in front the posterior fang-alveolus
of the antepenultimate, and behind the empty cavity of the
unformed last molar. The crown presents distinctly sixteen
principal ridges, with front and back talons, the dimensions
being :—

Length of crown, 9°5 in. Width in front, 2°4 in. Greatest width, 3-0in. Height
at fifth ridge, 5:0 in.

The five anterior ridges alone are affected by wear.

This specimen is designated in the Osteological Catalogue
of the collection the last true molar; but the form and
dimensions prove it to be penultimate.

A detached penultimate left lower molar in the same
Museum, No. 2825, presents a crown composed also of sixteen
principal ridges, with front and back talons. Hleven of the
ridges are worn. The dimensions are :—

Length of crown, 9:0 in. Width of ditto at seventh ridge, 3:2in. Height of
ditto at eleventh ridge, 5-5 in.

This specimen is described in the Catalogue as the last molar,
but it presents all the characters of a penultimate.

No. 2824 of the same collection, a lower ramus, left side,
contains the antepenultimate and penultimate wm situ, the
former well worn and reduced to the discs of the eight pos-
terior ridges, the latter nearly in germ, the three anterior
ridges alone being slightly abraded. The penultimate in this
instance also presents sixteen principal ridges, with talons.

In the Ipswich Museum there is a fine specimen of a de-
tached penultimate molar of the lower jaw, left side, presented
by Mr. C. Bree, which presents sixteen ridges, besides talons,
in a length of crown of 9:5 inches. Another specimen of a

1 Ostéographie: ‘ Des Eléphants,’ tab. 7, fig. 5 ¢.

156 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

left inferior penultimate in the Museum at Taunton has a
crown composed of sixteen principal ridges, with front and
back talons. The twelve anterior ridges are worn. The di-
mensions in this case are :—

Length of crown, 11:0in. Width of ditto at third ridge, 2°3 in. Width of ditto

at eighth ridge, 3-lin. Height of ditto at twelfth ridge, 6-0 in.
It is not meant to be insisted that the cipher 16 absolutely
and constantly determines the number of ridges in the pe-
nultimate molar, upper and lower, of the Indian Elephant. I
believe that exceptional cases occur in which they range as
high as twenty in the lower penultimate in very large indi-
viduals. But, taking the great majority of instances, the
prevailing number is seen to be sixteen.

The last true molar, both in the upper and lower jaws, is
subject to a considerable difference of size in different indi-
viduals ; but it is readily distinguishable, both by the modi-
fication in the form, and by the circumstance that the ridges
constantly either attain or surpass twenty in number. Where
the crown is complete, and all the ridges are present, the
last upper molar ordinarily presents twenty-four ridges, and
the last lower about twenty-seven. 'The posterior ridges in
the upper molar are proportionally much less elevated than
in the penultimate, the crown in profile, when unworn, pre-
senting an outline that is nearly triangular, but prolonged
backwards in the last lower molar; the posterior ridges,
besides being very low, have their apices incurved upon the
crown, and they diverge towards their bases somewhat in a
fan-shaped manner ; while, in the penultimate, the ridges are
of a more uniform height from front to rear, and depart but
slightly from parallelism in their general disposition.

As examples may be cited the cranial specimen No. 2662,
Cat. Mus. Coll. of Surgeons, which contains the last upper
molar in situ, in fine preservation. On the left side the
alveolar wall is removed, to expose the tooth, which has a
crown composed of twenty-four ridges, of which only the
anterior five are worn. The dimensions are :—

Length of crown, 13°5 in. Height at the sixth ridge, 71 in. Width of ditto in

front, where greatest, 3°2 in.
Another last upper, in a more advanced stage of wear, and
yielding an excellent illustration of this tooth, is presented
by the specimen No. 566 of the Cat. of Foss. Mam. Mus.
Coll. of Surgeons. De Blainville has given a figure (Ostéo-
eraphie: ‘Des Hléphants,’ tab. 7. fig. 6) of a deformed last
upper molar, composed of about twenty-seven ridges.

Of the last lower molar in the Indian Elephant a longitu-
dinal section is represented, half the natural size, by fig. 2 b of
Pl. I. of the ‘ Fauna Antiqua Sivalensis.’ The entire length

E, (EUELEPHAS) INDICUS. 157

of the crown is about 15 inches, including in all 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 lower molar from the penultimate. De
Blainville has given, in fig. 6 of Pl. IX. of his great work, a
beautiful representation of a perfect specimen of the same
tooth, composed of twenty-seven ridges. Another very fine
example of a last lower molar is presented by the specimen
No. 557 of the Cat. of Foss. Mam. Mus. Coll. of Surgeons,
there described as being of the Mammoth, but which I regard
as being of the existing Indian Hlephant, for reasons which
will appear in the sequel. The crown is composed of about
twenty-seven ridges. In the formula given in the note p. 10
of the preceding part, the numbers assigned to the true molars
Ue rilee eae
14418 +4 24—27 °
definition of the subgenus, the increments in the intermediate
molars are expressed by 12+14418. The formula was framed
thus to embrace the range of variation in excess which is met
with in nature, and to eschew the imputation of straining
facts for a numerical harmony that certainly is not absolute.
But if the ridge-formula in this species is to be framed
upon the prevailing ciphers exhibited in a large number of
teeth, it will run so :—

in the Indian Elephant are, and in the

Milk molars. True molars.
4+8+412 12+16+ 24 }
44+8+12 12+16+ 24-27’

thus presenting two terms of progressive increments, the
one ranging from four to twelve in the milk molars, and the
other from twelve to twenty-four in the true molars, the same
cipher being common to the last milk molar and to the first
true molar, in accordance with what is seen in the other sec-
tions of the Proboscidea. This last circumstance is that in
which my observation on the succession of the molar teeth in
the existing Indian Elephant differs most from the results
arrived at by previous observers.

There is no good evidence of the existing Indian Elephant
having as yet, anywhere in India or in Hurope, been met with
in the fossil state. The specimens attributed to it by Trimmer,
Mantell, and others, are referable to H. (Huelephas) antiquus.
But undoubted fossil remains, now preserved in the British
Museum, have lately been found in America, which indicate
either a distinct species closely allied to the Indian Hlephant,
and intermediate between it and the Mammoth, or merely a

158 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

well-marked variety of the former. In either view the case
is one of high interest in its paleontological and systematic
relations. This form is provisionally designated EH. Arme-
niacus in the Synoptical Table p. 14 of the first part of this
essay. The molar teeth combine the closely approximated
and attenuated ridges of the Mammoth with the highly un-
dulated enamel-folding or ‘ crimping’ which is so character-
istic of the Indian Elephant.'

3. Elephas (Huelephas) primigenius.—In a strictly me-
thodical order, H. antiquus would follow next among the
European fossil species for description. But it will better
suit the objects of this essay first to dispose of EH. primigenius,
the Mammoth properly so called, since most of the disputed -
points involved in the question of distinct species or varieties
only of a single form turn upon the exact determination of
the characters of the Mammoth.

Whatever may have been the approximation previously
made by Merk or Blumenbach towards a distinction of the
Mammoth from the two living species, Cuvier was undoubtedly
the first to characterize the extinct species with exactness, in
his joint memoir with Geoffrey, under the name of Hlephas
Mammoth, in the year 1796.2, In the same year he read a
memoir at the first public meeting of the ‘ Institute,’ but
which was not published until 1806, in which the diagnostic
marks are very pointedly expressed under the designation of
Elephas Mammonteus : ‘ Maxilla obtusiore, lamellis molarium
tenuibus, rectis,’ as distinguished from Llephas Indicus:
‘Fronte plano-concava, lamellis molarium arcuatis, undatis.’
Cuvier connected these dental and mandibular distinctions
with others yielded by Messer Schmidt’s figure of the skull
of the Mammoth, and combined the whole in the extended
specific definition of the extinct form, which appeared in his
memoir of 1806—‘ L’Eléphant 4 crane allongé, a front concave,
a trés-longues alvéoles des défenses, 4 machoire inférieure ob-
tuse, 4 macheliéres plus larges, paralléles, marquées des rubans
plus serrés.’ He abandoned the name EH. Mammonteus of his
memoir of 1796, and adopted the designation of Hlephas pri-
migenius, proposed by Bluemenbach,’ in 1803, which is that
now generally accepted among paleontologists. To this
normal form, as already stated, Cuvier referred all the fossil
remains of Elephants found over the whole of Europe, in
Northern Asia, and. in North America, however much at
variance with the terms of his definition; and to the last he

clung to the specific unity of the ‘Eléphant fossile ’ with the

1 See Memoir in Elephas Columbi— | * Mém. de l'Institut, Ie Classe, tom. ii.
[ Ep. ] 3 Voigt’s Mag. 1803, Band vy. p. 16.

E. (EUELEPHAS) PRIMIGENIUS. - 159

jealous partiality of a discoverer for the earliest results of
his most cherished labours.

The distinctive characters in the molars of the Mammoth,
as compared with those of the existing Indian Elephant, upon
which Cuvier relied, may be expressed in the following
terms :—

1. Great narrowness or compression and approximation of
the crown-ridges, involving both a larger number in the same
length of crown and in triturating use at the same time.

2. Tenuity of, and absence of crimping in, the enamel-
plates.

3. Greater width of the molar-crowns, both absolutely and
relatively to their length.

These peculiarities, when combined, are very constant in the
Mammoth. Exceptional cases have been admitted by Cuvier,
and adduced by others; but, when closely examined, they
have proved either to belong to other extinct species or to be
disguised molars of the existing Indian Elephant.

Taking the molars of the Mammoth in succession from first
to last, they yield the descriptions which follow :—

a. Upper Milk Molars.—Of the milk molars of the upper
jaw, the antepenultimate or most anterior, from its rudimen-
tary form, appears to have been shed at a very early period, and
itis consequently but rarely observed in situ in the fossil state.
_ It is inferred to have been composed of four ridges, with
talons like the corresponding rudimentary tooth of the Indian
Hlephant.

The penultimate milk molar (or second in appearance) is
much more common, especially in cave collections. I observed
in the Taunton Museum no fewer than eight worn penul-
timates, upper and lower, in the collection formed by the
Rey. D. Williams, from the Mendip caverns. There are several
also in Mr. Beard’s collection at Banwell, and one in the col-
lection of the Geological Society, from Kent’s Hole. The
displayed part of the collection in the British Museum contains
a few examples of this tooth referable to the Mammoth, and
it exists also in the collection of the College of Surgeons.
The crown, as in the corresponding tooth of the Indian Ele-
phant, is composed of seven or eight ridges, with talons. A
fine specimen, in the Museum at Taunton, from one of the
Mendip caverns, in perfect preservation, with the fangs present
and the crown worn, presents seven principal ridges, besides
front and hind talons.

The dimensions are :—

Length of crown, 2°3in. Width of crown at second ridge, 0°9 in. Greatest
vadth behind, 1-4 in,

From the dimensions it will be seen that the crown is narrow

160 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

in front and broad behind, yielding somewhat of an ovate
outline. The specimen in the collection of the Geological.
Society, from Kent’s Hole cavern, is a penultimate upper
milk molar of the right side, with the crown much worn and
the anterior portion ground out. The discs of the six pos-
terior ridges remain.

The dimensions are :—

Length, 2°2in. Width behind, 1°3 in.

The specimen (No. 583 of the Cat. Foss. Mam.) in the
Museum of the College of Surgeons is a left upper maxillary,
containing the penultimate milk molar far advanced in wear.
The crown in this case is also much worn, presenting the
discs of six principal ridges and a hind talon. The specimen
is reputed to be from the Drift-beds at Ilford.

The last milk molar, or third in succession, of the upper
jaw of H. primigenius, abounds in English collections, both
from the caverns and from the Drift-beds. It is readily dis-
tinguished from the same tooth in the other species, fossil
or recent, by the broad squat form of the crown and the
closely approximated ridges and uncrimped enamel-plates.
A fine illustration of this tooth is presented by the Hunterian
specimen (No. 585, Cat. Fossil Mam.) in the Museum of the
College of Surgeons, from Hinton, Somersetshire. The crown
is composed of eleven principal ridges, with talons, the
anterior part being slightly worn, showing the discs of five or
six ridges; the posterior ridges are intact.

The dimensions are :—

Length of crown, 3°6 in. Width in front, 1‘5in. Greatest width, 1-8in. Height

at sixth ridge, 2°6 in.
The ridges are closely approximated, and the attenuated
layers of enamel free from crimping. In the descriptive
Catalogue (p. 140), the crown is regarded as being composed
of twelve plates, the last being here considered the posterior
talon.

Another illustration of the same tooth, a right upper, may
be cited in a British Museum specimen, No. 156 of the
Paleontol. Cat. The crown is composed of eleven principal
ridges, besides talons ; the six anterior ridges are worn.

The dimensions are :—

Length of crown, 3°7 in. Width in front, 1°4in. Greatest width, 1-9in. Height
at sixth ridge, 3-0 in.

A third illustration is afforded by a germ-specimen of a
left molar from Kent’s Hole cavern, in the Museum of the
Geological Society. The crown is composed, besides talons,
of twelve principal ridges, of which the first alone is abraded,
the rest being intact.

E, (EUELEPHAS) PRIMIGENIUS. 161

The dimensions are :—

Length of crown, 4"Gin. Greatest width at second ridge, 1‘8in. Height at
second ridge, 4:0 in.

Numerous other examples of this tooth might be cited,
presenting either eleven or twelve ridges, with talons. De
Blainville describes it as being composed of eleven, and
Professor Owen of from twelve to fourteen ridges, the talon-
plates in the latter case being probably taken into the
reckoning.

b. Lower Milk Molars.—Of the lower milk molars, the ante-
penultimate, or most anterior, is exceedingly rare in col-
lections. An illustration of it is furnished by the specimen
figured and described by Kaup, under the name of Oymato-
therium antiquum. Like the corresponding rudimentary
tooth in the Indian Elephant, it is inferred to be composed
ordinarily of few ridges. There is a specimen in the British
Museum (No. 33,403), from Mr. Layton’s collection,! which
contains the sockets of the two anterior milk molars; but
the crowns are wanting.

Of the penultimate milk molar of the lower jaw, there is
a fine specimen in the Taunton Museum, from one of the
Mendip caves, in perfect preservation, with the fangs present
and the crown worn. It is composed of seven principal
ridges, besides front and hind talons; the latter is so large
that the crown may be regarded as comprising eight prin-
cipal ridges without a hind talon; the grinding surface
presents no inequalities in the shape of raised macheerides,
the cement, ivory, and enamel being on a uniform level, as
if polished.

The dimensions are :—

ee of crown, 2°3in. Width at second dise, 0-9 in. Greatest width, behind,

1-4 in.
From these dimensions it is seen that the crown is narrow in
front and about half an inch wider behind, yielding some-
what of an ovate outline. Other illustrations might be cited,
in which the crown of the penultimate lower milk molar
presents eight ridges, besides talons.

Of the last milk molar of the lower jaw (third in the order
of appearance), a very fine example in situ is afforded by a
east of a mandible in the Museum of the College of
Surgeons. Both rami are complete, with the exception of
the articular surfaces of the condyles. The last milk molar,
well worn but perfect, is present on either side, with the

* This specimen is probably from Happisburgh, and has evidently been in the
sea

VOL. II. M

162 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

empty sockets of the penultimate in front and of the first
true molar behind.
The dimensions of the last milk molar, left side, are :—
Length of crown, 3-9 in. Width in front, 1-2 in. Greatest width behind, 1-7 in.

The crown is composed of twelve ridges, with talons closely
approximated. The original of this specimen is reputed to
have been found in the superficial deposits of the valley of
the Rhine.

Another example of the last milk molar of the lower jaw,
detached, may be cited in the specimen In the collection of
the British Museum, No. 21,315, from Ilford, Essex. The
crown is composed of twelve principal ridges, with talons, the
anterior six being worn and the rest intact; the ridges are
closely approximated, and the discs of wear form parallel
transverse bands, with no tendency to expansion in the
middle, and with the plates of enamel attenuated and free
from crimping.

The dimensions are :—

Length of crown, 3°7 in. Width of crown in front, 11in. Greatest width
behind, 1:5 in. Height at the seventh ridge, 2°3 in.

Numerous other examples might be cited; but these two
suffice to indicate the ordinary characters of the tooth.

The third milk molars in the Mammoth, upper and lower,
are distinguishable with facility from those of EH. (Loxod.)
meridionalis and from HE. (Hueleph.) antiquus by the duodenary
cipher regulating the crown-ridges, and by the tenuity of the
enamel-plates ; but the antepenultimate and penultimate are
much less easily discriminated.

1 [In the Woodwardian Museum at Cambridge is a superb fragment
of the right ramus of the lower jaw of E. primigenius, containing the
last milk molar zn situ, quite perfect. The crown exhibits the normal
number of twelve collines, of which four only are worn.

In the Museum at Turin, mixed up with specimens of molars from
St. Paolo, I found an entire lower milk-molar of EL. primigenius. It
bears a loose label as being Piedmontese, and is probably the specimen
referred to by Sir Charles Lyell as a doubtful case of an Italian £.
primigenius. The loose label possibly belongs to another specimen ;
and the specimen is in exactly the same mineral condition as another
specimen found in the same collection from Federburg (see p. 173),
which differs from that of the ordinary character of the Piedmontese
specimens; and as the latter is avowedly German, it is probable that
the same is the case with the former. So far as the certain evidence
goes, I have seen no proof hitherto that E. primigenius has been met
with anywhere fossil in Italy ?—so far, at least, as is shown by the col-

1 The paragraphs in small type within 2 Subsequently Dr. F. found satisfae-
brackets are extracted from entries made | tory evidence of the existence of Z£. pri-
in Dr. Faleoner’s Note-books,—{ Ep. ] migenius in Italy (see pp. 170, 173, and

5 241).—-[Ep.]

E. (EUELEPHAS) PRIMIGENIUS. 163

lections at Florence, Pisa, and Turin; all of which, at the first and

last places, I have most carefully examined, with a view to determine

the point (1859). Dimensions of the lower left third milk molar :—
Extreme length, 5-2 in. Width in front, 1-9 in. Width behind, 2-4in. Height

of crown, 3°5 in.]

Taking the numbers yielded by the examples above given,
it is seen that the ridge-formula of the milk molars in EH.
primigenius is identical. with that of the existing Indian
Elephant, and liable to the same variation as regards the
antepenultimate, upper and lower, as is met with in that
species, namely, the ridges varying from seveu to eight. The
A+8+12
448412
progression by successive increments of four.

e. Upper True Molars.—The circumstances which render it
difficult to determine with precision the ridge-formula of the
true molars in the existing Asiatic form apply equally to
those of the Mammoth; and, in consequence, the ciphers of

formula may be expressed thus :— , exhibiting a

_ the antepenultimate and penultimate, being the two posterior

intermediate molars, have heretofore been but vaguely ascer-
tained. Cuvier had not advanced sufficiently far in the
investigation of the subject to attempt to determine them.
De Blainville attributes from fifteen to sixteen ridges to the
antepenultimate, and eighteen or nineteen to the penulti-
mate, in the upper jaw. Owen considers the antepenulti-
mate (fourth in succession) to have been subject to consider-
able variation in the number and proportion of the ridges,
which he estimates as ranging from twelve to sixteen, the
greater number being usually in the lower molar. Of the
penultimate he describes the ridges as ranging even from
sixteen to twenty-four. Upon the examination and com-
parison of a very large number of specimens, I have been led
to the conviction that, ordinarily, the antepenultimate upper
true molar repeats the duodenary cipher of the last milk

_inolar, and that the penultimate, as in the Indian Elephant,

_ of the College of Surgeons (No. 620, Cat. Foss. Mam. p. 158)

err re

advances by an increment of four ridges.

First, in regard of the antepenultimate upper, or fourth in
the order of horizontal suecession. A very fine illustration
of this tooth in situ is presented by a specimen in the Museum
comprising the palate with a molar on either side, and in
front of it the empty fang-pits of the last milk molar which
had been shed. The crown of the antepenultimate is worn
to the last ridge, but quite perfect, and presents the discs of
twelve principal ridges, with talons; it is very broad in
relation to the length, and, when compared with the corre-
sponding tooth of the existing Indian Elephant, it looks short

M 2

164 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

and squat; the outline is nearly a parallelogram, of which
the length is less than twice the width ; the discs of wear are
closely approximated, forming narrow transverse bands; the
enamel-plates are very thin, with a slight tendency to minute
irregular undulation, nowhere amounting to crimping.

The dimensions are :—

Length of crown, 5‘lin. Width of crown in front, 2-4 in. Width of crown
behind, 2°4 in. Greatest width of crown, 2°75 in.

This specimen is of North American origin.

1[T have also examined two specimens of the first true molar, upper
jaw, of Elephas primigenius, in the collection of the Rev. 8. W. King,
from the Norfolk Coast section, near Cromer. The specimens were
carefully compared with a molar from Ilford, in the valley of the
Thames, belonging to Mr. Prestwich (see p. 165).

The specimen first to be noticed, marked No. 3, is labelled ‘ Picked
out of blue clay and black gravel on beach, solid level, after scouring
from shoot of cliff, in March 1860. (Anderson, Cromer.)’

It is a very perfect example of an antepenultimate true molar, upper
jaw, right (t.m. 1), presenting the crown quite perfect. The crown
is composed of twelve principal ridges, with front and back talons; the
nine anterior ridges are worn, the rest intact and enveloped in cement.
The most anterior ridge is confluent in its disc with the adjoining
talon disc. The second disc is confluently transverse, somewhat irre-
gularly reniform in its contour, the convexity being directed back-
wards. The third is composed of three semi-detached, oblong discs,
the worn digitations not having become quite confluent. The four or
five posterior ridges are but very slightly affected by wear. The
ground surface of the crown is broad relatively to the length of the
molar; the ridges are high, thin, and closely compressed. The plates
of enamel are decidedly thin, presenting no appearance of crimping or
primary undulation, the only plicature shown being that produced by
the confluence of the discs (i.e. secondary undulations). In all these
respects this molar bears a very close resemblance to Mr, Prestwich’s
Ilford specimen. The fangs are all broken off below, where, in the inter-
stices, the matrix is distinctly seen, in some places penetrating into the
hollow cores of the fangs. This matrix consists of a very ferruginous,
fine sand, containing small pebbles, and closely agrees with the matrix
of the ‘ Elephant-bed’ at Mundesley. There are also some patches of
blue clay, resembling that of the laminated blue beds. There is no
appearance of a disc of pressure behind, but this is intelligible from
the semi-worn condition of the tooth. The fresh broken surfaces of
the ivory of the fangs presents a dull chocolate or pale sepia colour,
like that of the Mammoth molars from the ‘ Big-bone Lick’ of
America: it burns black and yields a strong odour of ammonia, prov-
ing abundance of gelatine. All this is against the remote age of the
fossil, and would indicate that it was yielded rather by some of the

1 The paragraphs in small type within | subsequently to the date at which the
brackets have been compiled from en- | memoir was written.—[Ep.]
tries in Dr, Faleoner’s Note-books, made

E. (EVELEPHAS) PRIMIGENIUS. 165

upper gravel-beds, or blue clay below the boulder-clay, than that it
came out of the ‘ Elephant-bed.’

The dimensions are :—

Length of crown at top, 5:5 in. Length below, above the fangs, 4°55 in. Width
of crown in front, 2-4in. Width in middle, 2°28 in. Width behind, 2:1 in.
Length occupied by eight anterior discs, 2°9 in. Extreme height of crown at ninth
ridge, 5°6 in.

I regard this specimen as being an antepenultimate of Klephas pri-
migenius. N.B. The posterior talon is a little abraded behind.

The second specimen, labelled ‘No. 2, from blue clay and black
gravel, beach, Cromer; scoured down after shoot of cliff, under light-
house, March 1860. (Anderson, Cromer.)’

This specimen in a general way very closely resembles No. 3, just
described; like it, the crown is very perfect, and composed of twelve
ridges, with front and back talons. The grinding surface has extended
over the eight anterior ridges. The front ridge is confluent with that
of the dise of the anterior talon, which has partly disappeared from
pressure. This front disc has the enamel-plate surrounding it some-
what folded, but without crimping; the folds beiug secondary undula-
tions, arising from the confluence of the digitations.

The second disc is transverse, and somewhat folded in a similar
manner, but without true crimping. The third, fougth, and fifth ridges,
present each three distinct and non-confluent flattened discs. The
sixth and seventh show about five flattened annular rings. The rest of
the ridges, back to the talon, are quite entire and enveloped in cement.
There is no indication of a disc of pressure, and the talon is quite
perfect.

This tooth closely resembles that above described; it is somewhat
smaller, and belonged to the left side. It does not appear, however, to
have belonged to the same individual. The crown-discs are somewhat
wider and more open, with less appearance of compression, but not to
a greater extent than is compatible with individual variation. The
specimen agrees, in colour and character of the matrix impacted in the
fang interstices, with No. 3. The fresh ivory fracture yields the same
sepia discoloration; and when burnt it gives a strong odour of am-
monia (burnt blanket), proving abundance of gelatine. I regard this
_ specimen as also of E. primigenius.

The dimensions are :—

Extreme length of crown, 5°7 in. Extreme length above fangs, 4:4 in. Width
of crown in front, 2°1 in. Width in middle, 2°3 in. Width behind, 2:1 in. Length
occupied by eight anterior discs, 3°3 in. Extreme height of crown at eighth ridge,
5:0 in.

Although presenting the ferruginous matrix of the ‘ Elephant-bed,’
this specimen, like the former, is inferred to have been yielded by one
of the superior gravels.

Mr. Prestwich’s specimen (which is labelled ‘Ilford, 1861’) is a well-
worn penultimate true molar, upper right (t.m.2). The crown is
composed of fifteen ridges, with a posterior talon. The twelve anterior
ridges are more or less affected by wear; the discs of the two front
ridges are worn very low and confluent into an irregular, scooped de-
pression. The dise of a semi-ridge is interposed at the outer half, be-
tween this common dise and the next adjoining it. The third, fourth,

166 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

fifth, sixth, and seventh discs are transverse, narrow, and somewhat un-
dulated from secondary flexures, but with no primary undulations or
crimping of the enamel. The remaining ridges, on to the twelfth, are
but slightly affected by wear, while the last three are intact. The
ridges are compressed and closely compacted together, with thin plates
of enamel and narrow cement at intervals. ‘The crown as a whole
agrees very closely in character with the Cromer specimen, No. 3. The
cement is partly abraded from the sides; the specimen is uniformly
tinted of a ferruginous colour. The ivory burns black, yielding a
distinct smell of ammonia, and proving the presence of gelatine. The
fangs are all broken off, but the specimen yields no indication of having
been rolled.
The dimensions are :—

Length of crown, 6'4 in. Width in front, 2:7 im. Width in middle, 2°5 in.
Width behind, 2-4 in. Height of crown at twelfth ridge, 5-4 in. Length of space
occupied by eight of the anterior dises, 3:4 in.

I regard this as a characteristic specimen of E. primigenius.

P.S. The crown of Mr. Prestwich’s specimen is worn down low in
the front near to the fang; and on looking closely at the anterior end,
a smooth, highly polished, depressed surface is distinctly seen, being
the remains of the disc of pressure against the molar which preceded
it, proving beyond*question that the tooth is entire at this end, and
that the portion supported by the anterior fang is present, the fang
only being broken off or absorbed. This further proves the tooth to
be the penultimate, and not the last true molar. |

De Blainville remarks that the penultimate upper (or fifth
in the order of succession) in the Mammoth is rare in the
French collections. He was unable to include a figure of it
in the rich series of representations contained in the ‘ Ostéo-
graphie.’ In the descriptive details of the dentition (p. 189)
he cites, as a fine illustration of it, a specimen from Warsaw,
on the Vistula, having a crown still composed of eighteen or
nineteen ridges, although the most advanced of these are worn
out; and he states that the tooth was remarkable for its large
size. These circumstances throw great doubt upon the nu-
merical rank assigned to it, which is strengthened by the fact
that, in the references to the plates (p. 3857), De Blainville
mentions that he had no illustration of the penultimate except
a bad cast, and that it was therefore omitted. The Warsaw
specimen is probably a last true molar. Perfect specimens
of this tooth, furnishing the ridge-formula of the crown com-
plete, are also rare, so far as my observation goes, in English
collections, although mutilated specimens are as common as
those of the other teeth. The illustrations which I adduce
are chiefly taken from foreign specimens in the most perfect
preservation. The first is a very fine molar, in the Museum
of Darmstadt, which I was enabled to examine by the kind
permission of Dr. Kaup. It is a detached penultimate upper

bt be

E. (EUELEPHAS) PRIMIGENIUS. 167

of the left side, of the Mammoth, having the crown entire
and all the ridges present. It is composed distinctly of six-
teen principal ridges, besides a front and a back talon. The
five anterior ridges alone are affected by wear, the rest being
intact and perfect. The specimen yields all the distinctive
characters of a Mammoth’s grinder—namely, a broad crown,
very high ridges separated by narrow interstices, and atte-
nuated plates of enamel free from crimping. The dimensions
of this specimen, which was yielded by the superficial deposits
of the valley of the Rhine, are :—

; eth of crown, 8°0 in. Width of crown, 3:0in. Height of the eighth ridge,
‘26 in.

From the last measurement it will be seen that the height of
the ridges, in the middle of the tooth even, is nearly equal to
that of the length of the crown.

Another detached penultimate upper of the left side, in
the same collection, presents the crown equally perfect, and
composed of from sixteen to seventeen principal ridges, with
talons. It differs from the specimen just described in having
a proportionately broader crown, with the ridges less elevated;
the dimensions being, with a nearly equal length of crown—

Width, 3°25 in. Greatest height, 6:25 in.

In the Museum at Taunton there are two very instructive
specimens from the Mendip caverns—the one being an upper
penultimate of Hlephas antiquus, formerly in the collection of
the Rev. D. Williams, and reputed to have been procured
from Bleadon Cave; the other, a corresponding penultimate
upper of the right side of H. primigenius, of which the precise
cave locality has not been recorded. These molars are in
perfect preservation, and when put in apposition they show
well by contrast the distinctive characters of the two species.
That of the Mammoth has the crown composed distinctly of
sixteen principal ridges, besides the front and back talons;
of these the eleven anterior ridges are worn, the rest being
intact; the crown is very broad relatively to the length, and
the ridges are closely approximated, with narrow interstices;
the discs of wear form narrow transverse bands, with at-
tenuated unplaited enamel.

The dimensions are :—

Length of crown, 6°7 in. Width in front, 25 in. Width at the eighth ridge,

33 in. Height at the eleventh ridge, 5-7 in.
The length of the crown in this specimen is considerably less
than in the first Darmstadt specimen above cited; but the
difference is partly owing to the circumstance that it is in a
more advanced stage of wear, involving necessarily a reduction
in length.

168 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

I have seen no authentic specimen of an upper penultimate
of the Mammoth presenting more than sixteen or seventeen
ridges. That exceptional cases do occur, in which as many
as eighteen may be seen, is not improbable; but I believe
that, as holds in the existing Indian species, the prevailing
and normal number is sixteen. De Blainville (Ostéographie :
‘Des Eléphants,’ p. 195) describes as a penultimate upper
the cast of a molar in the collection of M. Duhamel de
Namvilliers, of which the crown presents not more than
fourteen collines; but he adds that the tooth is unusually
short, and that the ridges are thick. It is, therefore, very
questionable whether the rank which he has assigned to it as
a penultimate is correct, even if the molar belongs to the
species. Many of the specimens in the Paleontological
Gallery at Paris, which M. de Blainville has refer,ed to the
Mammoth, have been identified by me as belonging to Hlephas
antiquus and to EH. (Loxod.) meridionalis

Professor Owen has given a very beautiful representation of
an upper molar of a Mammoth from the Essex Till in figs.
91 and 92 of the ‘ British Fossil Mammalia’ (p. 237), in-
cluding both crown and side aspects. It is not specially
described in that work; but in the ‘Odontography’ he
states (p. 666) that the fifth (or penultimate), ranging in
length of crown from 8 to 11 inches, is composed of from
sixteen to twenty-four plates; and he refers to the figures
above cited as illustrations of a penultimate upper of a
Mammoth showing as many as twenty-four plates. The
specimen, judging from the figures, is of an old molar in an
advanced stage of wear; and the posterior ridges, although
of less height than is usually seen in the penultimate, are
comparatively high for a last upper molar of the Mammoth,
as that tooth is commonly met with; but the excessive
number of the ridges is, in my view, conclusive against its
being a ‘ fifth,’ and equally so in favour of its being a last
true molar deviating somewhat from the common form. De
Blainville has figured in the ‘ Ostéographie ’ (Tab. VIII. fig. 6)
a last upper molar of a Mammoth, from the Canal de ’Ouregq,
in a more advanced stage of wear, which, allowing for this
circumstance, does not differ much in form from the tooth
figured in the ‘ British Fossil Mammalia.’ !

The last true molar, upper, of H. primigenius is subject to
the same variation in the number of ridges as the corre-
sponding tooth of the existing Indian species. They range
from twenty-two to twenty-six, the prevailing number being
about twenty-four. These teeth differ also very remarkably

' Mr. Prestwich’s specimen (p. 165) is another example of upper t.m. 2 of £.
primigenius.—[ Ep. |

> aa

E. (EUELEPHAS) PRIMIGENIUS. 169

in size in different individuals; but the largest specimens
have not necessarily the greatest number of ridges, the
reverse being frequently seen. The tooth in outline resembles
that of the Indian Elephant, being triangular, very high in
front and low behind, where the last ridges gradually fall off
into an angular termination ; while in the antepenultimate
and penultimate they are usually sufficiently high behind to
communicate somewhat of arhomb-shaped form to the crowrs
in their vertical contour. Examples of this tooth are com-
mon in all great collections. A very fine illustration from
the Ohio is presented by the Hunterian specimen, a right
upper (No. 615, Cat. Foss. Mam. Coll. Surgeons), presented
by Dr. Caspar Wister, which yields all the typical characters
of the true Mammoth. The crown is broad in front, narrow
behind, and composed of twenty-six ridges, of which the an-
terior seventeen are ground down by wear. The discs of wear
form narrow transverse bands, closely compressed, with thin
unplaited macherides of enamel. The dimensions are :—

Length of crown, 12:0 in. Width of crown in front, at third ridge, 3°3 in. Greatest
width of crown, at eighth ridge, 4:0 in. Height of crown at seventeenth ridge, 5:3
in, Length of seventeen worn ridges at summit, 8°2 in.

Another fine example of this tooth, minus the fangs, is
furnished by a specimen formerly in the collection of Dr.
Mantell, and now in the Jermyn Street Museum of Practical
Geology. It is a last upper molar of the right side, bearing
a label of ‘ Sea-shore’; the crown is composed of twenty-seven
divisions, including the posterior talon, a small portion at
the anterior end being wanting, probably not more than the
anterior talon or a single ridge. The vertical outline is
triangular in a very pronounced degree, high in front, and
low, terminating in an angle behind. WHighteen ridges are
worn into narrow parallel transverse discs, free from median
expansion, and showing very attenuated enamel-plates de-
void of crimping. The posterior talon forms a narrow rudi-
mentary splent. The specimen is heavy, and tinged of a
reddish colour, like those dredged from the sea. The fresh
fracture is very adherent to the tongue.

1[In Mr. Prestwich’s collection from railway cuttings at Bedford,
there is a very odd-looking specimen of the last true molar, upper jaw,
right side, of a dwarf-sized H. primigenius. It comprises about twenty
plates, of which nine are worn.

In the Woodwardian Museum at Cambridge there is a superb
specimen of the last true molar, upper jaw, right side, of H. primigenius,
of very large size, and which bears all the marks of having died in
captivity, in the service of man, of the flint-knife period. The anterior

1 The paragraphs in small type within | sequently to the date at which the me-

brackets have been compiled from entries | moir was written.—[ Eb. |
in Dr. Faleoner’s Note-books, made sub-

170 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

part supported on the front fang is worn out, but there are about twenty-
four collines, eighteen of which are more or less worn; the hind part,
comprising about five ridges, is very much contorted and pushed on one
side, like the specimen figured in Owen’s ‘ British Fossil Mammalia.’!

In the same collection there is a fragment of an upper maxillary, con-
taining a last true molar in situ of E. primigenius, which bears all the
marks of having come out of the lochs of America, or a peat-bog in
England. It is of the left side, and there is no indication of the animal
having died in captivity.

The Woodwardian Museum also contains a superb specimen of the
last true molar, upper jaw, left side, of the pre-glacial variety of Ele-
phas primigenius, from the Norwich Coast. Dimensions :—

Extreme length of crown, 11°5 in. Width in front, 3-1 in. Greatest width, 4:1
in. Extreme height of crown, 70 in.

The summit of the crown presents about eighteen discs of wear, of
which the most anterior have been ground down to a common base of
ivory; the space occupied by fourteen of these ridges is 74 inches.
The enamel is slightly thick, but the plates are transverse and perfectly
free from any appearance of crimping. The characters of this speci-
men diverge widely from the ordinary form of E. primigenius in the
direction of the Indian Elephant, but still maintain all the distinctive
marks of true Elephas primigenius. The matrix is indisputably of the
forest-bed of the Norfolk Coast, showing in the fangs a greenish gritty
sand, full of sulphur, derived from the iron-pyrites so prevalent in the
forest-bed.

Many other specimens of the last upper molar of E. primigenius are
preserved in the Woodwardian Museum, and have been derived from
various localities, such as Chesterton, and the valley of the Danube near
Ratisbon. One fragment, comprising twelve or thirteen ridges, of
which six are worn, is of the true post-glacial character, and bears all
the marks of a gravel-matrix.

In May 1859 I carefully examined, for the second time, the only
specimen of a detached upper molar of the true E. primigenius in the
Florence collection. It consists of the last true molar, upper jaw, right
side, the posterior portion with fourteen plates, and a disc common to
two or three plates in front. All the plates except the last four are
worn. .The specimen has the cement of a reddish or chestnut colour,
part of it dislaminated. The enamel is pearly looking. The ivory of
the broken fangs is discoloured, like Siberian specimens, but is fresh-
looking. This specimen is undoubtedly of the Mammoth, but in
mineral condition and colour it differs entirely from the femur in the
same collection (see p. 144); and I suspect strongly that it is not of
the Val d’Arno. There is no exact knowledge of its origin.

In the Roman Museum (Sapienza) I found a fragment, comprising
the anterior half of the last upper molar, right side, of KH. primigenius,
in the state of germ. The anterior angle is slightly touched by wear,
but shows no characters. It comprises about twelve plates, very
straight, high, and compressed. :

Extreme length of fragment, 7-2 in. Height of ditto in front, 8-6 in. Height
of ditto at eleventh ridge, 8-0 in. Width at eleventh ridge, 3°8 in.

1 Fig. 90.—This specimen is referred to again by Dr. Falconer at p. 281—[Ep.]

E. (EVELEPHAS) PRIMIGENIUS. 171

This specimen adheres strongly to the tongue, and is covered with
Ponte Molle volcanic gravel.

In the same collection there is a fine specimen of the posterior three-
fourths of the crown of a last true molar, upper right, of #. primi-
genius, comprising the fifteen posterior plates and talon: the seven
last plates are intact, but the seven anterior are worn into transverse
discs with no expansion. There is very great grooving of the outer
surface of enamel, but no undulation.

Extreme length, 10:2 in. Height at tenth plate, 5°8 in.

It is from Monte Mario or Ponte Molle. |

d. Lower True Molars.—Of the antepenultimate (fourth in
order of appearance) a very characteristic example is fur-
nished by the Hunterian specimen No. 622 (Cat. Foss. Mam.
Mus. Coll. of Surgeons, p. 155), consisting of part of the right
ramus of the lower jaw, with one molar i situ, in perfect
preservation. The crown is composed of thirteen principal
ridges, besides front and back talon, all more or less affected
by wear. The discs form transverse narrow and closely com-
pressed bands, surrounded by thin plates of uncrimped
enamel. The outline of the summit of the crown yields a
short broad parallelogram, the length being less than twice
the greatest width, while in the corresponding tooth of the
existing Indian species the ratio is generally about three to
one. The principal dimensions are :—

Length of crown, 571 in. Width of crown in front, 2°1in. Greatest width of
crown, 2°6 in.

The specimen is labelled as being from the Ohio, and when
applied to the maxillary fragment No. 620 in the same col-
lection, containing the upper antepenultimate (described
antea, p. 163), the crown-surfaces fit so exactly, and the two
specimens agree so closely in size, relative progress of wear,
and in general appearance, that it is highly probable that
they belonged to the same individual. They both present the
black surface which is so common in the Hlephant and Mas-
todon remains from the Bone-licks of the Ohio.

Another illustration of the same tooth is seen in the young
mandible (Coll. Brit. Mus.) represented in the ‘ Fauna An-
tiqua Sivalensis,’ Pl. XIII. A. fig. 2, which contains the ante-
penultimate on both sides, well advanced in wear, but com-
plete, and the penultimate in germ behind. The crown of
the antepenultimate is composed of twelve principal ridges,
with talons, all of which, except the posterior talon, are
affected by wear ; it is broad relatively to the length, although
in a less degree than is seen in the previous specimens; the
discs of wear form closely compressed transverse bands with
attenuated plates of enamel. It is deserving of remark that

172 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

some of these plates differ from the ordinary type of the Mam-
moth in exhibiting a certain amount of irregular crimping,
but in no degree approaching that seen in the Indian Ele-
phant, the presence of this character being concurrent with
a less than the ordinary width of crown.

The dimensions of the tooth are :—

Length of crown, 5°38 in. Width in front, 1°85 in. Greatest width, 2°3 in,

In a specimen in the Museum at Turin the dimensions are:—
Length of crown, 5:2 in. Width in front, 1:9 in. Greatest width, 2°4 in.

In the Museum of Taunton, so rich in remains from the
Mendip caves, there is a finely preserved detached antepenul-
timate lower molar from ‘ Wookey-hole,’ found along with
teeth of the Siberian Rhinoceros, Cave Lion, and Hyena.
The crown, although worn to the extent of seven or eight
discs, is complete, and composed of twelve ridges, with front
and back talons; it is broad and squat-looking, with all the
usual typical characters of the Mammoth, 7.e, narrow trans-
verse discs with thin unplaited enamel.

The dimensions of this specimen are :—

Length of crown, 5°1 in, Width of crown in front, 2°3 in. Height at the eighth
ridge, 3°5 in,

Cuvier has given a representation! of a young lower jaw
discovered near Cologne.”

[In Kaup’s Museum at Darmstadt there is a fine lower jaw, left
side, of E. primigenius, from the Rhine, containing the fourth tooth, or
first true molar, entire and very characteristic, but well worn. The
penultimate is seen in germ behind it. The first tooth shows twelve
main ridges, with a front and back heel. It is much broader than the
third milk molar :—

Length of crown, 4°65 in. Width in front, 1°75in. Width at middle, 2°15 in.
Width behind, at twelfth or last ridge, 1-76 in.

As regards the penultimate lower molar of /. primigenius, there is, in
Miss Thomson’s collection at Ilford, a specimen from left side, in the
finest state of preservation, comprising sixteen ridges, of which the
anterior eight are more or less worn ; the rest intact. This specimen
bears abundance of a very dark ferruginous matrix, but not the shelly
sandy matrix which characterizes the H. antiquus bed of Ilford.

In Mr. Grantham’s collection there are some fine detached large
specimens of molars of EZ. primigenius. Among others there is a lower
jaw, right side, containing the penultimate molar, with sixteen ridges,
or thereabout. It is a most remarkable specimen in having a pro-
longed symphysial beak, such as I have never seen before.

In the Taunton Museum there is a mass of bones comprising a frag-

1 © Oss. Fossiles,’ tom. i. pl. v. fig. 5. | is compiled from extracts from Dr. F.’s
2 The remaining portion of the de- | Note-books (see page 76, note).—| Ep. ]
scription of Z. primigenius, to page 176,

E. (EUELEPHAS) PRIMIGENIUS. 173

ment of left lower jaw of EZ. primigenius, containing one perfect molar
in situ :—

Length of crown, 7'2in. Width in front, 1:9in. Width behind, not well shown,
2°7 in.

It shows distinctly sixteen principal ridges, with front and back
talons; the first six ridges worn, the rest intact. This is probably a
small kind of #. primigenius. It is evidently that species, but the tooth
is excessively small for a penultimate. Attached to the outer surface
are part of a second milk molar, and third milk molar of /. primige-
nius. The specimen is vastly broader behind than in front.

In Kaup’s Museum at Darmstadt there is a fine specimen of the
entire lower jaw of EL. primigenius, containing the penultimate true
molars on either side entire, but well worn. The sixth tooth is seen
behind it in germ. The crown of the penultimate presents sixteen
ridges and talons :—

Length of crown, 6°6 in. Width in front, 2°1 in. Width at middle, 2:8 in.
Width behind, 2°4 in.

The jaw is considerably larger than the jaw containing the antepenulti-
mate true molar already referred to (p. 172).

In the Museum at Zurich I examined a fine entire molar, probably
penultimate, from lower jaw, left side. It is supposed to have been
obtained at Canstadt. It shows fifteen plates, very thin and compressed,
with a front and back talon. Its dimensions are :—

Length of crown, 8in. Height at thirteenth plate, 5-lin. Width of crown at
seventh plate, 2°85 in.

In the collection of the Turin Museum I found, mixed up with speci-
mens from St. Paolo, two lower molars of true Eleph. primigenius, the
one probably a penultimate, the other probably a third milk molar
(see p. 162). On examining the former carefully, an old label was
found showing that the specimen came from Germany—from ‘ Feder-
burg,’ and that it was not a Piedmontese fossil. Dimensions :—

Penultimate lower, left, extreme length, 9:in. Width in front, about 2°7 in.
Width behind, 3°3 in.

It has about seventeen plates, much worn.

The Museum at Arezzo contains a detached penultimate lower molar,
left side, with the crown well worn. It shows an anterior large fang,
but about two plates are removed; there are twelve discs of wear and a
hind talon, all worn except the talon. The enamel is thin and straight,
and there is no mesial expansion. This is an undoubted typical ex-
ample of EL. primigenius, but there is no indication of its exact origin.
The bone adheres strongly to the tongue.

In the Mineralogical Gallery of the Museum at Rome there is a frag-
ment of an inferior molar, left side, comprising (in two pieces which
unite) the anterior half of the crown of a true Kleph. primigenius. The
fragment includes the ten anterior ridges and the front talon; the pos-
terior half lost by a recent fracture, Dimensions :—

Extreme length of fragment, 5:in. Length of space occupied by ten anterior
plates, 45in. Width of crown at second plate, 3:lin, Greatest width at tenth
plate, 5°7 in,

174 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

This is a most decisively marked specimen of E. primigenius, and is
either the last or the penultimate left—probably penultimate. The
talon is present in front, and the ten plates are all more or less worn ;
the first six plates are either transverse or in two discs, the outer of which
is the smaller; the eighth has three discs; the ninth, five dises. The
discs are a little flexuous, but transverse, without the least tendency to
expansion; the enamel is thin, and entirely free from the least ten-
dency to crimping. The matrix is a yellowish-grey fine sand, showing
distinct grains of pyroxene; the ivory is quite white, and adheres
strongly to the tongue. It is from Monte Sacro, near Ponte Nomentano.
Professor Ponzi has no doubt of its Roman origin.

The following notes refer to specimens of the last true molar, lower
jaw, of Elephas primigenius :—

In the Museum of the Geological Society there is a specimen of the
last true molar, lower jaw, right side, comprising about twenty-two
plates, from Walton, in Essex, presented by Henry Warburton, Esq.
The internal fracture is excessively fresh-looking, and the enamel does
not adhere to the tongue.

Mr. Gunn’s collection contains a noble specimen of E. primigenius,
dredged up at Margate. It consists of both rami of the lower jaw, the
horizontal rami perfect to the tip of the symphysis; the ascending rami
are also present and nearly perfect, with a broad leaf; the condyles
only are gone. The specimen is especially perfect on the right side.
It contains on either side the last true molar well worn. I have hardly
seen anything like it in England, so perfect. It is remarkable also in
being coloured superficially with the reddish chocolate tinge that im-
mersion in the sea gives; the anterior end is very high. It ought to
be figured.

In Mr. W. H. Newsted’s collection, Quarry House, Maidstone, is a
very fine specimen of a lower jaw, with both rami, of Elephas primi-
genius, containing on either side the last true molars quite perfect from
end to end, and in front of it the fang-pit on either side for almost 14
inch in antero-posterior diameter of the penultimate true molar, the
posterior part of which has dropped out. On the left side the hollow
alveolus is filled up with fine sandy matrix. The tooth on both sides
is composed of twenty-two ridge-plates, of which the sixteen anterior
are more or less affected by wear. The discs are very narrow, trans-
verse and parallel, after the usual plan of the /. primigenius pattern, and
the enamel macheerides as a general rule are straight, but here and there
is an occasional loop with a little flexuosity, yet nowhere any decided
crimping.

Dimensions, left side :—

Extreme length of crown, about 12‘0in. Length of summit of crown, occupied
by sixteen worn ridges, 7°0in. Greatest width in front, 3°5in. Greatest height
of plates, 5:1 in.

Giving an average of 0-54 per plate.

Memo.—The specimen has been broken in two in the line of the
symphysis, but the form is complete. This specimen deserves to be
figured. It is from the sandy gravel-pit near the church at Aylesford,
near Maidstone.

In the Woodwardian Museum at Cambridge there is a fine specimen

ee?) ae?) lee 2 ee

E, (EUELEPHAS) PRIMIGENIUS. 175

of a nearly perfect last true molar of the right, lower jaw, of Elephas
primigenius (No. 295). The enamel-plates are close, attenuated, and
bearing all the characteristics of EH. primigenius.

In Kaup’s collection at Darmstadt there is a specimen of the sixth
lower molar, right side, with twenty-three ridges, the length of the crown
being 13 inches. In the same collection there is a very fine specimen
of the entire lower jaw, from the Rhine, containing the sixth tooth,
partly worn on either side, and a portion of the fifth tooth in front.
The sixth tooth has nineteen ridges and talons, of which the first twelve
only are worn. The length of the crown is 11 inches.

In the Museum at Rome there is a lower jaw of Hlephas primigenius,
from Ponte Molle. It is most perfect on the right side, where the
coronoid process is seen, and also the edge of the diasteme and part of
the spout. The right ramus contains the last molar in situ and in
wear, but the rear portion is embedded in the alveolus, and with an
empty fang-pit in front filled with gravel. On the left side the whole
tooth is shown (the back part of the alveolus is broken), presenting
eighteen plates, and a section through the anterior large fang; which
would seem to show that the bridge portion, supported by the large
fang, is worn out. On the left side the crown shows the remains of
eighteen plates and a posterior talon, of which the thirteen anterior are
more or less worn, while the last five are intact and enveloped by ce-
ment. The discs are transverse, with not the least mesial expansion,
and only a moderate degree of undulation. The plates are more ap-
proximated than in E. antiquus, and the crown broader for its length.

Length of remaining portion of left molar, 10°in. Width of crown at fourth
ridge, 28in. Greatest width, 3°2 in.

Lastly—Plate XIII. A., fig 3, of the ‘Fauna Antiqua Sivalensis,’
represents the lower jaw of Elephas primigenius, with the last two true
molars on either side (see description, vol. i. p. 439).

“After the examination of a very large quantity of materials, I
believe the dentary formula in the Mammoth to be thus :—

Milk Molars, True Molars.
4, 8, 12 12, 16, 24
4, 8, 12 12, 16, 24

The plates advance by quaternary increments in each series, bearing in
mind that the first true molar, although of larger dimensions, commonly
repeats the number of ridges presented by the last milk molar, and
that the last true molar in all the Hlephants and Mastodons is more
composite than the others.’

Skull and other Bones.—In Kaup’s collection at Darmstadt is a fine
cranium of HL. primigenius, with only the frontal bosses broken. The
orbits and right zygomatic arch are entire; the left zygomatic arch is
nearly so. The occipital fossa is enormous, as in #. Namadicus.
There are sections of both, tusks in the alveoli; and the last true molar,
with twenty-one ridges, well worn, is present on either side.

In Mr. Swayne’s Museum at Erith, I found the greater portion of the
ulna of Z. primigenius, which is rare in English collections.

In the collection of the Rev. John Gunn, of Irstead, there is a pelvis,
probably of H. primigenius, which was found at Mundesley in the

176 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

Elephant-bed under the Paston Hill, close to the large os innominatum
of E£. meridionalis already referred to (p. 142). It is much smaller
than the corresponding bone of E. meridionalis. A comparison of the
two exhibits well the gigantic proportions of the latter. The right
ilium of EL. primigenius, where fractured, seems to be highly infiltrated
with iron. ]

4. Elephas (Euelephas) antiquus.

[The distinctive characters of the teeth of Hlephas antiquus may be
expressed in the following terms :—

1. Narrowness of the tooth in proportion to its length and height.

2. Great height of the plates. The height is more than double the
width of the crown.

3. Mesial rhomboidal expansion of the discs of wear.

4. Great crimping of the enamel-plates.

The dental formula? of £. antiquus is as follows :—

Milk Molars. True Molars.
3+6+10 10+12+16
3+6+10 10+12+16

a. Upper Milk Molars.—In the British Museum (Cat. No. 21,654)
there is a germ of a first milk molar, probably upper. It is a most
exquisite specimen, forming an unworn shell. The crown is composed
of three principal ridges, plus talons ; front talon only on one side. The
specimen is from Grays Thurrock, Essex, and was purchased from Mr.
Ball :—

Length of crown, 0°9 in. Width in front, 0°5in. Width behind, 0°72 in,

In the York Natural History Museum, among the Kirkdale Cavern
remains, there is a superb specimen, with fangs, of the second milk molar,
upper left, of Eleph. antiquus, entire at both ends, but well worn.
The crown presents only six principal ridges, besides talons; dise of
anterior talon partly confluent with first ridge. The first ridge has two
distinct discs; the second, third, and fourth discs are very expanded
and highly crimped ; fifth ridge little worn—crimped ; in sixth ridge,
tips of the digitations are barely touched. Posterior talon immersed in
cement. The crown is very broad, but this is partially a distortion, from
an attempted restoration with a blue clay cement in front, where the

1 The entire description of Elephas antiquus has been compiled from entries in -

Dr. Faleoner’s Note-books.—| Ep. |
2 Dental Formula of Proboscidea, extracted from Dr. Falconer’s Note-book for
August 25, 1862 :—

Milk True

Dinotherium . é - 142+ 3 38+ 2+ 2 = 13
Trilophodon Ohioticus . 1+2+ 3 3+ 3+ 4 = 16
Tetralophodon Arvernensis 2+38+ 4 4+ 44+ 6 oe
Pentalophodon . : . 8444+ 3 5+ 5+ 6 = 28
Stegodon insignis. See Orit 7+ 8+10—11= 40
Loxodon meridionalis +648. . - 8+ 9412 = 46
Euelephas antiquus . - 38 FOr 10) 7. - 10+12+16 = 57

“f primigenius . 4+8+12 . - 12416424 = iG

Fn Indicus . . 448412 . . 12416424 = 76

Se

E, (EUELEPHAS) ANTIQUUS. 177

cement had disappeared. Crown broad and oblong; looks very like
E. primigenius in form.
Length of crown, 2'4in. Width of crown, at 1st ridge, l-in. Width at 8rd ridge,

1-3in. Width at 4th ridge (greatest), 145in. Width at 5th ridge, 1:2in. Height
at 5th ridge, 1°3 in.

In the British Museum (Cat. No. 23,766) there is a specimen of the
second milk molar, upper jaw, left side, certainly belonging to L.
antiquus. The crown is composed of six ridges; five ridges are
worn and much expanded; front talon broad and confluent; back
talon disguised. Purchased from Mr. Ball.

Length of crown, 2°65in. Width in front, 1:09in. Greatest height at front,
1:25 in.

In the Taunton Museum there is a specimen of an unworn germ of
a second milk molar; doubtful whether upper or lower, but believed
to be upper. This tooth forms a shell; no fangs remaining; ridges all
formed, but no cement; shows distinctly seven ridges, with front talon
small and narrow, and hind talon a low splent of four digitations; very
narrow in front, broad and egg-shaped behind; marked ‘ Banwell,’ but
believed to be from Hutton; is a small tooth. Dimensions :—

Length of crown, 1°9in. Width at 2nd ridge, -7in. Greatest width of crown,
12in. Height of 6th ridge, 1-3 in.

Figs. 1 and 1 a of Plate XIV. A. of the ‘Fauna Antiqua Sivalensis’
represent the second upper milk molar, right side, of £. antiquus, with
five plates. The figure is taken from a Kent specimen in the Canter-
bury Museum.

Lastly, one of Scharf’s plates for M‘Enery’s projected work on the
Kent's Hole fossils represents three elephants’ teeth, all second milk
molars—two of them germs, the third worn, and evidently of £.
antiquus. The characters are very distinct. One of the germs has
the side rubbed, exhibiting the characteristic section of EH. (Hueleph.)
antiquus—t.e. wide plates; and the worn crown shows the plaited
enamel-plates and wide discs of the species very clearly. It is not a
little remarkable that, im the MS. index of the headings, M‘Enery
refers to it as being the ‘ Indian Elephant’! This is another illustra-
tion of how shrewdly he observed ; and the notes extracted from Cuvier
show that he tried to master the subject of the dentition of the Ele-
phant ; but the detailed descriptions of the portion on Elephant appear
to have been lost.

No. 21,301 in the British Museum collection is a superb palate
specimen of /. antiquus, from Grays Thurrock, purchased from Mr.
Ball. It contains the third milk molar on either side, well worn.
The crown is long and narrow. The crowns converge, and all the
plates are more or less worn. The crown is composed distinctly of nine
main ridges. The anterior talon is confluent with the first dise. The
hind talon is covered by cement.

Length of right molar, 5°in. Width in front at 2nd ridge, 1:7 in. Greatest
width behind, 1:9 in. Interval between teeth, front, 2° in. Interval behind at

_ Second ridge from last, top, 3° in.

In the Museum at Oxford there is also a specimen of the third left
VOL. II. N

178 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

upper milk molar of Elephas antiquus, showing ten principal ridges,
with talons ; five plates are more or less worn.

Length of crown, 5:3 in. Width in front, 2°05in. Width behind, 1-9 in.
Greatest width, 2-lin. Height at fifth plate, 4-6 in.

This is a very beautiful and characteristic specimen.

In the Museum at Saffron Walden there is a beautiful specimen of
third (upper?) milk molar, right side, of H. antiquus, presented by W.
G. Gibson, Esq.

Length, 4:25 in. Width in front, 1-6in. Width behind, at 7th ridge, 1°din.

It has ten ridges, and a heel minutely crimped; front plates very
wide; a tendency to expansion in the middle; all the plates in wear,
the last very little touched ; out of a soft clayey material.

The Bristol Museum contains a specimen of the third upper milk
molar, with ten ridges, well worn; from Durdham Down.

In the Town Hall collection at Colchester is a specimen, consisting
of the two maxillaries of a young Eleph. antiquus, containing the third
milk molar, very much worn on the right side. There is an empty
cavity on the left side. The right tooth behind shows seven plates, all
worn, with broad discs. The matrix is very ferruginous and crag-
looking. It does not adhere to the tongue, and has a dark sepia colour
throughout, and is sea-polished.

It was dredged off the ‘ West Rocks,’ Essex coast, and was pre-
sented by Mr. Bolton Smith, in 1847.

Figs. 2 and 2a of Plate XIV. A. of the ‘Fauna Antiqua Sivalensis’
represent the third upper milk molar, right side, of HZ. antiquus. The
crown hasten plates. The specimen was obtained from Southwold, and
is in the Museum of the Geological Society.

Length, 5°5 in. Width, 2°3 in. Height, 2°8 in.

Figs. 3 and 3a of the same plate show the same tooth, left side, also
with ten plates and a heel.

Length, 6° in. Width, 2°2 in. Height, 3°5 in.

b. Lower Milk Molars.'—In the British Museum Paleontological
collection there are several specimens of the second lower milk molar
of #. antiquus.

No. 18,810 is a second lower right milk molar. The crown is com-
posed distinctly of seven discs, including the most anterior, all worn.
It is narrow and concave.

Length of crown, 2°4in. Width in front, at 2nd disc, ‘9in. Greatest width,
135 in.

No. 21,655 is a larger specimen, long and narrow, less worn, with
large front fang; shows distinctly seven principal ridges, plus talon. It
is the second lower left. The seven ridges are worn, but not much.

Length of crown, 3:in. Width in front at 2nd ridge, ‘9 in. Greatest width
behind, 1°3in. Height of crown at 7th ridge, 1°6in.

No. 21,310, from Mr. Ball, Ilford, is a beautiful lower jaw, left
side, with symphysis, and containing second milk molar zn situ. The
crown is narrow, and composed of seven discs, which are broad, ex-

1 IT can find no note of the first lower milk molar of E. antiguus among Dr.
Falconer’s papers.—[ Ep. ]

eS

E. (EUELEPHAS) ANTIQUUS. 179

panded, and crimped; appears to be of Z. antiquus, although the
species is somewhat doubtful. The alveolus of part of third milk molar
is seen behind.

Length, 2°6in. Width in front,:95in, Greatest width, 1:2in.

In the Bristol Museum there are two specimens of this tooth. One
from the left side is very perfect and characteristic, and shows seven
plates, with front and back talons. The four front plates are worn. The
second specimen is Mr. Beard’s, from Bleadon Cave. It is also from
the left side, and quite entire. The crown shows seven plates, with
small front and back talons. The discs are thick, and the enamel
plaited. The length of the crown is 2:3 inches.

In the Oxford Museum there is an exquisitely preserved second
milk molar, lower jaw, right side, showing eight discs of wear besides
front and back talons. All the discs are more or less worn and dilated,
but there is no mesial expansion. The enamel-plates are very much
crimped, but coarsely and unequally. The fangs are present, and also
a disc of pressure behind. The tooth has its perfect coat of cement,
while the corresponding Bristol tooth is denuded of cement. It is very
narrow in front.

Extreme length, 2°65 in. Width at second plate, -85in. Greatest width,
1-2 in.

This is the tooth figured by Buckland in the ‘ Reliquie Diluviane,’
ana is from Kirkdale Cave.

In Major Wood’s collection, from Gower caves, there is a second milk
molar, lower jaw, right side. The tooth is narrow in front; the crown
has seven main ridges; the back talon has an additional splent; the
five anterior ridges are worn; the crown is beautifully perfect, but
there are no fangs; cement present. The specimen is believed to be
from Spritsail Tor.

Length of crown, 2°25 in. Width of crown in front, 38 in. Greatest width
behind, 1:°25in. Greatest height at 8th ridge, 1-45 in.

In the Museum at Taunton there are several specimens of this tooth
(lm.m. 2). One is of large size, and has the crown worn, with seven
plates; crimped enamel, flat with the surface of the crown; fangs
complete.

Length of crown, 2°3in. Width at second plate, -9in. Greatest width, 1:4 in.

Two other beautiful specimens have also seven discs, but are smaller,
and have the enamel-edges raised. They probably belonged to the
same animal.

In the Oxford Museum there is a beautiful specimen of the third
milk (or first true?) molar, lower jaw, left side; the crown quite entire,
showing distinctly eleven main ridges, with front and back talons;
seven ridges worn. The crown is narrow, and the enamel-plates are
expanded in the middle.

Length of crown, 5°6in. Width in front,1‘8in. Greatest width, 2°in. Height
at seventh plate, 4° in.

This specimen is from Hurley, Bolion, Oxford.

In the Bristol Museum there is a third milk molar, lower jaw, right

side, from Durdham Down, showing ten plates, with front and back
n2

180 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

talons. It is very perfect; and the loop and expansion,’ as in Z.
antiquus, are typical.

The collection of the Rev. John Gunn, at Irstead, contains a very
characteristic third milk molar, lower jaw, left side, of EH. antiquus
(East Coast, No. 2). It shows ten plates, all more or less ground, and
its dimensions are :—

Length, 4°9 in. Width in front, 1°5 in. Width behind, 2° in.

Lastly—Figs. 6 and 6a, and figs. 7 and 7a, of Plate XIV. A. of
the ‘Fauna Ant. Sivulensis,’ represent two specimens of the third upper
milk molar. Fig. 6 is taken from a specimen from Suffolk, presented
by Dr. Cooke to the Geological Society (No. 8,411); it is from the
right side, and has seven well-crimped plates, but is imperfect.

Length, 4-2 in. Width, 271 in. Height, 3: in.

Fig. 7 is also from the right side, and shows nine well-crimped plates
and a portion of a tenth behind, where the tooth is not quite perfect.
The tooth is narrow in front and broader behind. (See Plate IX. figs.
1 & 2 of this volume.)

Length, 5-4 in. Width, 2°in. Height behind, 2°65 in.

ce. Upper True Molars.—Among Mr. Dixon’s specimens in the
Paleontological collection of the British Museum, No. 28,512 is a
fine specimen of the first true molar, upper jaw, right side. It is entire,
with fangs. The crown is composed of twelve ridges; the nine anterior
ridges are worn. The tooth is excessively like one belonging to Mr.
Rupert Jones, from Suffolk, but is narrower.

Length of crown, 7°7 in. Width in front, 2°2 in. Greatest width, 2-7 in.
Height at 9th ridge, 5°4 in.

The corresponding tooth of opposite side has also twelve ridges.

Dimensions of Mr. Rupert Jones’s specimen :—

Length of crown, 86 in. Widthin front, 2°7in. Greatest width, 3°2in. Height
at 9th ridge, 7°5 in.

Another specimen in the British Museum (No. 18,759) of the same
tooth has ten ridges, seven of which are worn.

Length of crown, 5°9in, Width in front, 2-1in. Width behind, 18in. Height
at 7th ridge, 4°6 in.

In the Museum of the Geological Society there is a specimen of the
antepenultimate true molar, upper jaw, right side, of Huelephas anti-
quus, labelled No. 8,409, as being from Southwold. It is entirely free
from mineral impregnation, and I doubt its asserted Southwold origin.

In the Museum at Taunton, among the specimens collected from the
caves by the late Rev. Mr. Williams, there is a very fine molar, upper
jaw, right side; crown worn, but quite perfect, showing twelve prin-
cipal ridges, with talons. The front talon is confluent with the anterior
disc; the hind talon is unworn. The twelve ridges all worn into discs;
the dises are a little expanded; macherides raised; enamel irregu-
larly but well plaited; front fang and surface of pressure present. This
tooth is convex outside and concave on the inner side; the plane of
wear slopes from inside, out. It is not known which cave the specimen
is from, but it is supposed to be from Bleadon, most of Mr. Williams's
specimens being from Bleadon.

Length of crown, 7:°2in. Width of crown at 2nd ridge, 2°3in. Width behind,
at 8th, 2°8in. Height of crown at 9th, 5°6 in.

E. (BUELEPHAS) ANTIQUUS. 181

Among some specimens belonging to Mr. Charlesworth, which I have
been permitted to examine, there is a fine upper molar, right side, of
#. antiquus, showing twelve ridges, of which nine are worn. The
dises of wear are less open and expanded than usual, but the enamel
is thick and crimped, of the HL. antiquus pattern. The crown is narrow.
The specimen is reputed to be from Happisburgh; but it has no
Lepralia patches. (See antea, page 134, note.)

In the Museum at Chichester there is a beautiful specimen of first
true molar of left upper jaw, quite entire, showing twelve plates in all,
nine of which are worn. All the fangs are present. ‘The enamel and
plates are most characteristic. From Brackelsham Bay, near Selsea.

In the Museum at Rome there is a fine specimen of #. antiquus, from
Monte Verde. It is a skull of a youngish animal, which was found
nearly perfect, but is now partly mutilated. It contains on either side
three plates of the last milk molar, worn out; the first true molar in
full use, and the second true molar behind—four plates of it being
shown.

The first true molar is beautifully shown, the crown consisting on
left side of ten principal ridges, with front and back talons; all the
plates well expanded and crimped; nine of the plates more or less
worn. The palate is filled with a sandy conglomerate, with some black
grains of Pyroxene (volcanic).

Length of first true molar, 5-5in. Width of first true molar in front, 1:7 in.
Greatest width behind, 1‘9in. Interyal between molars in front, 2°-4in. Interval
behind, 3-lin. Interval in front, outside, 55in. Interval outside, behind, 6°in.

An imperfect specimen, from Southwold, of the first upper true
molar, right side, is figured in the ‘ Fauna Antiqua Sivalensis,’ Plate
XIV. A. figs) 4 and 4a. The crown has nine plates, and the dimen-
sions are :—-

Length, 5:5 in. Width, 2°6 in. Extreme height, 6°5 in.

In the Museum of the Geological Society (No. 10,664) there is a
second or penultimate upper molar, right side, of H. antiquus, well
advanced in wear; the anterior part is broken off; the nine posterior
ridges are present, all more or less worn with the exception of the last.
This specimen is of a dark iron-shot colour, and highly infiltrated with
ferruginous matter; it is extremely heavy, and bears all the marks of a
crag specimen ; locality, ‘Southwold.’ The enamel is thick, and the
worn discs exhibit the characters of the species in the most pronounced
manner. If the asserted locality be correct, this is the only known
species of Elephant molar found in the Red Crag. Specimens of £.
meridionalis have been found in the Norwich crag by Mr. Prestwich ;
and by Mr. Gunn, in the same crag where Mr. Layton found the Mas- °
todon teeth, a molar of H. antiquus was found. It is of the utmost
importance to determine accurately the origin of this specimen.

In Colonel Wood’s collection from Minchin Hole there is a superbly
characteristic specimen of the penultimate upper right molar of FH.
antiquus, showing fourteen plates and a posterior talon. The six
anterior plates are worn; the worn part might almost pass for the exist-
ing Indian Elephant; the tooth narrows very much backwards; the
crown is perfectly entire; the lateral mass of cement is enormously
thick ; the discs not much expanded, but beautifully crimped.

Extreme length of crown, 8-4in. Width in front, 3:in. Height at 6th plate,
6°6 in.

182 BRITISH AND EUROPEAN FOSSIL ELEPHANTS

°

In the Woodwardian Museum at Cambridge there is a fine specimen
of the last upper molar, left side, of H. antiquus. It is quite perfect,
consisting of sixteen ridges, of which the anterior eleven are more or
less worn. The discs of wear are expanded and opened, with thick
boldly crimped enamel, and exhibit all the characteristic marks of the
species in the most typical way. The specimen is probably the most
perfect remain of a last true molar of the species in the kingdom,
but unluckily the locality has not been recorded. (Dimensions not
taken.)

In the Rev. Mr. Gunn’s collection there is a fine upper jaw of E.
antiquus, containing the last true molar in situ, complete on both sides,
and on the right side the pits of a narrow penultimate. There are
sixteen plates which are high, and the tooth has all the characters of E.
antiquus, and ought to be figured. The teeth converge in front. It was
dug out of the beach from the Elephant-bed near Ostend.

Extreme length of molar, 9° in.

In the Museum of the Junior United Service Club a very fine and
characteristic specimen is preserved in a glass case of the last upper
molar of E. antiquus. The eleven anterior plates are worn; the pos-
terior plates are unworn. ‘The cement is cracked, but not crumbled as
if out of sand.

In the Marticelli collection in the University Museum at Naples there
is a magnificent specimen from Fregelle, between Rome and Naples.
It is a superior molar, bearing label, ‘ Molare Elefantino trovato presso
Vantica citta di Fregelle.’ It is the last upper left molar, the thirteen
anterior plates remaining. Disc of pressure in front and bourrelet.
The four or five anterior plates are worn; enamel thin and well-
crimped. Part of the tooth wanting behind; is certainly of Eleph.
antiquus; fangs wanting.

Length of crown, 83in. Height of 8th plate, 10-4in. Width of 4th plate,
3°7 in.

The Elephant’s tooth above described is of enormous size; and, in
the great height of the plates of the molar, surpasses anything that I
have seen elsewhere of Hlephas antiquus. Specimens of the same tooth,
but fragmentary, I have also found in the University Museums of
Syracuse and Rome.

Figs. 5 and 5a of Plate XIV. A. of the ‘ Fauna Antiqua Sivalensis’
represent the last true molar, upper jaw, right side, with fourteen
plates and a hind talon well-crimped. The tooth, which is rather
imperfect in front, is from the Forest-bed, Norfolk, and is in the British
“Museum (No. 16,229). Figs. 5 and 5a of Plate XII. D. show the
same tooth, with sixteen ridges and a small heel much worn. The
specimen is from Kent, and is in the Canterbury Museum, bearing the
label 9. Plate XJV. B. fig. 16, gives another illustration of the same
tooth, and has already been referred to (see page 138, and vol. i.
page 447). The crown is entire, and comprises from sixteen to seven-
teen ridges within an extent of 11 inches.

d. Lower True Molars.—In the British Museum Paleontological
collection there is a specimen, which formerly belonged to the Earl of
Aylesbury, of a fragment of the lower jaw of Elephas antiquus, containing

ot aie Fe iP),

Beer. NFR LD Ake

E. (EUELEPHAS) ANTIQUUS. 183

the first true molar on either side ; the crown is entire, and presents
twelve plates and a heel. The plates are worn and well crimped.

Length of right molar, 6°7 in. Width, 2°3 in.

The same tooth is shown in Plate XIV. A. figs. 8 and 8 a, of the
‘Fauna Antiqua Sivalensis.. The crown in this specimen has also
twelve plates.

Length of tooth, 8:3 in. Width, 2°5in. Height, 4:2 in.

In the Museum of the Geological Society there is a portion of an
antepenultimate true molar, lower jaw, right side, comprising six ridges ;
said to be from Suffolk.

In the Woodwardian Museum at Cambridge there is a fragment of
right ramus of lower jaw of Elephas antiquus, comprising the sym-
physial beak, the right diasteme, and a part of the antepenultimate
true molar in situ. Six ridges of the tooth remain, showing the dis-
tinctive marks of the species in the most perfect way, The specimen
is black, polished and covered with Balani, showing that it has been
dredged up and probably derived from the Happisburgh oyster-bank.
The diastemal ridge in this case inclines very gradually forwards, con-
trasting in this respect in a very marked way with the character shown
in the Mammoth.

Among the specimens found in Bacon’s Hole, Gower, there is a
beautiful and characteristic specimen of a right lower antepenultimate
true molar of Z. antiquus, with eleven plates, all worn.

In the Museum of Practical Geology, Jermyn Street, there is a quite
perfect specimen of the antepenultimate, or first true molar of EF.
antiquus, from a cutting in the Great Northern Railway. It has twelve
plates and a heel.

Lastly, Professor Ponzi’s collection at Rome contains a beautiful first
true molar, lower jaw, right side, of H. antiquus. 1t is quite entire,
and 7m situ inthe jaw. It shows distinctly all the ridges of the crown
worn. There are ten expanded plates, with talons.

Length of crown, 5°7in. Width in front, 15in. Width behind, 1°9 in.

It,is from Tor di Quinto, a continuation of Ponte Molle.

One of the most remarkable molars of #. antiquus I have seen was
dredged up at the West Rocks, Harwich, in 1852. It belongs to the
lower jaw, right side, and is probably the penultimate true molar. It
comprises twelve principal ridges, with a front and back talon. Only
eight of the anterior ridges are touched by wear.

Extreme length of crown, 10:8 in. Width of crown in front at second ridge,
2-9in. Width of crown at eighth ridge, 2°8in. Width of crown at eleventh ridge,
2°8in. Extreme height of crown at seventh ridge, 66in. Height of crown at
eleventh ridge, 5-9 in. Height of crown at third ridge, between first and second
fangs, 5°3 in.

In the Ipswich Museum there is a very fine specimen of a lower jaw
molar, probably the left penultimate of Elephas antiquus, showing
twelve plates, and a very insignificant talon of two points behind, the
anterior part of the tooth broken off; no fangs, but a good deal water-
worn below; six of the anterior plates worn, five plates unworn.
The specimen is reddish, and has all the appearance of a Crag specimen.
Crown very narrow; plates expanded in the middle; enamel thick,
but not like £. meridionalis in either respect. Dredged up off

184 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

Harwich, and presented by Hedworthe Jolliffe, Esq., 4th Dragoons.
Dimensions :—

Length, 8°5in. Greatest width of anterior plate, 2-5in. Width of dth plate,
2:3in. Greatest height, 5-in. Basal length of 10 plates, 7-3 in.

In Colonel Wood’s collection of bones from Minchin Hole there is
an extremely characteristic specimen of the second true lower molar,
right side, of H. antiquus. ‘The crown presents sixteen plates in all,
with talons. The nine anterior plates are worn. The plates show
beautifully the mesial expansicn.

Length of crown, 10‘lin. Width in front, 2°6in. Width in front at 10th
ridge, 2°55in. Width behind, 2°in. Height at 9th, 5-2 in.

In the Museum at Chichester there is a penultimate lower molar,
right side, the crown of which shows fifteen plates, somewhat rolled.
It is from Brackelsham Bay.

In Professor Ponzi’s collection at Rome there is a magnificent lower
jaw of E. antiquus, from Monte Verde (quaternary). Both rami are
present, but they are separate, and the fragments do not fit, owing to a
portion of the symphysis being wanting. The specimen is beautifully
marked. The right fragment is most perfect, comprising part of the
diastemal ridge, the whole of the horizontal ramus and part of the as-
cending ramus, as far up nearly as the base of the condyle; the coronoid
plate is broken off. It contains the penultimate true molar in full use,
and eight plates of the last molar in germ. It is of enormous size.

Length of diastemal portion, 5‘2in. Height of jaw at commencement of ditto,
9-4in. Height of jaw behind, 6-in. Length of penultimate to anterior fang, 10: in.
Length of penultimate, crown portion, 9'lin. Width of penultimate in front,
2-2in. Width behind, 3-lin. Number of plates, 12, and talon. Greatest width
of jaw behind, 7:5 in.

The penultimate has all the plates and the posterior talon worn. It
shows the characters of the species beautifully, with a distinct mesial
expansion on either side of each disc, and thick plaited enamel in good
relief. The tooth is very high and compressed in front, low behind.

Figs. 10 and 10 a of Plate XIV. A. of the ‘ Fauna Antiqua Sivalensis’
represent the second true molar, lower jaw, right side, of Hlephas
antiquus. There are twelve plates and a heel. ‘The five anterior
plates only are worn, The specimen is in the British Museum (No.
19,844). The dimensions are :— :

Length, 10: in. Width, 2°5 in. Height, 6° in.

Three specimens of the last lower molar of EF. antiquus are figured in
Plate XIV. A. of the ‘Fauna Antiqua Sivalensis.’

Figs. 11 and 11 a show a tooth from left side, with fifteen to sixteen
plates, but a portion in front is gone. ‘The specimen is from Saffron
Walden (see vol. i. p. 443). The dimensions are :—

Length, 12°3 in. Width, 3: in. Height, 5: in.
This illustration is reproduced in Plate IX., figs. 8 & 4, of this volume.

Figs. 12 and 12 a show the last molar from the right side. Only
the eleven posterior plates are present. They are very crimped and
bent. The specimen is from Happisburgh, and is now in the British
Museum. The dimensions are :—

Length, 10°65 in. Width, 3°4in. Height, 5-7 in.

DESCRIPTION OF PLATE IX.

ELePHAS (HUELEPHAS) ANTIQUUS.

Figs. 1 and 2. Views in profile and plan of third milk molar, lower jaw,

right side, from Kent. The drawings, one-third of the natural
size, are copied from those by Mr. Ford in the Fauna Antiqua
Sivalensis, Plate XIV. A, figs. 7 and 7a. The specimen is
described at page 443 of vol. i. (See also page 180 of this vol.)

Figs. 3 and 4. Views in plan and profile of last molar, lower jaw, left

side, from Saffron Walden. <A description of the specimen will
be found at page 443 of vol. i. The drawings, one-third of the
natural size, are copied from those by Mr. Ford in the F, A.5.,
Plate XIV. A, figs. 11 and 1la. (See pages 147 & 184.)

. Another last lower molar, left side, from Rome, described at

page 443 of vol.i. The drawing is one-third of the natural
size, andis a copy of one by Mr. Ford in the F. A. §., Plate
XIV. A, fig. 138 a. (See page 185.)

[Fig. 4 of Pl. XIII. A of F.A.S. has not been reproduced in this Plate as stated

at p. 440 of voli. See list of Errata in yol. i.]

VOL, II. b

Fig. De

W West imp.

Elephas (Huelephas) antiquus

GH Ford del. JDimkel, bth

E. (EUELEPHAS) ANTIQUUS. 185

Figs. 138 and 13a show a last lower molar, left side. There are
fourteen plates remaining, but some in front are missing. This speci-
men is from the Via Appia, Rome. It was formerly in the collection
of Cardinal Gualteri, and is now in the British Museum. (Plate IX.
fig. 5.)

A more perfect specimen is represented in Plate XIIL. A. fig. 4, of the
‘Fauna Ant. Siv.’ This drawing is taken from a very perfect lower
jaw in the Museum of the Geological Society, which contains a portion
of the penultimate and the last true molar on either side. The last
molar is composed of seventeen ridges; the talon portion alone is
wanting. The nine anterior ridges are worn, the rest are intact. The
tooth is remarkably curved, being concave on its outer surface. The
specimen corresponds closely in character with a specimen from Ilford,
in the Museum of the Royal College of Surgeons. The origin of the
specimen is unfortunately unknown.

In the Museum at Saffron Walden there is a superb specimen of the
right side of a lower jaw of a fossil Elephant, probably 4. antiquus. It
contains the last molar nearly entire, with eighteen plates, all worn except
the three last. The tooth contracts behind, and is concave on the outer
side. ‘The plane of wear is also concave. The plates present a loop in
the centre, with bold crimping, which diminishes towards the sides.
Some of the plates behind are probably wanting. The last plate in
height measures 4:7 inches; the eighth, 5:4 inches.

In April 1859, I examined, in the Museo dell’ Universita della
Sapienza at Rome, a superb specimen of the right and left rami,
detached, of the lower jaw of a fossil Elephant discovered in the
quaternary volcanic sands of Magliana, on the railway cutting to
Civita Vecchia. The fragments do not unite, the spout of the sym-
physis being wanting. The original fractured surface, covered with
matrix, is.seen on the left side; on the right, part of the sympbysial
portion remains, but a portion is seen below to have been broken off by
a recent fracture. Both branches comprise nearly the entire length of
the horizontal ramus and the greater extent of the ascending ramus,
along the posterior contour up to the neck of the condyle. The right
side is the more perfect, including a considerable portion of the
diastemal ridge in front of the molar, the whole of the ascending ramus
up to the neck of the condyle, and a considerable part of the coronoid
apophysis, near the base, the upper part of the lamina only being
broken off. The two rami are exactly alike in mineral condition,
being impregnated with a ferruginous or rusty impregnation like many
of the Sewalik specimens. The matrix, of which but a sparing quantity
remains, is a yellowish-grey fine sand, with specks of pyroxene, red, with
all the characters of the volcanic sand-deposits of the Valley of the Tiber.

Each ramus contains one molar ‘in situ, in the most beautiful state of
preservation, the crown surface well worn, and a thick layer of glossy
cement, with its surface as uninjured as in the living state. The
expansion of the discs and the flexures and undulation of the thick
enamel are shown in the most perfect manner, the cement being tinged
of a glossy-pale, rusty, or almond tint.

Only one tooth is present on either side, and so exactly alike in size,
form, and wear, that no doubt can be entertained of their belonging to
the same animal. The molar (right side) presents eleven collines and.

186 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

a posterior talon. The whole of the crown appears to be emerged, but
the posterior extremity behind the talon is but very slightly concealed
by matrix. The discs of the nine anterior ridges are transverse ; of the
two last and talon, the tips are only worn. The crown is comparatively
narrow in front (2°5 inches), but broader in the middle (3-3 inches).
The discs of wear in front are very broad, with angular expansion in
the middle resembling the African Elephant. The anterior plate of
enamel in the discs is somewhat concave, in the general direction, as in
the Grays Thurrock, Z. priscus, and convex behind, with the angular ex-
pansion most distinctly pronounced there, indicating the six anterior discs.

The angle at either side projects into a loop. The expansion of the
anterior discs is so great that the aggregate length of four amounts
to 4 inches, or 1 inch to each, as in the African Elephant. The
lateral horns of the discs are bent forwards, as in the Grays Thurrock
form of H. priscus. The enamel of the plates is very thick, and
strongly undulated. The digitations of the last emerged ridge
make four ring-discs and the talon plate three; the digitations are few
in number and very thick.

The jaw is high in front, at the commencement of diasteme, in rela-
tion to the height behind; and the diasteme descends very vertically, as
in the African Hlephant.

This is one of the finest specimens of a fossil elephant lower jaw that I
have ever seen; but Iam greatly perplexed to determine with confi-
dence what it is. The tooth on the right side appears to be perfect, as
also on the left, and there is no indication of a tooth coming behind.
It does not seem as if any portion supported on the large anterior fang
had dropped off; it is a little high there, but I failed to make out the
disc of pressure. There is no appearance of another molar coming
behind. In this view it would be regarded as the last true molar of
E. priscus, but then the discs are like an exaggerated form of E. anti-
quus, and the number of the plates would correspond with the penul-
timate lower of that species.

On the left side the tooth has eleven ridges, talon, &c., and exhibits very
considerable torsion of the plane of view. It slopes from inside out in
front, and then is twisted so as to slope from outside inwards. In
front a part of the tooth appears to have fallen off, as exhibited by the
contour in front, about 1 inch in advance of anterior disc. Upon
the diastemal ridge, near commencement, there is distinctly seen the
remains of another fang. It is very difficult to say whether the
crown extended as far as the fang, or whether the fang belonged to
a fallen-out tooth ; in the former case these plates must have fallen off,
making x 16 x. :

Extreme length of crown, left, 11:‘4in.; right, 12in. Width of Ist dise outside
enamel, left, 2-4in. Width of 2nd dise outside enamel, left, 2°7in. Width of
2nd dise with cement, left, 2°9in. Width of 5th dise outside enamel, left, 2-9 in.
Width of crown at 5th dise, with cement, left, 3°35in. Length of 4 first dises out-
side expansion, left, 4:in. Length of space occupied by the 10 anterior discs in
middle of crown, left, 9‘ in. Length of jaw from post. border of aseending ramus to
anterior edge of molar, 21:in. Height of jaw in front to edge of jaw, 8°6in. Greatest
transverse diameter near coronoid, 6°6 in.

7th May, 1859.—The final result at which I have arrived is, that the
tooth is the last of E. antiquus, and that a portion is gone in front,
which was supported on the large fang.

vio

toys

E. (EUELEPHAS) ANTIQUUS. 187

On examining the Museum at Turin in July 1856, I was at once
struck with the fact that the greater portion of the Piedmontese ele-
phants’ teeth were evidently not of the same species as those of the
Val d’Arno, and on comparing them minutely with the figures in the
‘Fauna Antiqua Sivalensis’ (especially Plate XII. D. and Plate XIV. A.),
I had no doubt that they were of the Chartres and Grays (in Essex)
species, viz. E. antiquus.

There is a specimen of a huge upper last molar, right side, the ex-
treme length of which is 13-75 in. to 14 in., and the greatest width of the
crown 45 in. This molar is not quite complete at the anterior end; the
large fang and probably the anterior talon, with one or two ridges, are
broken off; but what remains shows twenty-four plates of enormous
depth (8 inches!). The crown is very broad in front and narrow
behind. The first twelve plates are worn; the enamel is thin, and very
much crimped, especially in the middle, where there is a tendency to
loop expansion. The enamel is not nearly so thick as in E. meridionalis.
In every respect it closely resembles fig. 5 of Plate XII. D. of the
‘Fauna Ant. Siv.’ The apices of the digitations of the ridges, just
beginning to come into wear, are very round and distinct. This
specimen comes from a railway digging at St. Paolo or Nizza della
Paglia. It is soft and white, and adheres strongly to the tongue.

There are also four very fine specimens, consisting of the two last
upper and two last lower molars, also from St. Paolo ; but unfortunately
none of the teeth are entire, so that we are deprived of exact numerical
information as to the crown-ridges. But they all present the marked
characters of the species—viz. approximated plates of enamel, very
much crimped, the plates enormously high—z.e. the height more than
double the width of the crown. The lower jaw specimens also are
very much curved outwards, and with a contorted curve to the coronal
surface. All these four teeth probably belonged to the same animal—
an adult. ach of the upper teeth consists of nineteen plates, the rear
part being broken off.

Length of fragment, 10-in. Greatest width of crown, 3:5in. Greatest height of
plates, 7°5 in.

Of the lower molars, the right shows twenty plates, and is very
narrow for its height.

Length of fragment, 12in. Greatest width of crown, 3°in. Greatest height,
6:5 in.

Another specimen is the right side of the lower jaw, containing an
entire antepenultimate true molar, with eleven ridges and a talon. The
tooth is very narrow for its height and length.

Length, 7:5 in. Width of crown at third plate, 2-4 in.!

e. Bones of Skeleton and Lower Jaw.—In the Museum at Rome there
are a number of the bones of the skeleton (the Pignano skeleton), evidently
_ belonging, from the characters of the teeth, to Hlephas antiquus. The
skeleton consists of the skull and lower jaw, the vertebre and ribs, the
femur, tibia, and bones of the feet.

The Chichester Museum also contains many of the bones of the
skeleton of Elephas antiquus, and among others a scapula 38 feet and

1 The San Paolo molars were after- | meniacus. See memoir on EF. Columbi,
wards referred by Dr. Falconer to E. Ar- | page 250, and also page 192.—[Ep.]

188 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

3 inches long, eight vertebree, one humerus, tibias, fibulas, &e.; also a
tusk 8 feet long and 23 inches in circumference. ‘These specimens were
obtained from Brackelsham Bay.

In the Museum at Syracuse there is a tusk of Elephas antiquus,
7 feet long, and rounded in section. It was found 100 metres above the
mouth of the Amasses (?).

In the Norwich Museum there is a superb lower jaw specimen of
E. antiquus, with the anterior portion of both sides entire, and a tooth
on each side. It was obtained from the jetty at Cromer, 15 feet below
ground, and was presented by W. H. Windham, Esq.

Height of jaw in front to alveolar margin, 11°5 inches. Ditto behind, at rise of
coronoid, 8: inches. Height of symphysial aperture, 6°5 inches. Width of ditto,
above, 3° inches, expanding below to 3-4 inches.

It is truncated off very vertically in front, with the beak directed
down, but broken off. It has a thick rugous rim below the symphysial
aperture. Very large foramina, about one-third below the alveolus.

The anterior part of the tooth is worn off, the fang remaining. In-
cluding fang-portion, the extreme length of tooth of right side is about
11 inches; width in front, 3:5 inches. Contracts a little backwards;
shows thirteen or fourteen plates remaining, some three or four worn
off, The plates very much undulated, as in Mr. Folger’s specimen, and
no loop in the middle; the discs of uniform width across, but a little
flexuous towards the horns; very concave in plane of wear, and the
great wear on the inside. The teeth exhale a strong animal odour in
burning. |!

VII. Geotocicat AGE or Fossin HLEPHANTS.?

In the preceding details I have dwelt at great length on
the characters presented by the molar teeth in the fossil
Elephants, in comparison with those of the cranium or of the
rest of the skeleton. In the fossil mammalia generally, the
characters yielded by the skulls are of the highest import-
ance in defining the genera and species; and so they would
be among the Proboscidea, ifattainable. But while geologists,
everywhere in the gravels and fluviatile or lacustrine beds of
the newer Tertiaries, meet with the fossil grinders of Ele-
phants, crania presenting determinable characters are exceed-
inglyrare. This arises from the very incompact and cancellar
structure of the great mass of the skull in these animals,
whence it arises that while the bones of the head rapidly dis-
integrate and crumble to pieces, the hard and solid grinders
are preserved. Although many thousands of Mammoth
grinders, derived from numberless individuals, have been
discovered over the broad area of Hurope, in the superficial
deposits, I know only a single instance of a well-preserved
cranium of this species (Huelephas primigenius) occurring in
any of the great Huropean Museums out of Russia, namely,
at Mannheim, on the Rhine. Another cranium, but con-

1 See page 176, note 1.—[Eb.] ? See page 76, note 1.—[Ep.]

Se

THEIR GEOLOGICAL AGE. 189

siderably less perfect, has been acquired, by the indefatigable
energy of Dr. Kaup, for the Museum at Darmstadt. It is of
importance to geologists to have pointed out to them the
characters of the remains that are most frequently encoun-
tered, and which are of the readiest application and of most
significance in the distinction of the species. Hence the
undue proportion in which I have directed attention to the
dental character of the fossil Elephants in the present com-
munication.

I shall now proceed to the consideration of the bearing of
these fossil Elephants, regarded as distinct species, upon the
classification of the newer tertiary strata.

There is probably no locality in Europe so favourably situ-
ated for the investigation of the mammalian fauna of the
Pliocene period as the district of the Val d’Arno, in Tuscany.
Above the gorge of Incisa, the valley gradually expands on
to Arezzo, a distance of upwards of twenty miles, forming an
extensive area, which was evidently in former times occu-
pied by a large lake. In the lacustrine deposit, and in the
low hills which immediately surround it, the remains of a
great Pliocene fauna are met with in immense abundance,
and, regarded as a whole, in a better state of preservation,
for the same number of species, than has as yet been dis-
covered anywhere else in Europe, comprising species of
Mastodon, Hlephas, Rhinoceros, Tapirus, Equus, Hippopo-
tamus, Sus, Felis, Machairodus, Hyena, Bos, Cervus, Anti-
lope, Lagomys, and other small mammalia. Besides in-
numerable remains of Hippopotamus, entire skeletons and
crania of all ages of the Val d@’Arno Elephant, and skeletons
of Tetralophodon Arvernensis, have been met with, presenting
the most ample materials for the establishment and distinc-
tion of the various forms. Another circumstance, especially
favourable to the study of the Val d’Arno fauna as a Pliocene
association, is that, according to my observation, it is entirely
free from any admixture of the characteristic forms of the
Post-Pliocene glacial fauna, such as the true Mammoth, the
Siberian Rhinoceros, the Elasmotherium, and their Arctic
associates. The Tuscan fossil fauna has been cursorily or
partially examined or gleaned from by Cuvier, De Blainville,
Nesti, and others; but, unfortunately for science, it has not
yet been taken up as a whole, although the Grand Ducal
Museum at Florence contains a collection which might form
the subject of one of the most splendid and important mono-
graphs that has yet been produced on extinct mammalia.
This is the more to be regretted, as the perfect condition of
the Tuscan fossils would have prevented a great deal of the
confusion which has arisen from the imperfect and uncertain

190 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

establishment of species, in other European countries, upon
less perfect materials, involving, in numerous instances,
superfluous synonyms. As we move northwards towards the
Alps, the same association of mammals is presented in the
Sub-Apennine Pliocene alluvium of the Valley of the Po and
its affluents, with this difference, that in various parts of the
great plain of Piedmont and Lombardy, between the Alps
and the left bank of the Po, there is an intrusion of erratic
block phenomena, which has been referred to the transporting
agency of ancient glaciers, extending low down on the
southern side of the Alps. But, so far as I am aware, there
is no intercalation of the characteristic forms of the glacial
fauna above enumerated.

When we cross the Alps and descend upon the valley of
the Rhine, corresponding Pliocene alluvium is encountered
in certain parts of Switzerland, containing remains of at least
one of the characteristic fossil Elephants of Italy. This
stratified alluvium is overlaid by a mass of erratic drift of a
different age, the mammalian fossil remains when met with
being also different. Lower down the valley, from Basle to
Mayence, stretching on as far as Bonn, and running up the
valleys of the Neckar and the Main, is the widespread Post-
Pliocene fluviatile-deposits of the Lehm, containing the re-
mains of the true Mammoth, the Siberian Rhinoceros, and
other characteristic forms of the glacial period. In the
plains of Northern Germany, a corresponding association of
fossil mammalia occurs in the northern Drift. Sir Charles
Lyell has lately directed special attention to the case of the
Hill of Kreuzberg, in the suburbs of Berlin, where remains
of the Musk Ox have been found embedded in the Drift
along with the Mammoth, the Siberian Rhinoceros, and
species of Horse, Deer, and Ox. On the plains north of the
Alps, the mammalia of the Drift fauna are presented to us
without the complication of the Pliocene period; while on
the southern side, in Italy, the Pliocene mammalian fauna
is exhibited free from the complication of the Drift period.

On crossing the Channel to England, the phenomena, so
simple elsewhere, are presented under conditions of infinitely
greater complexity. On the Norfolk and Suffolk coasts, de-
posits of unquestionable Pliocene age are seen in the Crag
and Lignite, or submerged Forest-bed. They contain the
same association of mammalian species as is met with in the
Sub-Apennine Pliocenes of the Valleys of the Arno and of the
Po. But these British Pliocenes are overlaid by enormous
beds of boulder-clay and superficial gravels, containing
mammalian remains of a much later period. The principal
sections exposed along the coast are subject to the wasteful

ose eee

THEIR GEOLOGICAL AGE. 191

action of the sea; and as the cliffs are undermined, the re-
mains of different ages are mingled together on the beach
below. The waves wash away the incoherent matrix, leaving
the fossil bones in adventitious association; and they are
thus presented to the paleontologist, either in museums or
by collectors, under circumstances in the highest degree
deceptive. If we pass inland, say, tothe Valley of the

Thames, the ‘high’ and ‘low level’ gravels, the brick- -

earths, and other fluviatile or subaerial deposits are ob-
served by the geologist thickening here and thinning out
there, and apparently intercalated with such subtle compli-
cation that their stratigraphical extraction satisfactorily
has been admitted by leading English geologists to be a
matter of great difficulty. In some cases the remains of the
older epoch are exhibited at the high level, while the more
modern forms occur in the lower. The consequence of all
this has been that it has long been a point of accepted belief
among English geologists, that the long-haired and woolly
Mammoth of Siberia had lived back to be a contemporary of
the Mastodon in the Crag; while the Hippopotamus major
had lived on from the period of the submerged forest, if not
from the Crag, to be a contemporary of the Mammoth, the
Siberian Rhinoceros, and the Musk Ox, upon the superficial
gravel of the Valley of the Thames.

The obvious way of dealing with cases of this nature is
to examine the conditions where they are most simple, and
then to apply the results to the more complex instances. I
shall adopt this course in the remarks which I have to offer
on the European fossil Elephants; and first in regard to the
Pliocene forms.

The most instructive instance with which I am acquainted
of the occurrence of the greatest number of Proboscidean
fossil species in the same deposit, under circumstances that
_admit of no doubt as to their common age and association,
is one to which I have referred in the former part of
this communication, namely, that of the Mastodon (Tetra-
lophodon) Arvernensis described by E. Sismonda. The entire
skeleton of the animal, spread out, was disclosed by a
railway cutting between Dusino and Villafranca, at a depth
of about twenty-six feet below the surface. In the same
locality and in the same stratum, but a little apart from
the skeleton of the Mastodon, were found fossil grinders of
HE. (Loxodon) meridionalis, Rhinoceros leptorhinus,! with stags’
horns; and close upon the surface the skull of a Lagomys.
In the fluvio-lacustrine matrix, along with these remains,
were found species of Unio, Helix, Paludina, and Clausilia.

' Subsequently R. Ztruscus, Fule.—[Ep.]

Vy
oe

192 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

The three first Dr. Eugenio Sismonda refers to Unio pictorum,
Helix lactea, and Paludina lenta, as the nearest approxi-
mation; and the Clausilia he considers to be an undescribed
species, which he has named Clausilia Mastodontofila. By the
kind permission of Professor Angelo Sismonda, I was enabled
to examine the Elephant remains minutely. -The most im-
portant of these was a lower jaw showing both rami, and the
ereater part of the beak of the symphysis, with the penulti-
mate and last molars both present on the right side. The
last true molar exhibited only eleven plates, besides the front
and back talons, with the distinctive characters of Loxodon
meridionalis strongly marked, namely, the low cipher of the
ridge-formula, very thick unplaited enamel, and the ridges,
where intact, comparatively low, 7.e. their height not much
exceeding the width of the crown. At another point in the
same stratum, near San Paolo, the railway excavations
brought to light abundant remains of EH. (Huelephas) antiquus ;
among these were the last upper and last lower molars of the
same individual, a large adult. They all presented in a
marked manner the character of this species, namely,
numerous plates, high and approximated as in the Indian
Hlephant, the enamel much crimped, the discs of wear
showing a tendency to angular expansion in the middle.
The upper molars, although partly mutilated behind, still
presented nineteen plates, of which the intact ones attained
seven and a half inches of vertical height, with a width of
three and a half inches to the crown. The right lower molar,
although mutilated, still showed twenty plates. Another
luge detached specimen of the last upper molar, right side,
from another animal, although not quite entire in front,
presented twenty-four plates, with the enormous height of
eight inches, the worn surface of the crown showing all the
distinctive characters of the species.'_ These teeth agreed in

1 This specimen was carefully ex- | expansion into angular loops, but a little
amined and described by Dr. Falconer | widening, with more crimping; (d) the
in July 1856 (see page 187). Five years | discs not so wide as in Indian elephant
later, in April 1861, he again examined | or in 2. antiguus—not much wider than
this specimen, and made the following | in some varieties of E. primigenius; (e)

- note: ‘The huge last upper molar (No.7 | no marginal reflexion of the cornua of
blue ticket of Museum Catalogue), which | the discs; (/) the enamel-plates thicker
in my old notes I had taken for a certain | than in £. primigenius, but not so thick
E. antiquus, strikes me now with consi- | as-in £. antiquus. On the whole, this
derable doubt. The old description in my | specimen reminds me of some of those
old notes holds good generally ; but the | doubtful ones which I saw at Rome with
following points now strike me: (a) the | Sigs. Ponzi and Ceselli, and with Carlo
enormous width of crown—nothing like | Strozzi at Leghorn and Pisa, coming
it do I ever remember in £. antiguus;| very near to E. Armeniacus, and left
(5) enamel grooved on the cement side, | undecided’ On the same date, the fol-
but very little true crimping, and that | lowing note is entered in the Note-book :
only near the middle; (c) no true mesial | ‘ On going over the elephant molars and

Wa

THEIR GEOLOGICAL AGE. 193
every important respect, with large grinders of Huelephas
antiquus from Grays Thurrock, in the Valley of the Thames,
which are preserved in the Museum of Practical Geology in
Jermyn Street. In this case there is an unquestionable in-
stance of the occurrence of three fossil species of Probosci-
deans, in nearly contiguous points of the same Pliocene
stratum, and there is no ground to doubt that they were con-
temporaneous members of the same fauna. Near Florence,
in the same district of the Astesan, and in the same deposit,
tusks and molar teeth belonging to five or six individuals of
Tetralophodon Arvernensis, with jaws of Rhinoceros leptor-
hinus,'teeth of Hippopotamus and Tapir (?), were dug up mixed
with Helices, Paludinas, and Clausilias. Professor A. Sis-
monda, Dr. Bellardi, and Signor B. Gastaldi, all accomplished
geologists intimately acquainted with the country, assured
me that all these remains were yielded by the same Pliocene
alluvial strata. Teeth of another Proboscidean species, Tvri-
lophodon Borsoni, at first taken for the Mastodon of the Ohio,
to which it is closely allied, were described by Borson from
the same Pliocene alluvium, in the hills near Villanova, also
in the Astesan. I was enabled to examine the original speci-
men in the Museum at Turin. As stated in the previous part
of this communication (p. 14), the specific distinctness of this
form has been satisfactorily proved by the labours of Pomel
and other French paleontologists, from teeth discovered in
the department of the Haute-Sadne, and other localities in
France. Wherever it has been met with, the remains of this
species are exceedingly rare, and as yet they have only been
sparingly observed in Italy.

In none of the Astesan localities, so far as I am aware,
have any well-determined remains of Loxodon priscus been dis-
covered up to the present time. But a very fine and conclu-
Sive specimen of this rare species exists in the Natural History
Museum at Milan, which I had the opportunity of carefully
examining, through the obliging permission of Dr. Emilio
Cornalia (see p.101). It consists of a nearly entire last molar,
lower jaw, left side, in a beautiful state of preservation, being
deficient only in the anterior talon, and a portion of the first
ridge, borne by the anterior fang. The crown shows twelve

comparing them with my old notes (see
antea, p. 187), I am struck with the dif-
ference of my present impressions in
respect to the predominance of LF. anti-
quus, and have therefore thought it best
to go over the whole of the specimens
again for fresh investigation.’ Dr. F.,
however, was still convinced of the dis-
tinctness of the specimens from the

VOL, II.

Elephas meridionalis of the Val d’Arno,
&e. Two years later, in 1863 (in his
memoir on £. Colwmbi), he expressed the
opinion that many of the specimens in
the Turin Museum, and also at Rome,
belonged to H. Armeniacus. (See pp.
| 249-50.)—[Ep.]
1 Written in 1857.

Etruscus, Fale.—[{Ep.]

Subsequently 2.

O

194 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

plates, all more or less worn. The discs of wear have the broad
rhomboidal expansion in the middle, which is so character-
istic of the existing African Elephant, with thick enamel,
considerably grooved or crimped on the outer surface (that
which is in contact with the cement). The enamel-edges
project high above the cement, in some of the plates to the
extent of seven-tenths of an inch. The tooth measures twelve
inches in extreme length, including twelve plates within this
space, being an average of one inch to each plate. While
engaged on the examination of this specimen I had with me an
accurate drawing, of the natural size, of the Grays Thurrock
specimen of the same species, Loxodon priscus, figured in
the ‘ Fauna Antiqua Sivalensis,’ which enabled me to institute
a rigorous comparison, and I found that the two specimens
agreed in the closest manner, making allowance for their
different stages of wear. Taking into account the constancy
and significance of the rhomboid expansion in the African
Elephant, as a specific distinction, and the mcompatibility
of the character above indicated with either Loxodon meri-
dionalis or Euelephas antiquus, together with its close agree-
ment with the Grays Thurrock specimens, no doubt remained
in my mind that the Milan specimen belonged to Loxodon
priscus. The mineral condition was equally conclusive of
its being an undoubted fossil. That molar was discovered
on Monte Serbaro, in the district of Romagnano, and valley
of the Pantena, near Verona. In this locality there is a
rich deposit of Mammalian bones, of the Sub-Apennine
period, which has been described by Forbes. Elephant re-
mains have been discovered in it in great abundance. Cuvier
has figured a lower jaw of very large dimensions, which, so far
as can be determined from an imperfect drawing, would appear
to belong to Loxodon meridionalis (Oss. Foss., tom. i., Plate
IX. fig. 8), but it may have belonged to Loxodon priscus.
Cuvier states that the Monte Serbaro Elephant remains,
which exist in the Paris Museum, indicate an animal at least
fifteen feet high.

I have already noticed, in the earlier part of this paper,
the Elephant remains discovered by Cortesi on Monte Pul-
enasco, near Piacenza (p. 113). The principal palate-speci-
men is very badly represented in Cortesi’s figure; but I ex-
amined it minutely, fresh after a visit to the Museum at
Florence, and I found all the characters agree exactly with
those of the Loxodon meridionalis of the Val d’ Arno.

These various localities, in Piedmont and Lombardy, re-
present the same Pliocene alluvium of the Sub-Apennine
period. It would be expecting too much, and contrary to
what we find elsewhere in similar cases, to look for the occur-

THEIR GEOLOGICAL AGE. 195

rence of all the fossil forms throughout ; for the points where
they are disclosed are little more than specks on the common
area. But the facts would seem to me to be as conclusive
as most of those on which we do not hesitate to reason in
geology, that five species of Hlephantine Proboscidea were
co-existent at the same epoch in Piedmont and Lombardy ;
namely, two Mastodons, T'rilophodon Borsoni and Tetralophodon
Arvernensis, and three Hlephants, Loxodon meridionalis, Loxo-
don priscus, and Huelephas antiquus.

In the earlier part of this communication I have already
referred to the association of Pliocene Mammalia in the Val
d’Arno. The Proboscideans which are found most abundant
there being Tetralophodon Arvernensis and Loxodon meridio-
nalis, along with Rhinoceros leptorhinus,! Hippopotamus major,
and other forms, herbivorous and carnivorous, characteristic
of the Sub-Apennine period.

In the Roman States, Elephant remains have been met
with in strata of the same age, mingled with those of Rhi-
noceros leptorhinus, Hippopotamus major, and various other
Mammalia. The Hlephants have hitherto been referred by
all or most authorities to the mammoth, Hleph. primigenius.
But Cuvier, under that name, has figured and described a
lower jaw from Monte Verde, which I refer to Loxodon meridi-
onalis.2 Professor Ponzi, of Rome, has given an enumeration
of the genera and species in a communication to the Scien-
tific Association of Italy, in 1847. He refers all the Elephant
remains to Hleph. primigenius, and divides the fauna into
three groups, Pliocene, newer Pliocene and Modern. No
specimens of Mastodon remains, so far as I am aware, have
been described from this part of Italy.

The only other foreign locality to which I shall refer for
a term of comparison is one of great interest in France,
in the Beauce, near Chartres, brought to light by a railway-
cutting within the last few years. Chartres is situated on the
plateau which separates the water-sheds of the Seine and of
the Loire. The line of excavation passed through a fluviatile
deposit of gravel, sandy loam, and brick earth, abounding in
remains of Hlephants, Rhinoceros, Hippopotamus, and other
large herbivorous quadrupeds.

A fine collection of these remains has been formed in the
Duc de Luynes’ Museum at Chateau Dampierre, by Mons.
Gory, to whose obliging kindness and liberality I was indebted
for an opportunity of examining them in detail. The Ele-
_ phant remains are numerous, consisting of molars, tusks,
bones of the extremities, and other parts of the skeleton.

1 See note, p. 193.—[ Ep. ]
? Ossemens Fossiles, Ed, 3me., tom. i. p. 165, Pl. ix. fig. 3,
02

195 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

The molars were all of adult or well-grown animals, and
must have been derived from many individuals. I observed
no milk-teeth among them. Many of the teeth are in fine
preservation, thus admitting of easy specific identification.
With one doubtful exception they were all of Loxodon meridi-
onalis. The ridge-plates on the last lower molar, where
entire, never exceeded thirteen or fourteen in number. The
other characteristic marks of the Val d’Arno species are all
well exhibited; namely, thick distinct digitations to the
plates, with very thick unplaited enamel, and the height of
the intact plates not much exceeding the width of the crown.
The doubtful specimen is a fragment of a very old molar of
the lower jaw, worn down to the base, which in some of the
plates presented discs of wear having a rhomboid mesial ex-
pansion, four-fifths of an inch broad, very much like those
of Loxodon priscus. But the very advanced wear and. frag-
mentary condition of this specimen left the specific de-
termination of it uncertain. The associated herbivorous
mammals were Rhinoceros leptorhinus,! Hippopotamus major,
with large bovine and cervine ruminants, with terete-branched
antlers, of species which during my short visit I was un-
able to determine. Mons. Du Bois Villette, of Chartres, has
made an extensive collection of these remains, and some very
fine molars of Loxodon meridionalis presented by him are dis-
played in the Museum of the Ecole des Mines at Paris. One
to which I would refer is a superb specimen of a last upper
molar of the left side, showing eleven of the ridges worn.
The association of the mammalia, so far as it extends, in this
case is clearly that of a Pliocene Fauna; but the special
interest in its bearing on the English deposit les in this,
that while the mineral characters of the bed and its contained
fossils present a close general agreement with those of Grays
Thurrock, the prevailing Elephant is a species which, so far
as I am aware, has not yet been detected in the brick-earth
deposits of the Valley of the Thames.

T shall now consider the general conditions of the principal
deposits in which Elephant remains occur in England. In
the preceding part of this paper I have stated the grounds
for considering the Crag Mastodon to be a Pliocene species,
and the associated contemporaneous mammals, so far as they
have been well determined, as being likewise Pliocene. As
the Crag constitutes the lowest geological level in which
Elephant remains occur in England, it is of importance that
the species should be carefully identified. A characteristic
specimen of Huelephas antiquus, from the Red Crag of Felix-

1 See note, p. 193.—[ Ep.]

THEIR GEOLOGICAL AGE, 197

stow, exists in the Geological Society’s collection ; there are
others in the Museum of Practical Geology, and in various col-
lections in Norfolk which I have examined. Of Loxodon meri-
dionalis a very characteristic and decisive specimen is repre-
sented in the ‘Fauna Antiqua Sivalensis’ (Plate XIV. B. figs.
18 and 18a).! It was discovered in the Mammaliferous Crag
near Harwich, and consists of the last lower molar, right
side, having twelve ridges, along a length of twelve inches.
The ridges are comparatively low, their height, where intact,
not much exceeding the width of the crown. The digitations
are thick and distinct for a considerable distance below their
apices; and the plates of enamel very thick. The ridges are
so thick that they average one to an inch. In the aggregate
the characters approach very closely to those of the equi-
valent teeth in Loxvodon planifrons of the Sewalik hills. The
specimen is highly ferruginous, like most of the Crag bone-
remains. Two other specimens, which I have examined from
the Crag, presented by Miss Anna Gurney to the Paris
Museum, are figured by De Blainville. They are ‘ inter-
mediate molars’ of Loxodon meridionalis. Various other
illustrations might be cited, but I confine the references to
figured specimens. These Crag specimens respectively of
Loxodon meridionalis and Euelephas antiquus differ in no
material respect from specimens of the same species above
referred to as occurring in the Italian Pliocenes.

The highest level in which Tetralophodon Arvernensis occurs
in British strata has not yet been accurately ascertained. In
the remarkable case of the entire skeleton recorded by Mr.
Layton as having been discovered at Horstead, ‘ stretched
out between the chalk and the gravel,’ it has not yet been
clearly made out whether the bed is an equivalent of the
Crag or above it. The locality has been examined conjointly
by Messrs. Godwin-Austen, Prestwich, and Morris; and Mr.
Austen assures me that there is no doubt about the authen-
ticity of the case, and that the skeleton was certainly of a
Mastodon. The bones of the herbivorous mammals occurring
in the Crag are invariably presented in a condition more or
less fragmentary, as if they had been washed out of a pre-
existing Pliocene land and deposited in a reconstructed sea-
bottom. A slight oscillation of the relative levels of the land
and sea would admit of this being effected within a com-
paratively limited period. Having regard to the undisturbed
and compact state in which the Horstead skeleton lay, it may
be inferred that the deposit was either on a higher level than
the Crag, or that the breaching action of the Crag-waves had

1 See vol. i. p. 447, and vol. ii. Pl. viii. fig. 1—[Ep.]

198 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

not reached it. No Mastodon molars have yet been dis-
covered in situ in the fluviatile and lacustrine strata on the
Norfolk coast above the Crag; but after a careful perusal of
the various recorded instances where teeth have been met
with on the beach, below undermined cliffs, I am not satis-
fied that they were in every case derived from the Crag and
not from the superincumbent Clay or Forest-bed. The age,
so long assigned to the Crag Mastodon, naturally disposed
geologists and collectors to refer the doubtful cases to the
oldest strata.

_ The next beds in the order of superposition, yielding Ele-
phant remains, are the freshwater deposits composing the
mud-cliffs of Eastern Norfolk, which have been examined in
great detail by Sir Charles Lyell and other English geologists ;
while the mammalian remains have been investigated by Pro-
fessor Owen. I have had opportunities of studying a very
large number of the Proboscidean fossils of these beds, in the
collections at Norwich, and in the various collections, public
or private, in London, and I have lately been indebted to the
obliging kindness of the Reverend John Gunn for a fine
series which he has transmitted to me for illustration on the
present occasion. So far as my observation has gone, the
species have invariably proved to be either Loxodon meridi-
onalis or Huelephas antiquus, and among the associated pachy-
derms Rhinoceros leptorhinus and Hippopotamus major. A
vast number of Elephant grinders have been found along the
coast, near Happisburgh, Mundesley, and Cromer. The col-
lection of Miss Gurney has acquired a celebrity for its richness
in these remains. Molars of Huelephas primigenius, the Si-
berian Mammoth, are occasionally picked up upon the beach,
mingled with the other species, especially below the Cromer
cliffs ; but the admixture would seem to be merely adven-
titious, from the Mammoth grinders having fallen out of
the superincumbent drift or gravel. They are usually ina
different mineral condition from the more ancient remains,
and readily distinguishable.

The Elephant molars, from the Norfolk mud-cliffs, both of
Loxodon meridionalis and Euelephas antiquus, occur in two
very different states of mineralization. Those from the gravel-
banks of Happisburgh and Mundesley exhibit generally a
reddish-brown colour, from ferruginous infiltration; and
when they have been subjected to the action of the sea they
usually show more or less of a vitreous polish. They are hard,
heavy, and compact, with a specific gravity of 2°73,as compared
with 2-08 in the recent molar of the Indian Elephant, while
erinders from the blue clay of Cromer adhere to the tongue
and give out a strong ammoniacal odour when burnt. I have

THEIR GEOLOGICAL AGE. 199

as yet seen no well-marked remain of Loxodon priscus from
these beds, which I regard, from the mammalian contents,
as being of the same age as the Sub-Apennine Pliocene
alluvia.

The next Elephant-yielding bed which I shall notice is
the ‘ mud-deposit’ of Brackelsham Bay and Pagham Harbour,
described by Mr. Dixon, and ably investigated lately by Mr.
Godwin-Austen. It consists of a fine dark-coloured sandy
mud, of great firmness and tenacity, lying upon the Eocene
blue clay, and in some places attaining a thickness which Mr.
Austen estimates at eighteen or twenty feet. He enumerates
thirty-eight species of marine Testacea, and ‘the character of
the whole assemblage,’ he states, ‘is essentially southern and
western.’ He considers that the ‘mud-bed’ is either an old
estuary or an ‘enclosed salt-water lagon’ deposit, formed
under conditions of comparatively high temperature. The
only mammalian remains which it was supposed to yield
were those of Hlephas primigenius. But during a late visit to
Brackelsham Bay, in company with Sir Charles Lyell and Mr.
Godwin-Austen, the Rev. Robert Everest and myself, after a
few minutes’ search, picked up a well-marked tibia of a small
Equus and another bone. The dark-coloured bones are not
obvious in the dark mud, and the attention of the local col-
lectors has been habitually so much directed to looking out for
fine specimens of Pholas and Lutraria, that the bone-remains
have been to a great extent overlooked. When a closer
search is instituted abundant mammalian remains may be
expected. A considerable collection of Hlephant-grinders
from the ‘mud-deposit’ exist in the Chichester Museum.
They are mostly in fine preservation, and after a careful ex-
amination the whole of them proved to be of Huelephas anti-
quus (pp. 181-4). I found no teeth of the Mammoth among
them. In this case the mammalian evidence, so far as it
goes, agrees with the shells in indicating the ‘mud-deposit ’
to be of the Sub-Apennine age, and it may prove to be one
of the oldest Pliocene beds in England.

I shall next consider the fluviatile deposits of the Valley of
the Thames, in relation to their Elephantine remains. These
freshwater strata occur at various points along the valley,
such as Brentford, Grays Thurrock in Essex, and other
localities. ‘They consist of sand, gravel, and loam, from
60 to 100 feet thick, and often form a terrace on each side of
the valley, rising to a much higher level than a vast bed of
more modern gravel.’! This ochreous gravel, says Sir Charles
Lyell, whose description I quote, extends from east to west,

1 Annals of Nat. Hist. 1836, vol. ix. p. 261.

200 BRITISH AND EUROPEAN FOSSIL ELEPHANTS,

from above Maidenhead to the sea, for a distance of fifty
miles, having a width varying from two to nine miles, and
a thickness of five to fifteen feet. It is spread out over the
lower levels of the valley, containing in it the remains of
Arctic animals. Sir Charles Lyell considers it quite clear
that it was deposited after the fluviatile beds were formed.
The most instructive of these freshwater deposits, as having
yielded the largest number of mammalian remains, is that of
‘Grays Thurrock,’ described by Mr. Morris in 1836. The
section is exposed to a depth of about forty-seven feet, consist-
ing of beds of mould, loam, and at the bottom sandy clay
abounding in freshwater shells. Numerous Elephant remains,
with those of other mammalia, have been discovered in the
lowest beds containing the shells; they are rarely found above
it. Among these was a nearly entire skeleton of an Hlephant,
of which two large grinders were preserved. One of these
is now in the British Museum; the other was in the pos-
session of Mr. R. Meeson, the proprietor of the brick-field. The
first is represented in the ‘Fauna Antiqua Sivalensis,’ Plate
XIV. fig. 7.1. Unfortunately it is mutilated at the anterior
extremity, depriving us of direct evidence as to the complete
number of the crown-ridges. The crown was in full wear,
showing seven ridges and a talon. One or two ridges in front,
supported by the large anterior fang, are wanting. The
tooth is inferred to have been the penultimate or second true
molar, lower jaw, left side. It measures 7°8 inches, comprising
in all eight plates (talon inclusive), being an average of about
one plate to an inch. The discs of wear present a rhomb-
shaped outline, wide and angular expansion in the middle,
and thick enamel, closely resembling the same characters in
the African Elephant. The specimen was sawn up to expose
the internal structure, and the relative proportions of the
different constituent materials exhibit the same general re-
semblance to the African Hlephant. The mineral condition
of the specimen puts it beyond question that it is an un-
doubted fossil. I have stated above how closely I found the
Monte Serbaro fossil grinder correspond in all its characters
with the Grays Thurrock specimen. Another fossil grinder
from the same deposit is represented in the ‘ Fauna Antiqua
Sivalensis,’? Plate XIV. fig. 6. It is mutilated at both extremi-
ties, showing only six discs, less advanced in wear.’ All these
specimens I refer to Loxodon priscus. Remains of Huelephas
antiquus also occur in the same bed at Grays Thurrock, dif-
fering in no material respect from grinders of the same
species, occurring in the Astesan, in the Crag and in the Nor-

1 See vol. i. p. 441, and vol. ii. Pl. vii. figs. 1 and 2.—[Ep.]
2 See vol.i. p. 441.—[Ep.]

THEIR GEOLOGICAL AGE. 201

folk freshwater Clays. A superb specimen of the last upper
molar, left side, of this species, from Grays Thurrock, show-
ing sixteen of the anterior plates (the back portion being
wanting), is deposited in the Museum of Practical Geology.
Tt presents all the marked characters of the species. In
1846 I made an excursion along with the late Professor
Kdward Forbes to Grays, with the express object of examin-
ing the Mammalian contents of the bed, and on that occasion
I saw other specimens of Huelephas antiquus, which had been
recently exhumed. No remains referable to Lovodon meridi-
onalis from any of the fluviatile deposits of this age, in the
Valley of the Thames, have as yet come under my observation,
although it occurs in the Beauce at Chartres, along with the
same association of Mammals. A similar deposit occurs near
Brentford, the fossil remains of which were first brought to
notice by Mr. William Kirby Trimmer, in a paper in the Philo-
sophical Transactions for 1813. Excellent figures, by Basire,
are given of two fossil Elephant grinders, the one of the upper
and the other of the lower jaw, which Mr. Trimmer refers
respectively to the Indian and African Elephants. They are
certainly not of the Mammoth, and appear to me to belong
to Huelephas antiquus. There is an old specimen in the
Museum of the Royal College of Surgeons, No. 570 of the
Catalogue, presented by Sir Joseph Banks, from Brentford,
which belongs to the same species. The same collection
contains other remains of this species, from Walton in Essex,
, presented by Sir William Blizard, and from Ilford, by Mr. John
Gibson, besides specimens of which the localities are unre-
corded.

The larger Mammalia associated with the Elephants in
Grays Thurrock are Rhinoceros leptorhinus, Hippopotamus
major, Horse, Cervus, Bovine Ruminants, and Ursus; at
Brentford, Rhinoceros leptorhinus, Hippopotamus major, Ox,

‘and Deer. Mr. Morris, in a later paper, describes a section
near Brentford yielded by a railway cutting 100 yards north
of Kew Bridge. Hight layers or strata, more or less distinct,
are exposed, the five inferior contaming Mammalian remains,
of which Mr. Morris gives the following enumeration :

1. Hlephas primigenius. 5. Bos longifrons.
2. Rhinoceros tichorhinus. 6. Cervus Tarandus.
3. Hippopotamus major. 7. Cervus Elaphus.
4, Bison priscus. 8. Felis spelza.

I consider the association here indicated, assuming that
the identifications are exact, to be in the highest degree im-
probable, as of animals that were coexistent at the same

202 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

period. It would seem intelligible only on the supposition
that the section includes beds of different ages. I have never
seen a tooth of the true Mammoth among those of Loxodon
priscus and Huelephas antiquus, from the fluviatile deposits in
the Valley of the Thames.

It is beyond the scope of this communication to diseuss
the nature of the Elephant remains occurring elsewhere in
deposits more or less analogous to those of the Valley of the
Thames, such as in the Valleys of the Stour, the Medway, the
Arun, the Avon, the Severn, and other localities in England.
T have thought it sufficient to restrict the inquiries on the pre-
sent occasion to the range between the Crag on the one hand
and the Thames Valley fluviatiles on the other; and to confine
the comparison to a few well-marked and significant forms.
Regarded by the light of their common mammalian fauna,
they appear to me alike to belong to the same Pliocene age.

The detail, into which I have entered in the preceding
remarks upon the range and associated mammalia of the
other European fossil Elephants, relieves me from the neces-
sity of dwelling much upon the conditions under which the
last and most modern species, Huelephas primigenius, or the
true Mammoth of Siberia, is met with in British strata.
The thin, parallel, closely arranged, and wncrimped plates of
enamel, with the attenuation of the other dentary consti-
tuents, disposed like the teeth of a fine comb, readily distin-
guish, in the view here taken, the fossil grinder of the
Mammoth from those of the other species. And wherever
teeth presenting these characters are observed it will, I
believe, be invariably found that they occur in deposits of a
more recent geological date than the Pliocene fluviatile
beds; and that they are never mixed up, except adventi-
tiously, in the same fauna with the other species. In the
preceding observations I have referred to the association of
species which is presented by the wide-spread loess of the
Rhine, and the superficial drift of the plains of Northern ~
Germany, where Mammoth remains are found in company
with those of Rhinoceros antiquitatis, horse, musk-ox, rein-
deer, and other northern forms. In the plains of Northern
Russia and Siberia, to this group the rare Elasmotherium
is added, which, there are plausible grounds for believing,
ranged as far asthe Rhine. The characteristic Sub-Apennine
forms, such as the Rhinoceros leptorhinus,! Hippopotamus major,
or any of the three Pliocene Elephants, are nowhere dis-
covered among them. In like manner, the remains of the
Mammoth occur in the superficial gravels which are so

1 See page 206, note—l[Eb. |

THEIR GEOLOGICAL AGE. 203

generally spread over England, and in the erratic beds to
which the names of ‘ Northern Drift’ and ‘Till’ are applied.
The fauna would appear to consist of the same forms as in
the northern drift of Germany, namely, Rhinoceros antiqui-
tatis, equus, musk-ox, rein-deer, and other ruminants. Unlike
the Pliocene elephants, of which entire skeletons, with the
bones in more or less contiguity, have been observed in the
freshwater fluviatiles, the Mammoth remains generally
occur detached and scattered in the gravel, and I think it
worthy of inquiry whether a well-authenticated case is on
record of the bones of a Mammoth skeleton, collected toge-
ther in one locality in British Post-Pliocene deposits, and
if so, under what circumstances the case was presented.
Unrolled crania of the Mammoth have been procured from
the loess of the Valley of the Rhine. In one instance which
I have examined, at Mannheim, the skull is very perfect,
the bones fresh-looking and full of undecomposed gela-
tine. Buckland states, ‘ that at Kingsland, near Hoxton, in
1806, an entire elephant’s skull was discovered, containing
two tusks of enormous length, as wellas the grinding teeth ;’
but no particulars are given by which the species would be
determinable. Detached Mammoth remains have been found
in the Till in Scotland. At Clifton Hall, between Edinburgh
and Falkirk, a Mammoth tusk was discovered so fresh that
it was fit for the manufacture of chessmen. (Werner. Trans.
vol. iv. p. 58.) Another was discovered in a mass of diluvial
clay at Kilmaurs in Ayrshire, near the water of Carmel.
Mr. Smith, of Jordan Hill (in his Memoir on the ‘ Changes
in the Relative Levels of the Sea and Land,’ p.35), cites several
other instances of Mammoth remains from the Till in the
southern and western parts of Scotland. It would be tedious
to enumerate all the localities where they have been discovered
in England. I will merely mention a few cases of special
interest. One of these is the ‘elephant-bed’ of Dr. Mantell
at Brighton. I have examined a grinder from his collection
now in the Jermyn Street Museum; it is labelled as being
from the chalk-rubble, is very fresh-looking, and presents
the characters of the Mammoth in a marked manner.
Another locality of interest, and demanding great care in
the observation, is the series of gravels described by Mr.
Godwin-Austen, as lying upon the ‘mud-bed’ of Brackel-
sham and Pagham Bay. Here the teeth are constantly ex-
posed to being washed out of the gravel and embedded in
the mud-bed along with the Huelephas antiquus molars which
belong to it. I have seen very characteristic specimens of
Mammoth grinders from Kingsland.

As regards the relative abundance of remains of the four

204 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

fossil species, according to my observation, in England they
would run in the following order.

1. E. (Huelephas) antiquus.

2. H. (Huelephas) primigenius.
3. EH. (Loxodon) meridionalis.
4. KE. (Loxodon) priscus.

The teeth of Euelephas antiquus are. excessively abundant
in the freshwater deposits on the Norfolk coast. Of the
two thousand Elephant grinders which Mr. Woodward esti-
mates to have been dredged up within thirteen years, from
the oyster-bed near Happisburgh, I believe that by far the
largest number belonged to this species. The next, in point
of number, are those of the true Mammoth, from the wide-
spread drift and gravel-beds. Teeth of Loxodon meridionalis
are much less frequent. They occur in the Norwich Crag,
and at Happisburgh and Mundesley. The only inland speci-
men attributable to this species which I have as yet met
with is the Staffordshire specimen described by Parkinson,
and of which there is a beautiful figure in the British Fossil
Mammalia (fig. 93, p. 239). Professor Owen, in describing
it, states that he had ‘seen a very similar molar of the Mam-
moth from the Norfolk freshwater deposits, in the collection
of Mr. Fitch of Norwich’ (loc. cit. p. 240). The Elephant
remains found so abundantly at Cropthorne, and at Deptford
and Eckington, in the Valley of the Avon, and along with
bones of Hippopotamus, &c., and freshwater shells by Mr.
Strickland, have not yet been precisely determined. Sup-
posing the record by Parkinson to be exact, I shall not be
surprised if among the Avon remains teeth of Loxodon
meridionalis turn up, presenting a case analogous to that near
Chartres, already referred to. The rarest species of all is
Loxodon priscus, of which only one English locality, namely,
Grays Thurrock, is known; and the specimens which
have been found there are all but unique. If this exces-
sive rarity be considered as furnishing any grounds for
questioning the reality of this extinct species, as distinct
from the other, I refer to the parallel case of the solitary
molar of Trilophodon Borsom, found in the Astesan. This
form was never confounded with Tetralophodon Arvernensis,
but with Trilophodon Ohioticus of North America. It is now
universally accepted by the modern palzontologists of France,
where it occurs in most abundance, although still rare. The
presence of Loxodon priscus in British strata is of great
interest, and I would recommend it as an object of praise-

1 Geological Transactions, 2nd ser, vol. iv. p. 555,

ey

THEIR GEOLOGICAL AGE. 205

worthy research to English geologists and collectors, who
have access to local museums, in which the remains of the
Essex fluviatile deposits are preserved. (See page 251, note 1.)

Tt has been objected to me by naturalist friends of emi-
nence, and for whose opinions I entertain the highest respect,
that there is a prima facie improbability that so many
Proboscidean species should have co-existed in the same
Pliocene fauna, namely, two Mastodons and three Htephants.
My reply is: First, go and examine the evidence in the
Astesan, where abundant remains of three of the species
occur within a limited area in the same alluvium, mixed up
with the same freshwater shells; then examine the exten-
sion of the same deposits near Piacenza and Verona; then
the deposits of the Val d’Arno; and compare the association
of mammals in each. Secondly, the Miocene fossil fauna of
the Sewalik hills (exclusive of Ava and of the Nerbudda
Pliocene beds) presents a parallel case, where at least six
fossil species of Proboscideans are met with in beds of the
same age, in the same matrix, the same mineral condition,
and with the same associates; namely, two Mastodons,
Tetraloph. latidens and Tetraloph. Sivalensis; two Stegodons,
Steg. bombifrons and Steg. insignis; one Loxodon, Lox. plani-
frons; and one Euelephas, Huel. Hysudricus. Of five of these
forms, crania exist in the British Museum, besides the de-
tails of their dentition; and the evidence of their distinct-
ness, founded on these crania, is quite as conclusive as that
upon which paleontologists would distinguish the existing
Elephant from the Mastodon of North America, by the form
of their skulls.

From the consideration of the various facts which I have
passed under review, I have been led to the conclusion that
the same mammalian fauna ranged throughout, from the
Crag up to the Thames Valley fluviatile deposits, and that
any division of them into older and newer Pliocenes, as
Pleistocenes and Post-Pliocenes, is untenable. It would seem
to me that in all the investigations where the classification of
the newer Tertiaries is concerned, too much stress has been
laid on the shell-evidence, while that which may be derived
from the mammalia has been ina great measure either over-
looked or undervalued. Like the artificial arrangement of
Linnzus in plants, the shell-evidence has furnished a ready
and convenient means of arranging the strata conventionally
in successive order, regarded from one aspect; but it can
never satisfy the requirements of a natural philosophical
system, where the solution of the conditions of the higher
forms of terrestrial life constitutes the most important part
of the problem. The evidence must be taken from every side,

206 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

and weighed according to the relative value of the facts. That
which is furnished by the mollusca has been everywhere
abundantly investigated. JI shall now briefly examine the
mammalian aspect. In the Italian Pliocenes, where the fauna
is most concentrated, two Mastodons and three Hlephants,
with probably two Rhinoceroses and one Hippopotamus,
are met with as leading and characteristic forms. All the
Proboscidean forms are not uniformly distributed through-
out; sometimes three or four, and sometimes two, or only
one, of the species are met with, but in every instance accom-
panied by Rhinoceros leptorhinus, and Hippopotamus major.
The presence or absence of the leading species of Mastodon,
namely, Tetralophodon Arvernensis, does not appear to be of
positive significance, since it has either not been found, or
only very rarely, in the Pliocenes around Rome, where the
same fossil elephants, Rhinoceros and Hippopotamus, with their
associates, occur, as In the Sub-Apennine beds proper. In
the English deposits, taking the range from the Crag up to
the fluviatile beds of Grays Thurrock, four of these fossil Pro-
boscidea, with the same Rhinoceros leptorhinus and Hippopota-
mus major, are met with.' The two latter pachyderms occur
uniformly throughout from the level of the submerged Norfolk
lignite bed on to Grays Thurrock. Of the Proboscidean forms,
Huelephas antiquus occurs everywhere from the Red and Nor-
wich crags, through the submerged forest and Norfolk Lacus-
trine clays ; the Brackelsham ‘ mud-deposit ;’ and the Thames
fluviatiles at Grays Thurrock, Brentford, and other localities.
Loxodon meridionalis occurs in the Crag and in the Norfolk
clays along with Huelephas antiquus. But it has not yet been
met with in the Thames fluviatiles, although occurring in a
corresponding deposit at Chartres in France, and appa-
rentlyin some fluviatile beds in Staffordshire or the contiguous
central counties. Lowodon priscus has only been found at
Grays Thurrock in company with Huelephas antiquus; and
Tetralophodon Arvernensis, in the Red and Norwich Crag,
along with Loxodon meridionalis and Huelephas antiquus.
There is no obvious reason in nature why more importance
should be attributed to any one of these Proboscidean forms
than to another, as tests of geological age. The Crags con-
tain one Mastodon and two Elephants, common to them and
the Sub-Apennine beds of the Astesan; while Grays Thur-
rock contains two Elephants, with Rhinoceros leptorhinus and
Hippopotamus major, common to it and the beds of the
Astesan and Lombardy. These facts would seem to me to
be conclusive that all these Pliocene alluviums supported the

1 This was written in 1857, before the Author separated R. Htruscus and R.
hemitechus from the R. leptorhinus of Cuvier.—[ Ep. ]

THEIR GEOLOGICAL AGE. 207

same mammalian fauna, and that they belonged to one geo-
logical period. It is not meant to be implied that all the
species of the fauna ranged equally everywhere throughout
the area; some in all probability predominated in the south,
others in the north ; but Ican see no grounds for entertaining
the belief that they were broken up into groups in successive
periods, corresponding with an older Pliocene, Pleistocene,
and Post-Pliocene division.

In speculating about the probable climatal conditions on the
land in Europe during the Pliocene period, one of the fossil
pachyderms has appeared to me capable of throwing more
light than all the others, namely, Hippopotamus major. Two
living species of this genus are known, the one Hippopota-
mus anvphibwus, the other the small Hippopotamus Liberensis.
They are both found in the tropical or warmer parts of
Africa. In their habits they are strictly aquatic, plunging
into rivers during the day, and emerging at night to pasture
along the banks. They always hug the margins of the
rivers or lakes, and are not known to make inland journeys
away from them. When they migrate, they either float
with the stream, or, if moving against it, they walk along
the beds of the river, only leaving it for a short distance,
when their course is interrupted by rapids, and replunging
into the stream when the obstacles cease. Wherever they
are found they enjoy open water all the year round. Their
unwieldy heavy form and short limbs are admirably adapted
for their aquatic habits, but unfit them for journeying by
land.

There is no reason to believe that the huge European fossil
Species was in any respect less aquatic in its habits than its
living congeners. Wherever its remains have been discovered
in the greatest abundance and perfection, it has invariably
been along the margins of rivers or great lakes, such as the
Val d’Arno, where the bones of hundreds of individuals have
been observed. It appears to have been spread over nearly
the whole of the Pliocene area of England, since bones and
teeth have been described from the Valleys of the Severn, the
Avon, and the Thames, Kirkdale Cave, Kent’s Hole, and
Durdham Down. The general argument, so ably discussed
by Dr. Fleming, that we cannot predicate in many cases from
the known food and habits of existing species, what the food
and habits of extinct species of the same genus may have
been, will not apply to the fossil Hippopotamus major, which
must have had open water free from ice, if it lived the whole
year round in England. That it was capable of migration
by land more than the existing species we have no grounds
for believing; and if it is argued that there may have been

208 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

large rivers flowing from the south during the Pliocene
period, along the course of which the Hippopotami could have
migrated during winter, the argument might apply to the
population of one or two river-valleys, but it would hardly
extend to the Hippopotami spread over the broad area of
England. In balancing these various considerations, it seems
to me most probable that the Hippopotamus major was a per-
manent resident of the country during the Pliocene period.
This would involve a comparatively warm temperature
throughout the year as late as the deposition of the ‘ Grays
Thurrock’ beds, and the same would seem to be indicated by
the presence of some southern freshwater shells, which are
now extinct in England.

It is at the present time very generally accepted among
geologists that there are good grounds for believing in a con-
tinual refrigeration of climate during the Pliocene period in
Britain. But this can only be entertained as holding good
in regard of the sea. The terrestrial evidence furnished by
the general characters of the mammalian fauna gives no
countenance to a corresponding reduction of temperature on
the land. It may be possible to reconcile the opposed indi-
cations by supposing that while the removal of an ocean
barrier in the form of islands to the north admitted the
gradual migration of Arctic marine forms, and repelled
Mediterranean species to more southern localities, still the
climatal conditions of the land may not have been altered in
the same proportion. One cause of this may be sought for
in the unbroken continuity of land between England and the
Continent during the Pliocene period, which would have
contributed to counterbalance the effects produced by the
decreasing temperature of the sea. f

I shall now briefly review the opinions of some English geolo-
gists who have been more especially engaged in researches upon
the newer Tertiaries. Mr. Searles Wood, in the introduction to
his admirable monograph of the Crag Mollusca, after giving
the division, at the time generally prevalent, of three dif-
ferent periods in the formation, namely, the Miocene coralline
crag, Pliocene red crag, and Pleistocene mammaliferous craq,
states his opinion that, upon the shell evidence, the lacustrine
or fluviatile deposit of Grays, Clacton, and Sutton, were pro-
bably the freshwater equivalents of the Red Crag, while the
Copford deposit may be of a more modern date.

Mr. Prestwich, in his note on the gravel near Maidenhead,
after describing the lithological characters and range of the
mass, refers to the Brentford bed, and to the fossil bones dis-
covered there by Mr. W. Kirby Trimmer, in 1813, as pertain-
ing to it. With regard to the age of the gravel, he states

THEIR GEOLOGICAL AGE. 209

his opinion that ‘it belongs doubtfully to the same period
as the Kingston, Brentford, Kew, and London beds, which I
am inclined to consider as among the most recent of our
drift-deposits, and as posterior to the period of the great
boulder-clay of Norfolk and Suffolk; but the relation of these
beds is nowhere clearly seen :’ and he adds that this ques-
tion of relative age depends on a variety of collateral evi-
dence which he hopes to lay before the Society at a future
period.

It would be presumptuous in me to question the stratigra-
phical results of a distinguished practical geologist, so emi-
nent for his profound knowledge of the tertiary formations as
Mr. Prestwich. But if by the passage I have cited it is
implied that the Brentford fluviatile strata containing the
mammalian remains described by Mr. W. Kirby Trimmer and
other beds of the same age, in the Valley of the Thames, are of
a later age than the boulder-clay, the conclusion is entirely
at variance with the mammalian fossil evidence, and would
involve the startling corollary that, after the Norfolk lacus-
trine beds had been submerged for a long term of the glacial
period, during which enormous deposits of boulder-clay and
till were accumulated above them, the Pliocene mammalian
fauna again reappeared on the emerged surface with a
restoration of all the warm climatal conditions. I believe
that it will be found that, while the Thames Valley fluviatiles
at Grays Thurrock and elsewhere form a terrace on either
side of the valley on a higher level than the ‘valley’ or mam-
moth gravel spread out below at Brentford, or in its imme-
diate vicinity, the gravel-beds are in direct superposition to
the fluviatile strata, and that the mammalian remains found
' there belong to two distinct geological ages, according to the
depth below the surface of the bed from which they are ex-
humed. These upper gravel-beds I believe, with Mr. Prest-
wich, to belong to a date posterior to the boulder-clay and
til. The enumeration of the Brentford fossil mammalia
given by Mr. Morris includes both the Pliocene and the
Post-glacial fauna. I have seen grinders both of Huelephas
antiquus and of the true Mammoth from the Brentford
section.

Mr. Joshua Trimmer, in his memoir entitled, ‘ Generali-
zations on the Erratic Tertiaries of Norfolk,’ among other
results gives the following important conclusion :—That with
regard to mammalian remains, ‘ We have two elephantine
periods, one preceding the submergence of the erratic period,
and the other inhabiting the country at the close of the period
_ of elevation. To the former are to be referred the mammalian
Crag, and the remains of the bone caverns in general; to the

VOL. II. P

210 BRITISH AND EUROPEAN FOSSIL ELEPHANTS.

latter the freshwater beds of the Valley of the Thames, of the
Avon in Worcestershire, of Gaytonthorpe, and of the Biel-
beck in Yorkshire, together with the marine deposit of the
Valley of the Nar.’ This generalization is open to the same
grave objections just now stated—namely, that it is directly
opposed to the mammalian evidence, and would imply the
intercalation of the boulder-clay and till between the Norfolk
lacustrine and the Thames Valley fluviatiles, whereas the
mammalian remains and shells alike would seem clearly to
indicate that the Grays Thurrock beds were deposited ante-
riorly to any portion of the boulder-clay and till. Mr.
Trimmer’s general expression of there having been two Ele-
phantine periods is correct, but not so the positions which
he assigns to them. The Pliocene Elephants preceded the
glacial period, while the Mammoth and its associates inha-
bited the country after the emergence of the land, and in all
probability during the decline of the glacial period.

It was my intention to have entered in some detail on the
circumstances under which Elephant remains are met with
in the bone-caves properly so-called, and in the cavernous
fissures; but the length to which this communication has
already extended deters me from the discussion of so large a
subject on the present occasion. I shall merely observe now
that I believe the caves, like the lacustrine deposits and
gravels, furnish evidence of Elephants of two distinct faunas,
both in England and in France. In the Cefn and Kirkdale
caves, for example, remains of Huelephas antiquus, and in every
instance of young animals, have been discovered; while the
fissure cavern of Kent’s Hole has yielded grinders of the true
Mammoth, both of young and old animals. If the two faunas
have inosculated it is in some of those caves, which have been
inhabited during both periods, where the most decisive proofs
will be found.

With regard to the geographical range of the true Mam-
moth, my inquiries have led me to the conclusion that
although still of vast extent it is much more limited than the
area hitherto assigned to it. The result of my observation is
that it never extended to the southern side of the Alps, although
met with in the great valley of Switzerland. This result I
am desirous of stating with the diffidence and reserve with
which a negative fact involving any generalization ought
always to be expressed. The grounds upon which it is rested
are these :—That, after a very careful search among the col-
lections at Florence, Turin, Milan, and Pavia, with the
assistance of Messrs. Angelo Sismonda, Bellardi, Cornalia,
and Crivelli, respectively in these different cities, I never met
with a single grinder of Huelephas primigenius which was not

Soa i)

THEIR GEOLOGICAL AGE. 211

traceable to a foreign origin ;! whereas, on crossing the Alps,
Mammoth grinders are not unfrequent in the Swiss collec-
tions. In the Pliocene alluvium of Diirnten (alluvium of the
Swiss geologists) I found that molars of Huelephas antiquus
are met with. Well-marked specimens exist in the Museum
of Zurich, which I was enabled to examine by the kind per-
mission of M. Escher de la Linth. In the erratic bed, above
this alluvium, the remains of the true Mammoth occur.

In like manner I believe that the Mammoth remains found
in the United States of America are confined to the more
northern and central States. Twelve years ago I pointed
out to Sir Charles Lyell that some Elephant grinders which
he brought from Georgia belonged to an extinct species, dis-
tinct from the Mammoth. These remains occur in the New
Brunswick Canal, along with remains of Megatherium,
Mylodon, Mastodon, and Horse. I have seen other remains
referable to the same species from Alabama, Florida, Mexico,
and Texas. To this American fossil species I have assigned
the name of Huelephas Columbi.

1 The author subsequently found reason to modify this opinion. See pp. 178,
175, and 241.—[ Ep. |

212 ELEPHAS COLUMBI.

Ill. ON THE AMERICAN FOSSIL ELEPHANT OF
THE REGIONS BORDERING THE GULF OF
MEXICO (KE. COLUMBI, Falc.); WITH GENERAL
OBSERVATIONS ON THE LIVING AND EX-
TINCT SPECIKS.!

INTRODUCTORY REMARKS—DENTITION OF E. COLUMBI—RANGE OF HABITAT
AND GEOLOGICAL POSITION OF E. COLUMBI—ASSOCIATED FOSSIL MAM-
MALIA—SYNONYMY OF AMERICAN FOSSIL ELEPHANTS—RANGE IN TIME
OF THE MAMMOTH—ITS EARLIEST HEAD-QUARTERS—PERSISTENCE OF
ITS DISTINCTIVE CHARACTERS— UNITY OR PLURALITY OF SPECIES OF
EXISTING INDIAN ELEPHANT——ASSERTED OCCURRENCE OF MASTODON
IN AUSTRALIA—FOOD OF LIVING AND EXTINCT ELEPHANTS :—A. OF
THE INDIAN ELEPHANT—B. OF THE AFRICAN ELEPHANT—C. OF THE
MAMMOTH.

§ 1. Inrropuctory Remarks.

My first knowledge of this form dates from the year 1846,
when Sir Charles (then Mr.) Lyell submitted to me, for exa-
mination, some fossil mammalian remains, which he had
brought with him on his return from his second visit to
America.” They formed part of a collection which had been
exhumed in 1838-39, in digging the Brunswick Canal, near
Darien in Georgia. A selected series of these remains was
presented by Mr. Hamilton Couper, the discoverer, to the
Academy of Natural Sciences of Philadelphia, where they
were identified by Dr. Harlan, some of whose determinations
were corrected by Professor Owen, and those of the latter
more recently by Dr. Leidy. The locality and the details of
the case have been described, first by Mr. Hamilton Couper,?
and afterwards from personal observation by Sir Charles
Lyell. The specimens brought by the latter included some
fragments of the molars of a fossil Elephant, which, after
careful examination, I satisfied myself belonged to a species
wholly distinct from the prevailing fossil form of North Ame-
rica, namely, H. primigenius; my attention having been at
the time closely directed to the study of the Proboscidea,
1 This Memoir appeared in the ‘Na-| ? ‘Second Visit to N. America,’ 3rd
tural History Review’ for January 1863, | edit. 1855, vol. i. p. 347.
from which it is here reprinted, with * «Proceedings Geol. Society,’ 1843,

corrections made by the author, and ex- | vol. iv. p. 33.
tracts from his Note-books.—[ Ep. ]

ot ee ome

INTRODUCTORY REMARKS. 213

fossil and recent. This determination I communicated to
Mr. Lyell, who, naturally swayed in his decision by the opi-
nions then prevailing, and his estimate of the turning weight
of authority, adopted for the Brunswick Canal form the
name of H. primigenius, with the comment that the species
ranged from the Alatamaha in Georgia, to the polar regions,
and thence through Siberia to the South of Europe; while I
applied to it, in my notes of a systematic classification of the
Proboscidea, the designation of H. Colwmbi, after the great
discoverer ; purposely avoiding a geographical name, which
subsequent research, or more extended materials, might
prove to be restrictively vicious.

I was fully impressed with the importance of the result, in
proving the co-existence of a distinct species of Elephant
with the extinct Edentate Fauna of the Southern States of the
Union, and as furnishing a probable explanation of the state-
ments made by Humboldt, Cuvier,’ Von Meyer,” De Blain-
ville, and others, of Elephant remains occurring in Mexico,
Texas, and other of the Southern States of North America.
Early in the following year I became acquainted with a re-
markable series of Elephantine remains, added to the collec-
tion of the British Museum, by purchase, in the spring of
1847. They professed to be principally from the vicinity of
San Felipe, on the Brazos river in Texas; and I identified a
portion of them as being of H. Columbi. But I was prevented
from following up the subject by my departure to India at
the end of 1847, and I reserved it for future research.

Soon after my return to Hurope in 1855, Sir Charles Lyell,
at my request, placed the Brunswick Canal specimens at my
disposal, and I resumed the investigation of the fossil Ele-
phant of the Gulf of Mexico. During the sameyear I had
an opportunity, on the indication of my friend M. Lartet, of
examining, in the Paleontological Gallery of the Jardin des
Plantes at Paris, a small molar referred to by De Blainville,’
as having been brought by M. Le Clerc from Texas, which I
determined to be a milk molar of the same form. In 1856
I was enabled, through the courtesy of M. Humbert, to
examine in the Natural History collection of the Musée
Académique of Geneva, a series of molars of the same fossil
Elephant, brought from Mexico by M. H. de Saussure, a
grandson of the celebrated Swiss explorer of the Alps; and
by the kindness of my friends Mr. Charles Norton and Mr.
Guild of Boston, I was supplied with an excellent cast of the
Alabama molar, figured and described by Dr. Warren.‘ Early

1 «Oss. Foss.’ 4to edit. tom. i. p. 157. 3 «Ostéographio : Eléphants, p. 190.

2 Leonhard and Bronn, ‘ Neues Jahr- 4¢The Mastodon giganteus of North
buch’ 1838, p. 414; Jdem. 1840, p.581. | America,’ 1856, p. 162,

214 ELEPHAS COLUMBI.

in the following year (1857) I became cognizant of the most
perfectly preserved molar of the same form that I have yet
seen. It was discovered in Mexico, and presented to the
Museum of the College of Surgeons by Mr. Taylor. This
specimen, in conjunction with M. Le Clere’s milk molar, sup-
plied the means of determining the ridge-formula of the
entire set of molars, and of fixing the exact serial position
of the form among the Elephants.

The whole of these materials I found to be markedly dis-
tinct from EH. primigenius, and to partake of the characters
which are typified in the Georgian molars from the Bruns-
wick Canal. But to place the specific distinction from the
Mammoth beyond question, I resorted to the crucial test of
sawing up the principal molar of the Brunswick Canal series
longitudinally and vertically, in the manner figured in the
plates devoted to the Elephants, in the ‘Fauna Antiqua Si-
valensis,’ a procedure which commonly quashes at a glance
all doubts as to the specific distinctness, or otherwise, of Hle-
phant molars, in critical cases. The section yielded colliculi,
showing rather thick plates of enamel folded upon cuneiform
cores of ivory, of very considerable width at their base, and
separated by correspondingly open interspaces filled with
thick masses of cement. These characters were strongly in
contrast with the attenuated, parallel, and pectiniform dis-
position of the materials seen in molar-sections of H. primi-
genius; combined with the dilated outline of the ‘discs of
wear,’ and the decided crimping in the plates of enamel, they
led me to regard the form as occupying a place in the series
between EH. antiquus and H. Indicus, and as differimg more
from the Mammoth than does the latter from the existing
Indian Elephant. These facts were epitomized, but neces-
sarily in a very condensed shape, in the ‘ Synoptical Table of
the Species of Mastodon and Elephant,’ appended to the
memoir which I communicated to the Geological Society on
April 8, 1857.! In it, H. (Huelephas) antiquus, and E. (Hue-
leph.) Namadicus, respectively Nos. 10 and 11 of the list, are
included in the group (/) characterized by ‘ Oolliculi approwi-
mati, medio leviter dilatati, macheridibus undulatis ;’ while
E. (Eueleph.) Columbi (No. 12), and H. Indicus (No. 13), are
included in the next group (q), characterized by ‘ Colliculi
approaimaty macheridibus valde undulatis;’ and for the ha-
bitat of H. Columbi are given Mexico, Georgia, Alabama,
with a ‘ Post-Pliocene (?) age.’ Thus, the leading points of
the dental characters, and the precise place in the natural
series occupied by the species, were distinctly indicated, to-

* See p. 14, and Quart. Journ. Geol. Soc., vol. xiii. p. 319. Noy. 1857.—[Ep.]

Oe he er biyes

INTRODUCTORY REMARKS. 215
gether with its range of habitat along a stretch of nearly 20°
of longitude in the regions bordering the Gulf of Mexico.

In my second memoir, on the same subject, communicated
to the Geological Society on June 17, 1857, I entered into
further details on the fossil Elephant of the Gulf of Mexico,
adding that it was found in the fossil state along with spe-
cies of Mastodon, Mylodon, Megatheriwm, Equus, &c.'| An
epitome of the paper, with these statements, is given in
Leonhard and Bronn’s ‘ Jahrbuch’ for 1858, p. 379; and the
name H. Oolumbi is adopted by M. Lartet in his important
memoir, ‘ Sur la dentition des proboscidiens fossiles, * showing
that my determination of the species had not escaped the
observation of continental paleontologists.

After the communication of the memoirs above referred to,
in which, I believe, the first attempt was made to determine,
with precision, the nature of the fossil Elephant of the Gulf
of Mexico, I became acquainted with Mr. Bollaert’s speci-
men of an adult lower molar of the same species of fossil
Elephant, from the Brazos River in Texas, when it found its
way into the British Museum; and the ground having been
thus broken, the attention of paleontologists was speedily
attracted to the subject.

In September of the following year (1858), Professor
Owen, in his address to the British Association at Leeds,
while discoursing on the geographical distribution of ani-
mals, made these remarks :—‘ Geology tells us that at least
two species of Elephant formerly did derive their subsistence,
along with the megatheroid beasts, from that abundant
source,’ (i.e. the luxuriant vegetation of tropical America).
‘Nay, more; at least two other kinds of Elephant, Mastodon
Ohioticus and Elephas tewianus, existed in the warm and
temperate parts of North America.’? On this occasion, Pro-
fessor Owen gives no authority for the name FH. texianus,
although then announced for the first time, thus by the
established usage in Zoology, producing it as his own.’ But
in the second edition of ‘ Paleontology,’ published three
years later (1861), in referring to the occurrence of the Mam-
moth in North America, he adds, ‘ where it existed not only

1 See p. 211—[Ep.] Géol. de France, 1859. 2° Série, tom.

2 «Trois autres proboscidiens ont vécu
dans l’Amérique du Nord pendant la
période post-pliocéne ou quaternaire ; ce
sont l’ Hlephas Americanus que M. Leidy
considére comme étant distinct del’ £, pri-
migenius; V E. Columbi, Falc., des Etats
du Sud et du Mexique, et le Mastodon
Ohioticus que quelques auteurs sup-
posent avoir été contemporain des pre-
miers hommes qui se sont établis dans
cette region du globe.’— Bullet, Sociét.

xvi. p. 506,

2 Report Brit. Assoc. Leeds, 1858,
Address, p. Lxxxiv.

8 In the Leeds Address, Professor
Owen is so scrupulously careful on the
score of citation, that he gives in a
foot-note the names of the gentlemen to
whom we are indebted for having col-
lected the Purbeck Mammalia, (Address,
p. Lxxxix.)

216 ELEPHAS COLUMBI.

with the gigantic Mastodon Ohioticus, but also with a second
species of true Elephant (Hlephas texianus, Blake), the teeth
of which were adapted to a succulent vegetable diet..' Ina
foot-note to this passage, Bollaert’s ‘ Antiquities of S. Ame-
rica, 2nd edition,’ is cited as the authority for the second
species. The author of ‘ Paleontology’ omits on both occa-
sions to notice that I had previously determined the Ele-
phant of the shores of the Gulf of Mexico, under a different
and recognized specific name; and in defence of the new
name, he cites authority the existence of which I have failed
to trace. I have ascertained in writing from the publishers
(Messrs. Triibner), that no second edition of Mr. Bollaert’s
work has yet appeared (August 1862); and on consulting
the only impression published in 1860, I have been unable to
detect the occurrence of the name even of EH. texianus any-
where throughout the volume, or the name of Blake coupled
with any fossil Elephant therein. The sole reference to the
Texan Elephant is in a note, professing to be by the author,
in which he states that the Elephant-bones occurring in
Texas are fossil, and well silicified; adding, ‘I have de-
posited a grinder in the British Museum, which appears to
be of a new species, see my Paper on Mastodon Bones in

Chile.

1 « Paleontology,’ 2nd edit. 8vo. 1861,
p. 395.
2 Bollaert, ‘ Antiquarian, Ethnolog.

&e., 8vo. 1860, p.80.
statement, contained in a foot-note in
‘Palzontology,’ which demands an ob-
servation from me. In the remarks
upon Mastodon Arvernensis and Mas-
todon longirostris, the following sentence

occurs: ‘ Both belong to that section of |

Mastodon in which the first and second
true molars have each four transverse
ridges,’ (Foot-note: ‘ First demonstrated
by Kaup, ‘“ Ossemens Fossiles de Darm-
stadt, 4to. 1835,”) and for which Dr.
Faleoner proposes the name Tetralo-
phodon. (Op. cit. 2nd edit. p. 387.) I
challenge the production from the work
cited of any passage containing the de-
monstration asserted in the note; it is
certainly nowhere to be found there:
even the word section, or any other equi-
valent term, expressive of the idea of a
subdivision of the genus into groups
does not occur in it, and for the simple
reason, that published materials to sug-
gest it did not at the time exist. I was
the first to generalize on the subject,
and establish the constaney of the ¢er-
nary and quaternary ridge-formulz in

Geological Soc. Journal, 1857.’ ?

the Mastodons as a means of ranging all
the known species under the two na-

| tural groups of Trilophodon and Tetra-
and other Researches in New Granada,’
There is another |

lophodon; I further extended the same
principle of the ridge-formula to the
arrangement of the rest of the Probos-
cidean forms, or Elephants, under the
divisions of Stegodon, Lowodon, and
Euelephas, Until then, the species were
in extreme confusion; and nowhere was
this more signally evinced than in the
writings of Professor Owen on the fa-
mily. Kaup, with characteristic fair-
ness, recognizes the fact in reference to
the Mastodons. In proof, I refer to his
‘Beitrige, 3 Heft. (1857) pp. 1, 19.
What Kaup vehemently claims as his
special discovery is, that he was the first
to show the precise number of molar
teeth, that sueceed each other from back
to front in Mastodon (excluding pre-
molars, i.e. vertical successors); and
that his observation on that genus fur-
nished the cue for determining the same
series in the Elephants. The ‘ Osse-
mens Fossiles de Darmstadt’ is freely
quoted in the ‘Odontography,’ and in
the ‘British Fossil Mammalia,’ both
published ten years later; yet it is not
a little singular that the demonstration
asserted in the note did not prevent

ree

ee

Pear

INTRODUCTORY REMARKS. 217

At length, in November, 1861, Mr. Blake makes his ap-
pearance about this fossil Elephant. In a paper, ‘On the
Distribution of Mastodon in South America,’ the following
sentence occurs, in sequence to remarks on the remains of
E. primigenius in North America: ‘South of the 30th degree
of N. latitude it’ (the Mammoth), ‘ however, gives place to a
totally different species of true Elephant (Elephas Teaianus,
Owen, EH. Colwmbi, Falconer), the molars of which by their
less degree of complexity were more adapted to triturate the
soft succulent herbage of Texas and Mexico.’! Here it will
be observed, the name EH. Teaianus is, with propriety, so far
as published evidence goes, attributed to the author, who
had four years before become responsible for it. But in
February of the present year another paper appeared in the
same periodical, entitled, ‘On a fossil Elephant from Texas
(H. Texianus),’ by Mr. Blake, who now stands sponsor for that
specific designation himself, H. Colwmbi being quoted as a
synonym.’ Itis a nice point to decide to whom the credit
of the new name should be awarded. Professor Owen at
first produces it as his own, and then, after a long interval,
assigns it to Mr. Blake: while conversely, the latter, in the
first instance, unguardedly attributes it to Professor Owen,
and then takes it to himself. There is jactitation of the
name between the two naturalists, with hesitation and self-
denial on both sides; but it is clear that it is a joint pro-
duction; and there is a consistent harmony of ideas and
expression in their reasoning regarding the succulent food of
the fossil species. Indeed, the only difference which I can
detect is, that Professor Owen introduces the name with a
small initial, and Mr. Blake with a capital; by the canons of
nomenclature the younger naturalist has the advantage of

Professor Owen from confounding, under
the common designation of Mastodon
angustidens the dentition of three dis-
tinct species of Mastodon, one of them
belonging to the section Trilophodon,
and two to the section Tetralophodon.
Further, in the same work, ‘ Palzeonto-
logy, Sismonda’s figure of the Astesan
Mastodon is reproduced, as the principal
illustration of the genus, under the mis-
nomer of M. Turicensis, and described
as having ternary-ridgcd molars, like
M. Ohioticus, notwithstanding that Sis-
monda,* confirmed by myself after a
detailed examination, figures and de-
scribes it as quaternary-ridged: M. Tu-
ricensis (a synonym of M. Tupiroides of

the French palzontologists) being a
miocene species of the Dinotherian type,
and the Astesan Mastodon (M. Arver-
nensis) a pliocene species of another
type. These and other like errors, which
I could adduce, are brought out in a
work professing to be an exposition of
the science at the present day. But
with reference to the ‘Bollaert’ and
‘ Kaup’ citations here challenged, it is
necessary to direct attention to the re-
prehensible practice of citing known
works for matter, the existence of which
cannot be traced in them.

’ Geologist, 1861. Vol. iv. p.470.

? Op. cit. 1862. Vol. v. p. 57.

* «Osteographia di un Mastodonte Angustidente.’

10. Pl. i. fig. 2,

Turin, 1851, 4to. p. 23, line

218 ELEPHAS COLUMBI.

his senior. But the systematic names of natural objects are
not marketable commodities, or negociable instruments of
exchange, passable from hand to hand, at the option or
caprice of the holders; nor does the usage of science coun-
tenance such accommodating arrangements as those above
indicated.

Let us now see what intrinsic claims the Hlephas Texianus
of Messrs. Owen and Blake has upon the recognition of pale-
ontologists, Mr. Blake being their exponent :—

‘ As the British Association, in their rules for Zoological
Nomenclature, have authoritatively sanctioned the principle
that names not clearly defined, and likely to propagate im-
portant errors, may be changed, and as the name of LH.
Columbi lays itself open to the grave charge that it is not
clear whether it is named in honour of Columbus, or because
it is found in Columbia (Venezuela y Nueva Granada), I trust
that this name will not be accepted. That of H. Teaxianus,
founded upon a yet unimpeached geographical distinction,
if it has not the advantage of published priority, yet gives
a more lucid idea of the nature of the species which it
indicates.

‘The figure by Mr. Mackie gives a better idea of its ap-
pearance than any mere verbal description. I however define
it as Elephas Texianus, dentiwm molarium (m. 6) Colliculi
undulati, magis remoti quam m H. Indico. Its association
with EH. Indicus and Armeniacus, by Dr. Falconer, seems war-
ranted by its legitimate affinities.’ (Op. cit. p. 58.)

In reply to the first point, the author must permit me to
remark that the supposed ‘important error’ and ‘grave
charge’ are only the results of unripe knowledge and inex-
perience. The derivation of H. Colwmbi is so obvious, that
nothing I can say could make it plainer. No educated na-
turalist could apply the term to the geographical region of
Colombia, without giving it an adjective form, or supple-
menting the last vowel with an important terminal diphthong,
the requirements of the case being inexorable. Putting aside
the fact that Colombia was nowhere in question as a habitat
of the species, the British Association has not legislated
against Latin grammar.

As regards the force of the claim, put forward in the
second clause, every terrestrial mammal must have a regional
habitat somewhere ; but I fail to apprehend how the proposed
geographical name would convey ‘a more lucid idea of the
nature of the species.’ In natural history, the distinctive
characters of species are commonly founded on something
more intrinsic and tangible. Further, four years before, I
indicated that the fossil species had ranged from Georgia to

DENTITION. 219

Mexico, a stretch of nearly 20° of longitude. To restrict it
to the intermediate region of Texas would be a step of retro-
grade error.

In the proposed specific definition I fail to detect a single
term or character which is not either expressed, embodied, or
implied, in my Synoptical Table above referred to; with this
difference, however, that the author has mal-adroitly handled
terms of which he knew the meaning but imperfectly. The
colliculi, or constituent ridges of the unworn teeth, are not
undulated ; but the enamel-plates of these ridges are crimped,
_ and their worn edges in the abraded molar display the cha-
racter by their ‘ macherides undulate.’

On these grounds, I cannot acquiesce in the ingenuously
avowed aspiration, that the law of priority should in this
case give way, in order that H. Teaianus of Owen and Blake
might supplant the earlier name of H. Columbi. The reasons
assigned for the proposed change are so light and trivial,
that I should not have considered it necessary to notice
them, but for the fact that the paper is accompanied by an
illustration of the Bollaert Molar; for a figure of a new or
imperfectly known form will always carry with it a citation
of the name it bears, and of the paper in which it occurs,
however slight the latter may be.

§ 2. Dentition or EH. Colwmbi.

I shall now examine the principal remains of H. Columbi
that have come under my observation.

Of the milk-dentition, the only specimen which I have
seen is a penultimate milk molar (m.m. 3) im situ in a finely
preserved left ramus of the lower jaw of a very young Ele-
phant, contained in the Paleontological Gallery of the Jardin
des Plantes, and labelled (No. 77), as having been brought
by M. Le Clere from Texas. It attracted the attention of De
Blainville (who figured and described it in the dental series
of H. primigenius) as presenting unusual characters :—

‘Le troisiéme échantillon est plus intéressant, d’abord a
cause de son origine puisqu’il vient du Texas d’ou il nous a
été rapporté par M. Le Clerc, et ensuite parce qu’il semble
indiquer quelque chose de particulier.’ '! In this young molar
the anterior ridge, together with the front talon, are broken
off, the remaining part of the crown being composed of seven
ridges and a slight posterior talon. These ridges are quite
intact, and much more apart than in H. primigenius, agreeing
in this respect with H. Indicus. The digitations are well
marked at the apex, forming distinct points, and in the last

1 Ostéographie: Eléphants, p. 190, Pl. x. fig. 2 ¢.

220 ELEPHAS COLUMBI.

ridge their separation can be traced to a depth of nearly an
inch, a condition which ordinarily involves a high degree
of crimping of the enamel-plates. The crown is narrow in
front, and widens so abruptly behind as to have suggested
to De Blainville the term ‘ subdidyme’ to characterize it; he
describes it as resembling most the analogous tooth of the
existing African Elephant. This peculiarity is best expressed
by the dimensions, viz., length of crown 2:7 inches, width in
front at the second ridge 1:1 inch, width behind 1:7; the
length being to the extreme width in the ratio of about 3: 2.
The empty alveolus of the last milk molar (m.m. 4) is dis-
tinctly visible in M. Le Clere’s specimen. The penultimate
milk molar thus yields, for its term in the ridge-formula, 8
colliculi besides talons. The specimen, so far as mineral
condition is concerned, is well fossilized, like those from the
Sewalik hills and the ‘ Forest-bed’ of the Norfolk coast,
being hard, heavy, and weathered, and not adherent to the
tongue.

The specimen next to be noticed is a detached and very
finely preserved antepenultimate true molar (m. 1) of the
lower jaw left side, No. 741 a of the additions to the Cat. of
Foss. Mamm. of the Royal College of Surgeons. It is a
comparatively late acquisition (since 1855), and was brought
from Mexico by Mr. Taylor. The crown and body of the
tooth are quite perfect from end to end; the fangs are mostly
broken off, but a portion of them still remains. The crown
is composed of twelve colliculi, with front and hind talons.
Of these the eight anterior divisions are worn, the rest being
intact. The discs of wear are wide and open, wider than in
the ordinary varieties of the existing Indian Hlephant, and
nearly approaching the width commonly presented by H.
antiquus. But they differ from those of the latter species in
showing no angular expansion in the middle of the discs,
and no outlying loop at the angles. In this respect they
correspond more with the discs of the existing Indian Hle-
phant. The edges (macherides) of the enamel-plates are
highly crimped with numerous close set flexures; in this
respect also maintaining a resemblance to the Indian Hle-
phant, and differing from EH. antiquus. Notwithstanding the
distinctions here indicated, the aspect of the crown in the
Mexican molar bears a striking general resemblance to that
of typical specimens of the same age of HE. antiquus, the most
obvious difference being, that the crown in the former is
much wider in proportion to the length than in the latter,
in which the molars have narrow crowns like those of the
ae Hlephant. The specimen is represented by fig. 2 of

late X.

DENTITION. 221

A notable peculiarity in the Mexican tooth is, that the
body of the molar is very much bowed sidewise, i.e. concave
on the outer side, and convex on the inner. The amount of
arcuation is much greater than I remember to have seen in
any other species of Hlephant, fossil or recent, in molars of
corresponding age, viz., adolescent. Something of the same
kind is seen in Mr. Bollaert’s specimen, as figured in the
‘Geologist ;’ but in this case, in a minor degree, in conse-
quence of the anterior third of the crown having been worn
away. I believe that this peculiarity in the lower molars of
4. Columbi is a constant character of the species, and that it
bears a relation to the converging form of the rami of the
jaw, to be noticed in the sequel.

The dimensions of the Mexican molar are :—

Length of crown, 7-4in. Width of ditto in front, 2'3in. Greatest ditto, 2:din.
Height of ditto at eighth ridge, 4°3in. Space occupied by the eight anterior discs,
4-3 in.

It is hardly necessary to remark that the characters of the
molar above described ciffer entirely from those of the
common form of the Mammoth of North America.

Von Meyer, in the ‘Neues Jahrbuch’ for 1840,! briefly
notices some fossil remains of Mastodon and Elephant, con-
tained in the Mexican collection of Herr Uhde. Among
these are an upper and lower molar of a fossil Elephant, in
which the enamel-plates were wider apart than in HL. primi-
genius, in this respect having a closer resemblance to those
of HE. proboletes of Fischer de Waldheim, which Lartet con-
jecturally refers to H. meridionalis. The description would
agree with that of H. Columbi, from the same region.

Sir Charles Lyell’s Georgian specimen, from the Brunswick
Canal, upon which my first knowledge of E. Columbi was
founded, consists of the middle portion of the penultimate or
last true molar, probably the latter (m. 3), lower jaw, right
side, broken off both at the anterior and posterior ends. ‘The
fragment comprises ten complete ridges, with part of two
others, of which the anterior seven are more or less worn.
All the fangs are broken off, together with the basal mass of
ivory. The summit of the crown is concave from back to
front, and the tooth is also concave with a little torsion on-
the outside, and convex inwards, showing that it was con-
siderably arcuated laterally, like the specimen last described.
The discs of wear are of moderate width, as in the Indian
Elephant, with a tendency in some of them to expansion in
the middle. This is most pronounced in the second, where
the expansion nearly attains half an inch. The plates of

' Leonhard and Bronn’s ‘ Jahrbuch,’ 1840, p. 581.

222 ELEPHAS COLUMBI.

enamel are thicker than in the Mammoth, and about equal
to those of the Indian Elephant; they present a considerable
amount of parallel shallow plaiting, which is prominently
shown where they rise above the level of the cement. The
wear of the crown takes place in a succession of steps, from
the front backwards, which it is of importance to notice with
reference to the inferred food of the species. These steps
rise like a flight of stairs, each being composed of the whole
mass of cement of one of the valleys and the combined
enamel-plates and ivory of the ridge immediately behind it.
There are five of these steps in the Georgian specimen, the
posterior ridges being intact.

The dimensions are as follow :—

Length of crown, measured at the base, 9°5 in. Ditto ditto at summit of crown,
6°9in. Width of crown in front, 3:°2in. Ditto at fourth remaining ridge, 3-5 in.
Ditto behind, at widest part, 3°3in. Height of ditto, at eighth plate where un-
worn, 6°2in. Ditto of anterior worn plate, 2°5in.

Pl. X. fig. 1, represents a longitudinal section of the spe-
cimen, by which the specific distinction from the Mammoth
is best shown. The plates converge from the convex base to
the summit irregularly, but somewhat like the voussoirs of
an arch; so that the same number of plates diminishes from
9-5 inches at the base to about 7 at the crown. The ridges
are not so high as in the Mammoth, and their constituent
elements, 7.e. the enamel, ivory, and cement, are thicker. In
the interval between the 10th and 11th ridges the cement
attains, near the base, the excessive thickness of six-tenths
of an inch, being about twice as much as what is ordinarily
seen in the section of the Mammoth. For the contrasted
difference I refer to the sections, Pl. I. fig. 1, of the ‘ Fauna
Antiqua Sivalensis.? (See vol. i. Pl. V. fig. 3.)

The specimen next to be noticed is No. 33,218 of the MSS.
register, British Museum, Palzont. Gallery. It was acquired
of Mr. Bollaert, and it bears a record of having been procured
from San Felipe de Austin, on the Brazos River, in Texas.
It is figured in the ‘Geologist.’' This superb morceau
consists of the posterior three-fourths of the last true molar,
lower jaw, left side, well advanced in wear. The crown
presents the remains of the posterior fourteen ridges and
hind talon; the anterior portion had been ground down by
use, and has disappeared. The two anterior ridges are worn
to the base, and confluent in a common depression of ivory,
upon which a slight islet of enamel remains. Of the suc-
ceeding ridges, the next seven are worn down into transverse
discs, which are open, and bounded by highly crimped and
thick plates of enamel, bearing a close resemblance in this

1 Vol. v. p.57, Pl. iv.

DESCRIPTION OF PLATE X.

ELEepHAS (HUELEPHAS) CoLUMBI AND HiEpHas (HUELEPHAS)

ARMENIACUS.

The figures in this Plate have been copied from drawings
by Mr. Dinkel, which appeared in the ‘ Natural History
Review ’ for 1863, Plates I. & I.

Fig. 1.

Fig. 2.

Fig. 3.

Elephas Columbi. Section of the middle portion of an adult
lower molar, from the Post-Pliocene deposit of the Brunswick
Canal, near Darien, in Georgia, one-half of the natural size;
showing the disposition and relative proportions of the ivory,
enamel, and cement, as compared with corresponding sections of
E. Indicus and E. primigenius, contained in the Fauna Antiqua
Sivalensis, and in Plate V. of vol. i. (See page 222.)

Elephas Columbi. Represents the crown aspect of an ante-
penultimate true molar, of the lower jaw, left side (m. 1), one-
half of the natural size. No. 741 a, Mus. Coll. of Surg. The
eight anterior ridges are worn, the rest being intact. From
Mexico. (See page 220.)

Elephas Armeniacus. Represents the crown aspect of the last
true molar (m. 3), upper jaw, left side, of #. Armeniacus, from
a specimen in the British Museum, No. 32,250, procured by
Col. Giels, in the province of Erzeroum in Armenia. One-half
of the natural size. (See page 247.)

You. I.

Vol. il Plate 10.

Harrison & Sons

1 & 2. Elephas (Ruelephas) Columbi 6 E. (Euelephas) Armeniacus

DENTITION. 223

respect to the corresponding teeth of the existing Indian
Hlephant. Some of the plates show a considerable amount
of undulation in the general sweep of the macherides, but
there is no tendency to the mesial expansion, or outlying
loop, seen in EH. antiquus. The five posterior ridges are all
more or less affected by wear; the most of them present
distinct annular discs on the tips of the digitations, which
are seen to be of large size, as in H. planifrons and E.
meridionalis. The eighth ridge shows these annular discs
semi-confluent into a transverse depression. The ninth
presents five worn digitations; the tenth and eleventh,
six; and the twelfth, five. There is no mark of pressure
behind, proving the tooth to be the last of the molar
series. The space occupied jointly by the first six dis-
tinct transverse discs amounts to 5-4 inches, giving an
average of nine-tenths of an inch to each ridge. This is
considerably greater than that shown by the crown of the
Georgian molar, but it is to be borne in mind that the
difference is accounted for by the teeth being in different
stages of wear.

The principal dimensions are :—

Extreme length of crown, 12°5in. Width of ditto, at first transverse disc, 3-0
in. Ditto, at fifth ditto, 3-7in. Ditto, at eighth ditto, 3:8in. ‘Ditto, at ninth
ditto, 3-1in. Height of crown at ninth ditto where highest, 54 in. Space in
length occupied by the six anterior distinct dises of wear, 5:4 in.

Making allowance for the part of the tooth borne upon
the anterior fangs, which has been worn away, the entire
molar must have been of very large size; and it indicates a
species that attained colossal dimensions.

Other illustrations of H. Columbi are furnished by a col-
lection of fossil bones, part of which was purchased for the
British Museum in 1847. They are stated to have been
found by Mr. W. Huff on the banks of the Brazos River, near
San Felipe de Austin in Texas. One of the specimens, a
fine Bovine skull (Bison latifrons, Leidy), is identifiable with
a figure given by Dr. W. M. Carpenter, of New Orleans, who
published the first account of these remains. Among them
were numerous fragments of bones, said to have been of
Elephant and Mastodon; the teeth of Elephants, especially,
- predominating. A proboscidean tusk measured eleven feet
in length, and twenty-six inches in girth at the base; but
no details are given regarding the molars of Elephas.' In
the series, belonging to the National Collection, reputed to
be of this origin (Nos. 20,701—5 MSS. Register) is a superb
specimen (No. 20,702) of a last molar of the lower jaw, right
side, comprising the posterior three-fourths of the crown in

? Silliman’s Journal, 1846, 2nd ser., vol. i. p. 246.

224 ELEPHAS COLUMBI.

fine preservation, the anterior part having been lost by a
vertical fracture. Fifteen ridges are presented together with
a talon ridge. Of these, thirteen are more or less worn, the
seven anterior into continuous transverse discs, which, mea-
sured along the surface of the crown, occupy a length of 4°6
inches. The two next (8th and 9th) are divided towards
the outer side by a wide fissure into two unequal flattened
elliptical discs; the 10th yields three discs ; the 11th, four;
and the 12th, five thick annular depressions. The rest are
nearly intact, and present from four to five very thick digi-
tations. The enamel-plates of the seven anterior bands
present irregular secondary wavy curves, but they are free,
or nearly so, from crimping. In this respect, and in being
perceptibly thinner, they differ considerably from the other
Texan lower molar, No. 33,218 above described. Regarded
sidewise, the ridges look very thick and massive, and they
are retrofracted about half way up, by an abrupt flexure,
somewhat like the Porentrui molar figured by De Blainville
(Eléphants, p. 199, Pl. X. fig. 5a).

The principal dimensions are :—

Length of space occupied by thirteen worn ridges, 8:5in. Ditto by seven an-
terior ditto, 4°6in. Width of crown in front, 8°4in. Ditto in middle, 4°6 in.
Ditto greatest, 4.8in. Greatest height of crown, 7:2 in.

The dimensions above yield an average of about ‘7 inch to
the seven anterior discs, and ‘65 to the series of thirteen,
being considerably less. than in No. 33,218. The latter, also,
in the thickness and undulation of the enamel-plates, re-
sembles more the existing Indian Hlephant. Although
crimping is absent from the fossil under description, the
ereat thickness of the ridges and the limited number and
massiveness of the digitations remove it from EH. primigenius,
in which the digitations are slender and double the number.
The width of the crown is enormous, being nearly 5 inches,
agreeing in this respect with the Alabama molar to be
noticed in the sequel. Although with some doubt, I refer
the specimen to H. Columbi. It is well fossilized and adheres
strongly to the tongue.

Another specimen, of the same series (No. 20,702?), is a
fragment comprising the posterior two-thirds of a left lower
penultimate (m. 2), and including eleven ridges, the talon
being wanting. Of these the anterior six are partly worn,
but none of them into transverse discs; the first three are in
three divisions, each forming a flattened ellipse; the enamel
is thick, but does not show any considerable amount of
crimping. The ridge-plates are nearly vertical, and the
intact digitations of the hinder ones are thick. The crown
is well coated with cement.

DENTITION. 225

. The following are the dimensions :—

Length of fragment, 7°8 in. Width in front, at middle of plate, 3:4 in. Height
where intact at 7th ridge, 5:7 in.

The above dimensions yield an average of *7 inch to each
ridge. The specimen agrees very closely, in every respect,
with the corresponding molar of the Indian Elephant, and
with the characters of the lower molar, No. 33,218.

There are other Elephant molars in the Texan series,
which belong to a different species, to be noticed in the
sequel.

The late Dr. Warren, in his excellent monograph on the
‘Mastodon of North America,’ has figured three specimens,
selected by him as representing the principal varieties of
fossil Elephant occurring in the United States. One of
these, of which I possess a cast (op. citat. Pl. XXVIII. A.
p- 162) is the Alabama tooth, stated to have been found in
the neighbourhood of the Gulf of Mexico. It consists of the
middle portion of an enormous last upper molar, right side,
well advanced in wear; the anterior part supported on the
large front fang had been ground down by use, and the
posterior third is wanting. The fragment exhibits eight com-
plete ridges and the half of a ninth in front. Of these,
the three anterior are worn into continuous transverse discs,
which are open, but without mesial expansion. They bear a
close resemblance, in general contour and in the sweep of the
secondary curves of the enamel-plates, to those of the Bollaert
molar described above, as figured in the ‘Geologist.’ It is
difficult to measure the amount of crimping on these plates
from a cast. Dr. Warren describes the enamel-edges as
slightly undulating; but his figure represents them to be
distinctly and closely festooned as in the molars of the exist-
ing Indian Elephant. The fourth ridge has the digital
terminations semi-confluent into three distinct discs; the
fifth, into four; while the three last ridges are nearly
intact. The digitations of the latter are very thick, and do
not exceed four or five in number, while commonly, in EH.
primigenius, they are slender and numerous. In illustration
of the difference, I may refer to Pl. XXVIII. C. of Dr.
Warren’s work, representing a huge upper molar of HE. pri-
migenius from Zanesville, in Ohio, in which the correspond-
ing ridges exhibit the ringed tips of ten slender digitations.
The cement filling the valleys is partly removed by decay
and denuded on the sides of the crown in the Alabama
tooth, so that the character is somewhat disguised ; but the
discs of wear appear to rise in successive steps as described
above of the Darien molar; bearing in mind that the one is
an upper and the other a lower molar.

VOL. Il. Q

226 ELEPHAS COLUMBI.

The following are the principal dimensions supplied by the
cast :—

Length of the molar measured at base, 7-0 in. Ditto of the 8 posterior ridges
at base of crown, 6°6 in. Width of crown at 8rd ridge, 4-6 in. Greatest ditto
behind, 4:9 int Height of the last ridge (intact), 8-0 in.

The above dimensions give an average of eight-tenths of

an inch as the thickness of the ridges, a proportion which, I |

believe, is never attained in the true Mammoth. With the
reserve dictated by the defects of a cast, and balancing all
the characters, I am led to regard the Alabama molar as
being of H. Columbi. Dr. Warren appears to have considered
it as an extreme variety of HE. primigenius.

The only other dental remains of this species which I have
seen were some mutilated specimens of adult molars in the
Musée Académique of Geneva. They were brought from
Mexico by M. H. de Saussure. No account of them, so far
as I am aware, has as yet been brought out; and the notes
which I took were of a general character, without entering

_ into details. They agreed, in all their leading characters, with
the Georgian form from the Brunswick Canal in presenting
comparatively broad ridges, separated by wider intervals
than in the Indian Elephant, but attaining less elevation than
in the latter; the enamel-plates were well crimped, and the
discs of wear open. They impressed me, at the time, as being
distinct alike from the Indian Elephant and from EH. antiquus,
and still more distinct from the Mammoth. The same col-
lection contained the cast of a magnificent specimen of an
adult lower jaw of Mastodon Andiwm, invested with a very
massive and elongated incisive beak, deflected downwards
and retaining the basal section of one very large incisor.
The original was stated to have been found near Tlascala,
and it appears to be the adult mandible of the same form,
which yielded the younger specimen figured and described by
Laurillard in d’Orbigny’s ‘ Voyage.’

The materials above described supply only two entire
teeth, both being of the lower jaw, 7.e. a penultimate milk
molar (m.m. 3) from Texas, and an antepenultimate true
molar (m. 1) from Mexico, the former showing eight ridges,
and the latter twelve; but as these two agree in the number
of their ridges with the corresponding tooth of the Indian
Elephant and Mammoth, and as they exhibit the same series
of progressive increments, the complete ridge-formula is in-
ferred to have been thus :—

1 Dr. Warren states the length to be | of excess ; but the crown with its coat of
seven, and the width four and a half} cement must haye exceeded 5 inches in
inches. The cast may give a line or so } width.

DENTITION. 227

Milk molars. True molars.

FS SSS
ANS a Oto meow 202
AWTS 12 12s6 20"

The last true molars, above and below, commonly present
from twenty-two to twenty-four ridges in the Indian Elephant
and Mammoth. It remains to be seen, whether, taking into
account the greater thickness of these ridges in H. Columbi,
they ever exceeded twenty in that species.

The species clearly belonged to the group Huelephas, and
in so far as the dental characters go, its nearest affinity was
with the existing Indian Elephant, occupying a position in
the series between it and H. antiquus. It differs from the
latter by the absence of the constant mesial and sub-angular
expansion of the discs of wear, common to it in a minor, and
to the African Elephant in a major, degree. The difference
from the Indian Elephant is less considerable, consisting
chiefly in the enamel-plates being less strongly crimped, and
in the discs of wear being more open. Judging from the tri-
turating characters of the molars, so far as the analogy of the
living species will help us, the food of H. Colwmbi was pro-
bably like that of the Indian species, consisting of branches,
twigs, and leaves of certain trees, with reedy grasses and
other similar vegetable matters. To my apprehension, they
do not indicate succulent matters (i.e. in the botanical sense
of the word) to have been the staple food of the species, as
conjectured by Messrs. Owen and Blake, nor anything less
ligneous than the aliment of the Indian Elephant. The
grounds of this opinion will appear fully in the sequel, in dis-
cussing the general bearings of the question with reference
to the food of other fossil Elephants.

Of the cranium and other bones of the skeleton, nothing
is at present known, although it is probable that abundant
remains exist in the North American and Mexican collec-
tions. Silliman’s ‘Journal,’ for 1838, contains an account
of some elephant bones, discovered in Jackson County, Ohio.'
Among these was a lower jaw, of which a rude sketch is given,
along with that of H. primigenius. The rami are represented
as converging to a pointed chin and a very contracted sym-
physial gutter, totally different from the broad rounded chin
and wide gutter which are constant in H. primigenius. In
both respects, the figure agrees more with the mandibular
form presented by H. Indicus and E. antiquus. The bowed
Mexican molar, above described, would suggest that the man-
dible of H. Columbi was of a similar form. But the figure

? Silliman’s American Journ. 1838, vol. xxxiv. p. 847, e¢ seg. ‘On Mather’s
Report on the Geological Survey of Ohio.’
Q 2

228 ELEPHAS COLUMBI.

of the Elephant of Jackson County is too imperfect to be re-
liable for more than a conjecture. The figure in the same
plate of a detached molar represents a crown resembling
that of the Mammoth. The anonymous author of the com-
munication provisionally names the form H. Jacksoni, but that
this means nothing more than to serve the occasion, is im-
plied by the fact that he names the existing species, com-
pared with it, as H. recens, 7.e. the Indian Elephant.

The ‘ Huff Collection’ from Texas, in the British Museum,
includes (No. 20,705) a right ramus of the lower jaw, which
presents the outer shell of the bone entire, from the posterior
edge of the ascending ramus to the symphysis, but the inner
side broken off vertically along the middle of the alveolus
(the whole of the inner wall of which is removed, together
with the molar contained in it); the beak of the mentum is
also broken off. Being so mutilated, it is impossible to say to
what species it belonged. But the diastemal edge of the
symphysis slopes gently forwards, and with much less vertical
abruptness than is characteristic of the mandible of HE.
primigenius. It is therefore not unlikely that the specimen
belongs to H. Columbr.

Apart from the very numerous instances, recorded in the
American Journals, of the occurrence of Elephant remains in
most of the United States, and commonly attributed to the
Mammoth, there are two cases bearing upon H. Columbi
which require special notice.

The first from the same reputed Texan series, in the
National Collection, is an enormous fragment of a cranium,
composed of the right maxillary, part of the malar, and the
right half ofthe palate, and containing a stupendous last true
molar (m. 3) im situ, in fine preservation. The posterior half
of the alveolus is wanting, leaving a great part of the tooth
exposed. The anterior part of the crown, borne on the large
front fang, had been ground away, but its presence is dis-
tinctly indicated by the fang-pit of the inner division, in front
of the tooth. What remains of it is composed of twenty
ridge-plates, with a single talon-digitation appended behind.
The anterior fourteen or fifteen are more or less worn, the
hinder ones being intact. The general plane of the worn
surface is quite flat, as is usual in Mammoth molars, and is
free from any tendency to the terraced steps seen on the
crown of H. Columbi. The discs of wear form narrow closely
compressed bands, transverse and straight, with no mesial
expansion. The enamel-plates are thin, and as a general
rule unplaited, although some of them, as in the fifth ridge,
exhibit a certain amount of fine crimping. The enamel-

tte a4

Te a a

DENTITION. 229

edges (macherides) rise but very little in relief above the
ivory and cement. The ridge-plates are vertical and enor-
mously high, the fourteenth, which in the germ was not the
highest, measuring between 10°5 and 11 inches. The five
hinder plates fall off very rapidly in height. The crown is
very broad, being but a line or two short of five inches.

The following are the principal dimensions :—

Length of crown measured by a tape, over the summit from base of talon to
anterior fang-pit, 18:25 in. Ditto from hind talon to anterior fang, straight by
eallipers, 14:°2in. Width of crown in front, 3-4 in. Ditto at middle of ninth
ridge, 5-0 in. Ditto behind, 3°5 in, Height of fourteenth ridge, 10-7 in.

The ridges are so condensed that the joint length of the
posterior twelve, having worn discs, amounts only to 6°7
inches, being an average of about half an inch to each. The
Alabama tooth and this Texan molar agree in being of very
large size; but they differ throughout in the detail of dis-
tinctive characters. I can detect nothing by which the latter
can be distinguished from H. primigenius. It is of a colossal
size. The substance of the bone and tooth is like iron-shot,
and the matrix is a coarse ferruginous gritty sand, mixed
with fine gravel.

The second case is more remarkable and important, being
that ofthe fossil Elephant of the Pliocene Fauna of Niobrara,
an affluent of the Missouri River in Nebraska, the account of
which, by Dr. Leidy, has excited much interest and surprise
among paleontologists! According to that distinguished
naturalist, this extinct Fauna has already yielded 3 distinct
Oanide ; 3 distinct Felide ; 2 Rodents; 8 Ruminants, the ma-
jority of them new; 8 Hquide, belonging to six genera or
sub-genera; 1 large Rhinoceros (R. crassus, Leid.) ‘ which
appears to have had almost the same size and formula of
dentition as the recent Indian Rhinoceros, R. Indicus’; 1
* Mastodon of the sub-genus Tetralophodon, M. mirificus, Leidy,
and a huge Hlephant, H. (Hueleph.) imperator, Leidy. The
published descriptive details of this Hlephant are as yet but
very meagre. One specimen only is mentioned, being the
anterior portion of an upper molar, of larger dimensions than
any known to the author. The crownis stated to be ‘ within
a line or two of five inches in breadth, and within a space of
seven inches only eight enamel-folds or double plates exist.’
This would give an average of nearly nine-tenths of an inch
to each ridge, corresponding closely with the proportions
yielded by HE. Columbi. The ridges are described as becom-
ing worn into transverse strongly crenulated ellipses. Dr.
Leidy adds, ‘that the fragment of the tooth has been

1 Proceed. Acad, Nat. Scien. Philadelph, 1858, p. 20, e¢ seq,

230 ELEPHAS COLUMBI.

assumed to belong to an unnamed species, from the fact that
it was found in association with a fauna very distinct from
any previously noticed.’ !

§ 3. Rance or Hasirar anp GzoLogican Position oF
E. Columbi.

The precise localities where remains of the species occur
in Mexico, and the conditions under which they are met
with, are but imperfectly known. The best authenticated
site is the ‘ Barranca of Regla’ near Real del Monte, 60 miles
north of the city of Mexico. It is stated in Silliman’s
‘ Journal,’ that in that region they are found in some places
in beds overlaid by lava.? The fragments of Elephant molars
communicated by Humboldt to Cuvier are said to have been
found at Hue-huetoca, in the valley, and not far from the
city of Mexico.* Cuvier describes the remains from Hue-
huetoca, as detached plates of very large molars, compressed,
and with the enamel attenuated and barely crimped, as in
the Siberian Mammoth. Von Meyer states, that the remains
brought to Europe by Herr Uhde were met with, partly in the
valley of Toluca, near the Hacienda of Salceda, about 9,000
feet above the level of the sea, partly near the ancient pyramid
of Wilcox (sic), on the Chalco-Lake, at 7,500 feet above
the sea, and some others on the hills of Chapultepec, about
100 feet above the level of Mexico.4 M. H. Saussure, in a
communication, with which he has very courteously favoured
me, mentions that the remains of the fossil Elephant which
he brought from Mexico were met with in the deposits of

Puebla and on the slopes of Tacubaya, in the valley of Mexico ©

(antea, p. 213) ; and that the Mastodon remains occurred, some
of them near Xalapa, others at Atonilco el grande, near Real
del Monte, and the great mandible with the elongated beak
(antea, p. 226) near Tlascala. In reference to the stratigraphi-
cal nature of the deposits, he adds: ‘Je crois que tous les
terrains du plateau, composés d’argiles et de cinerites con-
tiennent les mémes espéces. Ce sont des amas de dejec-
tions voleaniques melées par les eaux qui ont rempli les bas-
fonds. Is ont une puissance de plus de 100 pieds.? The
remains which have been observed in Texas were discovered
on the banks of the Brazos and Colorado Rivers, at San
Felipe, Bastrop, &c., in. the prairie deposits.’ Cases of the
occurrence of Hlephant remains in the valley of the Missis-

1 Idem, p. 29. * Leonhard and Bronn’s ‘Jahrbuch,’
* Silliman’s American Journ. of Scien.| 1840, p. 581.
2nd ser. 1858, p. 283. ° Bollaert. Journ. R. Geograph. Soe.

* Cuvier. Oss. Foss. 4to, tom. i. p.|vol. xx. 1850, pp. 115-117.
157.

HABITAT AND GEOLOGICAL POSITION. 231

sippi, in the States of Louisiana, Arkansas, and Mississippi,
have been recorded, but not in sufficient detail to determine
the species. Dr. Warren mentions that he possessed thirty
molars of the fossil Elephant of Alabama, but he gives no
details regarding the conditions under which they were
found.! All the circumstances connected with the remains
occurring in Georgia have been carefully investigated by able
observers. Between the Apalachian Mountains and the At-
lantic there is a wide stretch of horizontal tertiary strata,
forming three terraces, each about twenty miles wide.” The
lowermost, or littoral, platform rises from ten to forty feet
above the level of the sea, and stretches at least 400 miles
northward to Newbern on the Neuse in Carolina. The de-
posit is fluvio-marine, resting upon Hocene strata ; although
mainly marine, it contains beds of freshwater origin, in
which the Mammalian remains occur. Lyell considers it to
be very analogous to the great Pampean formation of South
America, as described by Darwin, and to be of Pleistocene age.
The bones were found between four and six feet below the
surface, embedded in clay and resting on yellow sand, and be-
longed to Megatherium, Mastodon, Elephant, &c. The ascer-
tained range of H. Columbi, from Mexico to Georgia, includes
18° of Long. and 12° of Lat. between the parallels of 20° and
32°. But there are grounds for suspecting that it ranged
into South America. M. Lartet has recorded the fragment
of an Elephant’s molar, characterized by thick ridge-plates,
brought from Cayenne in French Guiana by Captain Perret,
and presented by him to the Museum of Marseilles.* What ~
makes this not improbable is the fact, that Dr. le Conte while —
in Honduras examined the Mastodon bed near the village of
Tambla, in one of the passes leading from the plain of
Comayagua to the Pacific, and satisfied himself that the spe-
cies found there was identical with M. giganteus (Ohioticus)
of North America.! It is therefore not unlikely, that the
fossil Elephant of Georgia may have ranged still further
south than Mexico, into Guiana.

§ 4, Assoc1aTeD Fosstn MamMatia.

Of the Mexican Mammal contemporaries of HL. Colwmbi but
very little has as yet been ascertained. Von Meyer, in his
notice of Herr Uhde’s collection, mentions a Mastodon
resembling M. angustidens of Koepfnach in Switzerland; a

1 On Mastodon giganteus, p. 62. 3 Bullet. Soc. Géol. de France, 2d sér.
2 Hamilton Couper. Geol. Proceeds. | tom. xvi. p. 500.
1843, vol. iv. p. 33; and Lyell’s Second| * Proceed, Acad. Nat. Scien. of Philad.
Visit to North America, 3rd edit. 1855. | 1858, p. 7.
Vol. i. p. 347.

232 ELEPHAS COLUMBI.

phalangeal bone of a Pachyderm, bearing some resemblance
to that of the genus Rhinoceros ; and upper and lower molars
presenting the characters of the existing genus Hquus.' The
collection formed by M. H. de Saussure included Mastodon
Andium, as distinguished by the French palzontologists from
M. Humboldtw. In Texas,” as already stated, H. Columbi was
found along with remains of Tapirus Americanus (Leidy),
Bison latifrons (Leidy), a species of Mastodon not named ;
bones supposed to be of a species of Mylodon, and probably
also a colossal form of H. primigenius.

The Georgian remains from the Brunswick Canal have
been severely examined by different paleontologists. Among
them, besides Elephant, there were Megatheriwm mirabile
(Leidy), Mylodon Harlani (Ow.), Mastodon Ohioticus, Bison
latifrons (Leidy), Equus Americanus (Leidy),*? together with
bones of a large Chelonian, Chelonia Cowpert (Harlan). It
was supposed, at the time of their discovery, that the Darien
fossil Fauna included Hippopotamus and Sus ;* of these the
former was negatived by Prof. Owen, and the tusk fragment
upon which it was founded referred to Mastodon, while the
latter has passed through many phases of nomenclature.
First named Sus Americana by Harlan (loc. cit.), it was then
regarded by Prof. Owen as the type of a new genus inter-
mediate to Lophiodon and Toxodon, which he first described
under the designation of Lophiodon bathygnathus,’ and after-
wards as Harlanus Americanus;® but the specimen upon
which the opinion rested has been satisfactorily determined
by Leidy to be nothing more than a lower jaw of his Bison
latifrons.’ At Skiddaway Island on the Atlantic shore, near
Savannah, the same genera and species have been met with;
and Dr. Leidy mentions the association, on the shores of the
Ashley River in South Carolina,* of remains of the same
Megatherium, with those of Hlephas, Mastodon, Equus, Ta-
pirus, Dicotyles, Hipparion Hydrocherus, &c.° On all occa-
sions, until lately, where this Elephant has been named in
the American memoirs, it has been cited as H. primigenius.
Generalizing approximatively, as far as the ascertained data
will admit, it would appear that where fossil Elephants occur
in the States east of the Mississippi, those found to the north
of the Apalachians belong chiefly to H. primigenius; and

1 Leonhard and Bronn’s ‘ Jahrbuch,’ 5 Cat. Foss. Mammal. &c. Mus. Coll.

1840, p. 581. of Surg. p. 197.

2 W. B. Carpenter. Silliman’s Jour-| © Proceed. Acad. Nat. Scien. Phila-
nal, 1846. 2nd ser. voli. p. 245. delph. vol. iii. 1846, p. 94.

5 Leidy. Extinct Sloth Tribe of North| 7 Proceed. Acad. Nat. Scien. Philad.
America, 1855, p. 54. 1854, vii. p. 89.

* Harlan in Silliman’s Journal, vol.| * ‘ Extinct Sloth Tribe, &e.’ p. 51,
xlii. 1842, p. 143. ® Leidy, op. cit. p. 58.

ihe

ASSOCIATED FOSSIL MAMMALIA. 233

that H. Colwmbi is the predominant although not the sole
species, in the littoral States south of the chain, as far north
as Newbern, near Cape Hattras. Huge extinct Hdentata
accompany both ; but Dr. Leidy has found no authentic evi-
dence of Megatheriwm having ranged beyond the maritime
portion of Georgia and South Carolina. Mylodon Harlani is
said to occur north and south of the chain.

Knowing, as we do, what an important feature the large
extinct Edentata constitute in the Newer Pliocene Fauna of
the littoral regions both of North and South America and of
the interior of the United States, it is not a little remarkable
that neither in the Lower Miocene Fauna of Nebraska, nor
in the Pliocene Fauna of Niobrara, both of which have been
so ably investigated by Leidy, has a single Edentate form
been discovered, although in the latter, as already men-
tioned, both an Elephant and Mastodon occur.! The great
number of Hquine forms found in the Niobrara deposit,
coupled with the antiquity of some of the genera to which
they have been referred (e.g. Anchiteriwm), is equally remark-
able, and suggestive of reflexion with reference to some of
the great problems which now occupy naturalists, regarding
the derivation and spread of species and the former conti-
nuity of continents which are now severed by wide oceans.

Of two asserted facts, which it was of the utmost import-
ance to determine with accuracy, one appears to have been
clearly established: namely, that the extinct Hdentate and
Proboscidean Fauna of the United States existed long after
the deposition of the Northern Drift. This was put beyond
doubt by Lyell many years ago; the bones of Mastodon
Ohioticus, which are commonly associated with H. primige-
nius, were found along with existing shells in Tennessee in a
swamp ‘in a cavity of the boulder-formation, so that the
animal must have sunk after the period of the Drift, when a
shallow pond fed by springs was inhabited by the same
species of freshwater mollusca as now live on the spot.’ The
same result was arrived at in the freshwater deposit on the
right bank of the Niagara, near the Falls. The Drift between
Lakes Erie and Ontario was inferred to be of much higher
antiquity than the gravel containing the bones of Mastodon
at the Falls.”

But the evidence in support of the inference that the same
extinct Fauna existed before the deposition of the ‘ Drift, in
the same region, is not equally conclusive. It has been

1 Proceed. Acad. Nat. Scien. Philad.| by the later observations of Professor
1858, p. 20, et seq. Ramsay. Quartly. Journ. Geol. Soe.
2 Proceed. Geol. Soc. 1843, vol. iv. p. | 1859, vol. xv. p. 214.
86. These results have been confirmed |

234 AMERICAN FOSSIL ELEPHANTS.

asserted that Dr. W. M. Dickeson discovered in Mississippi,
east of Natchez, remains of Megalonyx, Mylodon, Mastodon,
Equus Americanus (Leidy), Bootheriwm, Cervus, Ursus, Tapirus,
&e. ‘in a tenacious blue clay which underlies the diluvial
Drift.’1_ The precise determination of the geological age of
the deposit in this instance is of the very highest importance,
since, at a depth of about two feet below the skeletons of
Megalonyx and other extinct genera, it yielded, according to
Dr. Dickeson, the greater part of an os innominatum, which
was identified as being of an adolescent human subject, and
which proved to be strictly in the same fossil state, as regards
colour, density, and mineral condition, as the bones of Me-
galonyz and the other associated animals. Had both the
asserted facts been satisfactorily established, the antiquity of
the human race would have been carried back in America to

a period infinitely more remote, than has anywhere as yet ~

been claimed for it through recent investigations in Hurope.
But the term ‘diluvial drift’ appears to have been used, in
this instance, in the vague comprehensive sense in which it
was generally applied by geologists, before the labours of
Prestwich, Godwin-Austen, and other observers had dis-
criminated the relations of the true ‘ glacial drift’ from those
of the Valley Gravels and Loess deposits. I am informed by
Sir Charles Lyell, that the bed of material overlying the ‘ blue
clay’ at Natchez is of the nature of a Loess or Lehm deposit,
and that the clay itself is probably of a more modern date
than the ‘ true drift.? A shade of suspicion has been cast on
the identification of the fossil itself, as having been derived
from a human subject, by a remark made by Dr. Leidy, who
has examined and identified the associated remains.? Other
instances have been appealed to in proof of the Probosci-
dean and Edentate Fauna having preceded the ‘ drift’ in the
United States; but in no case as yet, has the evidence been
as conclusive, as it can be shown to be with regard to the
Mammoth in Europe.

§ 5. Synonymy or AmeERIcAN Fossin ELEPHANTS.

The Elephant molars, occurring so abundantly along with
Mastodon Ohioticus in the United States have, since the time
of Cuvier, been almost universally referred to H. primigenius.

1 Proceed. Acad. Nat, Scien. Phila-|6.) The authenticity of the specimen
delph. 1846, vol. iii. p. 107, and Leidy,/and of the identification is strongly
op. cit. p. 6. maintained in Morton’s ‘ Types of Man-
_ ” ‘With these specimens, and present- | kind,’ by Nott and Gliddon, p. 843. Sir
ing the same general appearance of|Charles Lyell no longer doubts the
colour, compactness, &c., was discovered | authenticity of the bone as contempo-
the so-called fossil human innominate | raneous with the associated remains.
bone’ (Leidy, ‘Extinct Sloth Tribe,’ p.

SYNONYMS. 235

But an impression existed among paleontologists in Ame-
rica that there might be distinct species. Dr. Warren, in
1855, expresses it thus: ‘ We still remain without any de-
cisive fact calculated to determine whether the American
varieties differ specifically from the European, and whether
these varieties differ specifically from each other.’?! In 1852,
Leidy, in the introduction to the ‘ Fossil Fauna of Ne-
braska,’? designates the fossil Elephant of North America
EH. Americanus, as distinct from H. primigenius. But he gives
no diagnostic characters, and assigns no reason for the
change, which appears to be founded mainly on the sup-
posed improbability of the same species having ranged from
Europe to America. His H. imperator he assumes to be
distinct, ‘because it was found in association with a Fauna
very distinct from any previously noticed’ (antea, p. 230).
He separates the Megatheriwm of North from that of South
America, avowedly on these grounds; and it would appear
that he distinguishes his Hquwus fraternus and EH. complicatus
from the European forms called H. primiyenius (sic) ? and H.
plicidens, on the same principle.* But the practice is open
to grave objections. It assumes a difference on theoretical
erounds, where the direct evidence, so far as it goes, indi-
cates the contrary ; and its general adoption would tend to
arrest, on the threshold, the investigation of the means
through which remote geographical forms, presenting com-
mon characters, may have started from a common origin.
The separation of some of the American Pleistocene Horses
and Bisons from the European fossil species may prove, on
the comparison of sufficient materials, to be well founded.
But as regards the true Mammoth, H. primigenius, exclusive
of FH. Columbi, I am satisfied that it rests on no sufficient
erounds. The well-known characters, upon which Cuvier
established his definition of the species, have been confirmed
by the general observation of palzontologists up to the
present day, namely: the cranium long, with a concave
forehead, and very elongated tusk-sheaths; the lower jaw
rounded, with a rudimentary beak; the molar teeth very
broad relatively to their length, and the constituent layers
of ivory, enamel, and cement very thin and condensed. The
definition was good, in consequence of its including so many

1 On ‘Mastodon Giganteus,’ p, 161.| drift, which from recollection he believed
Agassiz, at the Cambridge (U. 8.) meet- | to be different from the European Mam-
ing of the American Assoc. for the|moth (p. 100).

Advancement of Science, 1849, described | ? Op. cit. p. 9.

the molar and tusk of an Elephant, 3 FP. fossilis.

found at Vermont in digging the Rutland| ‘4 Proceed. Acad. Nat. Scien. of Phila-
and Burlington Railway, and said to |delphia, 1858, p. 11.

haye been from below the Erratic boulder

236 AMERICAN FOSSIL ELEPHANTS.

elements of undoubted authenticity, presented together in
the same individuals. But entire skulls are very rare, and
it is only under favourable circumstances that entire man-
dibles are met with; while the dense, hard, and more
durable molars occur everywhere. Practically, therefore, the
identification of the species, in most instances, rests upon
them: and the characters which they yield are so constant
and well marked, that, with care on the part of the observer,
they are perfectly reliable and sufficient for the purpose. In
the London collections, taking those of the British Museum
and College of Surgeons together, abundant materials exist
for the comparison of molars of the American Mammoth with
those of the Siberian and European forms. The Hunterian
collection contains a fine series of palates with teeth, lower
jaws, and detached molars of the Mammoth, from different
localities in the United States. The vast collection in the
British Museum includes numerous remains of the species
from Eschscholtz Bay and Siberia, accessible for ready com-
parison with British specimens. They all present, in the
main, the same characters: a uniform ridge-formula; the
same obtuse form of the lower jaw, and the same broad-
crowned molars, composed of closely compressed colliculi,
with numerous digitations and attenuated uncrimped enamel-
plates. The space within which the present communication
is necessarily limited prohibits my entering into the details
of the comparison. One of the most essential points is to
determine the constancy of the ridge-formula, which, after the
examination of a very large quantity of materials, I believe in
the Mammoth to be thus:

Milk molars. True molars.
i > aa aN
4, 8, 12 : 125 6 24
4) G8 2 CD Ge

the consecutive ciphers indicating above and below, the num-
ber of colliculi which normally enter into the composition of
the antepenultimate, penultimate, and last milk molars in the
first groups, and in the second, those of the three true molars.
The plates advance by quaternary increments in each series,
bearing in mind that the first true molar, although of larger
dimensions, commonly repeats the number of ridges presented
by the last milk molar, and that the last true molar in all the
Elephants and Mastodons is more composite than the others.!

1 Itake this opportunity of indicat- | Geol. Society, 1857, p. 315. Instead of
ing a correction in the ‘ridge-formula’ | the ciphers

of the subgeneric group Huelephas, given Milk molars. True molars.
in my memoir ‘ On the Species of Mas- SF ee
todon and Elephant,’ contained in vol. 4, 8, 12,: 14, 18, 24,

xiii, of the Quarterly Journal of the 4, 8, 12, 14, 18, 24-27,

SYNONYMS. 237

The formula in the North American Mammoth is identical
with that of the Siberian and European forms. Exceptions
are occasionally met in which an unusual number of plates is
presented. For instance, Dr. Warren figures and describes a
last upper molar from Ohio, in which, including talons, the
tooth presents thirty-two ridges.! But Mastodon Ohioticus,
in which the dental characters are very constant, occasionally
exhibits a similar numerical excess in the ridges in the last
true molar.’

There is one peculiarity, however, in the molars of the
North American Mammoth which is so constant, that, I
believe in most instances, by means of it, they can be dis-
criminated in a mixed collection of European, Asiatic, and
American specimens, namely, that the ridges and their con-
stituent elements are more attenuated and condensed. For
example, in the Museum of the College of Surgeons there is
a palate-specimen (No. 620 of Cat.) containing the ante-
penultimate true molar, on either side perfect but well worn.
On the left side the tooth measures 5 inches long by 2:7 in.
wide, and presents the discs of 14 ridges, inclusive of the
two talons, being an average of but °36 inches to each. A
superb specimen of a last upper, in the Hunterian collection
from Ohio, presented by Dr. Casper Winter (No. 615 Cat.),
exhibits 17 discs of wear in a length of 7°7 inches, giving an
average of :46 to each ridge; while a last upper molar from
Siberia, presented by Dr. Rogerson, gives an average of *54
for each ridge. Taken singly, the difference seems incon-
siderable, but when uniformly repeated over a length of crown
comprising sixteen or twenty-four ridges, it is perceptible at
a glance, and gives a certain amount of distinctive phy-
siognomy to the molars of the North American Mammoth.
The same character is seen in specimens from Eschscholtz
Bay, e.g. in a palate (No. 24,585) Paleont. Gallery, British
Museum. But I do not regard it as indicating more than a
slight geographical variety, for the other characters, such as
the form of the lower jaw, &c., remain constant to the true
Mammoth type.

It has been asserted that the kind of molar upon which
E. meridionalis is founded occurs not only in England, but
as far north as Eschscholtz Bay in Arctic America; and the
figures given by Buckland, in the voyage of the ‘ Blossom,’

the series should have been : below, characteristic of the antepenulti-
hist, Una aeeieee ' mate and penultimate true molars, the
aa Or discrimination of which always presents

4, 8, 12, 12, 16, 24, the greatest difficulties.
4, 8, 12, 12, 16, 24-27. 1 On ‘Mastodon giganteus,’ p. 168,

The correction, it will be observed, | Pl. xxviii. fig. c. ;
applies solely to the ciphers above and| ? On ‘ Mastodon giganteus,’ p. 79.

238 AMERICAN FOSSIL ELEPHANTS.

are cited in illustration.! But the assertion has arisen from
a hasty and superficial examination of the specimens. The
ivory-cores of the ridges in the Elephants are wedge-shaped
bodies, broader at the base and thinning upwards. The
plane of abrasion intersects these wedges obliquely, so that,
when far advanced in wear, the discs of the same ridges are
much wider below than the width of their base, and than
they were in the earlier stage at the apex. The Eschscholtz
Bay molar referred to is a last true molar, so far advanced
in wear that little more than the posterior half of it remains
in the lower jaw. In front, part is worn down to the base,
thus yielding the dilated appearance which has been mis-
taken for the character of H. meridionalis. The mandible
which contains it is preserved in the British Museum, exhi-
biting all the typical characters of H. primigenius. When
the materials are in a tolerably fair state of preservation I
have hardly ever seen a case where a molar of H. meri-
dionalis, or of EH. antiquus, could be confounded with that of
the Mammoth. Mutilated and fragmentary specimens are
frequently puzzling; simply because they are torsos of the
worst description, in which parts are not merely wanting,
but what remains is disfigured and disguised by abrasion.

In the view here taken there are, at the present time, but
two well determined species of fossil Elephant known in
North America.

1. EH. primigenius, Blumb. Syn. EH. Americanus, Leidy. The
name H. Rupertianus,? of Sir John Richardson, might have
been cited as another synonym, but for the fact that that
distinguished naturalist and Arctic explorer, with character-
istic firmness, withdrew it, as soon as he became aware, by
his own later researches, that it was untenable.

2. H. Columbi, 1857, Syn. EH. primigenius, pro parte, of the
American palzontologists; H. Texianus, Owen, 1858.

‘Whether the Cayenne specimen spoken of by Lartet (antea,
p- 231) belongs to this, or to a distinct species, remains to be
ascertained.

The same, with our present knowledge, must be said of
the EH. imperator of Leidy, from Niobrara. Until a perfect
molar is figured and described, no satisfactory opinion can be
formed as to what the species is. Dr. Leidy, as already
stated, assumed it to be distinct, and gave it the name upon
the assumption.

The same uncertainty applies to the specimens described
conventionally by the anonymous author, for the occasion,

1 British Foss. Mamm. p. 238 (see antea, p. 107, note).—[Ep.]
2 Zoology of the Voyage of the ‘ Herald, p. 102.

RANGE IN TIME OF THE MAMMOTH. 239
under the name of H. Jacksoni (antea, p. 228).! The detached
molar, by the figure, agrees with H. primigenius, while the
lower jaw, so far as the figure can be trusted, indicates a
different species.

§ 6. Rance in Time or THE Mamoru.

The geographical range of the species has been established
from Texas across the continent of North America to Esch-
scholtz Bay, thence from Arctic Siberia, across the steppes
of Russia, through Germany, France, and England, to Cen-
tral Italy in the neighbourhood of Rome. I carefully ex-
amined the collections at Naples, including that of the Uni-
versity, where every facility was afforded to me by Professor
Scacchi, and that of Signor Costa, but failed to detect a trace
of it there, or in Sicily, in the Museums of Syracuse, Catania,
Messina, or Palermo, the last of which contains a very con-
siderable number of molars and other remains of fossil Hle-
phants. There is clear evidence of the true Mammoth
having existed in America long after the period of the
Northern Drift, when the surface of the country had settled
down into its present form. It becomes a question of the
highest interest and importance to ascertain the first appear-
ance of the species in time. The data for its solution are
still so limited and imperfect, that the most we can do is to
indicate where it is earliest met with, as a starting point

1 Almost the last scientific work in
which Dr. Falconer was engaged, before
his fatal illness, was the commencement
- of a paper, entitled ‘ On the so-called Ele-
phas Jacksont of North America.’ Fresh
interest, he said, had beeninfused intothe
discussion by a paper, ‘On the remains of
fossil Elephant found in Canada by Mr. E.
Billings, F.G.S., the able paleontologist
of the Canadian Geological Survey. Mr.
Billings had figured and described some
well preserved remains of the lower jaw,
one of them including an entire molar,
found at the bottom of a gravel section
forty feet deep at Burlington Heights,
near the western extremity of Lake
Ontario, and at an elevation of about 60
feet above the lake. Although the dental
characters resembled those of the Mam-
moth, Mr, Billings found the form of the
symphysis so different, that he was led to
regard the Canadian form as a distinct
species, and proposed to resuscitate for
it ‘ the obscure and hardly ever recognized
name of £. Jacksoni, provisionally and
anonymously applied to certain Ohio
specimens discovered by Briggs and
Foster in 1838.’ The evidence adduced

by Mr. Billingsfailing to convince Dr.
Falconer that the Ontario remains dif-
fered in any material respect from the
Mammoth, he applied to Sir William
Logan, and forthwith obtained a cast of
the principal specimen, in order that it
might be compared with a sufficient
number of typical specimens of the
European Mammoth. The details of
this critical comparison Dr. Falconer
unfortunately never lived to tell, but the
result was a confirmation of his original
opinion, that the so-called . Jacksoni
was not a distinct species. As Dr. F.
observed, the question involved interests
of much higher importance than the
mere settlement of a disputed species,
viz.: whether an extinct European Ele-
phant of the Quaternary period, whose
remains had been found as far south as
central Italy, really ranged from Arctic
Siberiaacross Behring’s Strait into Arctic
America, and thence across Rupert’s Land
to the valley of the St. Lawrence and the
adjoining lake regions; and if so, under
what physical conditions this migration
had taken place? This question is con-
sidered further on (see p. 245).—[Ep. ]

240 ELEPHAS PRIMIGENIUS.

for future observation. That the Mammoth existed in
Europe long after its emergence from under the sea of the
Northern Drift has been clearly established by more than
one class of evidence. Abundant remains of the species have
been observed in the ‘ high ’ and ‘low level gravels’ of river
valleys in France and England, the nature and origin of
which have been so ably investigated by Prestwich and other
observers ; these valleys having been excavated either during
or after the rise of the drift-covered land, but mainly after
it, when the country was inhabited by the Mammoth, Rhino-
ceros tichorhinus, &c., during the decline of the Glacial period.
In the Apennine valley of the Chiana in Tuscany, H. prim-
genius existed so late as to have been a cotemporary of the
Trish Elk (Cervus ewryceros), Bos promigenius, and Bison pris-
cus, bringing down the period to the very modern date of the
superficial marly beds of the Isle of Man. The proofs of this
assertion will be given more in detail in the sequel. Setting
aside the cave evidence, on which I have dilated elsewhere,
there is a specimen of a last lower molar, left side, of a Mam-
moth, in the Natural History Museum of Torquay, presented
by Mr. C. E. Parker, which, Mr. Pengelly informed me, was
dredged up in Torbay, at no great distance from the shore,
and probably came out of the well-known submerged ‘ peat’
or ‘forest-bed ’ of that inlet. It is exceedingly fresh-looking,
with a slight tinge of smut, as if it had lain in a peat-bed,
and the surface is entirely free from any incrustation of
marine Polyzoa, with which it must have got covered, had it
lain long at the bottom of the sea. This peat-bed indicates
a subsidence of the land in Devonshire, then peopled with
Elephants, of a very modern date, and long subsequent to the
period of the raised beach, which is so boldly developed along
that part of the coast.

For a long time I was led to question the occurrence of the
true Mammoth in England, anterior to the deposition of the
‘Boulder Clay,’ in consequence of the questionable nature of
the evidence upon which the asserted instances rested. They
had either not been observed in situ, or were patched over
with recent Polyzoa, showing that they had been dredged up
from the bottom of the sea. But I have lately seen abund-
ant proof of indisputable authenticity in the collections of the
Rev. John Gunn, of Irstead, and the Rev. 8. W. King, of
Saxlingham, both in Norfolk, besides other cases, that H.
primigenius of the characteristic type existed in England be-
fore the deposition of the Boulder-Clay. Perfect molars, pre-
senting every element for rigorous identification, have been
found in the ‘ Forest-bed’ at the bottom of the section, be-
tween Cromer and Happisburgh,! on a horizon of fluviatile or

’ See anica, p. 164.—[Ep.]

RANGE IN TIME OF THE MAMMOTH. 241

lacustrine strata, which have yielded remains of H. meridio-
nalis, H. antiquus, Rhinoceros Htruscus, and Hippopotamus
major, &e. But nota trace of Mammoth has as yet icen
discovered in the ‘ Norwich’ or in the ‘ Suffolk crag.’ The
submergence of the Jand of the ‘ Forest-bed’ under the sea
is defined with the utmost precision ; the true Mammoth ex-
isted in England long before it, or at any rate during the
gradual refrigeration which preceded that event.

In the supplement to Sir Charles Lyell’s 5th edition of the
Manual of Geology (1857), it was stated on my authority
‘ that there is no well authenticated example of this species
(EH. primigenius) having ever been met with south of the
Alps. The specimens from Monte Mario and other local-
ities near Rome belong, according to him (Dr. F.) to EH.
antiquus, Fale., and H. meridionalis, Nesti, and those in
Piedmont and Lombardy to the same two species, together
with Hlephas priscus.’ But this opinion was negatived in
1858, by the fact that M. Lartet, whose verdict is of the
highest authority in all that relates to the Proboscidea, iden-
tified an unquestionable specimen of H. primigenius, received
from Professor Ponzi, by whom it was discovered in situ, in
the volcanic gravel deposit of Monte Sacro.! On visiting
Rome in the spring of 1859, I saw abundant proofs of the
accuracy of M. Lartet’s correction, in the rich private collec-
tions of Professor Ponzi and Signor Ceselli, in the Univer-
sity Museum of La Sapienza, and in Kircher’s collection in
the Jesuits’ College. The authenticity of the localities was
placed beyond question, by the volcanic matrix of the speci-
mens showing crystals of Pyroxene and nodules of decomposed
Leucite. As this is a point of weighty importance in refer-
ence to the geographical range of HE. primigenius, it may be
well to adduce some instances of the evidence in support of
it. The first is a fragment in the collection of Signor Ceselli
comprising the anterior two-thirds of an unworn penultimate
upper molar, presenting nine collines, very attenuated and
closely compacted, seven of them being presented within the
space of 3:2 inches, giving an average of about ‘46 inch to
each. The enamel is very thin, and the digital terminations
are slender and numerous, there being about nine to each
colline. This specimen is undistinguishable in its character
from a Mammoth’s tooth of the same age from Siberia, or an
English gravel bed. It was found in the volcanic gravel bed
of Ponte Molle, and the matrix abounds in Pyroxene and
Leucite. Another is a specimen presenting the anterior half

of a penultimate lower left true molar, with ten ridges, all

1 Bullet. Sociét. Géologique de France, 2e série, tom. xvi. p. 565.
VOL. Il. R

242 ELEPHAS PRIMIGENIUS.

more or less worn, within a space of 4°5 inches, yielding an
average of ‘45 inch to each. The typical characters of H.
primigenius are most distinctly shown in the thin transverse
attenuated plates of enamel, free from any tendency to crimp-
ing. The matrix is of a fine greyish-yellow sand, full of
grains of pyroxene. It was found on Monte Sacro, near Ponte
Nomentano, and is preserved in the Museum of La Sapienza.
The same collection contains several other specimens of the
same form, of which I find detailed descriptions in my notes.
But the great mass of the Elephant molars contained in the
collections of Professor Ponzi and Signor Ceselli, and in the
Roman museums generally, belong to Hlephas antiquus,}
which occurs alike in the older Pliocene deposits of Rignano,
and in the volcanic gravels of the Campagna. Here, there-
fore, we have unquestionable evidence that EH. primigenius
inhabited Central Italy, when the extinct Latian volcanoes
were in full action. There are no data for correlating with
precision this epoch with that of the Norfolk ‘ Forest-bed,’
but it reaches as far back nearly as the close of the Pliocene
period.

On the other hand, in the Alpine valley of the Chiana, in
Tuscany, I met with decisive evidence that EH. primigenius
had survived in Italy down to a comparatively modern period.
The Museum of Arezzo contains a mutilated cranium of this
species, presenting all the basilar portion from the occipital
condyles on to the incisive bones, both the maxillary bones,
together with the palate and the two last true molars in situ,
on either side, of an old animal. The specimen exhibits the
most typical characters of the Mammoth throughout. The
same collection includes lower jaws, detached molars, an
entire humerus, and a radius and ulna of the same form.
Some of these remains were very fresh-looking in colour,
although adhesive to the tongue. Along with them, in
the same turbary deposit, were found eight frontal frag-
ments with the horn-cores of Bos primigenius and three of
Bison priscus; and in the University Museum of Bologna, I
saw an undoubted skull of Cervus ewryceros (the Irish Ells),
from the same localities in Val di Chiana. It is worthy of
remark, that in no one of the Italian museums, from Naples
to Turin, did I detect a trace of Rhinoceros tichorhinus,
although with an eye specially directed in search of it. I
carefully examined Monti’s lower jaw, referred to by Cuvier,
as being of that species,’ and I can affirm, with confidence, ©

1 «T did not find a single tooth of #.; 7? Oss. Fossiles, 4to. tom. ii. p. 73.
antiquus in the whole paleontological | Tab. ix. fig. 10. Prof. Capellini oblig-
collection of the Florence Museum. | ingly gave every facility for the examina-
(Letter to M. Lartet, Florence, July 17,\ tion of the specimen, by remoying the
1856).—[Ep. | | enveloping matrix,

RANGE IN TIME OF THE MAMMOTH. 243

that it belongs to another extinct form. It is preserved in
the Museum of Bologna. With the exception of R. tichor-
hinus, the fossil fauna of the Val di Chiana exhibits all the
leading forms of the large Ungulata that accompanied the
Mammoth in the north of Hurope, before its final extinction.

Passing over the. superficial deposits of Central and North-
ern Kurope, I shall refer briefly to the Mammoth deposits of
Siberia and the Ural mountains. The nature of the accumu-
lations of the bones of Elephants and other northern quad-
rupeds at the mouths of the Siberian rivers is so well known,
through the writings of Pallas and other naturalists and
travellers since his time, that it is only necessary to allude
to one leading fact, namely, that besides the freshness of
condition in which they are preserved, the Siberian Fauna, as
a whole, agrees with that of the ‘low level gravels’ of the
river valleys and ‘ superficial drift’ of the last stage of the
Glacial period in Central Hurope, and that it has not yet
been shown to contain any of the older extinct species, like
the Hlephas antiquus, Rhinoceros megarhinus, or Hippopotamus
major, which are found along with the early form of the
Mammoth, in the pre-glacial ‘forest-bed’ of the Norfolk
coast, or in the volcanic gravels around Rome.

The authors of the ‘Geology of Russia’ have, in their
great work, investigated with much ability the nature and
origin of the auriferous gravels, in which Mammoth bones
occur on the flanks of the Ural mountains. They infer
that the species had existed for a long course of ages upon
the adjoining high lands, when the low region now skirting
the Polar Sea was submerged; that the vast quantities of
fossil bones found near the mouth of these rivers are the
result of the secular accumulation, during a long period, of
carcases floated by floods from the highlands into the great
estuaries; and that the last elevations of the Urals, which
led to the production of gold veins, were probably the chief
causes that conduced to the final destruction of the Mam-
moth in Siberia.! But the leading general fact, observed
with regard to the Siberian fossil fauna, holds equally good
of that of the auriferous gravel deposits, of local origin, on
the flanks of the Urals: Hlephas primigenius, Rhinoceros
tichorhinus, Bison priscus, Equus, &c., are the prevailing forms.
Not an instance has been adduced of the older associates of
the Mammoth, above-mentioned, having been found among
these remains. LH. primigenius has become extinct in the
swamps of North America and in the valley of the Tiber,
where auriferous gravels and Ural upheavements had no

? Geology of Russia in Europe, &c. p. 492, et seg.
R 2

244 ELEPHAS PRIMIGENIUS.
share in producing the effect. The disappearance of the
species must be sought for in causes of a more general scope.

M. Lartet, in his very able and suggestive essay ‘On the
Ancient Migration’ of the existing Mammiferous Fauna of
Europe,! takes the inference of these authors as his starting
point, and carries it further than appears to have been in-
tended by them. He avers that the remains of H. primige-
nius and Rhinoceros tichorhinus have nowhere in Hurope
been discovered, except in deposits of a more modern age
than the Northern drift ; and that these species did not make
their appearance-among us until after the emergence of the
drift-covered plains of western Russia, at the close of the
Glacial period; in short, that the Fauna was first tertiary
in the north of Asia, and then became quaternary in Kurope.
But this very ingenious argument is at once negatived by
the fact, that we have unquestionable evidence that the
Mammoth existed in England before the deposition of the
‘Boulder-clay,’ as the cotemporary of Mammalian species,
handed down from the Pliocene period.

On a review of the data which we possess at the present
_ time, it would appear that there is not a tittle of proof that
E. primigenius has been met with anywhere in Europe or
Asia, in deposits of an older date than the ‘ Forest bed’ of
the Norfolk coast. The Mammoth is emphatically entitled
to the significant name proposed by Geoffrey St. Hilaire, in
one of the bright inspirations of his latter years, of ‘ Dicyclo-
therium,’ as having by a ‘miracle of Providence’ survived
through two epochs.?. The geographical range of its asso-
ciate Rhinoceros tichorhinus was greatly more restricted. It
has never been observed in America, nor as yet in Italy.’
The same restriction appears to apply to its range in time;
I have seen nothing as yet to satisfy me that it existed in
the Fauna of the ‘ Forest-bed’ of Norfolk. The negative
evidence, in a case of this kind, is of little value, since it
proves nothing more than the limit of our positive know-
ledge up to a given time; but the asserted instances of its
occurrence ‘ are regarded by me as erroneous identifications,
and as belonging to a more ancient extinct species.

1 Comptes Rendus, tom, xlvi. Séance,
22 Février, 1858.

2 Op. cit. Séance, 23 Jany. 1837.

8 Extract from Dr. Falconer’s Note-
book.— May 8, 1859. In the Museum
of the Collegio Romano at Rome there
is the germ of an upper left molar of
R. tichorhinus and two detached lower

molars of the same species. They have
no labels and their origin is unknown, but
their yellow ochre matrix is exactly like
that of the Kent’s Hole specimens. In
specifie characters they differed from all
specimens found near Rome with which
I compared them.’—[ Eb.
4 Brit. Foss. Mam. pp. 347 and 364.

ITS EARLIEST HEAD-QUARTERS. 245

§ 7. Haruiest Heap-QuaRTERS OF THE MAMMOTH—WHERE ?

Another question comes up for discussion. On what an-
cient land was the first dwelling-place of the Mammoth ?
Whence did it radiate through the vast geographical area
included within its ascertained range of habitat? The pre-
vailing impression, at the present time, appears to be in
favour of the high land of Northern Asia. But I know not
upon what good grounds it can be sustained. That the spe-
cies existed there in great force, during a long lapse of time,
has been clearly established ; and it seems equally clear, that
it spread from that area into America, by Behring’s Straits or
the Aleutian Isles, before the severance of the two continents
took place, surviving, in North America, down to a date that
would correspond with the superficial lacustrine marls and
ancient peat-bogs of Hurope. But the cast of the North
Asiatic fauna, as shown above, is so entirely modern as to
have been regarded by Lartet as being that of the ancestors
of our existing Huropean mammalia. The Sub-Uralian depo-
sits have, as yet, supplied no consistent evidence of Pliocene
strata or Pliocene mammalia, by means of which the Mam-
moth-yielding and auriferous gravels may be synchronized
with, or differentiated from, the newer Pliocene strata of
England, in which the Mammoth occurs along with species
of an older age. At present there is a wide gap, in the for-
mations other than marine, between the Miocene strata along
the shores of the Black Sea, which, at Nicolajew in the Cher-
sonese, near Odessa, have yielded the greater portion of the
skeleton of a Mastodon Tapiroides,'! and the Ural gravels con-
taining bones of H. primigenius and its usual associates.

But there are strong grounds to suspect that Pliocene de-
posits exist on the western flanks of the Ural mountains, the
geological history of which still remains to be unfolded.
Pallas, in his ‘ Observatio de dentibus molaribus ignoti ani-
malis, &c., ad Uralense jugum repertis,’? which Lartet refers
to M. Borsomi,’ distinctly states that the two molars were
found in a horizontal stratum of indurated sandy and ochreous
iron-ore which is worked on the bank of the Schebysy, an
affluent of the Bjelaya, on the western slope of the Urals.
The Vergennes and Abbé Chappe molars, figured by Buffon
in the ‘Notes Justificatives,’ appended to ‘Les Epoques
de la Nature,’ the former procured from ‘Little Tartary,’
and the latter reputed to be from Siberia (or the Crimea ?),

1 Quart. Jour. Geol. Soe. Vol. xviii.| * ‘ Acta Pretropolitana,’ 3d Ser., 1777,
(Translations, &e.) p. 13. [See antea, p. | tom. i. p. 218, tab, ix. fig. 4.
65.— Ep. | 8 Bullet. Societ. Géol. de France, 2e

| Sér. tom, xvi. p. 484,

246 ELEPHAS PRIMIGENIUS.

confirm the statement of Pallas, both being of M. Borsoni ;
and Lartet tells us that he had identified with certainty, as of
BE. meridionalis, the fragment of a molar lately received by
M. Ravergie from St. Petersburg, adding that the specimen
is encrusted with the same ferruginous, sandy, and ochreous
matrix, as described by Pallas.! M. Borsoni is a constant
Pliocene species; occurring in Italy in the same Sub-Apen-
nine beds which yield EH. meridionalis; in France, below
them; and according to Pallas, in Russia, in beds at a lower
level than those which yield the Mammoth. The evidence,
therefore, slight and imperfectly defined though it be, gives
the forecast of the same order of succession upon the slopes
of the Urals as in Europe, namely, subaerial beds, contain-
ing remains of the Pliocene mammals of Italy, and above
them Mammoth-yielding deposits of the age of the Glacial
period.

It is now well ascertained that after the Miocene period
ereat alterations in the relation of land to sea took place in
the regions stretching eastward from the shores of the Black
Sea, beyond the Caspian and the Lake of Aral. We have
also undoubted evidence that the true Hlephantine Probos-
cidea, exclusive of numerous species of Stegodon and Mas-
todon, existed in India during the Miocene period. The
same fossil fauna has been traced from Burmah, north along
the foot of the Himalayahs to the frontier of Afghanistan,
and thence southward, along the Sooliman range to the pro-
montory bounding the estuary of the Indus to the west.
But from this point westward to the shores of the Black
Sea, and from the Hindoo-Koosh to the Caucasus, the entire
region, including Persia, Arabia, Turkistan, Armenia, and
Asia Minor, is almost wholly unexplored, so far as the extinct
mammalia of the Pliocene and Quaternary periods are con-
cerned. Is it not probable that when this vast tract is better
known, the fossil Elephants of Europe and Northern Asia
may be traced back towards their Miocene head-quarters in
India? Where the ground has been broken facts of nruch
interest have been yielded. During the Crimean war, Colonel
J. M. Giels, in passing through the province of Erzeroom in
Armenia, discovered, close to a village called Sharvoon, near
Khanoos, some remnants of a fossil Elephant which he pre-
sented to the British Museum. Major R. Jones Garden,
F.G.S8., being soon after on a tour in Asia Minor, and having
courteously offered his services, proceeded at my request to
the locality, to make further explorations. The remains in-
dicated that the skeleton of the animal had lain in the cliff

1 Op. cit. pp. 500 and 516.

ITS EARLIEST HEAD-QUARTERS. 247

of a ravine, about twenty-five feet in height, the section
consisting of alternate beds of clay and fluviatile sand, the
latter charged with fragments of Dreissena. The bones were
in a very friable condition, and the skull crushed and decom-
posed; but Major Garden was able to exhume some portions
of tusks six-and-a-half inches in diameter, which in desicca-
tion crumbled to pieces. The specimens presented by Colonel
Giels to the national collection consist of two last upper
molars in fine preservation, and a portion of a lower molar,
all apparently of the same individual. These molars strike a
practised eye, at the first glance, as presenting something
intermediate between the Mammoth and the existing Indian
Hlephant. The case is of so much interest, that I shall
venture on some of the details. The left upper molar (m. 3,
being No. 32,250 Museum Regist. Paleont. Gallery) is entire
from behind the large front fang, the portion borne upon
which had been ground down by protracted wear.! The
anterior part of the crown to the extent of 2°7 inches is
also worn out, presenting merely a smooth surface of ivory,
behind which there are seventeen ridges and a posterior
talon. Of these, fifteen are more or less worn. The anterior
nine form transverse narrow discs; the next six are divided
nearly equally by two rather wide longitudinal channels into
three divisions, consisting each of a flattened elliptical disc.
The transverse discs, in their general character, bear a close
resemblance to those of the Indian Elephant, the enamel-
plates being rather thick, with very pronounced close-set
crimping in the middle, but diminishing towards the cornua.
These discs are narrower than is commonly seen in the exist-
ing species, less open and less parallel across. The crown
is broad, and the enamel-plates are high. To render these
descriptive details more appreciable and available for com-
parison, I append the principal dimensions :—

Extreme length of crown, 11°75in. Length of crown-surface in use (partly worn
out), 9°5 in. Space occupied by the anterior ten discs measured at top of crown,
5-7 in. Ditto ditto at base of crown, 6:1 in. Width of crown at 8rd ridge (greatest),
41in. Ditto at 11th ditto, 3-7 in. Height of crown at 12th ridge, 7:1 in.

These Khanoos molars are intermediate in character be-
tween the Mammoth and the Indian Elephant, but more
nearly allied to the latter. The specimens are in a per-
fectly fossilized condition, the ivory being of a salmon
colour, with dark mottled patches, like those which ac-
company dendritic crystallizations, and they are strongly
adherent to the tongue. That they are true fossils is con-
firmed by the fact of Major Garden having found the tusks
im situ. Hlephant tusks, six-and-a-half inches in diameter,

1 Pl. x. fig. 8, shows the crown-view of the tooth.—[Ep.]

248 ELEPHAS PRIMIGENIUS.
are too valuable to have been left by man to decay along
with the skeleton of a domesticated Elephant. In the synop-
tical table appended to my Memoir on the ‘ Species of Mas-
todon and Elephant, &c.,’ the Khanoos fossil form is ranged
between H. Indicus and EH. primigenius, under the provisional
name of H. Armeniacus.' Captain Spratt, whose indefati-
gable labours along the shores of the Mediterranean and
Black Sea have been productive of such valuable and varied
results, ascertained that remains of a fossil Elephant had
been discovered on the banks of the Bosphorus; but the
species has not as yet been determined. I have entered on
such detail on this point to direct attention to an imperfectly
explored region, which promises to yield important results.”
The northern shores of the Black Sea and the Sea of Azof
have yielded indications of the remains of fossil Elephants,
the specific identification of which remains to be determined.
Lartet refers to molars of H. meridionalis? as having been
dug up in the trenches before Sebastopol;* and Huot
mentions the discovery of Elephant bones at Sympheropol,
which he assigns to H. primigenius, employing the term in
the loose comprehensive sense in which it used to be applied
to all fossil Elephants met with over the European area.*
The same remark applies to the Mammoth remains men-
tioned by the authors of the ‘Geology of Russia,’ as occur-
ring in the stratum of ‘clay-drift’ which rests upon the
Miocene steppe limestone at Taganrog, on the shores of the
Sea of Azof.* It is greatly to be desired that the species of
Hlephant occurring in these cases should be accurately ascer-
tained. The fact, that so eminent a Proboscidean authority

The authors

1 Quart. Journ. Geol. Society, 1857,
vol. xiii. p. 319. [See antea, p. 14.—Ep.]

2 The ‘Khanoos’ and Bosphorus
fossil Elephants appear to furnish an ex-
planation of the statements of Pausanias,
respecting the gigantic bones of Geryon,
and large horns (Elephant-tusks) found

near the banks of the Hyllus,.in Upper |

Lydia; and of the colossal bones of the
Indian Orontes, together with a gigantic
horn, brought to light by digging a deep
canal, when a Roman Emperor tried to
pass a fleet to Antioch up the Orontes.
For the former case, vide Pausan. Attic.
Lib. i. cap. xxxy.; and for the latter,
idem. Arcad. Lib. vill. cap. xxix. Also
Cuvier, Oss. Foss. 4to. tom. i. p. 152,
3rd Edit.

3 Bullet. Sociét. Géolog. de France,
3d. Sér. (1859) tom. xvi. p. 600.

+ Demidoft’s ‘Voyage dans la Russie
Méridionale,’ &c. tom. ii. pp. 457 and
564.

5 Op. cit. vol. i. p. 402.
of this great work appear to consider
the Taganrog deposit in question, which
they term ‘Clay-Drift, or ‘Mammoth-
Drift,’ to be identical with the ‘Mam-
moth-Drift’ of Central and Southern
Russia, and to have been a result of
submergence, like that of the Lowlands
of Northern Siberia, when Mammoth
bones were transported into its estuaries.

| But it still remains to be proved that

the Arctic Ocean of the Glacial period
ever invaded the Aralo-Caspian province
of which the Sea of Azof was a part.
We have the high authority of Woodward
for the fact that the Aralo-Caspian basin
contains only a single species ( Cardiwm
edule, var. rusticum) common to it and
the White Sea. (‘ Manual of Mollusca,’
p. 431). Huot considered the Crimean
deposits, yielding Elephant remains, to
be of freshwater origin,

ITS EARLIEST HEAD-QUARTERS. 249

as M. Lartet has approximatively referred the Sebastopol
remains to H. meridionalis, coupled with the occurrence of
M. Borsoni either in Siberia or the Crimea, is strongly pre-
sumptive of Pliocene beds, yielding Elephants of a much
more ancient date than the Mammoth-yielding gravels of
the Urals. The dentition of EH. meridionalis, in the ridge-
formula, is identical with that of the Miocene fossil Hlephas
plamfrons of the Sewalik hills, and the characters yielded
by the enamel-plates and discs of wear are also closely con-
formable; while H. Armeniacus, as stated above, approaches
nearer to the existing Indian species.

There is another point connected with distribution of fossil
Elephants over the European area, to which I am desirous
of directing the attention of paleontologists. I now enter-
tain a strong suspicion that remains of EH. Armeniacus, or of a
form closely allied to it, occur in Italy. This impression is
founded upon specimens which I observed in the Natural
History Maseum of Turin, in the University Museum of Pisa,
in the private collection of the Marchese Carlo Strozzi at
Leghorn, and in those of Professor Ponzi and Signor Ceselli
at Rome, the satisfactory specific identification of which
puzzled me greatly. They certainly cannot be referred either
to H. meridionalis or to H. antiquus, from the high numerical
expression of their ridge-formula, nor do they appear suscep-
tible of identification with H. primigenius, without straining
the distinctive characters of that species to a degree which is
not warranted by our experience of it elsewhere. The first
which I shall adduce in illustration are a series of molars,
discovered in the Astigiano, during the excavation of the
railway section between Alexandria and Turin. One of them
is a huge last upper molar, right side, in the finest pre-
servation and half-worn. The crown is not quite perfect in
front, the portion borne upon the large anterior fang having
been worn down and broken off. What remains of it presents
no fewer than twenty-four ridge-plates, including the hind
talon; and of these the twelve anterior ones are more or less
worn, the rest being intact. The crown is very broad in
front, and the plates, where unworn, are very high, as will
be seen by the dimensions annexed. The discs of wear are
transverse, without mesial expansion; they are not so open
as in the Indian Elephant, but wider than in the Mammoth,

except in specimens of the latter worn low down; and they
exhibit nothing of the retroflexion of the lateral cornua,
commonly seen in H. antiquus. The enamel-plates are flex-
uose in the middle with decided crimping there, which does
not extend to the sides; they are thicker than in FE. primi-
genius, but less so than in H. antiquus. In this respect they

250 ELEPHAS PRIMIGENIUS.

resemble most H. Armeniacus and H. Indicus. The space
occupied by the twelve discs of wear, measured along the
summit of the crown, is 7 inches, yielding an average of
about °6 inch to each, which comes very near that indicated
above in the H. Armeniacus of Khanoos, 7.e. 57.

The principal dimensions are :—

Length of crown (not quite entire), 13°75 in. Extreme width of crown, 4:5 in.
Height of crown at 12th ridge, 8 in. Space occupied by the 12 dises of wear, 7: in.

I have detailed notes of numerous other molars, exhumed
on the same occasion, from the same locality, St. Paolo, or
near it, ‘ Nizza della Paglia,’ which yield similar characters.
My first notes were taken in July, 1856, and in April, 1861, I
re-examined the materials, the interval having afforded me
ample opportunities of examining the molars of fossil Hle-
phants over the European area. With the reserve sug-
gested by the fact, that I have not been able to confront the
originals, or good drawings of them, I have been led to
identify the ‘ Khanoos’ and St. Paolo molars as being of the
same species, H. Armeniacus, and to consider that they are
not referable either to HE. primigenius or EL. antiquus'. The
same remark applies to specimens which I examined along
with Prof. Meneghini in the Museum at Pisa; to a specimen
of which I saw a cast in the possession of Marchese C.
Strozzi, the original procured from the Val di Mugello, an
affluent of the Sieve; and to specimens in the possession of
Professor Ponzi and Signor Ceselli, from the volcanic gravels
around Rome. I may further add, that I failed to distin-
guish from the existing Indian Elephant the last milk molar,
from the Grotta of San Teodoro, in the lower jaw figured by
my friend Baron Anca ;? and that I discovered in the Grotta
of Maccagnone a last upper milk molar, presenting similar
characters; neither is reconcilable with EH. primigenius or
with EH. antiquus. I dwell upon these facts in the hope that
the attention of Italian paleontologists may be attracted to
the subject, and that they may follow up the investigation.
We now possess, through the accurate researches of M.
Lartet, conclusive evidence that the existing African Hle-
phant formerly extended its range to Southern Europe; and
it would hardly be more unexpected to find that the Indian
Elephant, or a form closely allied to it, had ranged into Asia
Minor and Italy.

1 See antea, pages 187 and 192, note.—[ Ep. ]
2 Bullet. Sociét. Géol. de France, 2e Sér. tom. xvii. p. 684, Pl. xi. figs 8 and 8 a.

EUROPEAN FOSSIL ELEPHANTS.

§ 8. PreRsIsTENCE IN Time oF THE DisTINCTIVE CHARACTERS
OF THE Huropean Fossin HLEPHANTS.

Having long enjoyed the privilege of intimate intercourse
with Charles Darwin, I have been for many years familiar
with the gradual development of his views on the Origin of
Species ; and I have been included by him in the category of
those who have vehemently maintained the persistence of
specific characters. My attention has, in consequence, been
closely directed to the evidence yielded by the Pliocene and
Quaternary deposits of Hurope in its bearing on the question,
in so far as the fossil Mammalia are concerned.

Commencing with the older Pliocene strata of the Sub-
Apennines and of the Val d’Arno, and ascending to the
superficial gravels or quaternary deposits of comparatively
modern origin, at least four well-defined species of fossil
Elephant have been ascertained to have existed in Europe,
namely, H. (Loxodon) meridionalis, H. (Huelephas) antiquus, E.
(Huelephas) primigenius, and HE. (Loxodony Africanus fossilis.!
A vast number of remains of the three first named of these
species have been exhumed over a large area in Europe;
and, even in the geological sense, an enormous interval of
time has elapsed between the formation of the most ancient
and the most recent of these deposits, quite sufficient to test
the persistence of specific characters in an Elephant. Do
then the successive Elephants, occurring in these strata,
show any signs of a passage from the older form into the
newer? or what light do they throw on the general question ?

It is obviously beyond the scope and limits of the present
communication to enter at length on the details of this great

1 IT omit from the list E. (Loxodon)
priscus (vide Synop. Table, Quart. Journ.
Geol. Soe, 1857, vol. xiii. p. 319*), which
I now regard as being a form of E. anti-
quus, and £. Armeniacus, or the fossil
Elephant of Sicily and Italy, which is
closely allied to the existing Indian
species, in order to relieve the argument
of any elements which may not be con-
sidered as being at present established
on sufficient evidence. I omit also an
undescribed fossil Elephant from the
ossiferous caves of Malta, which is in
some respects the most remarkable and
unexpected form that has yet been dis-
covered, fossil or recent. The conception
of an Elephant is associated in the
mind with the familiar idea of colossal
size. J. Melitensis, the name which I
haye applied to the new species, was the
pigmy form of the order. I am in

possession of a last cervical vertebra of
an adult animal, the body of which does
not exceed 2°8 inches in vertical dia-
meter, and 0°95 in thickness, with a
humerus of a younger, but nearly adult
individual, the entire length of which
was not more than 10 inches. The
species was discovered through the re-
searches of Capt. Spratt, C.B., of H.M.
ship ‘Medina,’ to whose indefatigable
labours in the Mediterranean Science is
so deeply indebted. The discovered
remains, now entrusted to my charge,
include nearly the entire dentition, from
the new-born calf up to the adult animal,
of numerous individuals. In the Syste-
matic Series, it belonged to the sub-
genus Loxodon, and in size, stood be-
tween a large Tapir and the small
unicorned Rhinoceros of Jaya.
* Antea, p. 14.--[Ep.]

252 EUROPEAN FOSSIL ELEPHANTS.

argument. I shall confine myself briefly, and with diffidence,
to the results to which one observer, whose -attention has
been earnestly fixed on the subject, has been conducted.

If there is one fact, which is impressed on the conviction
of the observer with more force than any other, it is the per-
sistence and uniformity of the characters of the molar teeth
in the earliest known Mammoth, and his most modern suc-
cessor. They maintain unchanged the same numerical for-
mula of the colliculi, in the successive teeth; the same great
breadth of crown relatively to its length; the same con-
densation of the constituent materials; the same narrow
parallel-sided transverse bands in the discs of wear; the
same general absence of crimping in, and tenuity of, the
enamel-plates; and uniformly the same flatness of the plane
of wear. It may be urged, that the observation is here
limited to the characters of a single organ, and that to
justify any well-founded generalization, the comparison
should be carried throughout the skeleton. The objection
would apply forcibly in the case of living forms; not merely
the skeleton, but the soft parts, and the texture and colour-
ing of the dermal appendages, would all require to be taken
into account. But with fossil forms this is manifestly im-
possible. The compass of a single life would hardly suffice,
even, for a rigorous comparison of the details of the skeleton
in all the geographical localities and geological deposits in
which the remains of the Mammoth have been found. The
observer is thus constrained to a selection. Through a wide
range of observation on living forms, we know the constancy
with which the characters of the teeth are maintained in the
same species; and having faith in the order of nature, we
extend the rule to extinct forms. The result of my observa-
tion is, that the ancient Mammoth of the pre-glacial ‘ Forest-
bed’? of the Norfolk coast differs less from the later form
occurring on the banks of the Lena, than does the latter
from the comparatively modern Mammoth of the superficial
bogs of North America, which I regard as being only a slight
geographical variety of the same species.

The same evidence, I believe, is borne by the organs of
locomotion; but the exposition of this part of the case is
beyond the limits of the present occasion.

Assuming the observation to be correct, what strong proof
does it not afford of the persistence and constancy through-
out vast intervals of time, of the distinctive characters of
those organs which are most concerned in the existence and
habits of the species? If we cast a glance back on the long
vista of physical changes which our planet has undergone
since the Neozoic Epoch, we can nowhere detect signs of a

PERSISTENCE OF CHARACTERS. 253

revolution more sudden and pronounced, or more important
in its results, than the intercalation and subsequent dis-
appearance of the Glacial period. Yet the ‘ dicyclotherian ’
Mammoth lived before it, and passed through the ordeal of
all the hard extremities which it involved, bearing his organs
of locomotion and digestion all but unchanged.

Taking the group of the four European fossil species above
enumerated, do they show any signs, in the successive
deposits, of a transition from the one form into the other?
Here again, the result of my observation, in so far as it has
extended over the European area, is, that the specific cha-
racters of the molars are constant in each, within a moderate
range of variation, and that we nowhere meet with inter-
mediate forms. The specific difference in the molars, be it
observed, rests upon a much more deep-seated foundation
than the superficial indication, merely, of ‘ thick-’ and ‘ thin-
plated’ varieties. This I shall endeavour to explain with
the help of figures. Taking Mastodon Ohioticus at one end
of the chain, and FH. primigenius at the other, the number of
ridges in the last milk mear and in the three consecutive true
molars yields, in the former, the ciphers 3: 3, 3, 4; while
in the latter, they rise to 12:12, 16, 24. The groups of
forms interposed between these extremes yield intermediate
numerical formule, which are very constant in each species,
within a moderate range of individual variation. Thus, the
Mastodon Arvernensis gives 4: 4, 4,5; Hlephas (Lox.) meri-
mionahs, 8: 8,9,12; H. (Lox.) Africanus, 7 : 7, 8, 10-11;
H. antiquus, 10: 10, 12, 16. We nowhere find in the suc-
successive deposits in Hurope, indications of a transition
from EH. meridionalis to H. antiquus, which could be repre-
sented by a formula between 8 : 8, 9,12; and 10: 10, 12,16;
nor between the latter species and H. primigenius by a
formula intermediate to 10: 10, 12, 16, and 12 : 12, 16, 24.
The difference is so great, that the penultimate upper true
molar (m. 2), which in EH. meridionalis does not exceed 9
ridges, attains in the Mammoth 16. And it is further to be
borne in mind, that these numerical distinctions in the divi-
sions of the crowns of the molars are accompanied by other
specific characters which are equally constant.

The inferences which I draw from these facts are not
opposed to one of the leading propositions of Darwin’s theory.
With him I have no faith in the opinion that the Mammoth
and other extinct Elephants made their appearance suddenly,
after the type in which their fossil remains are presented to
us. The most rational view seems to be, that they are in
some shape the modified descendants of earlier progenitors.
But if the asserted facts be correct, they seem clearly to

254 EXISTING INDIAN ELEPHANT,

indicate that the older Elephants of Europe, such as H. meri-
dionalis and EH. antiquus, were not the stocks from which the
later species H. primigenius and E. Africanus sprung, and
that we must look elsewhere for their origin. The nearest
affinity, and that a very close one, of the Huropean HE. meri-
dionalis is with the Miocene H. (Loxod.) planifrons of India ;
and of H. primigenius with the existing Indian species.
Another reflection is equally strong in my mind, that the
means which have been adduced to explain the origin of
species by ‘ Natural Selection,’ or a process of variation from
external influences, are inadequate to account for the phe-
nomena. The law of Phyllotaxis, which governs the evolu-
tion of leaves around the axis of a plant, is nearly as constant
in its manifestation as any of the physical laws connected
with the material world. Each instance, however different
from another, can be shown to be a term of some series of
continued fractions. When this is coupled with the geo-
metrical law governing the evolution of form, so manifest in
some departments of the animal kingdom, e.g. the spiral
shells of the Mollusca, it is difficajt to believe that there is
not in nature a deeper seated and innate principle, to the
operation of which ‘ Natural Selection’ is merely an adjunct.
The whole range of the Mammalia, fossil and recent, cannot
furnish a species which has had a wider geographical distri-
bution, and at the same time passed through a longer term
of time, and through more extreme changes of climatal
conditions, than the Mammoth. If species are so unstable,
and so susceptible of mutation through such influences, why
does that extinct form stand out so signally a monument of
stability 2? By his admirable researches and earnest writings,
Darwin has, beyond all his cotemporaries, given an impulse
to the philosophical investigation of the most backward and
obscure branch of the Biological Sciences of his day; he has
laid the foundations of a great edifice; but he need not be
surprised if, in the progress of erection, the superstructure
is altered by his successors, like the Duomo of Milan, from
the Roman to a different style of architecture.

§ 9. Unity ok PLURALITY oF SPECIES IN THE EXISTING
Asiatic HLEPHANTS.

This question has an important bearing on that of the
fossil species which we have just discussed. It is averred,
that from the time of Cuvier up to the present day, zoologists
have been commonly in error in regarding the Elephants of
Hastern Asia as all belonging to one species, H. Indicus;
that there are two well marked forms confounded under this
name, the one limited to Continental India, the other insular,

UNITY OR PLURALITY OF SPECIES. 255

ealled H. Swnatranus, inhabiting Sumatra and Ceylon and
probably extending also to the Trans-Gangetic portion of
the Continent. Let us see upon what evidence these asser-
tions are founded.

The opinion, so far as lam aware, was first broached, but in
avery general and conjéctural way, by Mr. B. H. Hodgson,
the eminent ethnologist and explorer of the zoology of Nepaul,
who, in a communication to the Zoological Society, in 1834,
suggested that there are two varieties, or ‘perhaps rather
species,’ of the Indian Elephant, the Ceylonese and that of
the Sal forests: the Ceylonese having a smaller and lighter
head, which is carried more elevated, and having also higher
fore-quarters; while the Elephant of the Sal forests has
sometimes five nails on its hinder feet.!

In 1847, Temminck brought out a work embodying a
general survey of the resources and productions of the Dutch
Hast India possessions, in which there appeared a brief
notice of a supposed new species of Elephant, named H. Su-
matranus.? As Temminck’s strength as a naturalist lay in
ornithology, the announcement did not carry with it the
weight of authority, when opposed to the opinion of Cuvier
and other eminent zoologists. But it now appears that the
inference originated with the distinguished Dutch zoologist,
Professor Schlegel, and that Temminck’s work was simply
the vehicle in which the results arrived at by the latter first
appeared.

In 1847 T visited Leyden, for the express object of exa-
mining the materials preserved in the Museum there, upon
which H. Sumatranus was founded; by the aid of Prof. Van
der Hoeven I was enabled to see them, although only in a
cursory manner, owing to the shortness of the time at my
disposal; but the inspection failed to satisfy me that H. Su-
matranus was distinct from the Continental Indian Elephant,
with which I had been familiar in its native haunts, during
many years.

In 1849, the late Prince of Canino (Charles L. Buonaparte)
made a communication to the Zoological Society, in which
he affirmed that H. Sumatranus of Temminck was inter-
mediate between the Continental Indian and African Ele-
phants ; and that the differences in the form of the skull
and in the teeth were so pronounced as to put an end with
certainty to the subgeneric distinction between Hlephas
proper and Loxodon.? But there are errors of statement in
the Prince of Canino’s brief notice which divest it of the

1 Zoological Proceedings, 1834, p.| sessions Neederlandaises,’ &c. 8vo. 1847,

96. tom. ii. p. 91.
2 «Coup Weil général sur les Pos- % Proceed. Zool. Soc. 1849, p. 144.

266 EXISTING INDIAN ELEPHANT.

authority of accurate or original observation. He even
asserts that the ‘undulated ribbons of enamel are nearly
quite as wide as those forming the lozenges of the African.’

Last year, Professor Schlegel, whose attention has been
continually directed to the subject since 1845, communicated
a paper to the Academy of Sciences of Holland, in which he
lays claim to the authorship of the opinion first put forward in
Temminck’s work, and maintains it upon extended observation.!

In order to facilitate their examination, I shall classify the
distinctions which have been adduced, from first to last, in
support of the view, although some of them have been
abandoned in the progress of the inquiry.

I. Heternal characters.— Small ears and general form,
both, as in the Continental Elephant; but the Sumatran
species more slender and more finely built; trunk longer
and more slender; extremity of the tail more dilated, and
invested with longer and stronger bristles, in this respect
reminding one more of the African than the Indian species.
(Schlegel.)

Il. Greater degree of intelligence and aptitude for instruction.
(Diard in Schlegel.)

IIL. Osteological characters.

(A.) General construction of the skeleton and form of the
cranium alike, but:

1. Free part of intermaxillaries shorter and narrower.

2. Nasal aperture more contracted.

3. Inter-orbital space narrower.

4, Posterior part of the cranium wider. (Schlegel in
Temminck.)

5. Form of skull intermediate between African and Indian.
(C. L. Buonaparte.)

(B.) Molar teeth.—Ribbons (discs of wear) in form like
those of the Indian species, i.e. the enamel-plates highly
crimped, parallel, and free from the rhomb-shaped expansion
of the African Elephant; but the ribbons wider (in the
direction of the long axis), and consequently less nwmerous
than in the Indian species; the difference being in the ratio
of 38 or 4:1 in the Sumatran, and 4 or 6:1 in the Con-
tinental Indian form (Schlegel in Temminck). Ribbons of
enamel nearly quite as wide as in the African Elephant.
(C. L. Buonaparte.) .

(C.) Vertebre and ribs.—The following numerical differ-
ences have been indicated by Prof. Schlegel; they vary in
some unimportant respects, according to the statements of
different dates :—

1 «Bijdrage tot de Geschiedenis van Olifanten, voornamelijk, Hlephas Suma-
tranus, translated by Dr. Sclater, Nat. Hist. Review, 1862, vol. 1. p. 72.

UNITY OR PLURALITY OF SPECIES. 257
African Elephant Sumatran Elephant Indian Elephant

Schlegel In Schlegel In Schlegel

Temminck! °°™°S* | Temminck| °°VS* | Temminck nee
Cervical vertebre . “f 7 7 7 7 7
Dorsal vertebree 21 21 20 20 19 19
Lumbar vertebre . 3 3 3 3 3 3
Sacral vertebrae 4 4 4 4 5 4
Caudal vertebre . 26 26 34 33 34 33
True ribs! 6 5 6 5 6 5)
False ribs 15 16 14 15 13 14
Pairs of ribs 21 21 20 20 19 19

According to this table, the Continental Indian Elephant
has only 19 dorsal vertebree and 19 pairs of ribs; while the
Sumatran species has 20 of each, the African Elephant
having 21; being differences which, if proved to be constant,
would be of considerable systematic importance.

The difference in external form between the Indian and
African species is so pronounced, that either can be told off
at a glance even from the stamp of a Greek or Roman coin.
Admitting the general form and small ears to be alike in the
Indian and Sumatran Elephants, Professor Schlegel has only
a slight difference in slenderness of the general proportions,
a more slender form of the trunk, and a larger terminal
fringe of bristles to the tail,? to rely upon. But even in the
Sal forests of North Western India, at the extreme northern
limit of the species at the present day, the difference of
slender-built and squat-built Elephants is well known, being
expressed by Corse, for the Bengal variety, under the de-
signation of ‘mirghi,’ or Cervine for the former, and ‘ Koo-
marea’” for the latter, or when the characters are combined
‘Sunkareah.’* ‘The trunk varies in a similar manner, being

1 Schlegel expressly states, ‘that the {
number of true ribs is alike in all the |
species, that is only five; but there is |
evidently a numerical slip in the ciphers |

other Sumatran. It must be admitted,
that the number of objects compared is
hardly sufficient to sustain the position.
The original passage in Schlegel’s

which he immediately afterwards assigns |
to the false ribs, namely, 15, 14, and 13
respectively, in the three different species,
which would give a total of 20, 19, and
18, instead of 21, 20, and 19, being the
asserted aggregate of pairs correspond-
ing with the assigned number of dorsal
vertebree in the different species. (Nat.
Hist. Review, vol. ii. p. 75.)

® The distinctions indicated were, |
according to the statement of Prof.
Schlegel, founded on the observation of |
Heer Westermann, upon two female
Elephants in the Zoological Garden at
Amsterdam, the one from Calcutta, the |

VOL. Il. tS)

| stronger hairs’ (Op. cit. p. 76).

' (Op. cit. p. 66).

memoir is thus:—‘ Dat het een’ langeren
en dunneren snuit heeft; dat de Staart
aan het einde meer afgeplat en met
langere, zware haren bezet is,’ &c. The
version given in the ‘Nat. Hist. Review,’
is ‘that the rwmp at the end is more
broadened, and covered with longer and
In
Tennent’s ‘Natural History of Ceylon,
the character of flattening with longer

_ hair is made to apply to the extremity

of the proboscis, instead of the tail
The version given in
the text is the correct one.

3 Philosoph. Transacts. 1799, p. 205.

258 EXISTING INDIAN ELEPHANT.

somewhat short and thick in some, and long and more
slender in others. The fringe of bristles to the tail is
variable in degree, according to the sex, age, and vigour of
the animal. A good fringe is seldom retained long in cap-
tivity; when present, it always enhances the price of the
animal in the estimation of the natives of India. That the
animal varies considerably in appearance, according to the
district in which he has been captured, has long been well
known in India. Aboo Fuzl, in his account of the Elephant
stables of Akbar, enumerates six varieties, distinguished by
form, different marks, or colouring;! and the experienced
mahouts attached to the Government Commissariat in Bengal
will tell, at a glance, the district where a recently caught
Elephant has been bred;? whether the Sal Forests of the
North-West Provinces, Assam, Silhet, Chittagong, Tipperah,
or Cuttack. The distinction, therefore, founded upon the
external characters of H. Swmatranus, completely fails.

I believe that the same could be shown, as regards the
asserted difference of intelligence and aptitude for imstruc-
tion; but as this is not a tangible, specific character, I leave
it undiscussed.

The Osteological distinctions in the skull, which Professor
Schlegel advanced in Temminck’s work, he has since seen
reason to abandon. But the identity of form is a strong
argument in support of the unity of species. Not only is the
general form of the cranium alike in both, but the relative
proportions and connections of the constituent bones are
the same in the wild Elephant of the North-West Provinces
and in that of Ceylon. The difference of variety, implied by
the terms ‘ Mikna,’ ‘small-tusked,’ and ‘ Dauntela,’ large-
tusked, necessarily involves a proportional degree of dif-
ference, in the development of the intermaxillary bones, in
the depth and breadth of the trough between the tusk-
sheaths, and in the amount of development of the occipital
bosses. But the connections of the bones remain the same ;
and all the leading modifications of form and proportion, so
clearly indicated by Cuvier, as distinctive of the Indian from
the African form, are maintained in the Continental and
Ceylon Elephants, within a range of variation which is
common to both.

The metropolitan collections furnish excellent and au-
thentic materials for testing the accuracy of this statement
in two magnificent skulls of adult wild Elephants, both
killed in combat by gunshot wounds. The one (No. 2656,

1 Ayeen-Akberry, translated by Gladwin, vol. i. p. 126.
* Hooker, ‘ Himalayan Journals,’ vol. il. p. 302.

UNITY OR PLURALITY OF SPECIES. 259

Osteol. Cat.) is preserved in the Museum of the College of
Surgeons, the other in the Zoological Department of the
British Museum. The former is of a large Ceylon Elephant,
which bears the open canals (one of them nine inches
deep) of three bullet wounds, of old date, that had been
repaired by nature, in addition to its recent death wounds;
the latter, of a most destructive Solitary, or ‘ Goondah’ wild
Hlephant, which for a long time was the terror of a district
near which I then resided. It was killed in the jungles, on
the banks of the Ganges, at no great distance from Meerut,
in May, 1833, by a party of four experienced sportsmen, who
went out for the express purpose of killing it. The savage
animal made no fewer than twenty-three desperate and
gallant charges against a battery of at least sixteen double-
barrelled guns, to which it was exposed on each occasion,
and fell, after several hours, with its skull literally riddled
with bullets. Besides the shot-holes of its last engagement,
the frontal plateau alone bears, above the nasals, the healed
canals of at least sixteen bullet-wounds received in previous
encounters, exclusive of those effaced by the confluent fissures
of its latest wounds. Meerut is in lat. 29°, close to the ex-
treme northern limit of habitat of the Indian Elephant. If
the two skulls, from localities so wide apart, are compared,
they agree in general form and proportions, and also in the
details of the pyramidal summit, long concave frontal plateau,
inial fossa, occipital bosses, nasal aperture, position of the
orbits, form and connections of the lachrymary, length of
incisive sheaths, &c.

On the other hand, in all the well-determined species,
fossil or recent, of which perfect crania are known, we in-
variably find that the latter yield strongly-marked distinctive
characters even when molar teeth are similar. In illustration
I may cite H. primigenius, EH. Indicus, EH. Hysudricus, EH.
Namadicus, H. planifrons, HE. meridionalis, and H. Africanus,
in no two of which are the crania alike; while in the
Ceylon and Indian Elephants they are so closely similar,
that, in a museum, without a record, the mere form will not
instruct the observer whence the specimen came—whether
continental or insular. The statement made in the Zoological
Proceedings of 1849, as to the amount of difference, is clearly
an exaggeration (antea, p. 255).

As regards the molar teeth, it is stated in Temminck’s
‘Coup d’cil,’ in reference to the discs of wear, that, ‘ ces
rubans sont de la largeur de ceux qu’on voit a la couronne
des dents de ’Eléphant d’Afrique; ils sont conséquemment
moins nombreux que dans celui du continent de l’Asie.’ In

$2

260 EXISTING INDIAN ELEPHANT.

Professor Schlegel’s later communication, the statement is
modified as follows :—-

‘The lamin of the teeth afford another distinction which
however is less apparent to the eye than that taken from the
number of vertebre. These lamine, or bands, in H. Suma-
tranus, are wider (or if one may so say, broader in the direc-
tion of the long axis of the teeth) than in E. Indicus. In
making the comparison, one may remark that the distinction
is less evident in younger individuals, and that there are met
with, in all species of Elephants within certain definite
limits, remarkable individual differences in respect of the
width of these lamine.’ (Nat. Hist. Rev. 11. p. 75.)

Here, it will be observed, the distinction is propounded
subject to so many qualifications, as to render it elusive for
any practical use. I have ascertained, after the examination
of a very large quantity of materials in India and Hurope,
that the ridge-formula in the Indian Elephant runs
thus :—

Milk molars. True molars,
eaten > C SS

4, 8, 12 12, 16, 20-24.
2S oT 1, 16, 2022s

The increase in the number of the ridges of the successive
teeth takes place as in EH. primigenius, by increments of 4,
repeated in two series, the first of which terminates with the
last milk molar. The second series commences with the
antepenultimate or first true molar, which constantly repeats
the number presented by the last milk molar, 7.e. 12; the
penultimate (m. 2) shows an increment of 4, the number of
its ridges being normally 16. The last true molar never
shows less than 20, commonly about 22, but sometimes, in
the lower jaw, attaining as many as 27 ridges. This liability
to variation in the last true molar is well known, and runs
more or less through all the species of Elephant and Mas-
todon.!. But the ciphers, shown above, are very constant,
in the three intermediate molars (7.e. the milk molar
and the antepenultimate and penultimate true molars),
namely, 12: 12,16. Ido not mean to affirm that they are
absolute and invariable ; but that the above formula is a fair
exponent of the results yielded by a great majority of in-
stances, on the comparison of a very large quantity of ma-
terials. For example, the penultimate milk molar (m.m. 3)
occasionally presents only 7 ridges; while the antepenul-
timate true molar of the lower jaw, in some cases, exhibits

' For illustrations of the fact in| in the family, ede Lartet, Bullet. Gévl.
Mastodon Ohioticus, vide Warren, Op. | Soc. de France, 2e série, tom, xvi., note,
cit. p. 79; and forits general occurrence | p. 498.

UNITY OR PLURALITY OF SPECIES. 261

as many as 14; the variation being dependent, partly on the
ereater or less development of the talon-ridges, which are
very inconstant, and, as I have elsewhere stated, partly on
the race, sex, and size of the individual.! I rest the more
stress upon the importance of the ridge-formula, since,
whenever the element of quantity can be shown to hold in
the animal organization, it becomes a powerful aid to re-
search and a criterion to test the accuracy of observation.
In the fossil H#. antiquus of Europe, the dentition of which I
have been able to determine with precision, the formula for
the three intermediate molars and the last true molar, above
and below, is 10: 10, 12, 16, being nearly intermediate
between the Indian and African Elephants. If then, as
asserted, the number of bands (i.e. ridges) is less in the
Sumatran and Ceylon form than in the Continental Indian,
the ridge-formula ought to show a lower series of ciphers.
Professor Schlegel tells us that he has had the advantage of
examining at least seven skeletons, including young indi-
viduals, besides several skulls of H. Swmatranus, farnishing
ample materials for determining the number of ridges in the
different teeth. Yet neither he, nor any of the other ad-
vocates of distinctness of the species, has as yet attempted to
show, by adduced instances, that the numbers are less; and
until that is done, the general and therefore vague assertion
of the fact cannot be admitted as of sufficient weight. In
the skull of the large Ceylon Elephant above referred to (No.
2656. Osteol. Cat. Coll. Surg.) the last true molar above and
below shows 22 ridges; a penultimate upper right molar, in
the collection of Mr. Prestwich, and of undoubted authen-
ticity as having been imported from Ceylon, still shows 15
ridges, although the most anterior portion is worn out, with
the loss of one ridge; while the penultimate lowest of a
Sumatran skull figured by De Blainville ? distinctly shows 16
ridges, besides a hind talon. These instances prove, so far
as they go, that the ridge-formula is the same in the Ceylon
and Sumatran form as in the Indian.

Next, as regards the width of the bands (discs of wear).
This is a most deceptive character if merely regarded per se,
since it varies very considerably, even in the same molar, at
different stages of detrition: Ist, because the ivory-cores of
the ridges being wedge-shaped, the discs of wear are neces-
sarily narrower at their apex than at their base; 2nd,
because, as already stated (supra, p.238), the plane of abrasion,

1 Quart. Jour. Geol. Soc., 1857. Vol. | fig. 6. De Blainyille numbers the tooth
xiii. p. 315. (Antea, p. 5.—Ep.) as a 6th or last, but it is manifestly a
2 Ostéographie: Eléphants. Pl. ix. | 5th or penultimate.

262 EXISTING INDIAN ELEPHANT.

instead of being perpendicular to the axis of the wedges, cuts
them obliquely, the obliquity increasing with the advance of
wear and constantly tending towards the horizontal. The
consequence is, that the width of the discs is always ex-
agegerated in a tooth worn down to the base, and that the
anterior discs are wider than the hinder ones. The only
accurate method of ascertaining the number of ridges within
a given space is to measure the crown, not at the summit,
but along the base where the enamel-plates are reflected ; the
product will then give the average width of each ridge. The
skull of the Ceylon Elephant (No. 2656, Coll. Surg.), supplies
excellent and readily accessible materials for testing the
value of the alleged character, in the so-called EH. Sumatranus.
It contains, above and below, the penultimate and last true
molars in action; the former in advanced wear, the latter
coming into use, and in the upper jaw barely abraded. The
right upper penultimate is worn low, with a loss of the
anterior portion. The crown presents the discs of eight
distinct ridges, together with a denuded base of ivory in
front, corresponding with two ridges that have been worn
out. These discs are wide, with highly crimped enamel
macherides, both being of the H. Sumatranus pattern. The
following are the principal dimensions :—

Length of crown measured at summit, 8° in. Space occupied by the 7 last dises
of wear, 58 in. Greatest width of crown, 3:0 in.

In this case the discs are very open, with an average width
of about ‘83 inches to each; but in the progress of wear, the
ridge-plates have become so reclinate, in relation to the
plane of detrition, that in the middle of the crown, from the
causes above assigned, the grooved enamel-plates are ex-
posed nearly horizontally, to the extent of nine-tenths of an
inch. The width of the bands, in this instance, is an ex-
aggeration arising from the obliquity of the section yielding
them, in a tooth far advanced in wear.

The last true molar of the same skull makes a different ap-
pearance. Although partly extruded it is hardly touched by
wear, and the outer wall of the alveolus was removed to
expose the concealed hinder portion. The crown is composed
of 22 ridge-plates, of which 18 are consolidated, the 4
hindmost being loose. It yields the following dimen-
sions :—

Extreme length, measured diagonally, from apex of front ridge to base of the

last ditto, 13 in. Length of ditto, measured along the base of the ridges, 11°5 in.
Greatest width, 3-1 in.

Instead, therefore, of :83 yielded by the penultimate, the
ridges in this case give only an average of *52 inches; and in
weighing the result it should be borne in mind that the

UNITY OR PLURALITY OF SPECIES. 263

three last true molars not only increase successively in the
number of their component ridges, but that the latter are
proportionally thicker in the older teeth, being an adaptation
of nature, to suit the long term of use which the last molars
have to serve. Here then are two consecutive molars of the
same skull, which, if detached and introduced into a museum
without a knowledge of their origin, might be cited—the
penultimate as a typical illustration of H. Swmatranus, and
the last of H. Indicus.

Nor is this width of the bands, in worn molars, confined to
the Southern Elephant. I have now before me two grinders,
picked up by Sir Proby Cautley, in the swamp of Azufehur,
a habitual resort of wild Elephants, in the Terai of the ‘ Sal
Forests,’ at the foot of the Himalayahs north of Meerut,
which present the characters of the discs of wear attributed
to the Sumatran Elephant. There is, doubtless, a certain
amount of difference to be met with in the teeth of different
Elephants living in the same forests; but it is common to the
Northern as well as to the Southern form, and as yet there
are no good grounds to believe that it ever attains the
importance of a specific distinction. The discs of wear in
the Ceylon and African Elephants never present a similitude,
except when the slightly abraded crown of the latter is con-
fronted with the worn out and torso crown of the former.

The most important part-of Professor Schlegel’s case re-
mains to be considered, namely, the number of the dorsal
vertebree and ribs. Here, also, I find my observations at
issue with the conclusions of this distinguished zoologist.
He avers not merely that the number of the former differs in
the supposed three living species, namely, 21 in the African,
20 in the Sumatran, and 19 in the Indian, but thinks that he
has detected a curious inverse relation between these num-
bers and the thickness of the lamine of the molars; where
the latter are most attenuated the number of dorsal vertebree
is least. Ifthe inference were well founded it would be of
high interest. I quote the passage containing it in eatenso:
‘Tf we take into consideration at once the size of the laminz
of the teeth, in the different species of Elephant, and the
number of the ribs and dorsal vertebree, we obtain the
remarkable result, that as the latter numbers decrease the
laminze become narrower. In H. Africanus these lamine are
widest, and here we find the greatest number of dorsal ver-
tebree and pairs of ribs: H. Swmatranus, in which the laminz
are narrower, has twenty dorsal vertebra and pairs of ribs:
H. Indicus, in which they are still narrower, only nineteen.
In the Mammoth, EH. primigenius, where they are narrowest
of all, the number of dorsal vertebrve and ribs appears to be

264 EXISTING INDIAN ELEPHANT.

only eighteen.’! Extending the comparison to the Masto-
dons, and finding that M. Ohioticus has only twenty dorsal
vertebree and an equal number of ribs, while its crown-ridges
are reduced to three or four, he concludes that the Mastodons
form not a diverging, but a parallel series with the Elephants.
The case, therefore, concerns not merely the Continental
and Sumatran varieties of the Indian Elephant, but is a
vital question ‘ pro aris et focis,’ affecting the whole of the
Hlephantide, fossil and recent. For this reason I must be
permitted to examine it in some detail.

And first as regards the asserted number in the African
Hlephant. Professor Schlegel twits Cuvier with having
neglected to compare skeletons of the different species of
Elephant, and having thus deprived himself of the merit of
the discovery of the third living species. Is the reproach
well founded? The only skeleton of the African form which
existed in the Parisian collections when Cuvier died, and
even when De Blainville wrote upon the family in 1844,? was
that of an adolescent female which lived for some time in
the menagerie of Louis the Fourteenth at Versailles.? It
was imported from Congo; and we have the expressed or
implied authority of four most eminent and experienced
French comparative anatomists, namely, Daubenton, Cuvier,
Laurillard, and De Blainville, that it had only 20 dorsal
vertebrae and 20 pairs of ribs. Perrault, who dissected the
animal, assigns the same numbers; and the accurate Dau-
benton, who enumerates the dimensions of all the bones in
such minute detail, says—‘ I] y a vingt vertébres dorsales, et
vingt cdtes de chaque cété.’4 He assigns the following num-
bers to the different divisions of the column :—7 cervical, 20
dorsal, 3 lumbar, 3 sacral, 31 caudal vertebree, and 20 pairs of
ribs, of which 7 are true and 13 false. Laurillard, as is well
known, stood to Cuvier in the same relation of aid as
Daubenton did to Buffon, although he never was formally
recognized as his collaborateur. When Perrault’s skeleton
passed into their charge, Cuvier could only state the number
which they saw, and finding the dorsal vertebree to be the
same as in the Indian skeletons which he had dissected,
namely 20, he naturally assumed that to be the normal
number in both the living species, as Peter Camper did on
the same grounds.

Skeletons of the African Elephant are very scarce in Eng-

1 Bijdrage, &c. Hleph. Sumatranus; | Hist. des Animaux, 1734, Part iii.,
vide Translation by Dr. Sclater, ‘ Na- | Pl. xxiii.
tural Hist. Review,’ vol. ii. p. 78. 4 Buffon’s ‘ Hist. Natur.’ 4to tom. xi.
2 ‘Ostéographie :’ Eléphants, p. 5. p. 1138.
3 Perrault, Mémoire pour sery. a

UNITY OR PLURALITY OF SPECIES. 265
land. I know of two only, to be found in public collections :
the one of an adolescent animal in the Museum of Saffron
Walden, mounted, and therefore less reliable; the other in
the Osteological department of the British Museum, not set
up, and in the most favourable state for examination. It is
of a young adult, in which the epiphyses are not yet united,
imported from the Cape of Good Hope; sex unrecorded.
The bones are still covered with the periosteum and shreds
of ligament, having not yet undergone the preliminary ope-
ration of cleaning. The vertebral column is in masses of
from three to five vertebrae, united by ligaments, while others
are free. On three different occasions, specially with a view
to the present investigation, has the vertebral column been
put together by me, scrupulously examining all the surfaces
of juncture from the sacrum to the atlas, and the following
results were yielded :—7 cervical v, 20 dorsal v, 3 lumbar v,
4 sacral v, 26-30 caudal vertebree, and 20 pairs of ribs, one
of the last pair being wanting.' The precise number of the
caudal vertebre could not be determined, as the terminal
portion of them is still embedded in the tail. But there are
at the least twenty-six.

_ We have thus two instances, the one South African and
the other from Congo, in which the Elephant of that con-
tinent shows only twenty dorsal vertebre. Cuvier is thus
relieved from reproach, in so far as this species is con-
cerned.

The skeleton belonging to the Museum at Saffron Walden
is that of a young but nearly adult male, which was imported
from Algoa Bay.? It was carefully examined, with reference
to the question now under discussion, jointly by Mr. W. H.
Flower and myself, and yielded the following numerical
results : cervical vert. 7, dorsal vert. 21, lumbar vert. 3, sacral
vert. 3, caudal vert. 30, pairs of ribs 21. The dentition was,
at the same time, minutely examined, and I can affirm that
the characters agreed exactly with those of the skeleton
belonging to the British Museum. The skull and other de-
tails of the bony frame were also alike. The evidence is
of the more weight, as both skeletons were derived from
Southern Africa; excluding the plea which might have been

1 Jn order to put the statement be- | affords an excellent illustration of what

yond question, as resting upon the testi-
mony of a single observer, I requested
my friend, Mr. W. H. Flower, the able
Conservator of the Museum of the Royal
College of Surgeons of England, to exa-
mine the skeleton closely, and he arrived
at the same numerical results.

2 The Museum at Saffron Walden

may be done by a small provincial town
to promote the cultivation of science. It
possesses two mounted skeletons of
large Pachyderms, which cannot be
matched by any of the Metropolitan
collections. The museum reflects great
eredit on the locality.

266 EXISTING INDIAN ELEPHANT.
urged, that they were possibly of distinct species, if they had
been procured from different parts of the Continent.!

The cases above adduced appear to establish the fact
beyond question, that the African Elephant varies in the
number of dorsal vertebree from 20 to 21.

Next, as regards H. primigenius, what reliable authority
has Prof. Schlegel for the conjectural assertion that the
Mammoth had but eighteen dorsal vertebrae and ribs? The
solitary skeleton,? reputed to be nearly perfect, of that species
known up to the present time, is the famous Adams-skeleton,
preserved at St. Petersburg, and of it there is but one
original description extant, namely that of Tilesius, who
distinctly states that it possessed nineteen dorsal vertebre
and as many ribs: ‘ Vertebrarum thoracis 19 tantum nume-
ravi, totidemque costas utriusque lateris, at plurimas e ligno
fabrefactas.’> If the statement could be trusted it would be
conclusive against Prof. Schlegel’s argument. But there are
errors of observation in the account given by Tilesius, which
divest it of authority. He describes the neck as being built
up of six cervical vertebrae: ‘Collum ex 6 vertebris com-
pressis et coarctatis compositum.’ The seventh he appears
to have transferred to the dorsal series. In the large and
finely engraved figure which he gives of the skeleton, Pl. X.,
21 vertebre are indicated by spinous processes, jointly to the
loins and thorax, and 7 to the neck. Allowing three of these
to be lumbar, 18 would be dorsal, as conjectured by Professor
Schlegel. But grave imputations have recently been cast
upon this celebrated skeleton that, like that of the Mastodon
Ohioticus of the British Museum, it is a make-up derived
from more than one individual.t Professor Piazzi Smyth

1 Prof. Schlegel throws out a conjec-
ture, that there may be more than one
kind of African Elephant ; and in sup-
port of it, he refers to two figures of
skulls in the ‘ Ossemens Fossiles,’ Plate
iy. figs 2 and 10 of vol. i., as indicating
differences of length and width; but 1
believe that they are both of the same
cranium ; fig. 10 representing the front
aspect, drawn toa scale of one-twelfth,
and fig. 2, the basal aspect on a scale of
one-fifteenth. In the latter, the inter-
maxillary bones are necessarily fore-
shortened, from the position in which
the skull has been placed, causing a de-
ceptive appearance of short tusk-sheaths.

2 Of the Mammoth-careases which,
according to the statement of Midden-
aorf, have subsequently been discovered
in Siberia, no osteological account, so
far as I am aware, has been published.

(Bullet. Acad. Petersburg. Class. Phys.
ii. p. 150.) The observations of Gle-
boff refer to the structure of the pre-
served soft parts. (Bullet. Soc. Imp.
Mose. 1846, xix. pp. 108-134.)

8 Mém. Acad. Impér. des Science. de
St. Pétersburg, 1815, tom. v. p. 503.

4 The mounted ‘Koch’ skeleton in
the National Collection, presents the fol-
lowing vertebre: 7 cervical, 19 dorsal,
4 lumbar, 8 sacral, and 19 pairs of ribs.
It was constructed according to this for-
mula, but the careful observations of
Dr. Warren, upon materials of well esta-
blished authenticity, indicate the follow-
ing numbers: 7 cervical, 20 dorsal, 3
lumbar, 5 sacral, and 20 pairs of ribs.
(Warren, Op. citat. p. 25). The lumbar
region appears to be built up of bones of
different individuals,

UNITY OR PLURALITY OF SPECIES. 267
examined it, in company with Professor Brandt, and states
that the ribs and other parts are restorations made of deal.
He sums up his account thus: ‘ The head, with much of the
skin hanging upon it, some cervical vertebre, a whole fore
leg, and more than one foot are, we believe, the genuine
Adams-Mammoth.’! There is therefore, as yet, no trust-
worthy evidence to show that the Mammoth had only
eighteen dorsal vertebrae and ribs; every presumption is in
favour of its having had at least nineteen.

Next, as to the number of dorsal vertebre in the Indian
Hlephant.—Skeletons, reputed to be Indian, abound every-
where, but strange to say, authentic materials for settling
this part of the question are rare, in consequence of the
particulars respecting their origin not having been carefully
recorded. Until lately, no one doubted that they all belonged
to the same species, and the precise locality from which they
came was considered to be unimportant.

Prof. Schlegel states that the Sumatran Hlephant has
constantly 20 dorsal vertebree and 20 pairs of ribs. That
this number does occur in the Ceylon animal also is placed
beyond question by the careful dissection of a great ana-
tomist, Peter Camper,? and by the observations of Cuvier
and De Blainville upon the skeletons of two known Ceylon
Hlephants brought from Holland to Paris in 1795. The
distinguished Dutch zoologist further states, that all the
Indian Elephants which he had examined had, without ex-
ception, only 19 dorsal vertebree and 19 pairs of ribs. That
this is occasionally or even frequently the case is beyond doubt,
from the corroborative evidence of Patrick Blair, Meckel,
Warren, and others. For, as remarked by Camper, it is highly
improbable that a dorsal vertebra should have disappeared
in macerating the bones, preparatory to setting them up.
But it is by no means equally certain that the number is
constantly limited, in the Indian form, to nineteen. Prof.
Schlegel cites the case of the Duvaucel skeleton, forwarded
from Bengal to Paris, in which there are twenty dorsal
vertebre.? But he tries to get over the difficulty of this
exceptional case by the hypothesis that the live animal may

1 «Three Cities of Russia,’ vol. ii. p. | stored in wood and gypsum; and, as re-
222. Iam indebted to Dr. J. EK. Gray | guards the vertebra, he writes: ‘ Verte-
for a knowledge of this passage; but | bree omnes genuine ossez, ideoque car-
there are good grounds to believe that | tilagine exsiccato inter omnem vertebram
the statement is unintentionally oyer- | instructs, robustiores Elephantinis.’
charged by an astronomer, giving an | (Op. citat. p. 504.) The vertebrae, ex-

opinion on a question of comparative
anatomy. For Tilesius, whose account,
however defective, shows no signs of
partiality to Adams, but the reverse,
enumerates the parts that have been re-

cepting those of the tail, were less liable
to be separated than any other part of
| the skeleton. H

2 « Anat. dun Eléphant Male,’ p. 63.

3 Nat. Hist. Review, ii. p. 74.

268 EXISTING INDIAN ELEPHANT.

have been imported from Ceylon into Bengal. I will men-
tion in the sequel the reasons, founded upon many years’
residence there, why I consider the assumption to be in the
highest degree improbable. It is rare to find the skeleton of
an Asiatic Elephant in England, the pedigree of which is so
well authenticated as to be beyond conjectures of this kind.
But there is one in London, the antecedents of which are well
known, being the skeleton of the celebrated male Elephant,
‘ Choonee,’ preserved in the Museum of the College of Sur-
geons. The young animal was imported from Bengal in the
year 1810, on board the EH. I. C. ship ‘ Astell,’ by Captain
Hay ;' and in 1826 it was shot, in the menagerie at Exeter
Exchange, in consequence of its violence from sexual ex-
citement. It bore an Indian name, ‘ Choonee,’ and on the
occasion of its slaughter it obeyed the word of command to
lie down given to it by its English keeper, in the language
of Hindostan. All the antecedents are here consistent in
proof that the animal was of a Bengal stock. I have exa-
mined the skeleton closely, and find that it has 7 cervical,
20 dorsal, 3 lumbar, 4 sacral vertebree, and 20 pairs of ribs.
The last pair have been lost, or omitted in mounting the
skeleton. The twentieth dorsal vertebra presents costal ar-
ticular cups, which are unsymmetrical and small: that on
the right side not much exceeding the size of a silver six-
pence. But the vertebra is distinctly present. This case,
coupled with the Duvaucel skeleton in the ‘Jardin des
Plantes,’ seems to establish, without searching for others,
that the Continental Elephant of Northern India varies in
the number of its dorsal vertebre from 19 to 20, as the
African varies from 20 to 21.?

The hypothesis entertained by Professor Schlegel, upon
the statement of Diard, that Ceylon Elephants are frequently
imported into Bengal is, Iam satisfied, untenable. Under
the pressure of the Great Mutiny of 1858, the Indian Govern-
ment brought Elephants by sea from Pegu and the adjoining
Tenasserim provinces to Calcutta, but none from Ceylon.
The occurrence up to that time was so rare there that the

1 Griffith’s ‘Animal Kingdom,’ vol.
ii. p. 848; and Hone’s ‘ Every-Day
Book. &c.’ Vol. 11. p. 322.

? The ingenious view advanced by
Prof. Schlegel regarding the inverse re-
lation between the number of laminein
the molars and the number of dorsal
vertebre in the different species (swpra,
p- 263), does not appear to be tenable
against the evidence adduced above, of
the numerical variability in the living
species. Nor can I assent to the infer-

ence founded upon it, that the ‘ Mastc-
dons form not a diverging, but a parallel
series with the Elephants.’ The Indian
fossil species, which have been ranged
under the designation of Stegodon, esta-
blish, through their molar teeth, a mani-
fest and nearly unbroken passage from
the Mastodons into the true Elephants.
[ Vide Quart. Journ. Geol. Society, 1857,
vol. xiii. p. 314, and antea, pp. 9 and 82.
—Ep.|

UNITY OR PLURALITY OF SPHCIES. 269

debarcation of the animals, slung through the air, was
figured in the ‘Illustrated London News’ of the day as a
remarkable event. Young Hlephants are, I believe, never
imported from Ceylon to Calcutta ;! and ‘Choonee’ was ex-
ported thence as a very young animal. Ceylon Elephants
are exported to the adjoining peninsula; but they are com-
monly reserved for the priests of the pagodas, for the chiefs
of Southern India, and for the commissariat demands of the
Madras and Bombay Presidencies. I doubt if the Ceylon
Elephant could endure the winter cold of the North-Western
provinces, exposed in the open air. I have been on the back
of an indigenous Hlephant, in the valley of Deyra, in the
North-Western provinces, which is constantly resorted to
by herds of the wild animal, when the thermometer stood
before sunrise at 22° Fahr.; and Sir Andrew Waugh informs
me that during the measurement of the base-line for the
Trigonometrical Survey, in Chuch, the temperature fell to 15°
Fahr., with Hlephants in the camp, exposed to the open air.

On a review, therefore, of the whole case, the evidence in
every aspect appears to fail in showing that the Elephant of
Ceylon and Sumatra is of a species distinct from the Conti-
nental Indian form. Having had opportunities of observing
the animal along a range of habitat, which rarely fall to the
lot of a single naturalist, I have felt called upon to express
an opinion on the moot question. These embraced a resi-
dence during many years at Suharunpoor, in lat. 30°, near
the extreme northern range of the species, close to jungles
where wild Hlephants abound, and which my duties led me
frequently to explore. In 1832, I was present at the ‘Koom,’
or great Religious Fair, which takes place at Hurdwar, on
the Ganges, after each cycle of twelve years.2 Vast multi-
tudes of devotees, and others of all ranks and castes and of
both sexes assemble there from the most remote parts of
India and the surrounding countries, all the wealthiest
classes bringing Elephants with them.? On that occasion I

1 Mr. Blyth, in a late number of the Madras and Bengal, showing the vast
‘Journal of the Asiatic Society of Ben- | number of Elephants occurring in the
gal, received since the above remarks | forests of the Trans-Gangetie provinces,
were made, confirms the statement. By a | and the adjoining districts of Siam.
return received from the Military Com- | (Op. cit. 1862, No. ii. p. 174.)
missariat Office, at Calcutta, it appears * *Kumbha Mela.’ Duodecennial,
that 826 Elephants were imported there, | when Jupiter is in Agwarius, and the
from Moulmein and Rangoon, in the | sun entering into Aries. (Vide Raper.
years 1857 to 1859. ‘No Elephants,’ | Asiat. Research. Vol. xi. p. 456.)
it is added, ‘were received at Caleutta| % General Hardwick, who was pre-
from Ceylon.’ A communication from | sent at the ‘Koom’ Fair of 1796, esti-
Col. Phayre mentions that in the seven- | mates the number of human beings then
teen months, from Dee. 1857 to April | assembled to have exceeded two and a
1859, no fewer than 1,034 Elephants | half millions! doubtless an exaggeration.
were shipped from the same ports to | Five hundred devotees of one sect were

270 EXISTING INDIAN ELEPHANT.

endeavoured, through the native officers connected with the
administration of the Fair, to ascertain the number of Ele-
phants then crowded within a small area, and the return
made was about eleven hundred, derived from all parts of
India, the majority of which passed under my eye. I have
seen the Elephants of Pegu and Siam im the forests of the
Tenasserim provinces, and the Ceylon Elephant in its native
island. The only geographical forms of the Asiatic species
which I have not examined alive are those of Cochin-China,
Borneo, and Sumatra. The result of this range of observa-
tion, combined with long osteological study, has been to
establish the conviction in my mind that there is but a
single species of Asiatic Elephant at present known, mo-
dified, doubtless, according to his more northern or southern
habitat, but not to an extent exceeding that of a slight
geographical variety.

It is the more necessary that the subject should be tho-
roughly investigated, since upon the hasty assumption that
the Elephants of Ceylon and Sumatra belong to a distinct
species, a speculation has been put forward which seeks to
explain it by means of a former direct continuity of land
between the two islands.’ But the inferences of physical
geography and of geology are alike opposed to the conjecture.
The range of low hills which forms the spine of the Malay
Peninsula, and which is separated by a narrow interval only
from the Islands of the Archipelago, can be traced north,
increasing in height and development till it jos on with
the Himalayahs ; while Ceylon, as has often been remarked,
presents all the physical characters of being a severed portion
of the distinct mountain-system of the Western Ghats. With
certain exceptions, the Mammalian fauna, as a general rule,
confirms this view, as do also recent investigations on the
Flora of the mountainous regions of the adjoining Indian
Peninsula, near its extremity. That a connection formerly,
and at no very remote epoch, existed between the Malay
Archipelago and the continuous main land, is clearly indi-
cated by the species of the large Mammalia common to both,
including Rhinoceros Sumatranus, R. Sondaicus, Bos Sondaicus,

killed in an affray by the Sikhs. Ido
not vouch for the aceuracy of the num-
ber of Elephants reported to me, on

and Tanjore, &c., from the necessity of
taking a cortege of Elephants with them,
when they attend the Koom Fair in
future.

the occasion above referred to; but I
believe it to have been under the truth,
rather than above it. I mention this,
the more especially, as probably no such
assemblage of Elephants will ever again
be seen at Hurdwar. The facilities of
railway travelling will relieve the Princes
of Southern India, such as Travancore

(Hardwick, Op. cit. vol vi. p.
312.) During the ‘Koom’ of 1760,
eighteen thousand Bairagis (Fakirs of one
sect) are said to have been slaughtered
by the Gosains, another sect. (Op. cit.
vol. xi. p. 455.)

1 Tennent. ‘Nat. History of Ceylon,’
1861, pp. 61-67.

ASSERTED OCCURRENCE OF MASTODON IN AUSTRALIA. 271
Tapirus Malayanus, Styloceros Muntjac, Nemorhedus Swma-
trensis, Ursus Malayanus, &c., the majority of which range as
far north as Pegu, or further.!| That the Indian Elephant
should have participated in the same common range is thus
relieved from any plea of improbability ; while the specula-
tion that Sumatra was in direct continuity with Ceylon
within the period of the existing fauna is beset with insur-
mountable difficulties. In the view here taken it is also
needless, since the species may have spread southwards from
a common centre, on both sides of the Bay of Bengal, and on
its eastern shore into the promontory which formerly forked
the Indian Ocean. The speculation here controverted
appears to rest upon grounds as fallacious as those which led
De Blainville, mainly upon spurious Proboscidean evidence,
to conjecture that Australia was formerly a dependency of
the American continent.?

§ 10. AsserTED OccURRENCE oF MastTopon In AUSTRALIA.

Next to the Hquide, the Prodboscideans are among the
most cosmopolitan and widely distributed of Ungulate Mam-
malia. Dinotherium occurs alike in the Miocene deposits of
India and Europe; while species of Mastodon and Elephant,
extinct or living, have been found over the whole surface of
Hurope, Asia, and Africa, and in both divisions of the
American Continent. On the other hand, as has often been
remarked before, Australia has a living fauna, so low and
backward in the scale of organization, that it has struck
those who have reflected on it in the light of being an
arrested fragment of an older world, in which progress was
suspended, whilst in the other continents it was being
steadily sustained by the appearance of higher and higher
forms. With the exception of the Dingo, which is believed
to have accompanied man,’ and of a certain number of in-
digenous Rats, the existing mammalian fauna of Australia
is exclusively restricted to marsupial forms. The extinct
fauna, which has been so ably investigated by Professor
Owen, so far as it goes, bears the same character. The

1 Cantor, ‘ Journ. Asiat. Soe. of Ben-
gal,’ 1846, vol. xv. p. 275.
? Ostéographie: ‘ Dinotherium:’ p.

8 Prof. McCoy, ina recent comparison
between the ancient and modern natural
history of Victoria, states that he had
identified remains of the Canis Dingo in

the bone-caverns lately opened beneath |

the basalt-flows at Mount Macedon.
They were found associated with those of

Macropus Titan, and of recent species
of Hypsiprymnus and Hydromys. He
infers from this and other arguments
that the Dingo is an indigenous animal.
| But there is no evidence that man may
/not have then been an inhabitant of
Australia, and the Dingo introduced
| along with him. The latter still stands
out, a symbol of isolation (Annals and
Mag. of Nat. Hist. 1862, 3d Ser. vol.
| ix. pp. 145, 147).

272 MASTODON.

colossal Diprotodon and Nototheriwm, with the carnivorous
Thylacoleo, have died out, and the giant Kangaroos, &c.,
have dwindled down into their small-sized living representa-
tives. But except in bulk and in the extinction of certain
types, there is no indication that the modern fauna has
degenerated from a higher to a lower grade of organization.
To this general rule there is only one asserted exception,
which, however, is of a very important order, being the so-
called Mastodon Australis of Prof. Owen. I have long enter-
tained doubts regarding the authenticity of the solitary molar
tooth, upon which the conclusion mainly rests. These I have
already advanced in an abridged form ;' but as the assertion
has since then been repeated by its author, it is full time
that the case should be either established or confuted, more
especially as the asserted exception, coming forth under the
authority of so eminent a name, has been commonly adopted
by paleontologists.

In 1843 Professor Owen published the description and
figure of a fossil femur of large size, discovered by Sir
Thomas L. Mitchell in Darling Downs, SW. of Moreton
Bay, in Australia.? It was compared with the corresponding
bone of Mastodon giganteus, and inferred to be of a Masto-
dontoid quadruped. But a perfect femur of Diprotodon
Australis, acquired within the last few years for the British
Museum, along with an entire cranium and other fine
remains, places it beyond doubt that the Darling Downs
specimen, now preserved in the Museum of the College of
Surgeons, is of the Marsupial Diprotodon, and not of any
Proboscidean form.

In the following year (1844), the same author published a
figure and description of ‘a fossil molar tooth of a Mastodon
discovered by Count Strzlecki in Australia,’ which he pro-
visionally named M. Australis; and he describes it as bearing
a close resemblance to the molars of M. angustidens of
Hurope.? In his report ‘ On the Fossil Mammalia of Austra-
lia,’ communicated to the British Association in 1844, the
following paragraph occurs :

‘T cannot conclude, without adverting to the singular excep-
tion which the Mastodon forms to the continental localization,
not only of existing but of Pliocene and Post-Pliocene extinct
genera of mammalia above briefly dwelt upon. The solitary
character of the exception helps rather to establish the
generalization, at least I know of no other extinct genus of
mammal which was so cosmopolitan as the Mastodon. It

1 Quarterly Journ. Geo. Soc. vol. xiii. | _* Annals and Mag. of Nat. History,
p. 319, Synop. Table. [See also vol. i. New Ser. vol. xi. p. 8, fig. 1.
p- 106.—Eb. | | 8 Op. cit. vol. xiv. p. 268.

ASSERTED OCCURRENCE IN AUSTRALIA. 273

was represented by species for the most part very closely
allied, if actually distinct, in Europe, in Asia, in North and
South America, and in Australia; it is the only aboriginal
genus of quadruped in that continent which was represented
by other species in other parts of the world.’! Here is an
exception, the importance of which, if sound, can hardly be
overrated, in reference to the laws which governed the dis-
tribution of the extinct mammalia of the Australian con-
tinent. The identification upon which it rests has not yet
been withdrawn, so far as I am aware, by the author; and
in his inaugural address to the British Association at Leeds,
he re-affirms it twice, in the remarks upon the geographical
distribution of animals.? As Professor Owen has not pub-
lished others, it is presumed that the evidences there referred
to are derived either from the remains received from Sir T.
L. Mitchell, or from the molar tooth brought by Count
Strzlecki. The former being of Diprotodon, the onus pro-
bandi now rests with the latter, which, also, is preserved in
the Museum of the College of Surgeons. The specimen
consists of a very perfect and intact germ-of a back molar.
The enamel-shell is completely formed, but the pulp-nucleus
had only been partially calcified, so that the ivory is limited
to a thin layer below the enamel, upon which the re-
entering angles of the transverse ridges are distinctly visible
underneath. No part of the ivory base, or fangs, had been
formed, nor is any trace of cement visible upon the crown-
surface. The specimen is entire, with the exception of a
slight fracture at the top of the inner tubercle of the front
ridge, which is decurrent to the base in a vertical fissure of
old date, being filled up with matrix. The tooth is the
penultimate true molar (m. 2) of the lower jaw, left side ;
the crown is composed of three very distinct transverse
ridges, divided in the longitudinal direction by a distinct
bipartient fissure into an outer and inner division, each
composed of a pair of high and obtusely conical thick points.
The outer division of each ridge throws out, both in
front and behind, a solitary outlying tubercle, attaining a
lower elevation than the principal points. These tubercles
of the contiguous ridges are connate, so as to form a
bridge connecting each outer pair of mammille and block-

1 British Association Report, 1844, ‘In the formation of these recent
pp. 223, 239. tertiary periods, and in the limestone

2 «T have received evidences of Ele- | caverns of Australia, abundance of
phantine species from China and Aus- | Mammalian fossils haye been found,
tralia, proving the Proboscidean Pachy- | and, with the exception of the single
dermsto have been the most cosmopolitan | tooth of a Mastodon, every one of them
of hoofed quadrupeds. (Brit. Assoc. | has proved to be marsupial species.’
Report, 1858, Address, p. ixxxvi.) (Idem, p. Ixxxyiil.)

VOL. II. ab

274 MASTODON.

ing up that part of the valley which lies between them,
while the inner pair of points, belonging to each ridge,
is free from accessory tubercles, thus leaving the portion of
the divided valleys between the inner points open. A small
anterior talon, running outwards, descends around the base
of the outer tubercle of the front ridge, and a larger posterior
talon, composed of two or three tubercles, is appended to the
posterior end of the outer division of the last ridge, but free
from any connection with the inner division of the same
ridge. The tooth, therefore, belongs to the sub-genus Trilo-
phodon, and to that section of it which may be characterized
by ‘Colliculi obtusi, vallicule interrupte. The necessary
consequence of the form of the crown, as above described, is
that in the progress of wear, when the ridges are ground
down, the outer pair of points with its outlymg appendages
must yield a trefoil or complex pattern to the disc of
abrasion; while the inner pair, being simple, would yield an
elliptical and transverse disc, free from any complication.
This is the character which distinguishes Mastodon Andiwm
from Mastodon Humboldt. As these species are but imper-
fectly known in England, and one of them still more imper-
fectly represented by specimens, a few remarks upon them
may not be considered out of place on the present occasion.
Although these names were vaguely imposed by Cuvier, we
are indebted to Laurillard’ for the first accurate definition of
their distinctive characters, which has been confirmed b
Gervais, from the fine series of remains brought by Weddell
from Bolivia.? Both belong to the subgenus Trilophodon. In
M. Humboldtii, both the inner and the outer divisions of
each ridge are flanked by outlying tubercles, so that the
valleys are blocked up, and the mammille being channelled
vertically, a very complex pattern is yielded by the discs of
wear ; two trefoils are produced, separated by a cleft, some-
what as in the molars of Hippopotamus, or, as it has been
happily expressed by Gervais, ‘ Deux figures en tréfle adossées
par leur base.’* The valleys are covered with a thick coat
of cement, and the lower jaw is destitute of an incisive
beak. In M. Andiuwm there is but a single trefoil, accom-
panied by an elliptical transverse disc to each ridge, the
valleys are sparingly invested with cement, and the symphy-
sis of the lower jaw is produced into a long massive and
deflected incisive beak, as in Mastodon angustidens. This
beak, as shown by the young animal, is figured by Laurillard
in d’Orbigny’s ‘ Voyage; ’ and I have seen at Geneva the cast

} Alcide dOrbigny’s ‘ Voyage dans | Fossiles de l’Amérique Méridionale.’ Ex-
VYAmér Méridion.,’ Géol. p. 144, Pls. x. | pédition de Castelnau, 1855, p. 14,
and xi.; and Dict. Univers. d’Hist. | Pl. v.

Natur. tom. viii. p. 30. 3 Op. cit. p. 18.
2 «Recherch. sur les Mammiféres

ASSERTED OCCURRENCE IN AUSTRALIA. 275
of it, as presented by the adult animal. The specimen was
brought by M. H. de Saussure from Mexico (antea, p. 226),
and the beak in this case bore the base of a very large
incisor on one side.! Mastodon Humboldtii is found in Co-
lombia, Buenos Ayres, and Brazil: M. Andiwm, chiefly in
Chili, Bolivia, and Peru; the valley of Tarija, in particular,
abounds in remains of this species. The reputed Australian
molar agrees so closely with specimens of M. Andium,
brought by Weddell from Tarija, which I have studied in
the Paleontological Gallery of the Jardin des Plantes at
Paris, that I have failed to detect any sufficient character by
which to distinguish them.? They agree also in mineral
condition, and in the dark brown glossy colour of the enamel,
where denuded of matrix.

As regards the history of the reputed Australian specimen,
unfortunately it was not seen in situ by Count Strzlecki.
That enterprising traveller, whose explorations embraced
North and South America, Australia, the Javanese Islands,
&e., states that he ‘bought it from a native at Boree, the
sheep-station of Captain Ryan, through the agency of the
overseer of that station. The native, in giving the bone,
stated that similar ones, and larger still, might be got
further in the interior ; but that, owing to the hostility of a
tribe upon whose grounds the bones are found, it was im-
possible for him to venture in that time in search for more,’
&c.? The account given in Professor Owen’s paper differs
in some respects: being to the effect, that the specimen was
brought by a native to Count Strzlecki when exploring the
ossiferous caves of Wellington Valley; ‘ and that the native

1 Laurillard inferred from d’Orbigny’s
figure, that, although the beak was elon-
gated, from its tenuity there were no
incisors to the lower jaw in this instance;
or that, if ever present, they were rudi-
mentary, and had been shed early in life.
In the specimen figured by Dr. Wyman,
showing the lower jaw of an adolescent
animal, the symphysis is somewhat
elongated, but blunt, and there is no ap-
pearance of its having heldincisors. The
possession of mandibular incisors may
haye been a sexual distinction. (United
States’ Australian Expedition, Pl. xii.
figs. 1 and 2.)

2 On the oceasion where I first ques-
tioned the authenticity of the reputed
Australian Mastodon, I was led to
identify it with M. Humboldtii, instead
of M. Andium (vide Quart. Journ. Geol.
Soc. 1857, vol. xiii. Synoptical table,

* Antea, p. 14.—[Ep.]

p. 319*). On the same occasion (op. cit.
p- 313f) I called attention to the ex-
ceptional character of certain specimens
of M. Andium, as if hesitating between
Tetralophodon and Trilophodon. 1 be-
lieve the species will prove to belong to
the latter group. [As early as 1846,
Dr. F. pointed out that the tooth of the
so-called Mastodon Australis did ‘not
furnish characters sufficient to dis-
tinguish it from M. Andiwm? See vol. i.
p- 106.—Ep. ]

° Stazlecki, ‘Physical Description of
New South Wales,’ &c., 1845, p. 312.

* Boree Creek, an affluent of the
Lachlan River, is in the Ashburnham
District, north of Canobalas Mountain ;
while Wellington Valley is on the
Macquarie River, an affluent of the
Darling. A considerable tract intervenes.

t Antea, p. 8, note 2.—[Ep.]

T 2

276 ASSERTED OCCURRENCE OF MASTODON IN AUSTRALIA.

stated that it was taken out of a cave further in the interior.’
But the fossil bones of the ossiferous breccias of that valley
are covered with a bright coloured ochreous clay, not a trace
of which is to be seen on the fossil molar, the matrix of
which is of an entirely different character. Settlements have
been pushed into the interior since 1843, far beyond Boree
and Wellington Valley; while the country has been pene-
trated in various directions by exploratory expeditions.
Stupendous remains of Diprotodon and Nototheriwm, in the
finest state of preservation, have been discovered, and remains
also of the very remarkable form named Thylacoleo; but
during the twenty years which have lapsed, not a trace has
been detected, so far at least as published accounts go, of
anything confirming the inference of an Australian Mas-
todon. Where remains of the Proboscidea occur they are
commonly found in abundance ; and the colossal size of the
bones has led, in all ages and in all countries, to their
attracting lively attention. The absence of direct testimony
in the first instance, the conflicting statements regarding the
place and conditions of occurrence, the discordance of the
matrix, and the failure of subsequent confirmatory evidence,
coupled with the fact of the solitary molar having been
identified as being of a South American species, lay the
authenticity of the specimen open to grave doubts. If an
American form, how did this unique morceau get to Australia ? ”
If Australian, how has the Mastodon alone, of all the higher
placental mammals, broken through the barriers of marsupial
isolation, characteristic of the great southern island? Of
the alternatives, there are probably few paleontologists who
will be disposed to seek for an explanation in the naive con-
jecture of De Blainville, that Australia, to meet the require-
ments of the case, was in connection with America within the
Pliocene period.* It seems more probable that some un-
intentional error has got mixed up with the history of this
remarkable fossil; and until further confirmatory evidence is
adduced, of an unimpeachable character, faith cannot be
reposed in the reality of the asserted Australian Mastodon.*

1<Tt is partially mineralized and
coated with the reddish ferryginous
earth, characteristic of the Australian
fossils discovered in the Wellington
ossiferous caves by Sir T. Mitchell.’
(Annals and Mag. of Nat. Hist., vol.
xiv. p. 270.)

2 Prof. Morris authorizes me to say,
that when examining the Paleozoic
fossils, brought home by Count Strzlecki,
from Australia, of which he gaye an
account, he found certain specimens
accidentally mixed up with a series

which he suggested could not have been
obtained from the same locality, and on
his remonstrance to this effect, Count
Strzlecki assented to their omission from
the described list. Prof. Morris suggested
that the exceptional specimens were
probably of South-American origin.

3 The fact that such a hypothesis
was advanced shows the responsibility
involved in the publication of the data
which gave rise to it.

4 Since the above remarks were written
Professor Owen again brought forward

LIVING AND EXTINCT ELEPHANTS—THEIR FOOD. 277

§ XI. Foop or Livine anp Extinct ELEpHants.

The alimentary habits of the Asiatic Elephant, in the wild
and subjugated state, have been so carefully observed, that
there is, perhaps, no other pachyderm with which, in this
respect, we are better acquainted. But the same cannot be
said of the African species; the details of the vegetable
matters which constitute his staple food are only known in
a very general way, although it is certain, from the difference
of the vegetation of Southern Africa, where he now exists in
great force, and of Northern and Western Africa, near the
foot of the Atlas, where he abounded within the historical
period, that his food must vary within a considerable range
of species. The teeth of the Asiatic and African Elephants
are so differently modified, and the trees on which they
browse are so distinct, that the Asiatic species would pro-
bably be distressed for food, where the African finds it in
abundance, and vice versd. Both are represented in the fossil
state by species having molars constructed more or less after
the patterns respectively yielded by them, and I propose to
consider how far our knowledge of the former will assist us
in speculating regarding the alimentary habits of the latter.

(a.) Food of the Indian Hlephant.—The ‘Sal,’ or ‘ Tarai’
Forests, which stretch at the foot of the Himalayahs, from
lat. 30°, where the Ganges and Jumna escape from the moun-
tains, to the Brahmapootra, embracing a range of several
hundred miles, are here selected to furnish the chief illus-
trations which I have to adduce. They everywhere abound
with Elephants, southwards from lat. 30°, which may be
regarded as the extreme northern limit of the habitat of
the species at the present day. Forests presenting similar
physical characters extend along the continuation of the
same range, through Sylhet, Chittagong, Arracan, Pegu,
and the Tenasserim provinces, to the point of the Malay
Peninsula; they become more and more tropical in their
vegetation, and, as a general rule, the Elephants improve in
size, form, and vigour, according to their more southern
habitat. .

The Sal Forests are densely covered with arboreous forms
belonging chiefly to the following Dicotyledonous genera :—

the ease of the Australian Mastodon, as
a proof of the remarkable geographical
distribution of the Proboscidea, in a
communication which he delivered to
the British Association at Cambridge,
on Oct. 4, entitled, ‘On a tooth of
Mastodon from the Tertiary marls near
Shanghai.’ In the subsequent discussion,

he frankly abandoned it, in consequence
of the doubts then urged regarding its
authenticity. As the asserted fact has
taken deep root in systematic works, it
is still necessary that the refutation
here embodied should appear in the
records of Science. (Vide ‘ Parthenon,’
Oct. 11, 1862, p. 754.)

278 LIVING AND EXTINCT ELEPHANTS.

Vatica, Pentaptera, Terminalia, Conocarpus, Casearia, Dal-
bergia, Cedrela, Buchannania, Semecarpus, Boswellia, Spondias,
Odina, Garruga, Cathartocarpus, Bauhinia, Butea, Erythrina,
Acacia, Robinia, Moringa, Kydia, Sterculia, Bombax, Grewia,
Murraya, Glycosmis, Citrus, Nauclea, Hymenodictyon, Ronde-
letia, Schrebera, Hugenia, Careya, Ulmus, Gmelina, Premna,
Emblica, Rottlera, Briedelia, Ehretia, Tetranthera, COordia,
Wrightia, Holarrhena, Antidesma, Putranjiva, Trophis, Cochlo-
spermum, Batis, Diospyros, Bassia, Morus, Ficus, &e.

But of the large number of species belonging to these
genera, a very small percentage only of the aggregate mass
of forms enters into the food of the Indian Elephant; the
reason of this being, that some of the species, such as the
‘Sal’ (Vatica robusta) and other predominating trees, which
extend for miles nearly to the exclusion of other trees, con-
tribute nothing to the aliment of the animal. In fact, the
range of his arboreous selection is restricted within a narrow
circle, and mainly to the foliage and branches of trees that
abound in milky juice which is not acrid, belonging to the
fainilies of the Moree, Artocarpee, and Sapotacee, such as
species of Ficus, Batis, Artocarpus, Bassia, and Mimusops.! Of
these, by far the greater part of his staple food is derived
from the colossal fig-trees which abound in the forests of
India; such as Ficus Indica, the ‘ Bur,’ or Banyan-tree; F.
religiosa, ‘ Peepul,’ or ‘ Bodhi-drooma’ (Tree of knowledge) ;
F. venosa, ‘ Pilkhun’; I. cordifolia, ‘ Gujeena,’ or ‘ Assoud’;
F. glomerata, ‘Goolur’?; F. Tsiela, ‘Kuth-bur’; and in
Assam, Ficus elastica, or the ‘ India-rubber tree,’ besides
other more southern species of similar habit and properties.
The strong partiality of the Elephant for these trees is so
well known to the natives, that the ‘ Obees,’ or Pit-falls, for
entrapping the animal are invariably constructed in their
neighbourhood, and many of their old Sanscrit names con-
nect them specially with the Elephant.? He tears down their
branches, and crunches the twigs and leaves, stripping off
the lactiferous bark of the larger boughs. The Elephant of
the ‘Sal’ Forests also derives occasional food from the foliage
and fruit of Artocarpus Lakoocha, ‘Dhao’; Batis awrantiaca,
‘ Puneeala’ ; Bassia latifolia, ‘Muhowa’; and among others
from the fruit of Feronia Elephantum, ‘ Kuth-bel’; gle
marmelos, ‘Bael’; Diospyros tomentosa, ‘Teindoo’; and in the
Southern forests, from. the huge induviated fruits of certain
species of Dillenia, &c. Of aliment derived from the roots of

1 Also Mesua ferrea (Nat. Ord. | ‘Gujashun,’ and ‘Gujbhukshuk’; all
Clusiacee), on the authority of Tennent’s | being to the effect of ‘ food of Elephants.’
Nat. Hist. of Ceylon, p. 230. (Vide Madden, Journ. Asiat. Soc, Beng.,

2 “Nagbhundoo, ‘ Koonjurashun’; | vol. xvii. p. 380.)

THEIR FOOD. 279

Dicotyledonous trees and shrubs, such as the African Ele-
phant is said to affect, I know of but one form in the ‘ Sal
Forests’ which the Indian species is known to touch, namely,
the huge tuberous dilatation of the ligneous root of the
Scandent, Pueraria tuberosa, ‘Sural.’ The fruticose and
herbaceous Dicotyledons, the foliage and stems of which
may enter into his occasional food, I do not attempt to
enumerate.

Among the monocotyledonous families, a very large portion
of his habitual fare is derived from the Graminee, and more
sparingly from Palms; of the former, he luxuriates on the
young shoots and tender foliage of various species of Bam-
boo, which occur in vast abundance, together with the fleshy
albuminous fruit of Beesha Rheedii, found in the southern
forests. The ‘jhils,’ or swamps, to which he resorts, are
sheeted with the gigantic reeds of Arundo kurka, ‘Nul,’ the
young culms of which, together with the stems and leaves of
Typha EHlephantina, ‘ Patela,’ at certain seasons, constitute a
favourite food of the Indian Elephant. The open glades and
prairie lands are covered with species of Saccharum, form-
ing what is called ‘Grass Jungle,’ composed chiefly of S.
spontaneum, ‘ Kas,’ interspersed with S. fuscwm, ‘Tat,’ S.
Sara, ‘Surkura, or *Moonj,’ S. exaltatum, ‘Suroo,’ &e.
Clumps of these grasses are twisted up by his trunk, in his
journeys to and from the forests; they are beaten against
his legs to free the roots from sand, and then subjected to
mastication. The sand which still adheres to these grasses,
together with the large quantity of silica contained in the
leaves and culms of Saccharum spontanewm, the most cha-
racteristic species of the grass jungle, performs an important
duty in the economy of wear of the Elephant’s molar teeth.!
Palms, which are stated to occupy the first rank in the
favourite food of the animals in Ceylon,? are represented in
the ‘ Sal’ Forests by species which either do not, or hardly
at all contribute to it: being limited to Calamus Roylei,
Phenix acaulis, and Harina oblongifolia. But in the more
southern forests they are replaced by various genera and
species, the tender and farinaceous leading shoot of which,
as in Ceylon, is eagerly eaten by the Elephant. But com-
pared with the wild fig-trees, bamboos, and other grasses,
they constitute a subordinate part only of the food of the
wild animal. When he makes a raid into cultivated tracts

1 The excessive abundance of silicain ment is rounded off, as I have repeatedly
the culms and leaves of S. spontanewm witnessed.
is practically shown when it is attempted * Tennent, Nat. Hist. of Ceylon, p.
to mow it withan English seythe. After 230,
a few sweeps, the edge of the imple- |

280 LIVING AND EXTINCT ELEPHANTS.

he commits great havoc upon sugar cane, rice fields, plan-
tains, and many other cultivated plants ;1 but these incidents
form only interludes in his established alimentary habits.
His dung in the wild state commonly presents a large pro-
portion of contused and undigested woody fibre, in a stringy
form, mixed up with other vegetable tissues.

It is difficult to conceive of a mechanism better adapted to
the duty which they have to perform than is presented by the
molars of the Indian Elephant. Taking the three true
molars, which serve during the adult stage of the animal,
they are composed successively of 12, 16, and 24 ridges.
Each ridge has the core formed of a high wedge-shaped plate
of ivory; a continuous plate of enamel is closely folded over
these wedges, which are confluent at their base; and the
intervals between the ridges are filled up, each with a re-
versed wedge of cement, which is insinuated between the
erooves and inequalities of the enamel. When the crown is
in full activity of wear, the penultimate molar, consisting of
sixteen ridges, presents an unequal triturating surface, com-
posed of thirty-two plates of enamel, alternating with sixteen
thin wedges of ivory and as many of cement, making in all
sixty-four alternations, disposed within a length of from 83
to 94 inches. The disintegrating and bruising power of the
surface is further greatly augmented by the circumstance,
that in the Asiatic Elephant the plates of enamel are folded
vertically into a number of bold close-set zig-zags, or undu-
lations, which present a crimped edge during wear. If a
number of these plates were brought together, so as to place
their undulations in contact, an appearance would be pro-
duced analogous on a large scale to the engine-turning of a
watch-case, arranged in longitudinal lines. The three con-
stituent materials being of unequal hardness, the cement is
worn lowest, the enamel highest, and the ivory to a level
between the two. A constant equilibrium is maintained, in
the normal state, between the nature of the food, the waste
of the crown-surface, the absorption of the fangs, the forward
movement of the body of the tooth, and the replacement of
the worn-out portion by a succession of fresh plates, pro-
truded from behind.

This goes on in the wild state, but no sooner is the animal
kept in captivity than the balance is upset, and the whole
mechanism put out of gear. Instead of grass culms and
leaves charged with silicious crystals, or mechanically mixed
with sand, and of tough woody fibre and bark, requiring a
powerful process of trituration to fit them for deglutition,

1 Corse, Asiat. Researches, vol. ili. p. 229.

THEIR FOOD. 281

the animal is supplied with concentrated cereal food and
hay, with an admixture of nutritious roots and mashes, or
green fodder. The consequence is, that the crowns of the
active molars do not get worn down with sufficient rapidity
to make way for the tooth forming behind, and abnormal or
morbid results follow :—

Ist. The used surface of the crown, instead of being un-
equal and terraced, is worn smooth and flat, in some in-
stances even like a slab of polished marble.

2nd. The uncalcified back portion of the capsule of the
tooth in action, instead of remaining distinct, becomes, from
the undue pressure behind, united with the formative capsule
of the contiguous back tooth, the development of which is
not retarded, and the two separate molars are fused into one
unwieldy mass, covered by a continuous shell of cement. A
fine example of this state is presented by an adolescent cra-
nium in the Museum of the College of Surgeons (No. 2665,
Osteol. Cat.), in which two molars and apparently part of
the third in front are united into one; and the pressure has,
besides, acted so as to contract the palate, and bring the
opposite molars nearly into contact in front.

3rd. The anterior fangs of the tooth in action are gradu-
ally absorbed while the corresponding portion of the crown
remains unworn and is projected forwards, like a foreign
body, beyond the edge of the alveolus. I observed a very
remarkable instance of this morbid condition in the cranium
of a ‘Mukna’ Elephant, preserved in the Natural History
Museum at Florence. On the right side, in this specimen,
there are three molars in situ: the last in germ, the penulti-
mate partly worn, and agglutinated to it in front the ex-
truded body, without fangs, of the antepenultimate, which is
projected forwards and upwards across the diastemal inter-
val, so as actually to press against the palatine floor of the
maxillary bones. In this case the morbid pressure had caused
the absorption of the plate of bone forming the base of the
sheath of the incisor, which is indicated by a deep pit, and it
probably led to the death of the animal, with great torture.

4th. The capsule of the last molar being constrained for
room, by the undue resistance in front of it, there is not suffi-
cient space for the normal arrangement of all the plates as
they are successively calcified, and the hindermost become
distorted in position. A fine example of this malformation
is presented by the last lower molar, fig. 90, of the ‘ British
Fossil Mammalia.’! The tooth is there described as being

1 Op. cit. pp. 226 and 233, and Cat. | make the rectification here indicated,

Foss. Mam. &c. Coll. of Sur. No. 567, | since the figure has been copied by an
p. 134, It is the more necessary to } eminent French Palontologist, on the

282 LIVING AND EXTINCT ELEPHANTS.

of the Mammoth, but it is in reality a molar, disguised and
blackened by smoke, of an Asiatic Hlephant which had died
in captivity. The back plates, in this case, are pressed and
crowded upwards so as to have become nearly horizontal.
Similar instances are figured by De Blainville,' without his
having been aware of the nature and cause of the distortion.

The Elephants kept in the menageries in Europe are all,
more or less, in this morbid condition of the dental system.
They are fed on rations composed largely of turnips, carrots,
mangold-wurzel, and of mashes of boiled rice, bran, sea-
biscuit, and chaff, &c. The only hard and dry food issued
to them consists of a truss or two of hay, and the straw used.
for their litter. Ligneous food, such as they partly live upon
in the wild state, is denied to them, and the results are so
certain, that one can anywhere point out in a museum the
molar of an Elephant which has been kept in captivity. For
obvious reasons, the effects, although still discernible, are
less pronounced in the molars of Elephants which have been
retained in bondage in their native country.

The bearing of these observations upon the normal condi-
tion of teeth of the Mammoth, and its inferred alimentary
habits, will be shown in the sequel.

(b.) Food of the African Elephant.—The alimentary habits
of the Indian species are so well known, simply from the
fact, that being tamed one can observe from his back, in
beating through his native jungles, everything which he
selects and all that he passes by. The same close observa-
tion cannot be applied to the African form, as at the present
day he is nowhere in his native continent trained for the use
of man. Our knowledge of his food is, therefore, of a vague
and general character, being derived from the cursory obser-
vation of travellers, whose attention was not specially directed
to the subject.

The molar teeth of the African Elephant are intermediate,
in construction and triturating characters, between those of
the Huelephantes, or Elephants proper, and the fossil Stego-
dons. They present, in the three intermediate and last
molars for the ridge-formula, the successive ciphers 7: 7, 8,
10; while H. antiquus presents the ciphers 10: 10, 12, 16,
and EH. primigenius and EH. Indicus, 12:12, 16, 24. The
aggregate of the series of ridges in the first amounts only to
32; in the second to 48; and in the two last to 64; involving
a great difference in the triturating mechanism of the teeth.
In the African form the molars are also shorter, narrower,
authority of the work, as a characteristic | Pl. xv. fig. 9.) vi
specimen of £. primigenius. (Vide Mé- 1 Ostéographie: Eléphant, Pl. vii.
moires Acad. Montpell., tom. 1. p. 423, | fig. 6, and Pl. x. fig. 6.

THEIR FOOD. 283
and of less elevation, than in the Asiatic species. The discs
of wear, instead of the narrow transverse bands seen in the
latter, exhibit the well-known rhomboidal expansion charac-
teristic of the species. Instead, therefore, of being adapted
to contuse and triturate the branches and twigs of trees, they
are better suited for squeezing and crushing leaves, and suc-
culent stems or roots. The habits of the animal, as observed
by travellers, are in accordance with these indications. Be-
sides browsing on the foliage of the Mimosas and Acacias,
which abound in Southern Africa, they tear up the trees of
certain species of these genera by the roots, aided, according
to Pringle, by their tusk, used as a crow-bar (?), and they
devour the succulent parts of these roots in the inverted
trees.! Burchell mentions a small species of Prosopis, P.
Hlephantorhiza, as yielding a favourite food to the Ele-
phant;? and the succulent ‘Spekboom’ Portulacaria Afra,
or ‘Tree Purslane,’ is noticed by most travellers as yielding
another.

That the African Elephant, such as we now see it, for-
merly extended to the South of Europe, has been put beyond
question—Ist, by the researches of Lartet upon remains
found in the neighbourhood of Madrid; * 2nd, by the remains
discovered by Baron Anca in the cave of San Teodoro in
Sicily ;4 3rd, by a molar from Grotta Santa, near Syracuse,
described by the Canon Alessi,’ and identified by myself;
and lastly, by a molar exhumed by M. Charles Gaudin, in
1858, in a cave near Palermo. The last specimen has lately
been transmitted to me for examination, and it proves that
the African Elephant existed in that island as the cotempo-
rary of the two extinct species of Hippopotamus of the
Sicilian caves. The reputed cases of molars of the African
Elephant, from the Valley of the Rhine, described by Gold-
fuss, I believe to be spurious fossils, after having submitted
them to a careful examination.® Captain Spratt, R.N., the
indefatigable explorer of the Hydrography and Geology of
the Mediterranean, has, as already stated, lately discovered
in Malta numerous remains of a surprisingly small fossil

1 Cited in the ‘ Library of Entertain-
ing Knowledge.’ Menageries, vol. ii. p.
36

2 Acacia Hlephantina, Burch. ‘Travels
in South Africa,’ vol. i. p. 2386. Hile-
phantorhiza Burchellii, Benth.

% Comptes Rendus. 22 féy. 1858.
Tom. xlvi.

4 Bullet. Soc. Géol. de France. 2e
Sér. t. xvii. p. 684. Pl. xi. figs. 5 & 6.

5 Atti dell’ Accad. di Scienz, Natur.
tom. vii. p. 228.

6 Nova Act. Acad. Natur. Curios., tom.
x, Pl. xiv; cand) ‘tom sain el
lvii. fig. 1. A specimen of a reputed
fossil molar, of EZ. Africanus priscus, in
the Museum of Rudolstadt (Schwarz-
burg), direct testimony tothe authenticity
of which was borne by the finder when
the case was investigated on the spot by
Sir Charles Lyell, proved, on examination
in London, to be of a recent African
Elephant.

284 LIVING AND EXTINCT ELEPHANTS.

Elephant, of the sub-genus Loxodon, which I have named
EH. Melitensis.

Of the more ancient European fossil species, H. antiquus
is that which most resembles the African Elephant in the
mesial expansion of the discs of its worn molars. But the
character is shown in a much less degree, and the great
difference in the ridge-formula of the two species places
them in two distinct sub-genera. LH. antiquus, in the series,
is intermediate between H. Indicus and EH. Africanus, but
more nearly allied to the former. The crowns of its molars
indicate alimentary habits intermediate between those of the
two living species.

(c.) Food of the Mammoth.—In order to estimate the force
and value of the arguments which have been raised on this
head, it is necessary to institute a rigorous comparison
between the mechanical conditions of the molar-crowns of
the Indian Elephant and of the fossil species.

The ridge-formula is the same in both, being for the four
last teeth of the upper jaw 12:12, 16, 24. The number of
ridges in the three first of these is very constant ; the last,
as already stated, is variable within certain limits, twenty-
two being the most common number. Taking the penulti-
mate, as in the case of the Indian Elephant, the worn surface
of the crown would show sixty-four alternations of unequally
hard materials.

Although agreeing in this essential respect, there are |
important differences in the mechanical disposition of the
plates. In EH. primigenius the molars are shorter for the
number of their constituent ridges, and their crowns are
also, both absolutely and relatively, broader than in the
Indian species. The alternate successions of cement, enamel,
and ivory, are therefore more attenuated and more condensed,
and a larger number of them enter into the surface of that
part of the tooth which is in wear. Lartet fixes the number
of ridges that may be in active use at from twenty to twenty-
three in a length of about 93 inches (0°24 mét.) ; while in
the adult Indian Elephant the number of bands in the same
length is usually about sixteen. But the great difference
lies in the mechanical properties of the enamel-plates.
Instead of being thick and robust, with close-set and regular
undulations, or zig-zags, as in-the Indian species, they are
thin and parallel, the projecting edges running either
straight across, or if there is a tendency to undulation, it is
but slight, fine, and inconstant; occasionally, even, there is
irregular angular expansion, or flexuosity in the edges of
discs that are worn low down; but, as a general rule, the
plates are straight and free from waviness. It is this cha-

THEIR FOOD. 285
racter which involves the greater width of the molar-crowns
in the Mammoth ; if the undulations of the Indian Elephant
were unfolded, the crown-plates would in that species be as
broad as in the fossil one. Another difference is, that these
plates are higher in the Mammoth. In the Texan specimen
of an upper molar, mentioned above (p. 229), they attain the
enormous height of nearly eleven inches.

The triturating surface of the crown in the active molar
presents another and very significant difference. Instead of
the terraced inequalities, seen in the molars of H. Colwmbi
and H. Indicus, as described above, the worn surface in the
Mammoth is nearly flat; the enamel-edges rising but a very
little above the ivory and cement. This is a constant cha-
racter of Mammoth-molars of all ages and of all regions,
whether from the pre-glacial ‘ Forest-bed’ of the Norfolk
coast, from the volcanic gravels around Rome, from the
superficial gravels of England, from the frozen soil at the
mouth of the Lena, from Eschscholtz Bay, from the swamps
of the Ohio, or the prairie lands of Texas. In fact, the
normal condition of the molar-crown of the Mammoth
resembles that of the Indian Elephant, which has been fed
in captivity, but without the distorted arrangement of the
plates seen in the latter. This observation, so far as I am
aware, has not been made before; and the fact will explain
the reason why I have entered so much in detail into the
cause of the unnatural condition in the captive Asiatic
species.!

What, then, was the nature of the food of the Mammoth?
In speculating on this question, we have for our guidance :—
1st, the mechanical properties of the molar crowns as a dis-
integrating apparatus; 2nd, the analogy of the living species ;
3rd, the climate and implied vegetation of the habitat of the
extinct animal.

Regarded as an instrument for crunching and contusing
the woody fibre and tough bark of trees, the crown of the
molar in the Indian Elephant is manifestly much more
powerful than that of the Mammoth. The elements which
determine the ratio of force in the comparison are the

1 The distorted condition of the molars
of the subjugated existing species is
occasionally, although very rarely, seen in
the teeth of the Mammoth. A fine example
is presented by a last true molar of the
upper jaw, preserved in the Woodwardian
Museum of Cambridge, in which the five
last plates are contorted and crowded on
one side. It might serve for the molar
of a Mammoth which had been in bond-

age to man of the early ‘ Flint-knife’
period. But a natural cause of this
condition is intelligible, on the supposi-
tion that the molar which preceded it
was not opposed by a corresponding
tooth in the lower jaw; a deficiency
which is known to occur, from disease
or accident, both in living and extinct
forms. [See antea, p. 169, and vol. i.
p- 430.—Ep. ]

286 LIVING AND EXTINCT ELEPHANTS.

strength, projection, and number of the enamel-edges, the
ivory and cement being, in the mechanical aspect, but the
setting in which the plates are fixed. In the molar of the
Indian Elephant they are like the edges of thick plates of
corrugated iron, having a considerable amount of relief;
while in the Mammoth they are like the edges of thinner
flat plates of the same metal, barely elevated above their
level setting, but more numerous, in the same extent of
erinding surface, in the ratio of 5 to 4. In the former, the
tough and ligneous matters which it is known to select for its
food tell upon the triturating elements, as might be predi-
cated, in the ratio of their densities. The soft cement is
worn lowest, the plate of ivory forms a depressed band, and
the enamel-plates project over both—the wider intervals by
which they are separated contributing to facilitate the me-
chanical result required in the case. In the Mammoth the
plane of the setting remains flat, and the enamel-edges are
but slightly in relief above it. The molar in the palate of
a Mammoth from Eschscholtz Bay, in the Paleontological
gallery of the British Museum, may be cited in illustration.
If hard woody fibre entered more largely into the food of the
fossil than it does into that of the existing species it is
difficult to conceive why corresponding mechanical results
should not have followed, in the greater proportional erosion
of the cement.

It has been argued, and the reasoning has met with very
general acceptance,’ that ‘if we find in an extinct Elephant
the same peculiar principle of construction in the molar
teeth’ (i.e. as in the living forms), ‘but with augmented
complexity, arising from a greater number of triturating
plates and a greater proportion of the dense enamel, the
inference is plain that the ligneous fibre must have entered
in a larger proportion into the food of such extinct species.’?
But there are objections to the terms here used, as accurately
expressive of the difference, which are opposed to the infer-
ence. It is true that there is a greater number of thinner
enamel-plates, in the same extent of triturating surface,
which thus becomes more composite; but it is not so that
there is a greater proportion of dense enamel, nor that the
crown is more complex. The greater thickness of the plates
in the Indian species compensates for their more frequent
repetition in the fossil form; while their strong undulation

! The deduction here referred to has | very high estimate of its importance, as
been adopted by the distinguished] a result of paleontological research.
authors of the ‘Geology of Russia,’ in | Op. cit. vol. i. p. 497.
their disquisition on the ‘ Habitation and | * Brit. Foss. Mamm., p. 268.
Destruction of the Mammoths, with a

THEIR FOOD. 287

in the former necessarily renders the grinding surface much
more complex than in the latter. Let any one look at the
beautiful figure of the molar crown of the Indian Elephant
in the ‘ British Fossil Mammalia,’ cut 90, p. 233, and compare
it with cut 92, p. 237, of the Mammoth; the contrasted dif-
ferences are obvious at a glance. The latter is a mechanism
for finer disintegration; but the former, from its conjoint
properties of greater strength, complexity, and inequality of
surface, is a more powerful apparatus for crushing and con-
tusing hard ligneous fibre.

For these reasons I cannot assent to the soundness of the
asserted physiological inference, that a coarser kind of
vegetable food and a larger proportion of ligneous fibre must
have entered into the subsistence of the Mammoth than enter
into that of the living Asiatic species, or that there was any
necessary relation between the peculiar structure of its teeth
and the subarctic arboreous vegetation of Siberia, seeing
that the same structure holds in the molars of the pre-glacial
Mammoth of the Norfolk coast, and in that of Central
Italy. Professor Owen has taunted the great observers who
preceded him with having failed to follow up the inquiry
regarding the Siberian Mammoth to its legitimate conse-
quences :—

“It might have been expected that the physiological con-
sequences deducible from the organization of the extinct
Species, which was thus, in so unusual a degree, brought to
light’ (¢.e. the Adams-Mammoth), ‘would have been at once
pursued to their utmost legitimate boundary, in proof of the
adaptation of the Mammoth to a Siberian climate; but save
the remark, that the hairy covering of the Mammoth must
have adapted it for a more temperate zone than that assigned
to existing Hlephants,' no further investigation of the re-
lation of its organization to its habits, climate, and mode of
life appear to have been instituted; they have, in some
instances, indeed, been rather checked than promoted.’?

It is certainly unexpected to see it insinuated that it was
left to Pictet to point out, in 1844, that the long hair of the
extinct species appeared to fit it for sustaining a greater
degree of cold than that which the Indian Elephant now
bears. Nearly a century ago Pallas threw out the same
conjecture regarding Rhinoceros tichorhinus, upon the hair
with which it was covered; while Cuvier expressed his
opinion on the subject with characteristic precision. After

1 «La longue toison dont cet animal | qui convient 4 Véléphant de l'Inde’
était couvert semblerait méme démon- Pictet, Paléontologie, 8vo, tom. i. 1844,
trer, qwil était organisé pour supporter | p. 75.
un degré de froid plus grand que celui * British Fossil. Mamm., p. 267.

288 LIVING AND EXTINCT ELEPHANTS.

describing the nature of the hair of the Mammoth, he adds:
‘Par conséquent, il n’est pas douteux que l’éléphant fossile,
tel qwil se trowve en Sibérie, avait une fourrure d’animal de
pays froids.’' Again: ‘Ainsi non seulement il n’y a rien
d’impossible a ce qu’elle ait pu supporter un climat que feroit
pézir celle des Indes, il est méme probable, qu’elle étoit con-
stituée de maniére a preférer les climats froids.’? Here it
will be observed that Cuvier, with philosophic caution, limits
his argument to the extinct animal, such as it occurs in
Siberia, believing, as he did, that the species had also existed
in more temperate regions. But we now know that the
Mammoth roamed over Europe before the Glacial period.
Take the cases where its remains have been found in the
‘ Forest-bed’ of the Norfolk coast and in the volcanic gravels
around Rome. In the former, the vegetation, arboreous and
herbaceous, according to the determinations of Heer, closely
resembled that of the existing period, and the pre-glacial
Mammoth subsisted upon it, in association with Hlephas anti-
quus, Hippopotamus major, and Rhinoceros EHtruscus. The
Valley of the Tiber, between the Seven Hills, was formerly a
great lake,? more than 130 feet above the present level of
the river, receiving the volcanic ashes and other ejecta of the
surrounding active craters, and forming enormous beds of
travertine and gravels, in which remains of the true Mam-
moth occur associated with Hlephas antiquus, Rhinoceros
leptorhinus (megarhinus, Christol), and a species of extinct
Hippopotamus. No one, at the present day, will be hardy
enough to maintain that the Flora of Central Italy was at
that time identical with, or as limited in the number of
Arctic species as that of Siberia, where the wool-clad variety
of the north lived and pastured; for we have distinct proof
that the glacial refrigeration, which characterized the Alpine
valleys and plains of Hurope north of the Alps, was greatly
modified in intensity on the southern side of the chain. The
enormous glacier of the Valley of the Adige, after emerging
from the ‘Lago di Garda,’ melted away, leaving on the
margin of the Valley of the Po a vast mass of moraine. On
the southern side of the Apennines, glacial phenomena have
nowhere as yet been traced down upon the plains on their
flanks. Yet the Mammoth existed in Central Italy, either
before that period of refrigeration began, or when its effects
told, but inconsiderably, in that southern latitude. It would
therefore be as legitimate to detect a special relation between
the composite structure of the teeth, and the vegetation upon
which they were exercised, in the Mammoth of the South of

’ Oss, Foss. 4to Edit. tom. i. p. 196. 3 Hoffmann, Edinb. New Philos.
2 Idem. p. 200. Journ, 1829, vol. viii. pp. 85 and 96.

THEIR FOOD. 289

Hurope, as in the asserted case of the Mammoth of Northern
Asia.

Again, let us take the case of the Mammoth of Texas, and
of the other Southern States, bordering the Gulf of Mexico.
It will hardly be asserted, at the present day, that the same
arboreous vegetation extended from the upper parts of the
valleys of the Obi and Irtish across northern Asia, and from
Behring’s Straits across the surface of North America to the
warm latitude of the Gulf of Mexico. Granted that the
refrigeration of the Glacial period extended so far south, it
must have been greatly modified in intensity by the southern
latitude, as it was in the south of Europe; and that modification
was incompatible with a tree vegetation restricted to pines,
birches, poplars, willows, and junipers. We further know,
that when the Mammoth pastured along the margins of
the great swamps of Ohio and Kentucky, the vegetation then
was nearly identical with what it is now, being very different
from that of Siberia.

An inconsistency of the advocates of the doctrine here
combated is worthy of notice. While so strongly insisting
on the special relation between the teeth of the Mammoth
and the leafless tree vegetation on which he fed during
winter, it was asserted that the variety of molar on which
EH. meridionalis is founded occurs not only in England but in
Siberia, and as far north as Eschscholtz Bay.! It is well
known that the teeth of the latter species possess characters
which are very different from those of the former; having
thick enamel-plates, which are few in number and wide apart.
The special adaptation, between the teeth and food, which
held in the one was therefore absent in the other, although,
under the view here referred to, they were both said to be
found in the same Arctic localities, where they must both
have subsisted on the same impoverished Flora.

The state of our exact knowledge, at the present time,
regarding the duration, geographical range, climate, habits,
and food of the Mammoth, appears to be thus. The species
existed before the Glacial period in Europe, and survived
long after it in Europe or America. The constitutional
flexibility, which is implied by its ‘ dicyclotherian’ term in
time, is equally evinced in its vast geographical range of
habitat ; extending from the valley of the Tiber to the Lena,
and from Eschscholtz Bay to the shores of the Gulf of Mexico.
Making due allowance for the interference of the glacial
phenomena, the extremes of north and south latitude, in
which undoubted remains of this ancient Hlephant have
been found, necessarily imply that his constitutional flexi-

1 Brit. Foss. Mamm., p. 2388.

VOL. II. U

290 LIVING AND EXTINCT ELEPHANTS.

bility was, like that of man, capable of adaptation to very
great differences of climate. In Siberia, he was ‘enveloped
in a shaggy thick covering of fur, like the Musk Ox, impene-
trable to rain or cold.’!. But we are not obliged to suppose
that in his southern habit he was thus clad. The dermal
appendages are very variable and adaptive, according to
climate. The fine silky fleece, from which the Cashmeer
shawls are wove, is abundantly developed at the roots of the
long hairs of the domestic goat in the plains of Tibet, at,
and upwards of, 16,000 feet above the level of the sea, where
a highly rarified atmosphere is combined with severe winter
cold. It grows also, on the Kiang, the Yak, Cervus Wallichiui,
the Brown Bear of high elevations in the Himalayah, and on
the Mastiff Dog of Tibet. But it disappears entirely from
the same Goat, and from the Dog, in the valley of Cashmeer.
The short crisp wool of the Siberian Mammoth, which seems
to have been the most protective portion of his fur, may, in
like manner, have disappeared from the variety that lived in
the valley of the Tiber, while the bristles and long coarse
hair were more or less retained; and it is in the highest
degree probable that the species presented varieties of ex-
ternal form, dependant on the nature of the dermal clothing,
far exceeding those which are seen in existing Elephants.
That the Siberian Mammoth migrated periodically, from the
more southern forests, towards the Polar Sea, during summer,
as his surviving cotemporaries the Musk Ox and Reindeer
now do, is also highly probable ; ? but we have no grounds to
believe that the Mammoth of Southern Europe ever made
migrations to the north of the Alps.

The same constitutional elasticity which enabled the
Mammoth to endure such a variety of climates, and to spread
over such a vast geographical area, necessarily extended to
his alimentary habits. I have already called attention to
the remarkable constancy in the specific characters of the
molar teeth, alike in the pre-glacial and post-glacial, in the
extreme northern and the extreme southern forms. Their
adaptation was not special to the vegetation merely of
Siberia, but general to that of every region over which the
species spread; and up to the present time not a plausible
conjecture even has been offered as to the class of vegetable
matters which they most affected. The question of the food
of the species has not been, in the least, advanced since the
discovery by Adams of the ice-preserved carcass on the
banks of the Lena in 1803, or since the philosophic doubts
were expressed by Fleming on the subject in 1829.3 Wher-

1 Fleming. Edinb. New Phil. Journ. 1828, vol. vi. p. 286.

2 Richardson. Polar Regions, 1861, pp. 275 and 296.
8 Edinb. New Phil, Journ. 1829, vol. vi. p. 285.

THEIR FOOD. 291

ever a certain result has been arrived at regarding the
alimentary habits of the extinct Mammalia of the Glacial
period, it has only been by discovering the remains of the
food itself in some of the organs of digestion. We have the
authority of Brandt for the fact, that he extracted from the
pits of the molar teeth of the Rhinoceros tichorhinus, of which
the carcass was obtained by Pallas from the banks of the
Wiljui, part of the albuminous seed of a Polygoneous plant,
portions of Pine leaves, and minute fragments of coniferous
wood, characterized by the distinctive porous cells.!_ In like
manner, four cases have been described in North America,
where the contents of the stomach and intestines of Mastodon
Ohoticus appear to have been preserved along with the
skeletons; and the facts recorded by different observers are
so much in accordance as to leave little room for doubt on
the subject. Broken pieces of branches, varying from
slender twigs to boughs half an inch in diameter and about
two inches long, were found mixed up with more finely
divided vegetable matter, like comminuted leaves, in one
case to the amount of from four to six bushels. We have
the authority of Gceppert for the fact, that twigs of the
existing coniferous Thuwia occidentalis were identified in the
stomach of the New Jersey Mastodon; and of Professor Asa
Gray and Dr. Carpenter, both eminent microscopical ob-
servers, that the stomach of the Newburgh Mastodon con-
tained fragments of the boughs of ‘some coniferous tree or
shrub, and probably some kind of spruce or fir (Gray); and
also fragments of a quite different kind of wood (not coni-
ferous), which from its decomposed and carbonaceous state
was not determinable (Carpenter).? But these observations
do not, in the slightest degree, advance our knowledge as to
the probable food of the Mammoth; residuary bits of stick,
half an inch in diameter, are reconcilable with the mastica-
tory operation of the rude open valleys and Trilophodont
ridges of the molars of the American Mastodon; but in the
highest degree improbable as a result of the multiplex divi-
sions of the flat molar crown of the Mammoth. We must be
content to remain in the dark on this question, until the
same kind of observation is applied to the contents of the
stomach of the latter in Siberia,’ as has been so successfully
effected with the allied genus in North America.

’ Leonhard and Bronn’s ‘Jahrbuch,’ | which the details have been published,
1846, p. 878; and Bronn’s Lethza | the remains of the brain, muscles, ten-

Geognost., band. iii. p. 855. dons, and periosteum have been micro-
* Warren. ‘ Mastodon Giganteus,’ p. | scopically examined, but not the contents
166. of the stomach. (Vide Gleboff, Bullet.

% In the researches upon the latest | Sociét. Impér. de Moscou, 1846, xix. 2,
discovered Mammoths in Siberia, of | p. 108, e¢ seg.)
u2

OSSIFEROUS CAVES OF MALTA.

IV. ON THE FOSSIL REMAINS OF ELEPHAS ME-
LITENSIS, AN EXTINCT PIGMY SPECIES OF
ELEPHANT; AND OF OTHER MAMMALIA, &c.
FROM THE OSSIFEROUS CAVES OF MALTA.!

Amone the most interesting of the fossils from the Zebbug
Cave is a series of molar teeth and fragments of tusks, re-
ferred to in Captain Spratt’s papers as belonging to a fossil
Elephant. The molars comprise specimens ranging from the
first milk molar of very young animals up to what appear to
be adult teeth, and they are at once characterized, besides
other differential marks, by the singularly small size of the
species which yielded them. Warned by the great blunders
committed by Nesti, Fischer de Waldheim, and other palzon-
tologists, in having been misled by the characters of milk
teeth, to identify them as the remains of pigmy species of
Elephant, I have been chary in admitting the convictions,
which the specimens forwarded by Captain Spratt forced
upon me when [I first examined them. One of the most
characteristic of these is an upper molar of the left side,
bearing the following label. ‘Dente che si conserva nella
Pubblica Bibliotéca di Malta, e trovato in Novembre 1859 in
Malta.’ As this specimen is about to be returned to Malta,
at Captain Spratt’s request, it is necessary to make an
accurate description of it, to accompany the figures drawn
by Mr. Dinkel. (Pl. XI. figs. 1 and 1 a.)

The tooth is a well-worn upper molar of the left side,
perfect so far as the crown goes, with the exception of the

1 This description of the teeth of
Elephas Melitensis formed the substance
of a communication by Dr. Falconer to
the meeting of the British Association
at Cambridge, on October 6, 1862. From
the appended letters and notes, however,
it will be seen that this as well as other
fossil remains, found by Capt. Spratt in
the caves of Malta, were identified by
Dr. Falconer as early as July 1860.
Dr. Falconer had drawings taken from
the original specimens by Mr. Dinkel,
some of which are here reproduced.
They included illustrations of the verte-
bre, pelvis, scapula, humerus, femur,
&c. These bones were subdivided by
Dr. Falconer into three classes, viz.:

old, young, and feetal, but the only
description of them left by him is con-
tained in the appended extracts from his
note-books. This defect, however, will,
I believe, be amply remedied by Mr.
Busk, in a memoir which will shortly
be published in the ‘ Zoological Trans-
actions,’ and which contains the account
of a second species of Elephant, dis-
covered by him among the fossil remains
from Malta, and designated by him
Elephas Falconeri. Capt. Spratt’s ac-
count of the caves whence the fossils
were derived will appear in an early
number of the ‘Quarterly Journal of
the Geological Society. —[ Ep. ]

ELEPHAS MELITENSIS. 293

front portion supported upon the large anterior fang, which.
had been worn away by continued grinding action. This is
distinctly proved by the circumstance that the grinding
plane of the crown intersects the most anterior of the extant
fangs. The rest of the fangs from this point backwards to
the posterior talon are all present, but more or less fractured
or abraded. The molar is vertically fractured across through
the middle, involving the loss of the greater part of one
colline, but as the fragments fit at the base this circum-
stance does not interfere with the precise appreciation of the
crown characters. What remains of the crown is composed
of ten ridges, of which nine are more or less worn, the rest
being intact. The posterior talon consists of a single flattened
gibbous digitation appended to the last ridge, which is com-
posed of three or four digitations. The most anterior disc
of wear is vertically divided through the middle, so that the
posterior half of it only is present. The seven anterior discs
form oblong transverse depressions, bounded by parallel
bands of enamel; there not being the slightest tendency in
any of them either to crimping or to digital subdivisions
forming secondary undulations. These discs are nearly of
uniform width across, parallel, and without any indication
of the retroflected cornua at the sides, such as are commonly
seen in Hlephas antiquus. The most striking character about
these discs is the nearly entire absence of anything ap-
proaching crimping (or primary undulations) upon the edges
of the enamel-plates, as they are shown in relief upon the
surface of the crown. There is a slight appearance of ver-
tical grooving upon the cement aspect of these enamel-plates,
but considerably less than is exhibited by the molars of any
species of Elephant, fossil or recent, with which I am
acquainted. The enamel of the plates is rather thick, quite
as thick in proportion as in the existing Indian Elephant or
Hlephas antiquus. There is the slightest possible tendency
to mesial angular expansion in some of the anterior discs,
but it is barely appreciable, while in some others of the
specimens this character is somewhat more pronounced.
The talon consists of but a single flattened digitation, and
there is this remarkable circumstance about it, that it does
not anywhere bear the slightest indication of any dise of
pressure upon it, arising from the protrusion of another
molar advancing from behind. The last or tenth ridge of
the specimen I have reckoned as such, and not as a talon
appendage, from the fact that it is continued vertically
down into the large posterior fang, and distinctly within its
bearing. The crown, in proportion to the height of the
plates, is narrow. The discs of wear are much abraded

294 OSSIFEROUS CAVES OF MALTA.

in front and are in close contact, the enamel-plates nearly
touching each other, but they are well separated backwards,
and the whole of the crown is enveloped by a coat of cement,
which at the sides is seen to be of considerable thickness.

I have reckoned that what remains of the crown is com-
posed of ten ridges, and taking into account that the most
anterior portion, supported upon the large front fang, had
disappeared by age, and that it probably was composed of at
least two ridges, this would yield for the ridge-formula of
the molar a total of twelve collines, exclusive of talons.

What was the age of this molar in the dental series of the
animal? At the first glance it might be supposed from its
size to be a third or last milk molar, but this inference is at
once negatived by the fact already remarked on, that the
posterior talon bears nowhere upon it, or upon the end of the
tooth, the slightest indication of a depression arising from
the pressure of a tooth advancing behind it. As the same
result is yielded in a still more decided fashion by the inferior
molars, to be noticed in the sequel, I see no alternative to
the inference that it was an adult tooth of a dwarf species of
Hlephant. The following are the dimensions :—

Extreme length of crown measured from back talon to anterior edge, exactly
4:in. Width of ditto at 2nd ridge, 1-4 in. Ditto at 8rd ridge, 1:5 in. Ditto at
6th ridge, 1-4 in. Greatest width of crown, 1°55 in. Width at 10th ridge (greatest),
1:2in. Ditto at last ridge, 1-in. Greatest height of crown taken at reflection of
10th ridge, 2-95 in. Length occupied by the five discs from 2nd to 6th inclusive,
1-8 in. Width at middle of 3rd dise taken between the enamel-edges, 023 in.

With reference to the alimentary characters, the dises of
ivory and the cement hollows between the enamel-ridges
are but slightly excavated; in fact, the most anterior portion
of the crown exhibits the flat and nearly uniformly smooth
surface which is commonly presented by Elephants reared
in the domestic state and fed upon potatoes or other soft
food. The inference drawn from this is, that the food of the
Malta species was more herbaceous than woody. Mr. Din-
kel’s figure of the crown-surface of this tooth is not quite
satisfactory in the details of the enamel-edges, but on the
whole is a good representation of the tooth.

Upper True Molars.—The specimen next to be noticed
(Plate XI. figs. 2 and 2 a) is a very beautiful and finely-pre-
served molar of the upper jaw, right side, complete in every
respect, with the exception of the ends of the fangs, which
are more or less broken off. The crown is composed of nine
principal ridges, together with a front and a back talon.
The front fang is distinctly present, and is seen to support
the two front ridges and talon. Of these, the eight anterior
ridges are more or less affected by wear, the rest being intact.

fat = ar aes
= SD Gee ee

Vt]

DESCRIPTION OF PLATE XI.
Eiepuas (Loxopon) MELITENSIS.

The figures in this Plate are of the natural size, and have
been copied from drawings executed by Mr. Dinkel for Dr.
Falconer, from the original specimens forwarded from Malta
by Capt. Spratt.

Figs. 1 and 1 a. Views in plan and profile of iast upper true molar, left

side. Described at page 292.

Figs. 2.and 2a. Views in plan and profile of penultimate upper true
molar, right side. Described at page 294.

Figs. 8, 3a, and 36. Three different views of a very perfect specimen
of the milk tusk. Described at page 296.

Figs. 4 and 4 a. Views in plan and profile of a germ specimen of the
penultimate upper milk molar, or m. m. 3. (See page 297.)

VOL. Il.

years * = ‘ aay ;

eho pare = = Sages ee ae a ae at
STS WO4TOTAL seydeyiy “URL TEP aC

ELEPHAS MELITENSIS. 295

The front talon appears to be composed (what remains of
it) of a couple of digitations, the greater part of it having
already been ground away, or become confluent with the
front ridge. The three anterior discs are transverse, the next
five are only slightly affected by wear, showing the tips of
the digitations abraded into annular detached discs, or in
three divisions. The anterior disc is expanded in the middle,
and narrows at either side, and presents only two or three
flexures in the enamel-plate, without crimping. The second
and third are nearly of a similar form, with uncrimped
enamel, and narrowing at the sides. The fourth, fifth, and
sixth are each in three divisions, and the seventh and eighth
only show the tips worn across. The enamel-plates are
decidedly thick for the size of the tooth, and the ridges are
very high relatively to the length. The layer of cement at
the anterior end has been removed, and with it all appear-
ance of a disc of pressure. The hind talon forms a gibbous
projection beyond the vertical plane of the posterior fang.
There is no distinct disc of pressure upon the crown portion
of the talon, but there is an obscure depression at the basal
part near the fang, which may be of this nature. The fol-
lowing are the dimensions :-—

Extreme length of crown, 2°9 in. Width in front, 1:35 in. Ditto in the middle,

13in, Ditto behind, 1:1 in, Length of surface occupied by the eight anterior dises
of wear, 2°2in. Extreme height of crown at unworn portion, 9th ridge, 2°8 in.

From the above dimensions, the contraction of the crown
backwards and the considerable height of the ridges rela-
tively to the length of the crown are well shown.

Had this specimen been discovered isolated, little or no
hesitation would have been entertained by the Paleontologist
in referring it to the age of a milk molar of some species of
Hlephant. But when regarded as part of a series in con-
nection with the undoubted milk molars (fig. 4 of Plate XI.,
and figs. 1, 2, and 3 of Plate XII.), the whole of which are
of such unusually small proportions ; and when further com-
pared with the adult molars of the lower jaw, of which one is
figured in Plate XII. figs. 4 and 4 a, and the upper molar
(figs. 1 and 1 a of Plate XI.), it is manifest that it maintains
its place consistently as a true molar of the same series.
Iam at present unable to decide with confidence whether it
had best be regarded as a penultimate.!

Of the antepenultimate upper true molar (m. 1) no perfect
specimen is to be found in the collection. One fragment,
inferred to be a part of this tooth from its size, form, and
proportions, comprises the two anterior ridges, together

' At page 298 it is spoken of as the penultimate.—[Ep.]

296 OSSIFEROUS CAVES OF MALTA.

with the large fang that supports them. The correspond-
ing molar of the lower jaw is equally wanting as an entire
specimen; but there are fragments referable to it also.

Milk-Tusks.—The Zebbug series fortunately includes a very
perfect milk-incisor, which confirms the line of specific
affinities indicated by the molars. The specimen is repre-
sented of the natural size by Plate XI. figs. 3, 3a, and 3 b.
It differs from the permanent tusk in having the crown and
fang portions distinctly defined. The crown forms an obtuse,
flattened, rounded, and irregularly indented body, invested
with a thick shell of enamel, and supported on a long cylindro-
conical fang, part of which is broken off, near the end. From
the diameter of the broken end and of the central canal, it
is manifest that at least a third of the entire length of the
fang is wanting. The specimen was compared with the cor-
responding tooth of a foetal African Elephant belonging to
a skull transmitted to the British Museum by Dr. Living-
stone, in which the milk molars are quite unworn. The
two agree very closely in the dilated blunt form of the crown,
the investing shell of enamel and the defined fang. The
chief difference detected between them was in the form of
the fang, which in the young African Elephant forms a
rather short compressed cone, terminating in a sharp slender
point; while in the Zebbug fossil the fang is stouter, more
cylindrical, and much more elongated.

The dimensions of the fragment are :—

Joint length of crown and fang, 14 in. Length of crown, 0°6 in. Width of
ditto, 0°4 in. Thickness of ditto, 0°3 im. Length of fang portion (imperfect), 0-9 in.
Diameter at the collar, 0°3 in. Ditto at broken extremity, 0°25 in.

These minutiz are given in so much detail, in order to
mark the affinity which the Malta fossil bears throughout its
dentition to the African Elephant. A shell of enamel has
not yet, so far as I am aware, been detected in the milk
incisors of any species of the subgenus Huelephas. It occurs
in those of the African species, and I have detected it, forming
a sheath upon the points of the young permanent tusks of
E. insignis, belonging to the group Stegodon.

Permanent Incisors.— The collection contains numerous
fragments of Elephant tusks, for the most part amorphous
pieces or splinters of the outer layers, many of them bearing
distinct marks of having been gnawed, but indicating tusks
of very considerable diameter, and out of all proportion to
the small Zebbug molars. These fragments, which appear
to indicate another and larger species of fossil Elephant, will
be noticed in the sequel. There is only one determinable
specimen which will admit of being referred to the smaller

ELEPHAS MELITENSIS. 297

form, and that only conjecturally. It consists of a portion
of the dental end of a slightly curved tusk, about five inches
in length. The greater part of the outer layer, which is
weathered, of a greyish tint, has disappeared by dislamination.
The butt end yields a round section, slightly compressed at
the sides. The outer layer is smooth, and throughout a
line of thickness shows no appearance of engine-turning.
Beneath it the ivory-surface is very distinctly channelled
longitudinally and regularly, and the section inwards to the
core exhibits very distinct engine-turning, more pronounced
even than is commonly seen in Proboscidean tusks, the in-
equalities being nearly as marked as upon a sailor’s thimble.
The specimen tapers to a conical point.

The dimensions are :—

Length of fragment, 5-0 in. Vertical diameter at butt end, 1:15in. Transverse
ditto, 1:0 in.

The tusk here described would correspond in size with
the true molar of the Malta form.

Ridge-formula.—It now remains to consider how the data
furnished above by the molars bear upon the determination
of the ridge-formula, which of all the characters is the most
significant in pointing out the affinities of the species.

Milk Molars.—The antepenultimate milk molar (m.m. 2)
_ is seen by Pl. XII. figs. 1 and 1 a, to have been composed of
three collines, like the corresponding tooth of the African
Elephant; while in EH. primigenius, EH. Indicus, and other
species of the sub-genus Huelephas, it presents four collines.

The penultimate (m.m. 3) is clearly proved by the upper
germ specimen (Pl. XI. figs. 4 and 4a), and by the lower
(Pl. XII. figs. 2 and 2a), to have had five collines, besides
front and back talons. In the African Elephant it is com-
posed commonly of five ridges in the upper jaw, and six in
the lower; while in the species of Huelephas it ranges from
seven to eight ; seven being the complement of FH. antiquus,
and eight that of the Indian Hlephant and Mammoth.

Of the last milk molar (m.m. 4) the specimen represented
in Pl. XII. figs. 3 and 3 a, fortunately presents the crown
_ of an inferior ‘tooth, in perfect integrity, composed of eight
ridges in addition to a front and hind talon. The African
Elephant commonly yields the same number; while in
FE. antiquus the number is ten, and in EH. primigenius and
H. Indicus it amounts to twelve.'

1 Extract from Dr. Falconer’s Note- | 18,810 and 21,655) of the last milk
book.—‘ British Museum, July 26, 1860. aan ar of Elephas antiquus from Grays,
Compared the perfect specimen of last |in Essex. In No. 18,810, the crown is

milk molar of lower jaw of Elephant | more worn than in the 7, ebbug tooth, and
from Zebbug with two specimens (Nos. | there is a very large dise of pressure

298 OSSIFEROUS CAVES OF MALTA.

True Molars.—Of the antepenultimate true molar (m. 1),
there is no perfect specimen in the collection. But as in all
the species of Elephant and Mastodon this tooth invariably
repeats the composition of the last milk molar, we have no
difficulty in fixing the normal number of its ridges to have
been eight, besides talons. In EH. antiquus the number is ten,
and in the Mammoth and Indian Elephant twelve.

Of the penultimate true molar there is no entire specimen
of a lower tooth ; but we have the upper beautifully preserved
in Pl. XI. figs. 2 and 2a, and exhibiting a crown distinctly
composed of nine ridges besides a front and a hind talon.
In the African Elephant the same tooth is commonly made
up of nine ridges. In EH. antiquus the normal number is
twelve, while in H. primigenius and E. Indicus it amounts to
sixteen.

Of the last true molar (m. 3) there are fortunately speci-
mens both of the upper (Pl. XI. figs. 1 and 1 a) and lower
(Pl. XII. figs. 4 and 4a) jaws; and although the portion
supported on the anterior fang is wanting in both, as that
constantly corresponds in all the species of Elephant with
what is borne upon the same fang of the penultimate, we
have little difficulty in restoring the missing part of the
teeth. The upper molar (Pl. XI. figs. 1 and 1 a) exhibits
the remains of ten ridges, and adding two for the part cor-
responding with the anterior fang we get a complement of
twelve ridges for the crown of the last molar. In the
African Elephant the same tooth in the upper jaw ranges
about ten ridges, and in the lower from ten to twelve. In
E. antiquus the number is sixteen, and in EH. primigenius
and E. Indicus they reach twenty-four.

From the above data the ridge-formula of the molar-series
is deduced to have been thus:

Milk molars. True molars.
Bae aN
38+5+8 8+9+12.
3+5+8 84+9412.

This formula at once brings the small Zebbug species
within the subgeneric group of the Elephants, which I have
called Loxodon, along with EH. Africanus; the affinity of the
fossil to the existing species is further clearly indicated by

behind; the enamel also is much more Zag 8 E. antiquus.
plaited and the crown much broader ; it ad oS a ns
has seven ridges and talons, the anterior Length of crown . . 2-2 355 29

talon being worn low, and the bounding | Width at second dise °7 85 6-9

shell of enamel not quite complete in | Gyeatestwidth behind °7 1:4 1°3
front. No. 21,655 has also seven ridges | Height of crown at

‘and talons, all of which are more or less seventh ridge . . 7 14 = —
worn; it is covered with an enormous [Ep.]

mass of cement.

DESCRIPTION OF PLATE XII.
Euepxas (Loxopon) MELITvTENSIS.

The figures in this Plate are of the natural size, and have
been copied from drawings executed by Mr. Dinkel for Dr.
Falconer, from the original specimens forwarded from Malta
by Capt. Spratt.

Figs. 1 and 1a. Views in plan and profile of antepenultimate milk
molar, m.m. 2. (See page 297.)
Figs. 2 and 2 a. Views in plan and profile of penultimate lower milk

molar, orm. m. 38. (See page 297.)

Figs. 3 and 3a. Views in plan and profile of the last lower milk molar,
or m. m. 4, showing the crown perfect. (See page 297.)

Figs. 4 and 4 a. Views in plan and profile of last lower true molar, or
t.m. 3. (See page 298.)

VOL. IL.

dr ys0y, ‘sisueqreyy seydapy TAT BP PMI LC

CL MTT TTA

ELEPHAS MELITENSIS. 299

the narrow crown and by the tendency to mesial expansion
of the discs of wear, together with the point already alluded
to of the milk-incisors being invested with a layer of
enamel. But although allied, the two forms are speci-
fically very distinct. Besides the signal difference of size,
the form of the discs of wear, although belonging to the
common pattern, presents broad marks of distinction. In
the African species the discs are angularly dilated into
rhombs in the middle, and the angles terminate in a round
loop, caused by a single outlying digital element, which in
the progress of abrasion becomes confluent with the disc of
the ridge to which it is appended. In the Zebbug form
there is never a trace of this outlying loop, and the discs,
although open, exhibit only but a very slight tendency to
angular expansion. The character is most pronounced in the
two milk-teeth (Pl. XII. figs. 2 and 3), while it is entirely
wanting in the penultimate upper molar (Pl. XI. fig. 2). In
fact, this tooth differs more from the ordinary type of the
African species than does the corresponding molar of H. an-
tiquus. ‘The amount of agreement and of difference in the
molars of the two species is best appreciated by comparing
the last lower molar (Pl. XII. fig. 4) of the Zebbug form, with
the corresponding molar of the African species.!

Elephas Melitensis is the name which I have applied to the
new species. It was the pigmy form of the order. In size,
it stood between a large Tapir and the small unicorned
Rhinoceros of Java.

APPENDIX.

CONSISTING OF MEMORANDA, ETC., BY DR. FALCONER, ON FOSSILS FROM
THE CAVES OF MALTA.

IJ—Exrracts or Letrers rroM Dr. Fatconer To Capt. Spratt, C.B.?

July 21, 1860.

‘Your last researches in the Zebbug Cave are to me of very great
interest. J am now occupied with them. The Elephant remains are of

™ See Plate vi. fig. 1.—[Ep.] of July 1860. The collection included

2 In May, 1860, Capt. Spratt de- | the tooth of Hephas Melitensis, already
spatched from Malta to the Geological | described as belonging to the public
Society of London the fossils which he | library of Malta, which Dr. Falconer
had discovered in the caves of Zebbug, | returned to Malta in May 1862. Since
&e., and their examination was under- | Dr. Falconer’s death a statement has
taken by Dr. Falconer at the beginning | been published in a Malta paper to the

300 OSSIFEROUS CAVES OF MALTA.
the highest importance. They belong, not to the Mammoth, but to
quite a distinct species.’

August 2, 1860.

‘The cave fossils (from Malta) are turning out to be of great interest.

‘From Krendi, the larger teeth-specimens are almost entirely of
Hippopotamus Pentlandi, the Sicilian species. The teeth and small
bones in the breccia are almost entirely of a new species of Myoxus
or Dormouse, Myowus Melitensis, of very large size, as big im com-
parison to the living Dormouse as the Bandicoot Rat to the Mouse.!

‘From Zebbug there is a small species of Proboscidean, I believe
quite new and allied to the Elephant, but not much bigger than a large
Tapir. There are two fossil birds among the Zebbug bones. The best
marked is an extinct species of Swan, which I propose calling Cygnus
Melitensis. It was very much larger than any of the known existing
species.2_ There are also some Reptilia not yet made out. But strange
to say I have as yet detected no big Carnivora, although from the con-
tracted tunnel form of the cave it is clear that the bones must have

: ae 5
been dragged in by large Carnivora.

effect that he was not the first to
identify the Elephas Mclitensis, and
to place it in the subgenus Loxodon,
on the ground that in a memoir
communicated to the Royal Dublin
Society in November 1861, but not
published till 1863, an opinion had been
incidentally expressed that a tooth in
the Museum of the Maltese University
presented markings on the crown dif-
fering from those of the Mammoth, but
approaching those of the African Ele-
phant. It may be well, therefore, to
state that careful tracings of this tooth
in the University Museum of Malta were
transmitted by Capt. Spratt to Dr.
Faleéner, who pronounced it as entirely
different from those of the pigmy Ele-
phant, but ‘allied to, if not identical
with, the African Elephant.’ —[ Ep. ]

1 Figures of the ‘ Myorus Miletensis
(Nobis),’ sic, accompanied a paper, en-
titled, ‘ Observations on the Fossiliferous
Caves of Malta.” By Andrew Leith
Adams, M.B., which was communicated
to the Royal Dublin Society on Novem-
ber 18, 1861, and published in the
Journal of the Society for 1863.—[ Ep. ]

2 The fossil remains of this species
of Cygnus were transferred by Dr.
Falconer to Mr. W. K. Parker, F.R.S.,
who communicated an account of them
to the Zoological Society on Dee, 12,
1865, from which the following passage
is extracted :—

‘The first species was a gigantic
Swan, nearly one-third larger than aver-

age individuals of the Mute Swan
(Cygnus olor). The head and more than
half of the long bones belong to this
kind. Its wings appear to have had
the same relative length as those of the
tame Swan; but the thigh-bone was
relatively shorter, the tarso-metatarso
(shank) was considerably longer in pro-
portion. The most remarkable dif-
ference, however, between this extinct
species (which I propose to call @.
Falconer) is to be found in the compara-
tive length of the phalanges ; for whilst
the proximal joint of the middle toe is
one-fourth thicker than that of the
Mute Swan, it is only three-fourths the
length ; so that whilst C. Fa/coneri was
between one-third and one-fourth larger
than the common kind, it stood on longer
legs, and had the comparatively short
toes of a goose.

‘The next species, that to which the
important sternal fragment belonged,
showing it to be a species of Hooper,
I have doubtfully named C. musicus (?).
Having recently examined the osteo-
logical specimens in the Museum of the
College of Surgeons, which are attributed
to C. ferus (see Cat. Mus. Coll. Surg.
vol. i. p. 233, Nos. 1241-1248), I am
strongly inclined to think, from the
extreme difference of size, that some of
the smaller ones belong to C. Bewickii.
The smaller bones from Malta may
either belong to small individuals (per-
haps females) of C. mwsicus, or perhaps
to the smaller C. Bewichit. —[Ep.]

ELEPHAS MELITENSIS. | 301

September 19, 1860.

‘The Hippopotamus of Krendi is the same as the Hippopotamus of
Sicily, viz. Hippopotamus Pentlandi, a species a little smaller than the
living African form. ‘The age of the animal in Sicily is certainly
newer than the older Pliocene, because it overlies the Pliocene con-
glomerates. The Sicilian fossil Hippopotamus was not, at least on the
northern side of Sicily, in connection with a great freshwater lake. It
occurs mixed up with marine Bryozoa; and upwards of two hundred
fossil Pliocene shells, all marine, occur in the deposit beneath it.
Graves got the same Hippopotamus from the Tertiaries of Candia. I have
examined the teeth in the College of Surgeons. Professor H. Raulin
of Bordeaux also met with them in Crete, in a “yellow sandy loam
of the Pliocene or newer Pliocene period, in the plain of Katharo, at
an elevation of 1,150 metres above the sea (say 3,750 feet). Katharo
is to the east of the large plain of Passith, and east of the town of
Kritsa, on the eastern side of the Gulf of Mirabello.”” Raulin con-
siders “that the plain of Katharo was formerly an intra-montane lake.”
Lord Ducie, however, got a tusk of a Hip. Pentlandi which he con-
sidered to be from a deposit of a Miocene age. The swimming hypo-
thesis for the Hyena will not do. Elephants will not swim across deep
sea channels, nor Hyznas either. The case is clear—that there must
have been continuity of land between Sicily and Malta, and Sicily and
Cape Bono, by Adventure Bank and the Skerki shoal.’

April 11, 1862.

‘ Although the Krendi Mammalia (the Hippopotamus) are the same
species as those of Sicily and Crete, and of a newer Pliocene age, I
cannot reconcile the Zebbug fossils as being of the same age, and I
have a suspicion that they may be upper Miocene. ‘The fossil Hle-
phant of Zebbug is an undescribed species of prodigious interest,
from its small size, &c. But there are no bones as yet of the skeleton,
only the teeth, and there are indications of another larger species. It
is of great importance that we should get a sight of all the materials that
are available. There are numerous bones in your Zebbug Cave collection
that are fiercely gnawed, and evidently by a large predaceous carnivore.
But with the exception of a single tooth, of which the crown is com-
pletely hammered and denuded of enamel, and therefore indeterminable,
there is not a single remain indicative of what this predaceous form
was, whether Hyzxna or something else ; and until this point is cleared
up, the history of the Malta caves will be incomplete.’

Il.—Extracts rrom Dr. Fatconer’s Notr-Booxs.

1. List of fossil Elephants Teeth, from Zebbug, in Captain Spratt’s
Collection.

‘No. la. Milk tusk, with an enamel-shell. 1. Antepenultimate milk
molar, probably upper, with a disc of pressure. 2. Germ of penulti-
mate milk molar; has three plates unworn; probably of lower; no
dise of pressure. 3. Penultimate lower, well worn, three discs and
fang; also front disc of pressure. 4. Paralaate lower (right ?) with
dise of pressure. 5. Penultimate upper left milk molar in germ; no
dise of pressure. 6. Last milk molar, lower ; very fine, eee he

302 OSSIFEROUS CAVES OF MALTA.

Last milk molar, upper fragment, showing four ridges. 8. Last milk
molar, lower? with section of three plates, fine. 9. Third milk (last)
molar, upper; six ridges in fragments. 9 a. Much worn fragment of
milk molar, three plates remaining. 9 6, Fragment of one semi-plate
of ditto. 10. Symphysial portion of right ramus of lower jaw, small,
and of young animal, no teeth; has three mentary foramina in vertical
line. 10a. Talon and last plate of milk molar. 11. Penultimate
upper right true molar; found in NW. branch of cave. 12. Last
lower molar, found in NW. branch of cave, in colour and mineral con-
dition like No.11. 18. Fragment of posterior half of last lower molar.
14. Last lower molar; fine specimen from Branch ‘A.’ 15. Tusk,
fragment from near end; found with No.14. 16. Five ridges of germ
of a lower true molar; one intermediate plate wanting to make pieces
fit correctly. 17. Anterior fang portion with two ridges of true molar,
upper jaw, probably antepenultimate. 18. Middle portion of another
true molar, probably antepenultimate. 19. Middle portion, comprising
two ridges of germ of lower true molar. 20. Posterior portion of
upper molar in germ state, comprising two ridges and talon, and
showing marks of gnawing. 21. Another portion of posterior molar,
more weathered. 22. Mutilated fragments of a larger sized molar than
any of the others.’

2. Vertebre of Elephas Melitensis.

‘ College of Surgeons, July 27, 1860.—Examined the small vertebra
of the Elephant from Zebbug, and compared it with the fourth, sixth,
and ninth vertebre of a young Indian Elephant. None of these have
the spinous processes so much reclined, or continued in the same plane
as the oblique surfaces. In the somewhat cordate shape of the spinal
opening, and in the proportion of the posterior costal depression con-
tributed by the transverse processes and centrum, it agrees best with
the ninth; but in the fossil the bridge between the anterior and pos-
terior depressions is not so broad. The transverse processes are not
quite so transverse as in the fourth and sixth, but are much less in-
clined upwards than in the ninth. The anterior articulating processes
are near together, and the intervening circular sinus is less marked
than in the fourth and sixth; in this respect the fossil closely resembles
the ninth. The sinus between the posterior costal cup and the pos-
terior oblique processes, seen sideways, is deep and marked exactly as
in the ninth; in the fourth and sixth it is open and shallow. The
posterior oblique articulating surfaces are approximated as in the
ninth, and not wide apart as in the sixth, or still more in the fourth;
but there is a raised ridge between them, and not a hollow, as in the
fourth, sixth, and ninth. The general form of the body of the fossil
is somewhat cordate as in the ninth, and the keel below the body is
very much as in the ninth. If the specimens in the College Museum
are properly numbered, the fossil is a vertebra intermediate between
the sixth and ninth, and is most probably the eighth; the spinous
processes at its base being much more like the ninth than the sixth.
The dorsal vertebra of the Zebbug Elephant closely resembles in form
that figured in Eichwald’s paper “De Pec. et Pachyd. reliq. fossil,”
(Nova Act. Acad. Nat. Curios., vol. xvii. part 2, p. 160. Tab. LIV.
figs. 3 and 4.), and referred by him to £. proboletes. The latter,

ELEPHAS MELITENSIS. 303

however, is about twice as large as the Malta fossil, but in the former
the vertebral foramen is much larger in proportion.’ (See Pl. XIII.
figs. 1 to 4.)

[ Among Dr. Falconer’s drawings there is a representation of another
vertebra of Hlephas Melitensis, which was identified as the seventh
cervical, and which has been already referred to. See antea, p. 251,
note.—Ep. |

3. Pelvis of Hlephas Melitensis.

‘Among the fossil remains from Zebbug there is also the right side of
a pelvis which agrees in the closest manner possible, so far as the
acetabulum goes, with that of the Indian Elephant in the College of
Surgeons Museum.’ (See Pl. XIII. figs 5 and 6.)

4. Humerus of Elephas Melitensis.

‘ College of Surgeons, July 21, 1860.—The humerus from Zebbug
is the left, and agrees most closely with that of Hlephas Indicus, No.
2654, which has not the epiphyses synostosed. All the leading points
of general form are exactly the same. It also agrees most closely with
the detached humerus, No. 2741 B., the chief difference being the two
lengthened depressions near the outer angle at the top, which are much
more pronounced in the fossil. Unluckily the head is gone in the fossil,
or the evidence would be complete. Although belonging to a nearly
adult individual, its entire length is not more than ten inches.’ (See
Plate XIV. figs. 1 to 5.)

5. Femur of Elephas Melitensis.

‘ College of Surgeons, August 14, 1860.—Examined the repaired
shaft of the Zebbug Proboscidean femur, and compared it with that of
Choonee’s skeleton.’ The general resemblance is excessively close.
There is the same bridge between the epiphysis of the head and of the
trochanter ; there is also exactly the same deep fossa (like a foramen)
below the epiphysis of the trochanter, in both; and the same general
form of the lower end. But the following differences are seen :—

‘1. The hollow descending, on the posterior side of the fossil below
the trochanter, is much more pronounced.

*2. The hollow in front (running in same direction) which isso great
in fossil is entirely wanting in the existing Elephant.

‘3. The edges of the base of the trochanter are more angular and
pronounced in fossil.

‘4. The outer edge, below middle of the shaft, is quite sharp in
existing species, but much rounder in the fossil.

‘5. On the inner side behind, leading to the condyle, there is a
longitudinal fossa in the fossil, which is wholly wanting in the existing
Elephant; but the outer posterior sides are alike.

‘6. The foramen of medullary artery in the fossil is exactly as in the
femur of Mammoth in the case in the British Museum; in Choonee’s
skeleton it is situated on the inner side, in a line contiguous with pro-
jecting part of head.’ (See Pl. XIV. figs. 6 to 10.)

© British Museum, August 9, 1860.—Examined a transverse section
from the middle of the shaft of the broken Zebbug left femur. The
centre is a compressed flattened hollow tube, the inner surface of which

1 See antea, p. 268.—[Ep.]

304 OSSIFEROUS CAVES OF MALTA.

is not smooth, but rugged and cancellated. Outline of section is ovoid.
Greater diameter of shaft, 1:2 in.; lesser ditto, at middle, °85 in.
Greater diameter of hollow tube, -53 in.; lesser ditto, ‘33 in.’

6. Fossil Halitherium, from Malta.

‘ British Museum, December 20, 1862.—Compared the fossil from
Malta, having three teeth, with Listriodon and Halitherium. In Lis-
triodon the ridges very much as in Tapir; outside in the upper jaw
they are nearly vertical, and the outer point high, with clean transverse
ridges. Further, in Listriodon, the ridges of the upper teeth wear ob-
liquely in front, the apex being intact, as in Dinotherium ; while in the
lower jaw the wear is reversed. In the Malta fossil the wear is hori-
zontal.

‘ The last tooth agrees very closely with figs. 16, 19 and 20 of Plate
1 of Kaup’s “ Beitriige,” on Halitherum. As compared with the skull,
the molars of the latter are much worn and very Hippopotamine-like in
the discs of wear. The fossil evidently is not of Listriodon, but of the
Dugong family.

‘ The cancellar tissue of the bone is very open in the fossil, as in the
Cetacea.’

7. Notes upon the Zebbug Specimens. March 1862.

‘Regarded as a whole, they differ in character from the Crendi speci-
mens, and appear to belong to an entirely different set of animals. The
most interesting of these are a series of Elephant molars, referred to in
Captain Spratt’s paper, as being of Mammoth. They consist of molar
teeth of all ages, from the first milk molar to the last, and many of them
are of remarkably small size. Three of them appear to be the last true
molar, lower jaw; of these, one, No. 14, is nearly entire, showing nine
plates much worn, and the posterior talon. No. 18 (see list at page
302), also a back molar, has the grinding plane of the crown of the
tooth scooped and obliquely distorted from being worn all on one side.
Two of the larger upper molars are nearly entire; one of them, No. 11,
is quite entire. There is also one very mutilated fragment of a molar
indicating a specimen of a larger size. Among these Elephantine speci-
mens there is a fragment consisting of the anterior portion of the right
ramus of the lower jaw. There is also a fragment of a tusk about 44 in.
long, and nearly 1 in. in diameter, comprising the terminal portion near
the point. Besides the entire molars there are eight or ten fragments of
molars consisting chiefly of plates which had not been cemented. One
weathered specimen of ivory indicates an individual of very consider-
able size. The next most interesting amongst the specimens are the
humerus and femur of a Proboscidean form of avery peculiar character
(Elephas Melitensis). To this form also appear to belong a pelvic
portion containing the acetabulum, a dorsal, and a lumbar vertebra,
besides the body of a vertebra, denuded of the processes. Belonging to
EE. Melitensis there is also the shaft of a very small humerus, without
epiphyses, evidently of a young animal.

‘Of Elephantine remains there is also a fragment of the anterior
portion of the lower jaw, comprising simply the symphysial por-

1 This fossil was not obtained from | bably obtained from the Miocene de-

the caves. It was purchased for Capt. | posits of Malta.—[Eb.]
Spratt from a collector, and was pro-

i

DESCRIPTION OF PLATE XIII.
ExepHas (Loxopon) MELITENSsIs.

The figures in this Plate are copied from drawings exe-
cuted by Mr. Dinkel for Dr. Falconer, from the original
specimens forwarded from Malta by Captain Spratt. Dr.
Falconer has inscribed on the drawings ‘ Zebbug Cave, Malta.
Captain Spratt, R.N.’

Figs. 1, 2, 8, and 4. Four different views of a dorsal vertebra, one-half
of the natural size, apparently that described at page 302.

>this

Figs. 5 and 6. Two different views of a fragment of the pelvis, showing
the acetabulum, probably that referred to at page 303. One-

third of the natural size.

a

VOL. It.

“CL FET TL TA

DESCRIPTION OF PLATE XIV.
EvepHas (Loxopon) ME.irensis.

The figures in this Plate are one-third of the natural size,
and have been copied from drawings executed by Mr. Dinkel
for Dr. Falconer, from the original specimens forwarded
from Malta by Captain Spratt.

Figs. 1, 2, 3, 4, and 5. Represent the anterior, outer, and posterior aspects

of the shaft, the lower articular surface, and a transverse sec-
tion of the left humerus, apparently that described at page 303.

Figs. 6, 7, 8, 9, and 10. Represent the posterior, inner, and anterior
surfaces and section of the shaft of the left femur, apparently

that described at page 303.

Vou. I.

jet <i eee ea ee

rie sist tree sceee

Su “STSTIOFT[O]L

“PLOT TTA

ELEPHAS MELITENSIS. 305

tion, the ridges bounding the gutter of the spout and the two mentary
holes, but much too mutilated to indicate the age or give any determin-
able character.

‘ There is also what appears to be a residuary portion of the smooth
ball of the humerus of an Elephantine form, but very much mutilated
and weathered ; also the iliac portion of a pelvis and three fragments of
the cranium of an Elephant, so inferred from the presence of large
diploe-cells. Among the other fragments are a metacarpal or meta-
tarsal bone of a pachyderm, the size of a large Tapir, also a scapula,
with the articular cup and spinous process nearly perfect ; also a pha-
langeal bone apparently of a Proboscidean. Fragments of spinous pro-
cesses of vertebra, both dorsal and lumbar, some of them indicating
animals of a very considerable size, as large at least as Bos primigenius.
Besides the mammalia, there are various fragments of the carapace
and plastron of two Chelonian forms, one of them of small size, and
indicated by small fragments ; thére are some leg bones without articu-
lar surfaces, apparently belonging to the same forms. Further amongst
the collection there is a series of specimens which bear very distinct
marks of having been gnawed by a large carnivorous animal. The
tooth marks upon some of these are very boldly pronounced. From
among these, nine principal specimens have been selected; one a cylin-
drical bone, is the most pronounced ; another is the gnawed and de-
tached globular head of a femur, about 1? in. in diameter, and another
apparently the head of a humerus. Many of the darker specimens ex-
hibit distinct marks of minute superficial burrowings or excavations,
which Mr. Rupert Jones tells me have been detected even in grave-bones
by Professor Henslow. It is not a little remarkable that amongst
these numerous indications of gnawing by a large carnivore, there is not
a single indication of what that form was. The core, with some enamel-
inequalities of one tooth was found, but with no part of the crown
remaining, and therefore wholly undeterminable. It would agree, in
size, with the carnassier tooth of a large Hyena, but it is too mutilated
to guess even at the order to which the form belonged.

‘A number of sharks’ teeth were found in the clay among the fossils
from the Zebbug Cave; these are the only objects that might possibly
have been used by man, no fragments of flint having been found in the
eavern. These teeth probably fell out from the yellow sandstone
deposit in which the cave occurs, and in which similar teeth are fre-
quently found.’

8. Notes of Captain Sprati’s Collection of Fossil Bones from Crendi
Cavern, Malta. (Mahlek Cave of Adams.)

* Geological Society, March 18, 1862.—The principal mass of the
collection consists of a very hard, compact, calcareous matrix, and
fragments containing solitary molars, fragments of molars of tusks and
incisors of Hippopotamus Pentland. They appear all to belong to a
single species Hipp. Pentlandi. About a dozen of the specimens, com-
prising fore molars and back molars, are tolerably perfect; they were
compared with some of Baron Anca’s specimens from Olivella, and
others found by myself in the Maccagnone Cave, with which they agree
entirely. Some of the molars agree with the larger Sicilian form of
the same cave and of the caves near Palermo. These have been put up

VOL. II. x

306 OSSIFEROUS CAVES OF MALTA.

apart for reference. There are also some fragments of tusks and inci-
sors, but none of them sufficiently perfect to figure.

‘ Besides these there is a series of amorphous fragments of bone,
variously splintered, and, therefore, indeterminable, and three deter-
minable specimens, No. 14 being the half of the head of a humerus,
divided vertically through the middle, the fractured surface clean, flat,
and smooth, as if effected through a fault-displacement. Another is
the articular end of a radius, and the third the basal part of the spinous
process of a vertebra.

‘ Next are three slabs of very compact dark grey-coloured stalagmitic
matrix. No. 1 contains in strata a great many small fragments of the
teeth and bones of a minute mammal ; the molar teeth belong to a large
extinct species of Myoxus (Myoxus Melitensis, Falc.).

‘ There is also one very large rodent incisor with various fragments
of ribs, the end of an ulna, &c. One very beautiful molar of Myoxus
shows the crown distinctly.

‘These remains are confined to the upper layers of the matrix,
through which they are very profusely dispersed ; the compact portion
of the stalagmite, about 1} in. thick, is devoid of fossils, or nearly so,
and the weathered surface would indicate that this is the upper surface
of the deposit.

‘ Slab No. 2 is of the same character, but the fragments of bones
contained in it are less perfect. It contains also one fine molar of
Myoxus presenting the crown-surface.

‘ Slab No. 3 is remarkable in showing the greater part of the shaft of
what appears to be a femur, but undeterminable from being crushed at
the articular ends.

‘ Besides these there is a quantity of detached and weathered minute ©

fragments, inferred to belong chiefly to Myowus; of determinable frag-
ments amongst these are a molar, part of a jaw, a metacarpal, or meta-
tarsal, and a phalangeal bone.

‘ One specimen consists of a small canine or outer incisor of a car-
nivorous animal, of the size of the Fox. The above includes all the
specimens deserving notice.

‘ The carnivorous tooth here referred to is the only indication of a
feline form being contained amongst the whole of the Crendi specimens.

‘So far as the comparison has been carried, the Crendi specimens
differ entirely from those found in the Zebbug Cave.’

9. Additional Note on the Fossils from Zebbug and Mahlek Caves.

‘ Geological Society, July 10, 1862.—Memorandum of supplemen-
tary despatch received from Captain Spratt yesterday.—A box con-
taining a mixed collection of fossil bones from the newly-discovered
cave, called by Captain Spratt the Hippopotami Cave, together with a
small box containing numerous fragments, mostly small, and supple-
mentary to his previous despatches. Among these are numerous frag-
ments of bird-bones; 2nd, fragments of carapace, &c., of Chelonians ;
3rd, a large quantity of fragments of ribs and other bones of mammalia,
mostly undeterminable ; 4th, a tin box containing fragments, more or less
perfect, of small molars of Elephants; and lastly, a pill-box containing
an undoubted tooth of a carnivore, but of small size, being the first
carnivorous indication in the shape of a tooth from the Zebbug Cave.

ELEPHAS MELITENSIS. 307

‘The great mass of the specimens contained in the box consist of
fragments of the extremities of a small species of Hippopotamus, the
most important of which are lower articular ends of femurs, lower ends
of humeri, and some of tibie. The bones are for the most part in a
very fragmentary condition, and much weathered or rolled. The
articular surfaces, as a general rule, are abraded or more or less muti-
lated ; one fine specimen consists of the shaft of a femur with the tro-
chanter present, but the articular head unluckily broken off and not
yet discovered in the collection; the lower portion has the articular
condyles broken off by an ancient fracture. It is not quite clear,
therefore, whether the bone belonged to an adult individual or not ;
but it indicates a smaller species than Hippopotamus Pentlandi, or any
of the other Hippopotamus remains sent previously by Captain Spratt.
This, however, will require further and more careful observation.
Besides these fragments of the bones of the extremities there have been
picked up by Captain Spratt, in a separate compartment, about a dozen
molars, some of them very perfect; the worn last molars with their
talons indicating a species of unusually small size, thus corresponding
with the indications yielded by the bones of the extremities. There
are also some fragments of incisors and some good specimens of
inferior canines, all of which will have to be compared with the
corresponding teeth from the Maccagnone and San Teodoro caverns,
belonging to Hippopotamus. ,

‘ Memo., July 31.—The molars and canines appear mostly to be of
the small species, Hipp. minutus, Cuv.’

- 1J.—Assrract or Communication By Dr. FaLconer To THE BritTIsH
ASSOCIATION AT CAMBRIDGE, ON OcToBER 6, 1862, ENTITLED ‘ ON
OssiFEROUS CAVES IN MALTA, EXPLORED By CAPTAIN SPRATT,
C.B., R.N., wira an Account or ELEpHAs MELITENSIS, A PIGMY
SPECIES OF Fosstz HLEPHANT, AND OTHER REMAINS FOUND IN
THEM.’

Reprinted from ‘ The Parthenon’ for October 18, 1862, p. 780.

‘ The author’s notice of this interesting discovery was prefaced by a
few remarks from Captain Spratt, explanatory of the former geogra-
phical connection of the islands of Sicily and Malta with the African
continent, as evidenced by their contained fossils and by the shallow
soundings which lay between these coasts and that continent. Three
eaves in Malta had yielded vast numbers of Mammalian bones of very
dissimilar character. The first, discovered while quarrying for a dock,
contained vast quantities of Hippopotamus bones, and was similar to a
cave previously discovered in Crete, The second, lying upon the side
of a ravine, three hundred feet above the sea, was partly filled with
clay and Elephant bones; the third contained bones of Hippopotami,
associated with remains of a large species of Dormouse.

‘Dr. Falconer, after corroborating the divisional line pointed out by
Captain Spratt, which, now represented by a line of roadway of shallow
water, formerly existed asa belt of land, dividing the Mediterranean into
two basins, and pointing out that the fossil fauna of Italy, Sicily, Malta,
and Crete, was that of the African and not of the European continent,
proceeded to describe the fossil remains new to science which Captain

x 2

308 OSSIFEROUS CAVES OF MALTA.

Spratt’s investigation had produced. The pigmy Elephant was an ani-
mal of remarkably small proportions; an adult individual could not
have exceeded the Indian Tapir in height and bulk, a creature not
much larger than a_full-grown Hog. Contrasted with the bones and
teeth of an adult African Elephant the difference in size of these por-
tions of its frame exhibited were most striking.! But though so small,
the skeleton agreed in every particular with the one of greatest bulk.
A series of harmonies ran through the two skeletons, one bone answer-
ing to another truly, and without ordinal or generic difference. The
author could refer it unhesitatingly to his subgenus Lowodon, in the
African group of Elephants. Captain Spratt’s researches had brought
to light many bones, tusks, and teeth, belonging to the young of this
pigmy species; the milk-teeth being about half an inch in length, and
the tusks corresponding in age not exceeding an inch in length.
Myoxus Melitensis, the fossil Dormouse, the remains of which had
been found in the Crendi Cave,? was as much a giant among the
Dormice as the Maltese Elephant was a dwarf. Other fossil remains
found in these remarkable caves were referable to two species of Swan :
one of colossal size, probably thrice as large as the bulkiest living
species. Many of the bones were gnawed, apparently by a powerful
felme animal, but no remains of the carnivore had yet been discovered
in association with the cave contents.’ $

1 The specimens exhibited comprised 2 See pages 300 and 305.—[Ep.]
the molars, tusks, lower jaw, vertebree, 3 See page 306, last line.—[ Ep. |
scapula, humerus, pelvis, femur, meta-
tarsal bones, and phalanges.—[ Ep. |

RHINOCEROS. 309

V. ON THE EUROPEAN PLIOCENE AND POST-
PLIOCENE SPECIES OF THE GENUS RHI-
NOCEROS.!

AFTER examining all the collections in England and Italy
and those of Lyons, Montpellier, &c., I have come to the con-
clusion that there were four distinct Pliocene and Post-Plio-
cene species of Rhinoceros, three of which have long been
confounded by Cuvier and other paleontologists under the
name of Rhinoceros leptorhinus. I have carefully examined
at Stuttgart the materials on which Kaup’s and Jiger’s Rhi-
noceros Merckii is founded. It is not a distinct species, but
is identical with the Grays Thurrock species, or Rhinoceros
leptorhinus (mihi). The R. Iunellensis of Gervais is founded on
a young jaw with milk-dentition, which is not to be depended
on for determining distinctions. So, also, the Rhinoceros
elatus of Croizet, and the R. mesotropus of Aymard, found in
Auverene, are not distinct species. JI have examined the
chief collections in Auvergne. The specimens in M. Pichot’s
collection and in the Museum of Le Puy are mainly R.
Ltruscus, while the R. mesotropus of Aymard comprises both
R. leptorhinus and R. antiquitatis. The four species may be
classified as follows :—
PLIOCENE.
I. No bony nasal septum.
1. Rhinoceros leptorhinus (Cuv. pro parte).
Syn. R. megarhinus of Christol.
Il. Partial bony septum.
2. Rhinoceros Htruscus (Falc.).
Syn. R. leptorhinus (Cuv. pro parte).
3. Rhinoceros hemitcechus (Falc.).
Syn. R. leptorhinus (Owen pro parte).
Post-PLIocENE.
III. Complete bony septwm.
4, Rhinoceros antiquitatis (Blumb.).
Syn. R. tichorinus (Fischer and Cuvier).

1 The introductory remarks have been | first time published. The notes on Phin.
compiled by me from two letters, ad- | feptorhinus, including the lengthened
dressed by Dr. Falconer in 1862 to Mons. | description of the Cortesi cranium at
Lartet, of Paris, and Col. Wood, of Stout- | Milan, on Rhin. Etruscus and on Rhin.
hall, Swansea, and from his note-books. | antiguitatis, are extracted from the

The important essay on Rhin. hemitechus | author’s note-books.—[Ep. |
was written in 1859, but is now for the

310 RHINOCEROS.

1. Rhinoceros leptorhinus.—This is the original and typical
Rhinoceros leptorhinus of Cuvier, founded on Cortesi’s Monte
Zago cranium. It is the species described by Christol as R.
megarlinus, and is the only Pliocene or Post-Pliocene European
species that had not a nasal septum. To this belongs the
celebrated Cortesi cranium in the Museum at Milan, which I
have carefully examined. With this species also I have iden-
tified the Rhinoceros remains found in the Sub-Apennine beds
of Piacenza, in the Val d’Arno upper beds, at Montpellier and
Lyons, and at Grays Thurrock in Essex. The Rhinoceros,
however, found in the Elephant-bed of the Norfolk coast is
different.

2. Rhinoceros Etruscus.—This species, like the following,
had an incomplete bony nasal septum, but it had a compara-
tively slight and slender form. It is met with, along with
Elephas (Loxodon) meridionalis and Mastodon Arvernensis, in
the lower beds of the Val d’Arno, and in the ‘ Submarine
Forest-Bed,’ or super-imposed blue clays of the Norfolk Coast,
immediately underlying the boulder-clay ; but, as yet, it has
been found in none of the ossiferous caves of Britain. With
this species, also, I have identified the remains of a Rhinoceros
submitted to me by Professor Ansted, which were found a few
miles from Malaga, in white marl, overlying Pliocene blue
clay abounding with shells.

3. Rhinoceros hemiteechus.—This species has been described
by Professor Owen as R. leptorhinus. It has the nasal septum
incomplete in the centre, and it differs from R. antiquitatis
(R. tichorinus) in other cranial characters, as well as in those
of the teeth. I am satisfied on this point, after examining the
entire dentition of both young and old animals. Rhinoceros
heniteechus accompanies Hlephas antiquus in most of the oldest
British bone-caves, such as Cefn, Durdham Down, Minchin
Hole, and other Gower caverns. It is also found at Clacton
in Hssex (e.g. The ‘Clacton Rhinoceros’), and in certain
beds in Northamptonshire. It is also met with in Italy.

From some of these localities, entire skulls and a great
portion of the skeleton have been obtained.

A, Rhinoceros antiquitatis (R. tichorinus).—This species had
a complete bony nasal septum. It is found in the newer Plio-
cene deposits of Kent, Surrey, and Essex, and associated with
Elephas primigenius in caverns of the same date. Hlephas
antiquus with Rhinoceros hemitechus, and Hlephas primigenius
with Rhinoceros antiquitatis, though respectively characte-
rizing the earlier and later portions of our period, were
probably contemporary animals ; and they certainly were com-
panions of the cave-bears, cave-lions, and cave-hyxnas, and
of some at least of the existing mammalia. There can be no

INTRODUCTORY REMARKS. 311

reasonable objection to the name Rhinoceros antiquitatis. South
of the Rhine, that is in Geneva, France, and Italy, all modern
palzontologists call the species Rhinoceros tichorinus ; but,
north of the Rhine, in Germany, Holland, Scandinavia, and
Russia, the most eminent authorities designate it Rhinoceros
antiquitatis. A name in science ought not to be a disputed
point of mere geographical predilection. Blumenbach named
it first Rhinoceros antiquitatis. Fischer de Waldheim, a pa-
leontologist of no great authority, changed the name into
Rhinoceros tichorinus, and Cuvier adopted Fischer’s name
without acknowledgment. Desmarest called it Rhinoceros
Pallasii. Blumenbach’s names of Hlephas primigenius and
Mastodon Ohioticus are now accepted by everyone; and there
is no reason why his Rhinoceros antiquitatis should be rejected
for a more modern name. Living neither north nor south of
the Rhine, I have no geographical predilections, and as an
impartial foreigner I accept the earliest name, viz. Blumen-
bach’s; besides, the name Rhinoceros tichorinus is faulty,
inasmuch as three species had a nasal septum.

T.—On RHINOCEROS HEMIT@HCHUS, AN EXTINCT SPECIES
PREVAILING IN THE GOWER Caves, SoutH WatEs.!

In two previous communications (Quart. Journ. Geol.
Soc. for Nov. 1857, and vol. xiv. p. 81),? I have attempted to
trace the distribution of the fossil Proboscidea, with some of
their constant associates, in the newer Tertiary deposits of
England, and in corresponding deposits on the continent of
Europe. One important branch of the inquiry concerns the
fossil remains of the ossiferous caves; but my examination
of the cave-collections was not, at the time, sufficiently ex-
tended to warrant well-founded conclusions on the subject.
I had seen undoubted evidence of the occurrence of Hlephas
antiquus and Hippopotamus major—both Pliocene forms—in
several of the English caverns ; but I was in doubt regarding
the associated fossil species of Rhinoceros. Since then I
have had opportunities of examining most of the great cave-
collections in the metropolitan and provincial museums, and
of investigating, on the spot, the conditions under which the
remains were associated in several of the most productive
caverns. Some of the results appear to be of sufficient interest
to warrant my bringing them before the Society,’ although
with less detail of evidence, and in a more restricted form,
than the nature of the case might seem to demand. But the
general subject is so extensive in its relations as hardly to

1 The MS. of this essay was found; *% The paper was evidently intended
among Dr. Faleoner’s papers, and is now | for presentation to the Geological

for the first time published.—[Ep.] Society.— [| En. ]
2 See antea, pp. 1 and 76.—[Ep.]

312 RHINOCEROS.

be susceptible of being embraced within the scope of a single
communication ; and the remarks which follow the descriptive
details forming the special subject of this essay will be con-
fined to the association, in some of the ossiferous caverns in
England, of the remains of certain of the fossil mammalia,
which I regard as positive indicators of the age of Pliocene
deposits, without reference to the altered physical conditions
of the caves at different periods, or to the agencies by which
the remains were introduced within them.!

I may premise that my inquiries have embraced an exami-
nation, more or less detailed, of collections from the follow-
ing caverns :—Kent’s Hole, Oreston, and other South Devon-
shire caves; Banwell, Bleadon, Hutton, Berrington, &c., in
the Mendip Hills; Paviland, Spritsail Tor, Minchin Hole,
Bacon’s Hole, and Bosco’s Den, in the peninsula of Gower,
in South Wales; Cefn, in North Wales; Kirkdale and
Wirksworth. The museums which have been visited in
search of materials are the British Museum and those of the
College of Surgeons and Geological Society, in the metro-
polis ; and in the provinces, Oxford, for Dr. Buckland’s very
extensive and classical series of cave-remains from British
and foreign localities; Bristol, for the interesting collection
from Durdham Down, formed and described by Mr. Stutch-
bury; Taunton, for the collection amassed during many
years by the Rev. D. Williams, from Bleadon, Hutton, and
others of the Mendip Caverns; Torquay and Plymouth, for
Kent’s Hole and Oreston ; Swansea, for the Gower Cave col-
lections ; and York, for that from Kirkdale. I have further
had the advantage of examining the private cave-collections
of the veteran Mr. Wm. Beard, at Banwell, from the Mendip
caverns; of Miss Talbot, at Penrice Castle, from Paviland ;
and, above all, the unrivalled collection formed at Stout Hall,
by my friend Colonel E. R. Wood, F.G.S., during the last
nine years, from the ossiferous caves of Gower. This last has
furnished more materials for the description of the extinct
Rhinoceros, which is the special subject of this paper, than
all the rest together.

Rhinoceros hemitechus.2—The species to which I have
assigned this name (for reasons which will more fully appear
in the sequel) is, avowedly, not a new accession, except by
name, to the Fossil Fauna of Britain. It has long been fami-
liar to geologists as the Rhinoceros leptorhinus of Cuvier,
according to Professor Owen, and described at great length

1 This portion of the essay was never ? From jusvs, half, and ro?xos, parti-
written; but the subject will be found | tion, in reference to the partial nasal
treated in the author's paper, ‘On the | septum, distinctive of the species.
Ossiferous Caves of Gower..—[Ep.]

RHINOCEROS HEMITQCHUS. 313

in the ‘British Fossil Mammalia.’ I have arrived at the
conclusion that it is essentially distinct from the original
Rhinoc. leptorhinus of Cuvier, which latter, however, I be-
lieve occurs in England, in beds, in some respects different
from those in which Rhinoc. hemitechus prevails, and to a
certain extent, with different associates. In this view, the
exact identification of the two species becomes in its geolo-
gical bearings a question of much higher importance than
the mere rectification of a specific name. Before entering
on the descriptive details, it will be necessary to revert to the
«origin of the name Rhinoc. leptorhinus, and to trace the suc-
cessive opinions which have been entertained by palzeontolo-
gists regarding it up to the present time; for there is not,
within the whole range of Mammalian Paleontology, an ex-
tinct species regarding which more has been written and
more opposed views advanced.

The great French anatomist, having conclusively demon-
strated the distinctness of the Siberian Rhinoceros from all
the species then known, framed his diagnostic character upon
the most obvious of its peculiarities, namely, the ossified
nasal septum, and designated it ‘le Rhinoceros a narines cloi-
sonées,’ or Rhinoceros tichorhinus. His attention was natu-
rally awakened to the probability of other species occurring
in the fossil state, in which the nasal septum would be found
to agree with existing species, in presenting the ordinary
condition of an unossified cartilage. Cortesi had discovered in
1805, upon Monte Zago, near Piacenza, the entire skull, in
fine preservation, of a fossil Rhinoceros, which he referred
with doubt to a young Rlin. bicorms.! A drawing of this
cranium, by M. Adolphe Brongniart, and thus carrying high
authority with it of a competent execution, was many years
afterwards forwarded to Cuvier from Milan. The figure
represented a cranium differing essentially in form and pro-
portions from that of the Siberian Rhinoceros, and most ob-
viously in the absence of the bony partition of the nostrils,
characteristic of the latter. Cuvier inferred that the Italian
form constituted a different species, which, in contradistinc-
tion, he named ‘le Rhinoceros a narines non-cloisonées,’ or
Rhinoceros leptorhinus. The specific distinctions which he
indicated for the latter were, that the cerebral part of the
skull was proportionally shorter than in the Siberian form,
and less projected backwards over the occiput; the position
of the orbit above the fifth molar; the termination of the
nasal bone by a free point having no connection with the
intermaxillaries through a bony partition ; and the abbrevia-

1 Sagei Geologici, p. 72, tab. vil.

314 RHINOCEROS.

tion and different form of the intermaxillaries. To these
cranial characters he added more slender proportions in the
general construction of the skeleton, inferred from Val
d’Arno specimens which he attributed to the same species ;
and he held that the Italian fossil form approached more
nearly to the Rhinoceros bicornis of the Cape than to any
other known species. He appears to have considered that it
had been invested with two horns. Upon the characters of
the molar teeth he furnished little beyond what was merely
conjectural; for, having founded his conception of the
Species mainly upon the characters furnished by Cortesi’s*
skull, without examining the molars in that specimen, he
took it for granted that all the lower jaws, molars, and other
remains, occurring in Italy, which did not admit of identifi-
cation with Rhinoc. tichorhinus, must of necessity belong to
his Rhinoc. leptorhinus. The subject was not at the time
sufficiently advanced, nor the materials in sufficient abund-
ance, to lead him to conjecture that there might have been
two or more Italian fossil species different from the Siberian
form. But there are now the strongest grounds to believe
that such is the case; and that Cuvier, as in the similar in-
stance of EHleph. primigenius, Eleph. antiquus, and EHleph.
meridionalis, confounded the remains of at least two Italian
fossil species of Rhinoceros under the common designation of
Rhinoceros leptorhinus.

Rhinoceros leptorhinus,as thus defined by Cuvier, met with
ready acceptance among palzontologists, and remained un-
disputed until the year 1834, when M. de Christol,' in a very
able and elaborate memoir ‘On the Characters of the Large
Species of Fossil Rhinoceros,’ broadly asserted that this sup-
posed species had no existence in nature, and that Cortesi’s
cranium belonged to the Siberian form, Kh. tichorhinus.
Christol, like Cuvier, had not an opportunity of examining
the original, which in the interval had suffered considerable
injury by fracture of the facial portion ; but, having received
from Milan fresh drawings of the specimen thus altered in
appearance, he erroneously interpreted as a bony septum a
shaded representation of the internal surface of the nasal
cavity of the left side of the head, viewed from the right
side, where the corresponding part was mutilated. Dr. Cor-
nalia, of Milan, so late as 1853, submitted Cortesi’s cranium
to a rigid examination, specially with a view to the deter-
mination of this point, and states in the most positive terms

that there is not a trace even of the supposed bony septum: —

‘ Cette cloison n’existe nullement. La voite de la cavité
nasale ne présente, le long de sa ligne médiane, aucun prin-
1 Annales des Scienc. Nat. 1835, 2me Ser. tom. iv. p. 44.

%

RHINOCEROS HEMITQCHUS. 315

cipe de cloison descendante qui aurait pu étre détruite.
Enfin je suis stir, et je vous assure que le crane que nous con-
servons n’appartient pas au R. tichorhinus, et qu’on a eu tort
de confondre les deux espéces. Le regard de M. Cuvier était
bien plus percant et tombait plus justement dans le vrai.’ !
Christol erased Rh. leptorhinus from the list of fossil species,
and at the same time proposed the name of Rhinoceros mega-
rhinus for the remains of a two-horned fossil species occur-
ring in the Pliocene Sands of Montpellier, and characterized
by the great length of the nasal bones; by the short in-
- terval between the nasal sinus and the orbits; by the slight
elevation of the pyramid of the vertex above the plane of
the brow; by the inconsiderable inclination of the occipital
plane, which is abruptly truncated at the vertex; by the
relative position of the orbits, and by peculiarities in the
teeth. Marcel de Serres had previously endeavoured to dis-
tinguish the same form under the name of the ‘ Fossil Rhino-
ceros of Montpellier’ (Rhinoceros Monspessulanus, De Blainv.) ;
but gave way to the dissent expressed by Cuvier, who iden-
tified it with his ‘ Rhinoceros a narines cloisonées.’ Christol
was further led to the conclusion that the Rhinoceros incisi-
vus of Cuvier was identical with his Rhinoc. megarhinus.

From a remark by Laurillard, it would appear that at a
later date Christol was convinced that his opinion respecting
Rhin. leptorhinus was erroneous ; but no formal expression of
this altered view having been published, the objections which
he had raised continued for a considerable time to influence
the opinions of palzontologists.

Croizet and Jobert, in 1828, described and figured remains
of a Rhinoceros from Puy-de-Déme, which from its general
slender proportions they designated Rhinoceros elatus. No
perfect cranium of this form has as yet been discovered in the
Velay; and the jaws and teeth at present known are not
sufficiently pronounced to determine with certainty whether
Rhinoceros elatus is distinct, or to what nominal species it
ought to be referred. De Blainville identified it with the
Miocene Rhinoceros incisivus! Laurillard doubted whether it
ought to be referred to R. megarhinus or to R. leptorhinus ;
Pomel refers it to his Atelodus elatus, which includes Rhinoc.
elatus, together with Rhinoc. megarhinus of Christol; and
Gervais hesitatingly refers it, together with Owen’s form of

1 The first authoritative correction of | explains the cause of the mistake. But
Christol’s statement was made in 1842, | the correction escaped the notice of
by Professor Balsamo Crivelli, the | European Palontologists until 18538,
curator of the Museum of Santa Teresa | when Dr. Cornalia of Milan, at the
in Milan, where the specimen. was pre- | request of Duyernoy, re-submitted the
served. He states, that the supposed | Cortesi cranium to a rigid examination,
partition was absolutely wanting, and

316 RHINOCEROS.

Rhinoceros leptorhinus from Clacton, also to Rhinoceros
megarhinus.'

Jager, in 1839, proposed the provisional name of Rhino-
ceros Kirchbergense for certain remains discovered in sand-pits
in the pleistocene (‘ Diluvial-boden’) deposits of Kirchberg
in Wurtemberg. The materials were limited to one lower
and two detached upper molars; and the comparison of them
was confined to corresponding teeth of Aceratheriwm incisiwum,
of the Rhinoceros tichorhinus occurring at Cannstadt, and of
the two-horned Rhinoceros of the Cape. No attempt was
made by Jager to distinguish the Kirchberg form from the lep-
torhine Rhinoceros of Cuvier, the R. elatus of Croizet, or the
R. megarhinus of Christol.2 The name proposed by Jager
has therefore strictly no claim to be regarded otherwise than
as a conjectural determination; and at a later period he
abandoned it, having adopted the opinion of Owen, that the
Kirchberg Rhinoceros was identical with the supposed Rhi-
noceros leptorhinus, discovered at Clacton, as described in the
‘ British Fossil Mammalia.’

In 1841 Kaup brought out, in the ‘ Akten der Urwelt,’ his
description of the same nominal species, but under the new
designation of Rhinoceros Mercku of Jiiger, who, at the
instance of his friend Kaup, consented to the substitution
of this specific name, both as less open to objection on the
score of local derivation, and as a tribute to the memory of
Merck, its earliest indicator, who, towards the close of the
last century, made the first important step towards the dis-
tinction of the Mammoth from existing species. Kaup col-
lected additional materials from various localities in the
valley of the Rhine, and extended their comparison, beyond
what was attempted by Jager, to supposed remains of the
Rhinoceros leptorhinus of Cuvier. The conclusions at which
he arrived were, that Rhinoceros Merckwi was a distinct species,
of the size of the two-horned Rhinoceros of the Cape; that
it belonged, jointly with Rhinoceros Africanus (R. bicornis)
and Rhinoceros leptorhinus of Cuvier, to a particular division
of the genus, characterized by the form of the molar teeth
and the absence of incisors; and that it had been a con-
temporary of the Mammoth, Rhinoceros tichorhinus, Rhi-
noceros leptorhinus, and other forms of the so-called Diluvial
Period.?

The next step of importance in the history of Rhinoceros

1 It was subsequently referred by Dr. 8 In his last work (Beitriige, 1 Heft.
Falconer to Rhin. Etruscus—I{Ep.] p-. 4), Kaup gives up Phin. Merchii for
* The lower jaw, in the reference to | hin. leptorhinus,
fig. 6, tab. xvi., is attributed by Jiiger to
thin. tichorhinus,

RHINOCEROS HEMITGCHUS. 317

leptorhinus dates from the publication of the ‘ British Fossil
Mammalia’ in 1846, when Professor Owen brought out his
elaborate and detailed description of the remarkable cranium
and other remains discovered at Clacton, in Hssex, by our
veteran Associate, Mr. John Brown, of Stanway. The skull
in question is chiefly notable from its presenting the well-
marked appearance of an incomplete bony partition connect-
ing the anterior half of the nasal bones vertically with the
osseous floor of the nasal cavity. (See Plate XV.) When
the specimen first came under the inspection of Mr. Owen,
he was induced to refer it, on account of this septum, to the
“Rhinoceros a, narines cloisonées, or Rhinoc. tichorhinus of
Cuvier, and it is quoted as such in his Report to the British
Association in 1843. But when submitted to a more rigorous
examination, at a subsequent period, the practised eye of this
eminent palecntologist detected in it important points of
difference irreconcilable with Rhinoceros tichorhinus; and
having faith in the accuracy of the confidently-expressed,
but erroneous conclusions of Christol, respecting the presence
of a septum in Cortesi’s cranium, he was naturally led to
identify the Clacton skull with the Rhinoceros leptorhinus of
Cuvier. This conviction was strengthened by the examina-
tion of a ramus of the lower jaw, also found by Mr. Brown
in the same deposit at Clacton, which Professor Owen con-
cluded was identical with lower jaws from Tuscany, referred
by Cuvier to his Rhinoceros leptorhinus (Oss. Foss., tom. ii.
Pl. TX. figs. 8 and 9); and with the lower jaw from the
Rhine, referred by Kaup to Rhinoceros Merckii. The Clacton,
Tuscan, and Rhenish specimens were included under the
common designation of Rhinoceros leptorhinus.

The great weight of Professor Owen’s authority was
evinced in the accounts given by other palzontologists of
Rhinoceros leptorhinus after 1846. De Blainville, in his ‘ Os-
téographie,’ although at variance upon some points of detail,
admitted the Clacton skull into his limitation of Rhinoceros
leptorlinus, with which he combined the Rhinoceros of Mont-
pellier, of Marcel de Serres, and the Rhinoc. megarhinus of
Christol. But he eliminated the Rhenish materials, referred
by Jiiger and Kaup to Rhin. Merckii, and referred them to
Rhinoceros incisivus, being in his view the male of the Miocene
Aceratherium incisivum of Eppelsheim! This portion of De
Blainville’s paleontological labours has met with severe
strictures from some of his own countrymen, and with stern
condemnation by Kaup.

Laurillard, in 1849, in his revision of the Fossil Species
of Rhinoceros, presents Rhinoceros leptorhinus in a manner
which attempts to combine the irreconcilable conceptions of

318 RHINOCEROS.

Cuvier and of Owen. He admits the partial bony septum
described by the latter, and even concedes three fossettes to
the upper molars, as in Rhin. tichorhinus, excepting only the
last true molar; while he attributes to it the slender pro-
portions inferred by Cuvier, and assigns for its habitat Italy
and the Pliocene formations of England. Laurillard ad-
mitted also Rhin. megarhinus of Christol, or Rhin. Mons-
pessulanus of De Blainville, as a distinct species. He refused
to accept the Rhin. Mercku of Jager and Kaup, and the
Rihin. elatus of Croizet he regarded as referable either to
Rhin. megarhinus or to Rhin. leptorhinus.

Gervais has devoted much study to the fossil species of
Rhinoceros, occurring in the Pliocene and Post-Pliocene de-
posits of Auvergne and the South of France. The results
are embodied in the ‘ Paléontologie Frangaise.’ He adopts
the Rhin. megarhinus of Christol, yet although that species is
described by all original observers, himself inclusive, as
devoid of a bony septum, he considers it probable that the
Clacton cranium figured by Professor Owen as of Rhin.
leptorhinus belongs, notwithstanding its septum, to that
form. On the other hand, he doubtingly admits the Rhin.
leptorhinus of Cuvier as a distinct species, occurring in Italy
and the Velay. He has applied the designation of Rhin.
Innellensis to the remains of a species discovered in the
Cave of Lunel-viel, first named Rhin. minutus by Marcel de
Serres, Dubrueil, and Jean-jean, under a mistaken interpre-
tation of the age of the teeth, and at a later date described
as being identical with the Rhin. Africanus. He has re-
peatedly directed the attention of palxontologists to the
important fact, that this fossil species of Lunel-viel is hardly,
if at all, distinguishable from the existing two-horned Rhi-
noceros of the Cape.

Pomel, in his ‘ Catalogue,’ published in 1854, after a study
of the remains occurring in Auvergne and the Velay, admits
Rhin. leptorhinus with a bony nasal septum, as defined by
Professor Owen, but under the designation of Atelodus lep-
torhinus; and gives for its habitat England, the Milanese,
and the valley of the Rhine. Under another name, Atelodus
elatus, he includes the Rhin. elatus of Croizet, and the Rhin.
megarhinus of Christol. A third species, exclusive of Rhin.
tichorhinus, he designates Atelodus Aymardi, and refers to it,
as a synonym, the Rhin. leptorhinus of Gervais.

Duvernoy, the successor of Cuvier and De Blainville in the
chair of Comparative Anatomy, attempted a revision of the
Fossil Species of Rhinoceros, in a very elaborate memoir
published in 1854. In the section devoted to the Pliocene
species, he maintains, with many details, that the Rhin.

pn J -j
f
:
3
:
: i ‘
,
x ioe
‘ ; ,
. , *
wi
i
h 4
:
'
_* i*
f
Pe Oa x
v. - ty
: a eaee
}
ey Aine
, we ay
t
;
|

‘ Pky “-
PS a) | eee, ie Yh

DESCRIPTION OF PLATE XV.
RHINOCEROS HEMITECHUS.

The figures in this Plate represent the ‘ Clacton Skull’ in
the British Museum (Cat. No. 27,836, not 132,133, as stated
in text), described by Professor Owen, in the ‘ British Fossil
Mammalia,’ as Rhinoceros leptorhinus. The figures have
been copied from original drawings executed for Dr. Falconer

by Mr. Dinkel, and are one-ninth of the natural size. (See
pages 317 & 351.)

Fig. 1. Profile view of cranium, showing partial nasal septum projecting
downwards.

Fig. 2. Under surface. The posterior portion is only drawn in outline.
At the anterior extremity is seen the nasal septum.

Fig. 3. Upper surface of cranium.

Fig. 4. Section of the nasal septum, one-third of the natural size.

VoL, Il,

VoLIL Plate 15.

J.Dinkeal del et lith W.West imp.
Rhinoceros hemitcechus.

(The Clacton Skull.)

Loe 5

cue

pee ok

RHINOCEROS HEMITGICHUS. 319

leptorhinus, as established by Cuvier, was a sound species;
and that Cortesi’s cranium was entirely devoid of a bony
septum, according to the positive evidence of Dr. Cornalia.
To this Rhin. leptorhinus he refers the Rhin. megarhinus of
Christol, and the Rhin. Monspessulanus of Marcel de Serres.
He regards the Clacton cranium, described by Professor
Owen, as wholly distinct from Rlin. leptorhinus, and, although
still different, as being more closely allied to Rhin. tichor-
linus. He proposes for it provisionally the specific desig-
nation of Rhin. protichorhinus, as an independent form. Not
the least remarkable result of Duvernoy’s inquiries was, that
he identified, as certainly belonging to Rhin. tichorhinus, the
representations figs. 8 and 9, Pl. IX. of the ‘ Ossemens Fos-
siles,’ which Cuvier adduced as typical illustrations of the
lower jaw of Rhin. leptorhinus, from specimens found in the
Val d’Arno; and upon which Professor Owen mainly relied,
in identifying the lower jaw from Clacton with the latter
species! A more signal illustration could not be adduced of
the diametrically opposite conclusions which may be drawn
by different paleontologists from the same evidence, when
presented in the form of imperfectly executed figures.

Lastly, Brandt, in his very complete and valuable mono-
graph of the Rhinoceros of Siberia, published in 1849, reviews
the figures and description of the Clacton skull given by
Professor Owen, and expresses the opinion that it does not
belong to Rhin. leptorhinus, but to an individual of Rhin.
tichorhinus, in which the septum was not completely ossified.
He gives a representation of a Siberian instance of this
nature, corroborating the cases previously cited by Pallas
and Collini.

In order to show at a glance the range and fluctuation of
opinion on this paleontological question, it may be useful to
summarize them in a few words :—

1. 1812. Rhinoceros tichorhinus, established by Cuvier as character-
ized by its bony nasal septum.

2. 1819. The ‘ Rhinoceros of Montpellier’ (Rhin. Monspessulanus, De
Blainv.), proposed by Marcel de Serres as a distinct form ;
identified by Cuvier (1822) with Khinoc. tichorhinus ;
tacitly abandoned by De Serres.

8. 1822. Rhinoceros leptorhinus, proposed by Cuvier upon Italian
specimens as destitute of a bony septum.

4, 1828. Rhinoceros elatus, of the Velay, proposed by Croizet and
Jobert.

5. 1834. The absence of a bony nasal septum in Cuvier’s Rhinoc. lep-
torhinus, denied by De Christol; the name regarded as
a synonym, merely, of Rhinoc. tichorhinus.

6. 1834. The‘ Rhinoceros of Montpellier,’ reproduced by De Christol
under the name of Lhinoc. megarhinus, as identical with

320 RHINOCEROS.

Rhinoe. incisivus of Cuvier, but distinct from Rhznoe. ti-
chorhinus and from the supposed Rhinoc. leptorhinus.

7. 1838. Rhinoceros primigenius, proposed by Bronn, in the ‘ Lethea ~
Geognostica,’ to include the Rhinoc. tichorhinus and
Rhinoc. leptorhinus of Cuvier, in conformity with the
views of Christol.

8. 1839. Rhinoceros Merckii (syn. Rhinoc. Kirchbergense, Jig.) pro-
posed by Jiiger (1839), and by Kaup (1841), as a distinct
form.

9. 1842. Rhinoceros de Filippi, proposed by Balsamo Crivelli, for
remains occurring in the Lignite of Leffe (Gandino) as
distinct alike from Rhinoc. tichorhinus and Rhinoc. lepto-
rhinus.

10. 1846. Rhinoceros leptorhinus of Cuvier, reproduced by Owen
upon British fossil specimens, but invested with a bony
septum, and Rhinoc. Merckii identified with it; Rhinoc.
megarhinus or R. Schletermacheri, held to be distinct.

11. 1847. Rhinoc. leptorhinus admitted by De Blainville, as with or
without a bony nasal septum; Rhinoc. megarhinus com-
bined with it; but Rhinoc. Mercki transferred to Rhinoc.
MMCiSivUs.

12. 1849. Rhinoc. leptorhinus of Cuvier, accepted by Laurillard, in
the view of Owen, as having a bony septum; Rhinoc. me-
garhinus of De Christol held to be distinct.

13. 1849. The Rhinoceros leptorhinus of Cuvier accepted by Brandt,
but the Rhinoc. leptorhinus of Owen identified with
Rhinoc. tichorhinus (!).

14. 1852. Undecided opinions entertained by Gervais, who adopts the
Rhinoc. megarhinus of De Christol, and leans to the view
that the Rhinoc. leptorhinus of Cuvier, founded on Cor-
tesi’s cranium, and of Owen, are of the same species; but
that the Rhinoceros remains of the Velay (Rhinoc. elatus,
Cuv.) and of the Val d’Arno belong to another distinct
form.

15. 1854. Rhinoc. leptorhinus, adopted by Pomel, in the view of
Owen, as having a bony septum, under the name of Atelo-
dus leptorhinus; and another species, besides Ehinoc.
tichorhinus, proposed under the name of Atelodus Aymardi,
as also having a bony septum; the Rhinoc. elatus of
Croizet, identified with Rhinoc. megarhinus, under Atelodus
elatus.

16. 1854. The Rhinoceros leptorhinus of Cuvier, reproduced by
Duvernoy as destitute of a bony septum; Ahinoc. mega-
rhinus identified with it; and the Rhinoc. leptorhinus of
Owen erected into Rhinoc. protichorhinus.

The above table suggests a grave and instructive comment
on the uncertainty of paleontological determinations, even
when guaranteed by names of the highest authority. The
point upon which hinged the discussion, protracted during
upwards of twenty years, was, ‘Had Rhinoc. leptorhinus an
osseous nasal septum, or had it not?’ The pendulum oscil-

RHINOCEROS HEMITGCHUS. 321

lated between septum and no septum. The array of autho-
rities on either side was nearly balanced, with the exception
of a discreet few headed by De Blainville, who followed the
convenient via media and argued that the character was of
little importance, being but a degree, more or less, of ossi-
fication of the nasal cartilage, and that, according to circum-
stances of age, sex, or vigour in the species, might, or might
not, have had the partition ossified. Considering that the
cranium upon which Cuvier relied has been deposited during
nearly half a century in one or other public museum in Milan,
on the high road of continental travel, it might have been
expected that the disputed point would have been speedily
settled by an appeal to the original specimen. But until
the appearance of Crivelli’s evidence in 1842, confirmed by
Cornalia in 1854, the Cortesi cranium, upon which the case
rested, does not appear to have been examined by any one of
the numerous paleontologists all over Hurope who took a
share in the dreary discussion.

It will be admitted that an essay to determine with pre-
cision a single form, out of such a class of confused synonymy
and perplexed opinions, will be of some service to Paleeonto-
logy. This I shall endeavour to do with the Clacton species,
hitherto described under the name of Rhin. leptorhinus; and it
has appeared to me to be better to give it a new specific
name, than to attempt to identify it conjecturally with some
one of the names that have already been proposed for forms
supposed to be different from the Rhin. leptorhinus of Cuvier.
The ad interim designation, suggested by Duvernoy, for the
Clacton species of Rhin. protichorhinus, is manifestly inadmis-
sible. Whether Rhinoceros hemitechus may not be identical
with some of the materials figured and described by Kaup,
under the name of Rhin. Merckii, Iam unable to determine
satisfactorily. The upper molars from Chagny and Crozes,
figured by Cuvier, which Kaup refers to that species, differ
materially in the form of the ‘ crochet,’ a character of great
significance, from those of Rhin. hemitechus. The same un-
certainty applies to the Atelodus Aymardi of Pomel, from the
Velay, so named in his ‘ Catalogue Méthodique,’ but without
figures or sufficient distinctive characters to establish the
species. Rhinoceros hemitechus certainly differs from Rhin.
leptorhinus of Cuvier, as founded on Cortesi’s cranium, which
I have examined, both in the dental characters and form of the
skull, and also in the general proportions of the skeleton ;
and it differs equally from the Rhinoceros megarhinus of De
Christol, skulls and other remains of which I haye examined at
Lyons and Montpellier. Ifthe distinctness of the species is
established, and its range in time and geographical distribu-

VOL. II. Y

322 RHINOCEROS.

tion over Europe are determined, my objectin this communica-
tion will have been attained. Itis left to systematic writers on
Paleontology to decide by what specific designation the form
here called Rhinocercs hemitechus shall hereafter be recog-
nized. In the meantime, the name now applied will be of
convenience to geologists in dealing with the Mammalian
remains of one period of the Caves, and of deposits of the
age of Clacton, and certain localities in Northamptonshire as
distinct, on the one hand, from the ‘ EHlephant-bed’ of the
Norwich coast, and on the other, from the superficial gravels
of the Glacial period.

My first acquaintance with the species dates from the
spring of 1858, when, on a visit to Plymouth, to examine the
remains of the Oreston caves, I saw in the possession of Mr.
Spence Bate a beautiful drawing (which he liberally placed
at my disposal) of a ramus of the lower jaw of a Rhinoceros,
discovered by Colonel HE. R. Wood in ‘ Bacon Hole,’ which
a cursory examination satisfied me differed alike from
Rhin. leptorhinus and from Rhin. tichorhinus. (See Pl. XX1.)
On proceeding to Swansea, in company with my friend the
Rev. Robert Everest, I compared the original of Mr. Spence
Bate’s drawing with a fine specimen of a corresponding
ramus of the lower jaw of a fossil Rhinoceros, from the EHle-
phant-bed’ of the Norfolk coast, belonging to the collection
of the Rey. John Gunn of Irstead,! which I had previously
inferred to be of Rhinoceros leptorhinus of Cuvier, as met with
in the valley of the Po and the Val d’Arno. In the Museum
of the Royal Institution of South Wales at Swansea, besides
the specimen in question, I found the right and left rami of
another lower jaw, containing on the left side the series of
the six posterior molars in beautiful preservation (Pl. XIX.),
together with a fragment composing four ‘consecutive molars
of the upper jaw, right side (namely the penultimate and
antepenultimate true molars, and the two posterior pre-
molars), and likewise some vertebree and fragments of bones
of the extremities. The whole of these remains were dis-
covered in 1850 in the cave of ‘ Bacon Hole,’ in Gower, about
six miles west of Swansea, during an exploration carried on
by Colonel E. R. Wood, of Stout Hall, by whom they were
presented to the Swansea Museum. The character of the
upper molars established to a certainty the distinctness of
the species. On communicating this result to Colonel Wood,
T was informed by him that in another of the Gower Caves,
named ‘Minchin Hole,’ the exploration of which he had
undertaken after exhausting ‘Bacon Hole,’ he had dis-

1 See antea, p. 349,—[Ep.]

onawt

ams

RHINOCEROS HEMIT@CHUS. 323

covered the entire cranium of an adult Rhinoceros with the
series of molars complete on both sides, the nasal bones
perfect to their tips, and a well-pronounced partial bony
septum connecting the antericr portion of the nasals with
the floor of the nostrils. This most precious and unique spe-
cimen met with a grievous accident, by which it was crushed
and destroyed whilst temporarily out of Colonel Wood’s hands ;
and all that now remains of it is the palate, with the line of
molars on either side, which, is deposited in the Museum
of the College of Surgeons. I found in Colonel Wood’s rich
collection at Stout Hall a very fine specimen comprising
the cerebral part of another cranium of the same species of
Rhinoceros, the facial portion of which appears to have been
broken and destroyed by the workmen during extrication
from the floor of the cavern. This fragment will be described
in the sequel.}

Colonel Wood, on being made aware of the important
nature of the result of his researches, immediately recom-
menced the operations, which had been temporarily sus-
pended, in ‘ Minchin Hole,’ and discovered a large number

- of additional remains of the same species. These I had an

opportunity of examining on a second visit to Gower during
last autumn, and the whole series has been liberally placed
at my disposal by Colonel Wood.

My attention was next directed to ascertaining whether
the Gower form occurred in any other of the ossiferous caverns
in England; and on proceeding to Bristol, I found in the
very interesting series of fossil remains, discovered by Mr.
Stutchbury, in Durdham Down, several upper molars speci-
fically identical with those of the Rhinoceros of Bacon Hole
and Minchin Hole.? The same result followed an exami-
nation of the Rhinoceros remains from ‘Oreston,’ near
Plymouth, described by Mr. Whidbey in 1817,° and now
preserved in the Museum of the College of Surgeons. They
all proved to belong to the same species. I next instituted
a comparison between the upper molars discovered by Mr.
John Brown, F.G.8., at Clacton and the Gower specimens,
with the same result.°

The materials available for the description of the species
are therefore very abundant, including specimens, more or
less complete, of at least four crania of different ages; five
upper jaws presenting the molars in different stages of wear ;
eleven rami of the lower jaw, young and old ; together with
fragments of most of the principal bones of the extremities.

1 See paper on the Gower Caves, and 8 Philosoph. Trans., 1817.
also pp. 850 to 352.—[Epb. ] * See Appendix, No. IX.,p.353.—[Ep. ]
See Appendix, No, II., p. 349.—[Ep.] ° See Appendix, No. V., p.351.—| Ep. |
Y¥ 2

324 RHINOCEROS.

With the exception of two of the skulls, all the specimens
here enumerated are the products of Colonel Wood’s zealous
and meritorious researches in the Gower Caves. On the
present occasion, I shall confine myself to the description of
such specimens only as are essential to establishing the spe-
cific distinctness of the form.

Characters of the Molar Teeth.—The crowns of the upper
molar teeth in Rhinoceros present a common pattern of great
complexity, but subject to modifications in the different
species that are very constant, thus furnishing good cha-
racters for distinguishing them. Cuvier gave such a clear
and complete analysis of the elements that enter into the
composition of the crown, and was so happy and simple in
the terms by which he designated them, that little was left to
his successors besides the application of these terms to the
new forms discovered after his time. De Christol followed
up and extended the observations of Cuvier with much
ability, in his Essay on the European Fossil Species, and
succeeded more especially in tracing the peculiarities of cha-
racter produced by the attachment of the distal end of the
‘ crochet’ to the contiguous parts, or by its remaining free.
The other points of principal importance to be regarded are
the number of fossettes on the worn triturating surface; the
presence or absence of an internal basal bourrelet to the
three last premolars; the form of the hind barrel of the last
true molar in respect of its being either simple and undivided,
as in most of the species fossil or recent, or divided by a
posterior figure or fossette, which is so distinctive a cha-
racter of Rhinoceros simus, among the living, and of Rhinoceros
tichorhinus among the extinct forms; and lastly, the relative
thickness of the coat of cement, a character the value of
which in the species of Rhinoceros has, in some measure,
been hitherto overlooked.

Fig. 1 of Pl. XVI. represents a fine fragment of the
upper jaw, right side, belonging to the collection of Colonel
Wood. It contains the five last molars in perfect preserva-
tion; 7.e. the penultimate and last premolars, with the three
true molars. The antepenultimate premolar (p.m. 2) has
been appended in outline, from a reversed figure of the tooth
on the opposite side of the same individual. The age and
relative stage of wear in the different teeth are such as to
present the characters in the most favourable manner. The
antepenultimate true molar (m. 1) is so far advanced in
wear, that the posterior fossette is reduced to a small oval
pit; on the penultimate (m. 2) the detrition is so little
advanced, that the same valley is not yet isolated, and the
peculiar form of the ‘ crochet,’ which constitutes one of the

DESCRIPTION OF PLATE XVI.
RHINOCEROS HEMIT@CHUS.

The figures in this Plate represent molars of Rhinoceros
hemiteechus found in ‘ Minchin Hole,’ Gower, formerly in the
collection of Colonel Wood of Stout Hall, but now in the
British Museum.'! The figures are two-thirds of the natural
size, and have been copied from drawings executed from the
original specimens by Mr. Dinkel for Dr. Falconer. They
are fully described at page 524, et seq.

Fig. 1. Shows the five last molars, ze. the penultimate and last pre-
molars, and the three true molars, of the upper jaw, right side.
The antepenultimate premolar (p.m. 2) has been appended in
outline from a reversed figure of the tooth on the opposite side
of the same individual. The crochet (@) is well seen in the
penultimate true molar. The specimen is fully described at
page 324, et seq.

Fig. 2. Shows the four last molars of the upper jaw, right side, but
considerably more worn than those shown in fig. 1. A descrip-
tion of these teeth will be found at pages 325 & 329. At
page 325, ‘left side’ has been misprinted for ‘ right side.’

Fig. 3. Represents a detached penultimate molar of the left side, being
the counterpart from the opposite side of the tooth (m. 2)
represented in fig. 1. The characteristic thick massive crochet
(a), forming an acute angle with the anterior margin of the
posterior barrel, is well shown. A description of the tooth will
be found at pages 3825 & 329.

' As these pages are passing through the Press, Col. Wood has presented to the
British Museum these and other specimens of Rhinoceros remains found in the
Gower Caves, which are described or figured in this work.

VOL. Il.

Vol. Il. Plate 16.

JDinkel del et idk W.West amp.

Rhimoceros hemitcechus.

RHINOCEROS HEMITCCHUS. 325

distinctive marks of the species, is well shown; while the
apex of the crown of the last molar had only come slightly
into use. The premolars are well worn, and in the normal
ratio, to the state of the antepenultimate true molar (m. 1).
All the teeth are invested by a very thick coat of cement,
which is denuded from the upper part of the anterior barrel
of the last molar (m. 3). The specimen was yielded by the
last excavations in ‘ Minchin Hole.’

Fig. 2 of Pl. XVI. represents another fine fragment in
Colonel Wood’s collection, composing the four last molars,
also of the left side of the upper jaw, but considerably more
worn, the crowns of the last premolar (p.m. 4) and of the
antepenultimate true molar (m. 1) being ground down to a
uniform surface, each enclosing two fossettes; while the last
true molar (m.3) shows the various folds of enamel, and the
form of the ‘ crochet’ in the stage of abrasion best suited for
exhibiting the characters. All the teeth in this specimen
also are enveloped by a thick coat of cement. It was yielded
by ‘ Minchin Hole.’

Figs. land 2 of Pl. XVII. represent a fragment of the right
side of the upper jaw, containing three consecutive teeth,
namely, the last premolar mutilated at the outer surface, and
the antepenultimate and penultimate true molars, the latter
having the inner side of the posterior barrel fractured. The
crowns are in a less advanced stage of wear than in the two
preceding specimens, and the last premolar presents a modi-
fication in the disposition of the fossettes, to be noticed in
the sequel. ‘The specimen belongs to the Swansea Museum,
and was discovered by Colonel Wood in ‘ Bacon Hole.’ The
enamel in all the teeth is invested with a very thick layer of
cement.'

Figs. 3, 4, and 5 of Pl. XVII. represent different views of
a detached germ of the last true molar, upper jaw, left side,
which has not yet come into use. It is free from any coat
of cement, thus presenting all the folds and depressions of
the enamel-shell in a perfect manner.

Fig. 3 of Pl. XVI. represents a detached penultimate
molar of the left side, being the counterpart, from the opposite
side, of the tooth (m. 2) represented in fig. 1 of the same plate.

These specimens are all drawn two-thirds or three-
fourths of the natural size, and taken together they furnish
a complete view of the characters of the upper molars, with

1 The dimensions of this specimen , Width of ditto, in front, 2‘2in. Length
are given in Dr, Faleoner's note-book | of penultimate, outer surface, 2°2 in.
as follows :— Width of ditto, in front, 2°3in. Length

‘Length of three molars, 5-9 in. Length | of last premolar, broken, 1°75 in” —[Ep.]
of last molar, outer surface, 2°9 in.

326 RHINOCEROS.

the exception of the small and deciduous first premolar,
which is rarely seen im situ.

Premolars.—The premolars (Pl. XVI. fig. 1, p.m. 2, 3,
and 4) in R. hemitechus belong to the series indicated by
Cuvier, in which there are only two fossettes produced by
wear on the grinding surface. The antepenultimate (p.m. 2)
presents a nearly square crown; and the median termination
of the transverse valley is reduced to a triangular fissure,
which on the inner side is not quite isolated, the anterior
and posterior divisions not been ground down sufficiently to
efface the intervening cleft. The posterior valley is isolated
and reduced to an elliptical fossette.

The penultimate premolar (p.m. 3), as is usual in the
genus, presents a sudden and very considerable increase of
size beyond the antepenultimate. The inequalities of the
crown are worn down to a common plane, the middle of
which is occupied by a large and irregular fissure, being the
isolated termination of the middle valley; and a round
fossette indicates the remains of the posterior valley. The
hinder boundary of the middle fissure forms a flexuous edge
composed of two projecting rounded lobes, being the remains
of a bifid ‘ crochet.’ Several small tubercles are seen rising
up from the bottom of the fissure.

The last premolar (p.m. 4) is presented in three different
stages of wear by the different specimens. In the ‘ Bacon
Hole’ fragment (Pl. XVII. figs. 1 and 2), the abrasion of the
crown (p.m. 4) is so little advanced that the posterior
valley is not yet isolated; the anterior and posterior barrels
are separated by a wide and deep valley, which is nearly
straight and of uniform width. Its posterior boundary is
undulated, but free from any considerable projection directed
from the posterior towards the anterior barrel. A portion of
the termination of the middle valley is already detached,
forming a third fossette. This, however, is an individual
variety, that is not uncommon in either the penultimate or
last premolars of species which have ordinarily but two
fossettes. It occurs occasionally in the premolars of R.
bicornis, and Gervais has figured an instance of the same
kind occurring in a premolar of R. megarhinus. The portion
of the crown corresponding with the outer or longitudinal
ridge is broken off, in this specimen; but the loss does not
interfere with the principal character.

A more advanced stage of wear and a different pattern
are seen in the same tooth (p.m. 4) as presented by fig. 1
of Pl. XVI. The posterior valley forms a large detached
oval fossette. The inner side of the crown is worn so low
that the barrels are almost confluent, and the commencement

DESCRIPTION OF PLATE XVII.
RHINOCEROS HEMIT@HCHUS.

The figures in this Plate represent molars of Rhinoceros
hemitechus found in the Gower Caves by Colonel Wood. The
figures are three-fourths of the natural size, and have been ,
copied from drawings executed from the original specimens
for Dr. Falconer by Mr. Dinkel.

Figs. 1 and 2. Show in plan and profile a fragment of the right side of
the upper jaw, containing the last premolar mutilated at the
outer surface, and the antepenultimate and penultimate true
molars, the latter having the inner side of the posterior barrel
fractured. The crowns are in a less advanced stage of wear
than in the specimens represented in figs. 1 and 2 of Plate XVI.
(See pages 825 & 382.)

Figs. 3,4, and 5. Represent three different views of a detached germ of
the last true molar, upper jaw, left side, which had not yet come
into use, and which presents all the folds and depressions of the
enamel-shell in a very perfect manner. Fig. 3, crown surface.
Fig. 4, inner surface. Fig. 5, outer surface. a, anterior colline ;
b, longitudinal colline; c, the continuation of the longitudinal
colline which is the homologue of the posterior transverse
colline; d, the crochet; e, the anterior barrel; /, the anterior
basal bourrelet; g, the posterior barrel; h, a small tubercle at
the posterior inner angle; 7, vertical groove of the anterior outer
angle; k, intercolumnar tubercle. (See pages 325 & 335.)

VOL. Il. c

— =”

Rhinoceros hemtoechus.

RHINOCEROS HEMITCCHUS. 327

‘of the middle valley nearly effaced. The central fissure
forms a very irregular chasm deeply indented by the salient
processes of a bifid ‘crochet’ thrown off in front of the
posterior fossette, and by a thick projecting plate given off
from the middle of the longitudinal outer ridge and con-
verging towards the top of the crochet. If during the
further progress of wear the points were to run together into
a common surface, a third detached fossette would be formed,
exactly as is seen in the preceding specimen, and the anterior
border of the posterior colline would present only a slight
amount of undulation, instead of the numerous salient plates
or denticulations yielded in its present state. These pro-
cesses are less conspicuous in the penultimate premolar
(p.m. 3), in consequence of its more advanced stage of wear,
which has led to their disappearance; but the two lobes of
the bifid ‘ crochet ’ are distinctly discernible in the latter tooth.

A third condition of the last premolar is furnished by the
anterior tooth (p.m. 4) of Pl. XVI. fig. 2. Here the abra-
sion of the crown has proceeded so far that the transverse
valley is reduced to a diagonal excavation, oblong in form,
with rounded ends and parallel sides. The enamel boundary
of this fossette is perfectly smooth and equal, the projecting
processes of the bifid ‘ crochet’ having entirely disappeared ;
and the posterior valley is reduced to a small round pit. On
the inner side the waste of the crown by grinding has gone
so far that no indication remains of its having been origin-
ally composed of two distinct barrels.

These three examples furnish an instructive series of
illustrations of the very different patterns which may be
presented in this species by the same tooth in different
stages of abrasion. In each case the tooth is fortunately
in place in the jaw in connection with other molars, which
determine its rank and numerical position with certainty.
Had they been found detached it would have been but con-
jectural to identify p.m. 4 of fig. 2 (Pl. XVI.) with the com-
plex crown of p.m. 4 in fig. 1 (Pl. XVI).

IT have seen other detached premolars of R. hemitcechus
from various localities, all presenting the same characters,
that is to say, the hind barrel projecting into the central
fossette a bifid ‘crochet,’ and an accessory parallel plate
emitted from the middle of the outer or longitudinal ridge,
forming together three ‘combing plates’ of a complex
pattern, as in p.m. 4 of fig. 1 (Pl XVI). Two specimens of
this nature, from the cavernous fissure of Durdham Down,
are preserved in the Bristol Museum. They are contiguous
premolars of the upper jaw, left side.’

1 See Appendix, No. II., p. 350.—[En.]

328 RHINOCEROS.

Another character of much importance, as a specific dis-
tinction, is that the premolars of the Rhinoceros of the Gower
Caves are constantly devoid of an internal basal bourrelet.

Having regard to the various points above indicated, the
premolars of R. hemitechus may be characterized :—

1. By the absence of an internal basal bourrelet.

2. By there being two fossettes only to the worn crowns.

3. By the middle valley being traversed by the processes
of a bifid crochet emitted from the posterior barrel,
and by a parallel combing plate given off by the
outer or longitudinal ridge.

4, By being invested, like the true molars, with a very
thick coat of cement.

The presence of only two fossettes instead of three at
once distinguishes these premolars from those of Rhin.
tichorhinus, while the absence of a basal bourrelet, besides
other characters, distinguishes them from Rhin. megarhinus
and Rhin. leptorhinus. Among existing species, Rhin. bicornis
resembles the Gower fossil form in the bifid crochet and
combing plate which project into the ‘cul de sac’ of the
middle valley; but it differs materially in the strongly
developed crenated bourrelet, which encireles the inner side
of the premolars.

De Christol has figured two varieties of the last upper
premolar in Rhin. megarhinus, in one of which (fig. 10 of
Pl. III.) there is a very pronounced basal bourrelet, while
the other (fig. 4 of Pl. III.) is entirely free from it (Pl. XVIII.
figs. 1 and 2 of this work).! The teeth correspond so exactly
in every minute detail of pattern in other respects, that it is
impossible to doubt that they are of the same species. The
tooth, fig. 4, agrees also with the last premolar of Rhin.
hemiteechus (p.m. 4 of fig. 1, Pl. XVI.), in the absence of a
bourrelet, in the ‘crochet’ being bifid, and in emitting a
single combing plate from the outer ridge. But on insti-
tuting a minute comparison, the following points of dif-
ference are discernible. In the premolars of Rhin. megar-
hinus and also of Rhin. bicornis the ‘combing plate’ (R. of
figs. 10 and 4 of De Christol) is emitted in a line with the
anterior outer angle, and converges diagonally to meet the
plane of the crochet (T.) nearly at a right angle; and the

1 Annales des Sciences Nat. 2me Sér. | and fifteen in Pl. III. The latter
tom. iv. Zool. In Dr. F’s MSS. the | figures, moreover, correspond with copies
figures cited from De Christol are ‘figs. | which Dr. F. had made from those
25 and 19,’ which correspond to figs. 10 | numbered 25 and 19. These figures
and 4 in Pl. III. of the memoir in the | have been reproduced in figures 1 and 2
‘Annales des Sciences, there being | of Pl. XVIII. of this work.—[Eb. |
seven figures in Pl. L, eight in Pl. IL.,

RHINOCEROS ILEMITCCHUS. 329

processes of the bifid crochet do not project much into the
valley. Onthe other hand, in Rhin. hemitechus, the ‘ comb-
ing plate’ (p.m. 4 of fig. 1 of Pl. XVI.) is given off from the
middle of the longitudinal or outer ridge, and is directed
forwards nearly parallel to the upper lobe of the crochet ; and
both these processes jut more into the valley and are more
massive. These alleged points of difference may be regarded as
minute and fine-drawn, but they have appeared to me to be
constant, and to run through the whole series of the molars.

T do not consider it necessary, on the present occasion, to
extend the comparison of the premolars of the Gower Rhino-
ceros with those of other species.

True Molars.—The distinctive characters of the teeth in
this species are still more pronounced in the upper true
molars. Fig. 3 of Pl. XVI. represents a detached penul-
timate of the left side, in the most favourable stage of wear
to show the characters. The posterior valley, not long
isolated, forms an irregular triangular pit with sloping walls.
The transverse valley at its commencement also forms a
triangular fissure with the apex pointing to the sinus between
the posterior barrel and the crochet; the valley next bends
forward in a sigmoid curve, and is very much contracted by
the advance of the crochet towards the anterior barrel; and
it then expands into a rounded cul de sac, the extremity of
which points backwards. During the progress of wear the
two valleys never form more than two fossettes, in the
manner exhibited by m. 1 of fig. 2 (Pl. XVI.), which presents
the antepenultimate or first true molar in a very advanced
stage of abrasion. This character, as in the case of the pre-
molars, at once distinguishes the molars of Rhin. hemitechus
when found detached from those of Rhin. tichorhinus.

But the character which best distinguishes them from all
other species lies in the peculiar form of the ‘crochet,’ or
promontory projected forward from the posterior colline into
the transverse valley. In all the species, fossil or recent, ex-
cepting Rhin. hemitechus, the crochet forms a plate which is
emitted at a very open angle with the posterior colline, and
directed more or less diagonally towards the anterior outer
corner of the crown. This is well seen in the figures given
by Cuvier in the ‘ Ossemens Fossiles.’* Pl. V. figs. 1, B. C.,
and Pl. Il. fig. 3, B. of that work exhibit the character
in the unicorned Rhinoceros of Java, where the margin of
the crochet is continued nearly in a straight line with the
anterior margin of the posterior colline. The same is seen
in the penultimate B. of fig. 1, Pl. XVIII., representing the

1 Fed, 3me. tom. ii. Rhinoceros.—[E.]

330 RHINOCEROS.

adult dentition of the two-horned Rhinoceros of the Cape.
For the other existing species, the beautiful figures given
by De Blainville in the ‘ Ostéographie * may be referred to
generally in illustration of the same character. In Rhin.
tichorhinus the crochet is given off at a very open angle, and
is united with the ‘ combing plate’ of the outer ridge, so as to
form the third fossette; the same occurs in the molars of
Rhin. simus, which in their general plan bear a close affinity
to those of Rhin. tichorhinus. In regard to the other fossil
species, there are but few specimens figured in the ‘ Ossemens
Fossiles’ that can be referred to in illustration. The molar
from Chagny (Département du Sadne et Loire), Pl. VI. fig. 6,
cited by Kaup, as an illustration of his Rhin. Merckii, is far
advanced in wear, but what remains of the ‘ crochet’ exhibits
the same very open angle in its offset from the posterior
barrel. Of the two molars from Crozes (Départ. du Gard),
also cited by Kaup as of Rhin. Merckii, and adduced by Pro-
fessor Owen as identical with his Rhin. leptorhinus of Clacton,
the specimen fig. 5 of Pl. XIII. is ground down so low
that the crochet has nearly disappeared, and it is therefore
hardly a suitable case for comparison ; but if it is compared
with m. 2 of fig. 2, Pl. XVI. of the accompanying illus-
trations, being a penultimate of Rhin. hemitechus which is
nearly in a corresponding state of abrasion, it is manifest
that in the former the curve of the crochet forms a much
less abrupt flexure than in the latter. The second Crozes
specimen (Oss. Fossiles, Rhin., Pl. XIII. fig. 4) is an abnormal
case, the nature of which has been clearly explained by
De Christol, in which the crochet is so produced as to be
concrete with the middle of the anterior colline, thus leading
to the early isolation of a third fossette, in a manner different
from what occurs, as an ordinary condition, either in the true
molars of Rhin. tichorhinus or in any other known species.
But although so little worn, that the posterior valley is not
yet isolated into a fossette, if the figure given by Cuvier
is compared with fig. 3 of Pl. XVI. of the accompanying
illustrations, it will be seen that the anterior edge of the
posterior colline does not form an acute angle and a re-
entering niche with the base of the crochet.

Of the European fossil forms from the Pliocene and more
recent deposits, Rhin. megarhinus is that of which the den-
tition is best known, after Rhin. tichorhinus. The excellent
descriptions and figures supplied first by De Christol, and
afterwards by Gervais, leave little to be desired in regard to
the cranial and dental characters of this species. In fig. 5,
Pl. III. of his memoir (reproduced in Pl. XVIII. fig. 3),
De Christol has given a fine illustration of the natural size

RHINOCEROS HEMITCICHUS. 331

of a penultimate upper molar, of which the crown is but
slightly abraded, and the ‘crochet’ well developed. In this
case, also, the crochet forms, at its offset, a very open angle
with the disc of the posterior colline. In fact, it is continued
in nearly the same line of diagonal as the latter, and points
to the anterior outer corner of the crown. Gervais (Paléon-
tologie Frangais, Pl. I. fig. 5) has given a beautiful illus-
tration of an upper molar (penultimate or antepenultimate)
of the same species, yielding precisely the same characters
(reproduced in Pl. XVIII. fig. 4); and I have, through the
kindness of M. Gervais, had an opportunity of examining a
considerable number of molars of the same species in the
Museum of the Faculty of Sciences at Montpellier, which
presented a constant agreement in the offset of the crochet
from the posterior colline, at a small inclination only.

Tf on the other hand the penultimate true molar in Rhin.
hemitechus (Pl. XVI. fig. 1, m. 2, and fig. 3) be examined,
the crochet (a) presents a thick massive body thrown straight
forward, and forming an acute angle with the anterior
margin of the posterior barrel. It is flat or concave above,
and convex below; narrow at the base, and thickening to a
blunt margin. In mass it bears a much larger proportion to
the dise of the hind barrel than in most of the other species.
The distal extremity is closely approximated to the anterior
barrel, but always remains detached, undivided, and free
from the hooked inflection, so common in the other species,
which suggested the name applied to this body by Cuvier.
The pattern presented by the stage of abrasion seen in fig. 3,
Pl. XVI., may be compared to a boot of which the disc of the
hind barrel forms the leg, and that of the ‘ crochet’ the foot.
In the corresponding molars of Rhin. megarhinus already
cited, namely fig. 5 of Pl. III. of De Christol’s memoir (re-
produced in Pl. XVIII. fig. 3), and fig. 5 of Pl. IT. of Gervais’
Paléontologie (reproduced in Pl. XVIII. fig. 4), besides the
difference of alignment in its offset from the hind barrel, the
section of the crochet is wedge-shaped, thinning from a broad
base to a sharp edge.

In the antepenultimate true molar, m.1 of fig. 1 of PI.
XVI., the same general characters are presented, but modified
by the greater age and more advanced abrasion of the crown.
The posterior valley is reduced to an oval pit. The discs of
the anterior and posterior barrels occupy much larger areas ;
the crochet being ground low down is greatly diminished in
projection, but it still forms a right angle with the anterior
edge of the posterior barrel. The cul de sac of the middle
valley is reduced in size, and a ‘combing plate’ or fold of
enamel from the outer longitudinal ridge juts into it, directed

332 RHINOCEROS.

forwards and parallel to the crochet, corresponding with
what was described above of the same process in the last
premolars. The origin and connection of this ‘combing
plate’ are explained by the mammillary processes seen above
the ‘ crochet’ in the terminal expansion of the transverse
valley in m. 2 of fig. 1 (Pl. XVI.). These denticuli are con-
nected with the bottom of the fissure and with its outer wall.
It is obvious that if the abrasion of the crown were carried a
little further they would run together into a continuous
plate, which would project into the valley parallel to the
crochet, reproducing the pattern seen in p.m. 4 of the
same figure, and in the last true molar, m. 3 of fig. 2
(Pl. XVI.). When this occurs a very complex pattern is
the result. Cuvier has figured no examples, but in the
additions to Vol. iii. of the ‘Oss. Fossiles,’ he refers to some
teeth procured by Mr. Pentland in Tuscany, ‘dont la colline
postérieure, au lieu d’un seul crochet, en donne plusieurs
petits en avant; ce que fait paraitre cette colline dentelée
vers sa base quand elle commence a s’user.’ He adds, ‘ ce
caractére pourra servir 4 reconnaitre cette espéce (referring
to Rhin. leptorhinus) par ces molaires.’ Professor Owen had
his attention directed to the same peculiarity in a fossil
which he describes ‘as the germ of the antepenultimate
molar of a Rhin. leptorhinus from Grays, in Essex, in which
many smaller processes are sent off into the principal valley,
in addition to the large promontory,’ but he was not disposed
to place much stress upon this as a specific character. In
Rhin. megarhinus, these ‘ combing plates’ are not directed
forwards, but converge from the anterior outer angle towards
the crochet. I have lately ascertained, by the examination
of the cast of a cranium with teeth contained in the Museum
at Pisa, that Rhin. hemiteechus occurred in the Fauna of the
Val d’Arno,' and the teeth so briefly yet pointedly noticed by
Cuvier in the passage cited above in all probability belonged
to this species. In the penultimate (m. 2) of the ‘ Bacon
Hole’ specimen (Pl. XVII. fig. 1), although not much ad-
vanced in wear, the denticuli of the ‘combing plate’ have
run together and it is projected forwards parallel to the
‘crochet,’ thus confirming the constancy, of the character.
The penultimate and antepenultimate upper true molars
differ so little from each other, except in dimensions and some
trivial details of proportion, that it is unnecessary to describe

1 This cranium of a Rhinoceros, with | jaw of Rhin. hemitechus in the Pisa
a partial bony septum, was subsequently | Museum. The existence of the latter
determined by Dr. Falconer to belong to | species in Italy is also mentioned in his
Rhin. Etruscus (p. 859). Mention, how- | letter to M. Lartet, already referred to,
ever, is made in his note-books of a lower | p. 309.—[ Ep.]

RHINOCEROS HEMITC@CHUS. 333

them separately. A very advanced stage of abrasion is pre-
sented by the antepenultimate or m. 1 of fig. 2 of Pl. XVI.
The posterior valley is reduced to a small pit, and the large
sinuous transverse valley to a diagonal fossette, from the pos-
terior wall of which every trace of a crochet or of a combing
process has disappeared. The penultimate (m. 2) of the same
figure, although less worn, has lost the greater part of the
mass of the crochet by the waste of abrasion, and the middle
valley, in consequence, forms a fissure of nearly uniform width,
much reduced in expansion at its extremity.

Next, in regard to the last true molar. Of all the grinding
teeth in the genus Rhinoceros, the last true molar of the
upper jaw is that which presents the greatest difference of
form and the most pronounced characters for distinguishing
the species. Fortunately we possess, in the series of the
Gower specimens, a complete set of illustrations, showing
this tooth in every stage, from that of the intact germ up to
the worn crown of the aged animal; and the modifications of
form which it presents are so peculiar, and of so much syste-
matic interest when considered in connection with the partial
bony septum, that I shall not hesitate to enter into more
detail in describing it than in the case of the penultimate and
antepenultimate. Thisis the more necessary, as De Christol,
the most original and weighty authority on the subject since
the time of Cuvier, has omitted the last molar in his elaborate
analysis, under the belief that it yielded no specific characters
of importance.'! In order to make the description clear, it is
requisite to refer to the general composition of the crown of
a true molar in Rhinoceros, as indicated by Cuvier. Taking
the penultimate as the type, the crown is nearly rectangular
in outline and bounded by four sub-equal sides; the outer
and inner, and the anterior and posterior, forming parallel
sides of the square. The outer side (a b of the teeth B and C
of fig. 1 Pl. XLITI.,* Cuvier’s Oss. Fossiles (supports a longi-
tudinal ridge or colline, from either extremity of which a
transverse flexuous ridge is given off at aright angle, forming
(a c) an anterior colline, and (b e) a posterior colline, parallel
to each other, but separated by a sinuous transverse valley.
The terminations or barrels of these collines constitute the
inner side of the square. The anterior side forms a straight
unbroken line, and in all the species presents nearly the same
uniform character, except in the greater or less amount of
development of its basal bourrelet. The posterior side is the
most subject to modification. It is shorter than the anterior,
deeply notched by an antero-posterior fissure, generally

1 De Christol, op. cit. p. 47.
? Corresponds to Pl. y. of Rhin. in vol. ii. of 3rd ed. 1825.—[Ep.]

334 RHINOCEROS.

triangular in form, separating the inner hind barrel from the
posterior termination of the outer ridge. This fissure forms
the posterior valley. All the species of Rhinoceros hitherto
described may be ranged under two heads: 1. Those in which
the last true molar has a posterior valley; 2. Those in
which it is wanting. To the former series belong Rhin.
tichorhinus and Rhin. simus, which further agree in the
common character of presenting three fossettes to the worn
crown of the last true molar, namely: one fossette, formed by
the posterior fissure ; the second, caused by the confluence of
the crochet with the combing plate intercepting a portion of
the transverse valley; and a third fossette, formed by the
remaining or open portion of the latter valley. To the second
series belong the unicorned and bicorned species of Asia, and
the African Rhin. bicornis, together with the European fossil
species, such as Rhin. megarhinus, Rhin. leptorhinus, Cuvier,
Rhin. Schleiermacheri, &c. They all agree in the common
character of the posterior valley or fossette being wanting,
but are susceptible of being divided into two subordinate
series, namely, those in which the last molar presents two
fossettes ; one formed by the confluence of the crochet with
the ‘combing plate’ intercepting the outer portion of the
transverse valley, the other, composed of its open or inner
portion. This series is exemplified by Rhin wnicornis among
living forms. The second subdivision includes the forms in
which the crochet is free from adhesion to the ‘combing
plate,’ and the crown, during wear, only exhibits a single
fossette, namely, the sinuous fissure of the transverse valley.
To this series belong the unicorned Rhinoceros of Java, Rhin.
bicornis, and the majority of the European fossil forms. The
last true molar may therefore be presented with one fossette,
as in Rhin. megarhinus (videGervais, Paléontol. Francaise, Pl.
Il. figs. 6 and 7); with two fossettes, as in Rhin. wnicornis
(vide Cuvier, Oss. Fossiles, Rhin., Pl. I. fig. 3); or with three
fossettes, as in Rhin. tichorhinus (op. cit. Pl. VI. fig. 4).

The presence or absence of the posterior or third fossette
entails an important difference in the form of the crown of
the last molar. When present (vide the fig. last cited), the
outline of the tooth is still four-sided, although the posterior
side is considerably reduced in width, and the separation of
the hind barrel from the end of the outer colline is distinctly
marked by an intervening fissure. But when the posterior
valley is wanting, the outline of the crown becomes triangular;
the summit of the anterior transverse colline remains as
usual, while the outer colline is directed diagonally inwards
and backwards, so as to make an acute angle with the former.
The result is that the summit of the crown, instead of being
rectangular, is V-shaped, and the posterior transverse colline

eS rtrt— _ ‘OCS

RHINOCEROS HEMIT@CHUS. 335

is confluent with and undistinguishable from the outer colline,
except by the offset of the crochet, and by the round or
barrel-shaped termination at the posterior inner angle. Not
a vestige even remains of the posterior outer angle. In fact,
the hind leg of the V is composed along two-thirds of its
length of the outer colline, the remaining third being made
up of the posterior transverse colline, with no mark of de-
marcation between them. No trace of a depression or groove,
corresponding with the posterior fossette, is left upon the
surface of the enamel. These characters are well shown by
the accompanying figures in Pl. XVIII. fig. 7, representing
the summit of the crown in plan, and fig. 6, the same from
the inner side, in a germ of the last true molar, left upper, of
Rhin. bicornis, drawn two-thirds of the natural size: (a) indi-
cates the anterior colline; (b), the longitudinal colline ; (c),
the continuation of the latter, which is the homologue of
the posterior transverse colline; (d), the crochet; (¢), the
anterior barrel; (f), the anterior basal bourrelet; (g), the
posterior barrel; (i), a small tubercle at the posterior inner
angle ; and (i), the vertical groove of the anterior outer angle.

Let us now examine this tooth as it occurs in Rhin. hemi-
techus. Figs. 3 and 4 of Pl. XVII. represent top and side
views of three-fourths of the natural size of an intact germ of
the left last molar, corresponding with the figs. 7 and 6 of Pl.
XVIII. of Rhin. bicornis ; and fig. 5, Pl. XVIL., gives an erect
view of the outer surface. The same letters of indication
apply to the different parts. The outline of the crown in
plan is triangular, exactly as in the Rhin, bicorms; and the
ridges (a and b) meet at an acute angle, yielding the same
V-shaped pattern, the outer and the posterior ridges (b and c)
being continued in the same line without interruption; the
anterior basal bourrelet (f) repeats the form presented in fig.
7, Pl. XVIIL., but is more salient. The crochet (d) is pro-
jected farther forwards across the valley, and when the erect
figure, Pl. XVII. fig. 4, is compared with Pl. XVIII. fig. 6, it
is apparent that in the former the crochet makes a more
acute angle with the posterior barrel. The niche of the
anterior outer angle (a) is more pronounced, and there is
an intercolumnar tubercle (k) at the mouth of the valley
which is not seen in the African species. This tubercle is
also present in m. 3 of fig. 2, Pl. XVI., and strongly de-
veloped in the detached specimen, fig. 5, Pl. XVIII, but
wanting in m. 3 of fig. 1, Pl. XVI. On the whole there is
a very strong general agreement in form between the last
true molars of Rhin. bicornis and Rhin. hemitechus; the
most obvious difference being the considerably greater di-
mensions of the tooth in the latter.

The transverse valley in Rhin. hemitechus is triangular at

336 RHINOCEROS.

its commencement, as in the penultimate, m. 2 of fig. 1, PI.
XVI., and is then reduced to a narrow cleft by the projection
of the free end of the crochet close to the anterior barrel.
The continuation of the valley, beyond the crochet, forms an
oblong and somewhat angular expansion, rising from the
bottom of which a line of denticular points is seen, con-
nected into a plate attached to the outer colline. This
‘combing plate’ is projected forwards parallel to the
crochet, repeating the pattern already described in p.m. 4,
m. 1, and m. 2 of fig. 1, and in m. 2 of fig. 3, PL XVI. A
similar disposition of these denticular points is exhibited in
m. 3 of fig. 1 of Pl. XVI. When the crown is ground down
by use, the effect is to produce the appearance presented by
m. 3 in fig. 2, Pl. XVI. of a double crochet projected across
the valley, one of the processes representing the ordinary
crochet, and the other the ‘combing plate.’ The constancy
of this character in running through the whole of the molars
proves its importance as a mark of specific distinction.

In some cases, the worn pattern of the middle valley is
still more complex. A fine example of this is presented by
the last true molar of the specimen No. 22,020 of the Pa-
leontological Catalogue, British Museum, purchased of the
late Mr. Ball, and reported to have been procured from the
fluviatile deposits of the Valley of the Thames. The five
posterior molars of the right side are presented in sequence,
the last being in full wear; and in this tooth, besides the
crochet and ‘ parallel combing plate,’ the termination of the
middle valley presents two additional processes ; namely, a
stout plate projected at right angles to the crochet, from the
anterior outer angle, and a short plate emitted from the
anterior colline, above the crochet, and directed backwards.
The valley, in consequence, presents a pattern of extreme
complexity, with plates jutting into it from three sides.!

In Rhin. bicornis, Pl. XVITI. fig. 7, the valley is of a similar
form, but its posterior wall is free from any combing plate,
or tendency to a double crochet. In Rhin. megarhinus the
crochet of the last molar is also single, and emitted at an
open angle from the posterior colline. In illustration, fig.
9 of De Christol’s plate, and figs. 6 and 7 of Pl. IL. of
Gervais’ Paléont. Franc. may be referred to.

An abnormal condition of the crochet in the last molar of

1 The specimen here referred to is | and 310), he includes the Grays Thurrock
described in detail in Dr. Falconer’s Note- | Rhinoceros under R. leptorhinus (RR. me-
book as ‘PR. hemitechus of Grays Thur- | garhinus). From some of his later notes,
rock,’ under date Oct. 1858. Even then | however, it would appear that he identi-
he noted important differences in the | fied R. hemitechus as also occurring in the
crochet, &c. from the Minchin Hole | lower brick-earths of Grays Thurrock.—

Molars ; and in his letters to M. Lartet | [Ep.]
and Col. Wood in 1862 (see pages 309

DESCRIPTION OF PLATE XVIII.

RHINOCEROS LEPTORHINUS (R. MEGARHINUS), RHINOCEROS

HEMIT@CHUS, AND RHINOCEROS BICORNIS.

Figs. 1 and 2. Represent the two varieties of the last upper premolar,

right side, of Rhinoceros leptorhinus (R. megarhinus, Christ.),
referred to at page 328 of the text, in one of which (fig. 1)
there is a very pronounced basal bourrelet, while the other
(fig. 2) is entirely free from it. These two figures have been
copied from the illustrations of De Christol’s memoir in the
Ann. des Sc. Nat. 2me. Sér. tom. vi. Zool. Pl. IIL, figs. 10
and 4. In both, the ‘combing plate’ (R) is seen converging di-
agonally to meet the plane of the crochet (‘T) nearly at a right
angle. The drawings are about two-thirds of the natural size.

. Represents the penultimate true molar, upper jaw, right side,

of Rhinoceros leptorhinus (Rk. megarhinus, Christ.), two-thirds
of the natural size. The section of the crochet is wedge-shaped,
thinning from a broad base to a sharp edge, and it forms a
very open angle with the disc of the posterior colline. This
specimen is referred to at page 331, and the figure has been
copied from Plate III., fig. 3 (not fig. 5, as stated in text), of De
Christol’s memoir above referred to.

. Penultimate upper molar, left side, of R. leptorhinus (R. me-

garhinus), yielding precisely the same characters as fig. 3. It
is about two-thirds of the natural size and has been copied
from the ‘ Paléontologie Frangaise,’ by Gervais, Pl. IL., fig. 5.
It is referred to at page 331.

. Last upper molar of Rhinoceros hemitechus, two-thirds of the

natural size, showing an abnormal condition of the crochet.
The specimen is believed to have been procured from Grays
Thurrock, and is now in the Museum of the College of Surgeons.
Dr. Falconer’s reasons for regarding this tooth as belonging to
R. hemitechus, rather than to R. leptorhinus, will be found at
page 337.

Figs. 6 and7. Two views of a germ of last upper molar, left side, of the

existing species, Rhinoceros bicornis, two-thirds of the natural
size. Fig. 6, inner side. Fig. 7, crown. a, anterior colline ;
b, longitudinal colline; c, continuation of longitudinal colline,
which is the homologue of the posterior transverse colline; d,
the crochet; e, anterior barrel; f, anterior basal bourrelet; g,
posterior barrel; /, small tubercle at posterior inner angle; 2,
the vertical groove of the anterior outer angle. (See page 335.) -
The specimen is now in the British Museum.

VoL. Il,

Vol. IL. Plate 18.
Fig. 2.

delethth 1] 2, 3,4. Rhinoceros leptorhinus (R megarhinus) a
5. Rhinoceros hemitoschus. 6,7 Rhinoceros bicornis.

RHINOCEROS HEMIT@CHUS. 337

Rhin. hemitechus is presented by a specimen in the Museum
of the College of Surgeons, of which the precise origin has
not been recorded, but which is believed to have been procured
from Grays Thurrock, or some other of the fluviatile deposits
in the Valley of the Thames. It is represented two-thirds of
the natural size by fig. 5 of Pl. XVIII. In this case the
crochet forms a wall across the valley, insulating its upper
portion and connecting the two barrels. It is united to the
middle of the anterior colline, and above it, a parallel, short,
stout, ‘combing plate ’ juts into the insulated fossette. The
general form, angular offset of the crochet, enormous coat of
- cement, and details of the characters prove it to be of Rhin.
hemitechus.}

That this peculiar confluence of the crochet with the ante-
rior barrel is abnormal in the true molar is proved by the
extreme rarity of the instances which have been observed of
it in any species of Rhinoceros. Cuvier has figured one (Oss.
Foss. Rhinoc. Pl. XIII. fig. 4), a penultimate, being the
Crozes specimen already referred to (supra, p. 330). I have
examined, in the Museum of the Faculty of Sciences of Mont-
pellier, other specimens from the Département du Gard, which
agreed with the figure of this specimen in every essential
respect except the irregular connections of the crochet, and
they appeared to me all to belong to the Rhin. megarhinus
of Montpellier. If the form of the crochet, its offset, and the
acute angle which it makes with the posterior colline in
m. 2 of fig. 2, Plate XVI., are compared with the same points
in the Crozes specimen, the differences are very obvious.
No other instance of a bridge-crochet in a true molar has, so
far as I am aware, been figured. In the milk molar it is by no
means of rare occurrence, and is often seen in those of Rhin.
bicorms. This appearance must not be confounded with the
cohesion between the crochet and the ‘ combing plate,’ which
gives rise to the third fossette so characteristic of Rhin.
tichorhinus.

The most significant peculiarity in the last true molar of
Rhin. hemiteechus remains to be described. From the marked
triangular outline of the crown in plan, and the V-shaped
confluence of the terminal ridges, it might have been ex-
pected that the posterior fossette would be entirely suppressed,
as in Rhin. bicornis and other species in the same category.
But at the posterior angle of the hind barrel, and dislocated
from its ordinary position in the other true molars, a well-
defined fossette is placed close to the base of the crown. It is
of a triangular form, with a gaping rim, which is deeply

* On the drawing of this specimon Dr. F. has written ‘22. megarhinus?’—[Ep.]
VOL. I. Z

338 RHINOCEROS.

emarginated behind, and it repeats, but with reduced dimen-
sions, the usual posterior fossette of the penultimate m. 2 of
Fig. 1, Plate XVI. The form of this fossetie is exhibited by
figs. 3 and 4, h, and its relation to the other parts by the ex-
ternal view, fig. 5, h, of Plate XVII., where a shallow and in-
distinctly defined channel, bounded on either side by a ridge,
is continued upwards upon the enamel-surface from the basal
fossette to the apex of the crown, but becoming more and
more indistinct as it ascends. This channel is the homologue
of the posterior fissure (Oss. Foss., Rhin., Plate VI. fig. 4) in
the last molar of Rhin. tichorhinus and Rhin. simus. On the
opposite side of the same figure (fig. 5, 7) a vertical groove is
seen descending from the anterior outer notch. In the last
molar of Rhin. bicornis (Plate XVIII. fig. 7), the small
tubercle (h) is the abortive representation of the rim of the
posterior fissure of fig. 3, h, of Plate XVII.

In consequence of the basal position of the cup of the
posterior fissure in the last molar of Rhin. hemitechus, the
abrasion of the crown cannot reach it so as to form an
insulated fossette till the last stage of use, and ordinarily it is
enwrapped by the very thick layer of cement, so as to be
only indicated by a protuberant gibbosity, as is seen in m. 3
of fig. 2, Plate XVI., and less distinctly in m. 3 of fig. 1 of
the same Plate. The channel, which is continued upwards
from the cup, remains usually inconspicuous.

The last true molar, therefore, in the Gower species exhi-
bits the remarkable combination of the following characters:
namely, a triangular crown with a V-shaped summit, and
two fossettes; one corresponding to the middle valley, the
other to the posterior fissure; the posterior barrel narrow
and compressed, and giving off a double crochet. In its sys-
tematic relations it occupies an intermediate position between
Rhin. bicornis and Rhin. tichorhinus.

In the description of all the molars, reference has been
made to the thick layer of cement. This dental constituent
is present in greater or less quantity on the teeth of all the
species of Rhinoceros. But in Rhian. hemitechus the mass of
the layer is so great as to become a character of specific
importance. The proportion which it bears to the shell of
enamel is best seen on the anterior barrel of m. 2 of fig. 1,
Plate XVII. It is there partly denuded, and the enamel
looks as if set in a casing of cement. In Fig. 2, Plate XVL.,
all the molars are completely enveloped by an enormous coat
of cement, through which the edging of enamel protrudes.
It is also most abundant in all the molars of Fig. 1, Plate
XVI. In the last true molar of Fig. 2, Plate XVI., the
cement is seen to form a thick layer, insinuated between the

RHINOCEROS HEMITGICHUS. 339

plates of the double crochet and lining the walls of the
valley. I have ascertained that it is equally abundant in
all the molars of this species from the caves of Oreston
and Durdham Down, and from the fluviatile deposits of
Clacton, and other similar localities. In the teeth of ex-
isting species, such as Rhin. bicornis and Rhin. simus, the
coat of cement cracks and dislaminates, by long exposure to
the weather. This accident will account for its absence in
certain teeth of Rhin. hemiteechus, in which the cement had
probably disappeared from weathering before they were em-
bedded in the matrix. When the matrix is a calcareous paste,
the layer of cement is apt to be detached from the enamel
along with it, as appears to have happened to the external
surface of the molars in the Bacon Hole specimen, figs. 1
and 2, Plate XVII. The shell of enamel is very much thinner,
in proportion to the other dental elements in this species,
than in Rhin. tichorhinus. In the latter the external surface
is very rugous, while in the former it is comparatively
smooth. The difference is so considerable that in many
instances the teeth of the two species can be distinguished
by this character alone.

De Christol has directed attention to the fact, that in
genera of the same families, the older forms have a less
coating of cement on their teeth than the newer types. Thus,
in Hipparion, the layer is much thinner in proportion than
in species of the genus Hquus, and in Aceratheriwm than in
Rhinoceros. The same difference apples to the Miocene
species of Rhinoceros as compared with the modern forms.
He has ingeniously attempted to give a general expression
to the observation, designating the older forms Acemento-
dontes, and the newer Cementodontes. Without accepting the
generalization as universally applicable, it is worthy of
remark that cement abounds on the teeth of Rhin. tichorhinus
and Rhin. simus, 4nd in the extinct form Rhin. hemitechus,
while it is comparatively scanty in the teeth of Rhin. mega-
rhinus and in specimens attributable to Rhin. leptorhinus.

Inferior Molars.—The molars of the lower jaw, in all the
species of Rhinoceros, present fewer and less appreciable mo-
difications of the general form than the upper ; and they are
in consequence of less avail in the distinction of the species.
For this reason, they would have been described, on the present
occasion, with much more briefness than the upper, but for
the fact that the materials for instituting a comparison be-
tween R. leptorhinus and R. hemitechus ave much more abund-
ant, in the shape of lower jaws and teeth, than of upper.
Cuvier, having omitted to pay sufficient attention to the cha-
racter of the upper molars in R. leptorhinus, during his journey

z2

340 RHINOCEROS.

in Italy, left it as a behest to the naturalists of that country,
to supply the deficiency. But nothing adequate to the de-
mands of the subject has as yet been accomplished by them,
and there is hardly extant a single good figure or description
of an authentic upper molar of that form, to serve as a
standard of comparison; while of lower jaws, besides the
figures in the ‘ Ossemens Fossiles,’ there exist in the Palzeon-
tological series of the British Museum several fragments
containing teeth, from the Val d’Arno, furnishing the desired
means in so far as the mandible is concerned.

Figs. 1 and 2 of Pl. XIX. represent the greater portion of
the horizontal ramus of the lower jaw, left side, of R. hemi-
techus, containing the full series usually seen in the adult, of
six molars. The crowns are in the stage of wear best suited
to show all the characters, the last true molar, although
abraded, having the divisions distinct. The specimen be-
longs to the Swansea Museum. Figs. 1 and 2 of Pl. XX.
represent a fragment of a left ramus of equal extent, showing
the five last molars in situ, and the empty alveolus of the
antepenultimate premolar. The wear of the crowns had
advanced so far in this specimen, that the four anterior
teeth are ground down each to a uniform disc of ivory. Both
specimens are from Minchin Hole, and belong to the collec-
tion of Colonel Wood.

Fig. 1 of Pl. XXI. represents a mutilated right ramus of
the lower jaw, exhibiting also the six posterior molars én situ,
together with a portion of the symphysial expansion. The
specimen is remarkable, in showing the abnormal condition
of two collateral teeth, for the last premolar. The crowns
are seen in the early stage of abrasion of the adult animal.
The specimen, discovered in ‘ Bacon Hole,’ was presented to
the Swansea Museum by Colonel Wood. Its dimensions are :—

Extreme length of 6 molars (very nearly), 10-0 in. Ditto of last three molars (to
base), 5°97 in. Ditto of 3 anterior ditto, 4-0 in. Ditto of summit of crown last molar,
inner side, from edge to edge of enamel, 1°8in. Ditto of crown near base, 2°18 in.
Ditto of penultimate crown, 1°95 in. Width of ditto behind, 1:25 in. Width of
ditto in front, 1-in. Extreme width of last molar, 1:3 in. Extreme length of
fragment, 12°5 in. Height of jaw inside, behind last molar, 4: in. Ditto in front
of last premolar, 3°1 in. Length of summit of antepenultimate true molar (first),
1°7in. Width of ditto behind, 1-2 in.

The first character that strikes the eye in the teeth of all
the three specimens is the very thick layer of cement. In —
Pl. XX. the last true molar is completely encased in it;
while the other teeth are more or less denuded, they show
by the fractured edging that this has arisen from accident.
The same appearance is presented by the molars of Pl. XTX.,
which are still more bared. The layer of cement, therefore,

DESCRIPTION OF PLATE XIX.
RHINOCEROS HEMIT@CHUS.

Figs. 1 and 2. Represent in plan and profile the greater portion of the
horizontal ramus of the lower jaw, left side, containing the full
series of six molars usually seen in the adult, in the stage ot
wear best suited to show all the characters. The specimen was
found in Minchin Hole by Colonel Wood. The figures are one-
half of the natural size, and have been copied from drawings
of the original specimens executed for Dr. Falconer by Mr.
Dinkel. (See page 340, et seq.)

VOL, I.

SNYOSo TUS sorecouryyy

THIL 99 [8p TexunT-

OL ETT TE TON.

Ad 2) RAs OT Ue th” Pes “ er -.
SAO id aint bh ' oe *
: in i ‘ Ly,
es me ‘
4 :
: j u
f y
‘
S
|
/
By \
a
f
.
4 i {
¥ — Uae
_. it “
. —
i ; ’ » s
S Tie :
: r ; vh
t ‘
, r ry i i { \
rl t , J y,
ee)
on ,
r _
‘
i : Ae tee
. 4 a
.*
i
fp Src ae
‘

DESCRIPTION OF PLATE XX.
RHINOCEROS HEMIT@CHUS.

Figs. 1 and 2. Represent in plan and profile the greater portion of the
horizontal ramus of the lower jaw, left side, showing the five
last molars in situ, and the empty alveolus of the antepenul-
timate premolar. The teeth are much further advanced in wear
than in the specimen shown in Plate XIX. The specimen was
found in ‘ Minchin Hole’ by Colonel Wood, and the figures, one-
half of the natural size, have been copied from drawings

executed for Dr. Falconer by Mr. Dinkel. (See page 340,
et seq.)

VOL. I.

- Eat 480M," jh STYO ad JIULEY soreoouryy

eee

UHL 328 Pp Pye

06 Sd TT TOA

RHINOCEROS ILEMITGiCHUS. 341

is present alike in the upper and lower molars in very great
thickness.

Another character, equally obvious, is the very consider-
able amount of concavity in the common grinding surface of
the teeth, in the antero-posterior direction, from the ante-
penultimate premolar to the last true molar. This concavity
is much more pronounced than in the jaw of either R. tichor-
hinus or R. megarhinus, with which I have compared it; and
that itis constant in R. hemitechus is proved by its uniformity
in the three jaws having teeth in different stages of wear.

Premolars.—The premolars agree very closely in form with
those of R. megarhinus, the principal difference being in the
proportion which their aggregate length bears to that of the
true molars. The antepenultimate (p.m. 2 of Pl. XIX.), in
horizontal section, is somewhat wedge-shaped, contracting
from behind forwards to a narrow edge, which is bent in-
wards. Its outer surface shows the vertical groove of divi-
sion between the two crescents, and on the inner side behind
there is a well-marked niche indicating the concavity of the
posterior crescent. In R. megarhinus, the antepenultimate
in the same stage of wear is free from any corresponding
indentation. The anterior edge in the Gower specimen
forms a convex projection. The tooth agrees in the closest
manner with the Clacton tooth figured in the ‘ British Fossil
Mammalia,’ Cut 136, p. 363, and there referred to R. leptor-
hinus. In the ‘Bacon Hole’ specimen (fig. 1, Pl. XXI.)
the antepenultimate premolar repeats the form presented by
p-m. 2 of Pl. XIX.

The penultimate (p.m. 3 of Pl. XTX.) has the crown ground
down to a common sinuous disc. The indentation between
the two crescents forms on the outer surface a deep niche
directed forwards. The remains of the hollows of the crescents
on the inner side show that they were deep and boldly de-
fined. The crown of this tooth in R. hemitechus is consider-
ably smaller, both in the actual dimensions and relatively to
the last premolar, than in R. megarhinus (vide Gervais,
‘ Paléontologie Frangaise’ (Pl. IT. fig. 8).

The last premolar (p.m. 4 Pl. XIX.) presents an oblong
crown, with two boldly pronounced crescents, which are
nearly of equal size. It is also very much larger in all its
proportions than the tooth which precedes it. Compared
with the corresponding premolar of R. megarhinus, the
following points of difference are observable :—

Ist. That the crown is much longer in relation to the
antepenultimate, and shorter in relation to the first true
molar, than in R. megarhinus.

2nd. That the anterior horn of the front crescent is much

342 RHINOCEROS.

more developed and more nearly of the size of the posterior
horn, in R. hemitechus than in R. megarhinus.

As regards the first of these characters, the penultimate and
last premolars in the latter species are nearly of equal size ;
while in R. hemitechus there is a progressive increase in
length of crown from the antepenultimate to the last. The
difference is shown by the subjoined comparative measure-
ments.

= on > a oi ein Ons
geanisa_|ae2| 2, f28saae | Bee
a eaeses|as | Sa SEES Se sa/ SR
S543 —O'°| FAs SE SR ER S| os
Bee scgslS =| oA |ESAl|F2 .°] as"
a" 6 Se a A 8a Hse | Pas
Length of 6 last molars 10 a eel Oey = ~- = —
Length of space occupied
by 5 last molars if thee alle tents, 93 | 85 | 815 —_ —
Length of 4 last molars
(1, 2 2,3m.,and 4p.m.).} — — 0 W488) 636 — 6-4
Length of 3 last molars .| 6° = 6 — | 53 —_— 51
Length of 2 lastmolars .| — 4-2 4-2 — — —_ —
Joint length of 1st and 2nd
true molars . -| — = 4: — | 34 33 --
Length of 3 premolars allt As -— +5 — -— — —
Length of last true molar,
over crown top 28 | 2-2 22 | 21 1°75 _— 19
Length of ditto, base Slee 2 — — 2-2 1:8 — —
Width of ditto, behind .| 1°3 14 — ise 11 — —
Length of penultimate, do.| 1:95) 2-1 2 21 1°85 1'8 —
Width of do., behind |. 1:25) 1°6 — 13 1:15 =
Width of do., in front at
base of crown 1: = _— — | 12 = —
Length of antepenultimate 1G, == — LOneiialees -- —
Width of do., behind| 1:2 — — 115 — =
Do. of do., in front | 1:1 — — — 11 — —
Length of last premolar | — —_ — 16 —— oo —
Do. of penultimate do,| — oa — 1-4 —- —_ —
Length of diasteme to in-
cisive border — = 4-4 —- — a —-
Height of jaw behind last
molar, inside ? 4: 48 — 4° 39 — —
Height of jaw in front of
antepenultimate - alc) = — — | 3:36 = —
Height of jaw to alveolar
marg. at middle of inner
side of p.m. 3 s -| — 35 — — — —_ —
Extreme thickness under
last molar. 5 -| 2°25) — — — | 23 —_— -—

The second character is well exhibited, on comparing
fig. 8 of Pl. II. of Gervais’ ‘ Paléontologie’ with Pl. XIX.,
annexed. In the former the anterior horn of the crescent,
in p.m. 3 and 4, and in m. 2 and 3, forms an insignificant
lobe, indicated by the anterior niche on the inner side of
See of these teeth; while in p.m. 3 and 4, and in m. 2 and

3 of Pl. XIX. of R. hemitechus the anterior horn of the
crescent 1 La as large a sweep as the posterior horn. The

DESCRIPTION OF PLATE XXTI.
RHINOCEROS HEMIT@CHUS.

Fig. 1. Representsa mutilated right ramus of the lower jaw, exhibiting |
the six posterior molars in situ, together with a portion of the
symphysial expansion. The specimen is remarkable in showing
the abnormal condition of two collateral teeth for the last
premolar. The crowns are seen in the early stage of abrasion
of the adult animal. The specimen was discovered in ‘ Bacon
Hole,’ and was presented to the Swansea Museum by Colonel
Wood. The figure is about one-half of the natural size and has
been copied from a drawing belonging to Mr. Spence Bate.
(See pages 340 & 349.)

Figs. 2 and 3. Represent two views in profile and plan, of a fragment
of the right upper maxilla with the milk dentition. The first,
second, and third milk molars are im situ, and part of the
alveolus of the fourth milk tooth is also seen. The specimen is
among Colonel Wood’s collections from the Gower caves, and is
believed to have been found in ‘ Minchin Hole.’ The figures
are about two-thirds of the natural size, and have been copied
from drawings executed for Dr. Falconer by Mr. Dinkel. (See
page 353.) Dr. Falconer was struck with the resemblance
which this specimen presents to a cast of the dentition of
Rhinoceros Lunellensis, sent to him by M. Gervais, and now
deposited in the British Museum. (See page 309.)

VOL, Il.

SS a ee ee “ep TS

MEE. ~

say

RHINOCEROS HEMITGCHUS. 343

common consent of paleontologists has pronounced against
the value of distinctive characters, derived from the lower
molars in the genus Rhinoceros; but the differences above
indicated are so constant and well marked, that I regard
them as being of specific importance.

Tn the ‘ Bacon Hole’ specimen (fig.1, Pl. XXI.), there are
two points connected with the premolars deserving notice.
In the penultimate (p.m. 3), the crown of which is well
worn, a distinct fossette is seen. This is unusual, and has
been caused in the present instance by the solution of a
portion of the valley between the horns of the posterior
crescent. The second point is that the last premolar is
double, and represented by two collateral teeth, the outer of
which is at a slightly lower level than the inner. The abra-
sion of the crowns of both these teeth, in relation to that of
the penultimate premolar in front, and of the first true
molar behind, proves that they are both of the second set,
and not a permanent premolar protruded excentrically along-
side of a retained milk molar.

True Molars.—The antepenultimate true molar (m. 1 of
Pl. XTX.) shows the remains of two well-marked crescents,
but being considerably worn it yields no distinctive cha-
racters. The crown is oblong, shorter than that of the
penultimate. Compared with the corresponding tooth of
R. megarhinus (Gervais, op. cit. Pl. II. fig. 8), it is narrower,
in reference to the length. The penultimate (m. 2) being
less worn shows the anterior crescent more pronounced; the
posterior crescent takes a very oblique antero-posterior direc-
tion, its front lobe terminating near the outer third of the
anterior crescent ; and it represents but a small degree of cur-
vature. The last true molar (m. 3) is the least worn of the
three, the posterior crescent being distinct from and still at
a lower level than the anterior crescent. Its anterior division
presents a very pronounced horse-shoe pattern, with equal
limbs. The posterior division is very oblique in direction,
and its worn surface exhibits but a small amount of curve.
The crown of this tooth is somewhat longer than that of the
penultimate.

The dimensions of the same tooth in the same lower jaw
vary not a little, according to the different stages of abrasion.
They are all inclined a little forwards, and the length of a
slightly abraded crown taken at the summit is less than that
near the base. In consequence of difference in measurement,
arising from causes like these, authors are not agreed in
regard to the relative proportions of the different teeth,
more especially the penultimate and last, which are the most
significant. Duvernoy positively affirms that in R. tichor-

344 ' RHINOCEROS.

hinus the penultimate true molar is smaller than the last ;
while in R. leptorhinus the last is smaller than the penul-
timate; the latter species in his view been represented by the
Rhinoceros jaws figured by Cuvier, from the Val d’Arno, and
by the R. megarhinus of Montpellier. Brandt distinctly
mentions, on two occasions, that in R. tichorhinus the last
molar is a little larger than the penultimate. On the other
hand, Professor Owen, in the table of comparative measure-
ments between the teeth of R. leptorhinus and R. tichorhinus,
given at p. 364 of the ‘ British Fossil Mammalia,’ makes
it appear that in R. tichorhinus the last true molar is smaller
than the penultimate, the reverse holding with the teeth of
the so-called R. leptorhinus, with which he compares them.
But I entertain grave doubts whether the Cromer specimen, '
assumed in this instance as an example of R. tichorhinus,
really belongs to that species. There are strong reasons to
believe otherwise. An undoubted specimen of a lower jaw
of R. tichorhinus,! from Lawford, is preserved in the Oxford
Museum, in which the last true molar is slightly shorter
than the penultimate. The dimensions of these teeth are
given in the subjoined table of comparative measurements.

Lawford R. hemitcechus Mr. Gunn’s*

Length of crown of last molar, at apex see ET p ise, 1°8 in. 1-75in,
Length of crown of last molar, below . pe Bis 2:2.) es eee
Length of penultimate, below : - 1°85 in. 1-95 in. 1°85in.
Length of antepenultimate, below - sop Uys Lies 150 55

In R. hemitechus, the teeth increase in length, uniformly,
although not symmetrically, from the antepenultimate pre-
molar to the last true molar, and the last true molar is
ordinarily considerably longer than the penultimate. The
relative proportions are best exhibited by the worn crowns of
PL XX. In R. megarhinus, the ratio of the length of the three
true molars to the three posterior premolars is as 6 to 4:5:
and in R. hemitechus as 6 to 4; the length of the whole series
being nearly equal in the two species.

it now remains to compare the teeth of the Gower species
with an important series of Rhinoceros remains, occurring
in the ‘ Hlephant-Bed’ or ‘Submarine Forest’ of the Nor-
folk coast, near Happisburgh and Mundesley, which, so far
as the evidence goes, constantly present well-marked dif-
ferences. The most perfect of these consist of rami of the
lower jaws with teeth. Upper molars are comparatively
rare, and such of them as have been met with have in most
instances been dispersed. No considerable fragment of a

1 See p. 401.—[Eb.] ® R. Etruscus. See p. 345.—[Ep.]

~~ s

ls

RHINOCEROS HEMITGCHUS. 345

cranium has yet been observed, nor an upper jaw containing
many teeth. The most conclusive description of evidence to
determine the species is, therefore, still incomplete. The
best examples of these remains are to be seen in the col-
lection of the Rev. James Layton, lately acquired for the
British Museum, or that of Mr. R. Fitch of Norwich; and in
the valuable collection of the Rev. John Gunn of Irstead.
Figs. 1 and 2 of Pl. XXII. represent a fragment com-
prising the greater part of the horizontal ramus of the left
side of the lower jaw, with the three true molars in situ, and
the empty alveoli of the three last premolars. The aggregate
length of the series of teeth is less in this specimen than in
either Rhin. megarhinus or Rhin. hemitechus, and the pro-

_portions between the teeth are different; the relative length

of the antepenultimate, penultimate, and last true molars
being in Rhin. hemiteechus nearly as 1:7. 1°95, and 2:2, and
in the Happisburgh specimen 1°5, 1°85, and 1:8. It belongs
to the collection of the Rev. John Gunn at Irstead, and was
found in the true Forest-bed, with roots of trees, &c., in situ.!

Another specimen from the collection of the Rev. James
Layton, now in the Paleontological series of the British
Museum, Cat. No. 33,326, is a corresponding fragment of
the lower jaw left ramus, containing the last premolar, and
the antepenultimate and penultimate true molars, together
with the anterior fang of the last molar im situ. The
ramus is mutilated in front through the anterior portior
of the penultimate premolar, and behind through the last
true molar. It is a trifle smaller in size than the previous
specimen, and the teeth are a little more worn; but the
form. of the jaw and the relative proportions of the teeth
correspond closely with those of the latter (Pl. XXIT. fig. 3).

There are two fragments of lower jaws in the British
Museum, presented by Mr. Pentland, from the Val d’Arno.?
The one (No. 28,802 MSS. Palzeont. Cat.) shows the upper
or alveolar portion of the left ramus, containing the last

1 Mr. Gunn has kindly forwarded to
me this specimen, to ke drawn by Mr.
Dinkel. Affixed to it is a label in Dr.
Faleoner’s handwriting, ‘ 2. /eptorhinus,
Cuy.’ But as above stated (p. 314), Cuvier
included under his hin. leptorhinus the
Rhinoceros of the Val d’Arno, which
Dr. Falconer subsequently separated
and designated Rhin. Etruscus. The
Rhinoceros of Messrs. Gunn and Lay-
ton’s collections was therefore hin.
Ktruscus. See pp. 310 and 355.—[Ep.]

2 Other specimens of the same species

in Mr. Pentland’s collection are also |

deseribed in Dr. Falconer’s note-books,
viz.: No. 28,804, a last or penultimate
true molar; No. 33,324, a last milk ?
molar, upper jaw, right side; and No.
33,323, a last milk molar, upper jaw,
left side. A tibia of Rhinoceros which
accompanies these teeth is described as
‘much more slender and considerably
longer than that of the Clacton species.’
The teeth are stated to have been from
the Conglomerate Sansino of the Val
d@’ Arno, and not from the Sabbone or
blue clay.—[ Ep. ]

346 - RHINOCEROS.

premolar and the three true molars in situ. The antepenul-
timate true molar is worn low, and the last tooth is well ad-
vanced in wear (Pl. XXII. fig. 4). In form, proportions, and
size, the teeth agree very closely with those of Mr. Gunn’s
specimen. The second Val d’Arno specimen (No. 28,803
MSS. Palzont. Cat.) contains the penultimate and last true
molars of the left ramus of the lower jaw. In form and size,
they are exactly the counterpart of No. 28,802. The follow-
ing are the comparative dimensions of the teeth in these spe-
cimens, contrasted with the same in Rhin. hemitechus.

FEL JUD “MET © 2 S
‘SOq “Ig ‘Ua pete ev call hrs boca | | de aS
‘SnUIYLOYO! “UL
L¥E & F9E “ddwaMKO 2 o
“wel Todo] Jeut0rg es a Vereen Ga | | lek fea ere | | |
8, Gon
MEPS EEE] Si Oo
pogoe se ou | ol eis cl Ls pl seo La
TRA Spuelyuog
Caq—'sno ~ = ae i
EanL ay Tete Lal | a 2 1 Tea
-10jdal "yy u0jAB'T
- ‘ 1D
(‘agqj—'sno 2 Oo 0 + com 0 SO wets
-snayg “Y) SnUtYy 2 So de & | (3S 2 See ee
-10}dey "GY s,uuny
N=) 12
snutyt Seley Seta leacc) 12 | | | | | |
-vZoml “YY JO ISeO Sete o tie st 1
snurqie3 Sy Gehl ee S Se (Se cord
-oul “YY ‘SBAIa4) Os SO Pte ON a | al |
UWOroaT[oo wWNASsnT > 00 ‘| | | | NANO 4 Oo |
“4LIg JO OFS‘6L “ON | oOo ' Ana a a
T9¢ “d — er) a at a ©
“UIE “20 “31 lo | | a ie Tos Seana
‘U0JOVID 8,UeaMO
snyoaqrunay "yy o : ee ew ies
yo aul So |e Io ol. ta ae ere
IIMOT VISUBMG 4
EP ORSe Bi a > eae
S38 9 8 8° 2 a5 0 cee
roy sy lek fe prot GS. Q g ° =| at
RSA og. ge Ee ae
aa tts 2 2 + p=t fe eras A,
SCwo Ht OA AE GS 3 eet eis =
oO Be Qy n a n
a = oe a, - 8
r= oOo « Ef a sai a
Ses ds wiwhs Ss 3 Some
iS) iS) o O88 © 2
a To adadced
eh 3g 8 5 S oS mA bo
ep eS Onon 3S Ord a rons a
e2ARAA A A, Ao xe Sao a)
H Foe 2

The agreement of the Happisburgh and Tuscan teeth so
closely, in size, form, and proportions, excites attention to

Figs. 1

Fig. 3.

Fig. 5.

DESCRIPTION OF PLATE XXII.
Rurnoceros Erruscus.

and 2. Represent in profile and plan the greater part of the
horizontal ramus of the left side of the lower jaw, with the three
true molars in situ, and the empty alveoli of the three last: pre-
molars. The specimen is in the collection of the Rev. John
Gunn, of Irstead, and was found in the Forest-bed of the Norfolk
Coast. The drawings are about two-fifths (.2,) of the natural
size (the length of crowns of three teeth at inner edges being
51 inches), and have been executed by Mr. Dinkel from the
original specimen which was forwarded to London for the pur-
pose by Mr. Gunn. (See page 3435.)

Represents a specimen also from the Forest-bed of the Norfolk
Coast, formerly in the collection of the Rev. James Layton, and
now in the British Museum (Cat. No. 33,326). It is a frag-
ment of the left ramus of the lower jaw, containing the last
premolar, the first two true molars, and the anterior fang of the
last molar. The drawing is about two-fifths of the natural size,
and has been executed by Mr. Dinkel from the original speci-
men. (See page 345.)

. Is a fragment in the British Museum (Cat. No. 28,802), from

the Val d’Arno; showing the alveolar portion of the left ramus
of the lower jaw, containing the last premolar and the three true
molars in situ. The drawing is about one-third of the natural
size, and has been executed by Mr. Dinkel from the original
Specimen.

Represents the first true molar, upper jaw, left side. The figure
is three-fourths of the natural size, and has been copied from
a drawing made for Dr. Falconer, and on which he had written :
‘Rhinoceros Etruscus—t. m. 1. 1. Happisburgh. The Rev.
J. Gunn. Coll. Yarmouth Museum.’

VOL, Il.

3 , Vol.IL Plate 22.

RHINOCEROS JTEMITCICHUS. 347

their other dental characters. Cuvier, in describing the
lower jaws of his species ‘a narines non-cloisonées’ of Italy,
refers to figs. 8 and 9 of Pl. IX., representing Tuscan speci-
mens, in proof that it had seven molars below in the adult
state, the pre-antepenultimate or first premolar, which is
suppressed in the Siberian Rhinoceros, being developed ;
and he seized upon this character as a distinctive mark of
his Rhin. leptorhinus. This pre-antepenultimate, although
present in the milk dentition, is suppressed in Rhin. hemi-
techus in the adult state, and it is also wanting in Rhin.
megarhinus. Thence, it becomes a point of the highest
interest to ascertain, whether it was present or suppressed
in the fossil Rhinoceros of the ‘ Elephant-Bed’ of Happis-
burgh. Professor Owen has described (Brit. Fos. Mam., p.347)
a fine specimen, comprising the greater portion of the hori-
zontal ramus of the lower jaw of a Rhinoceros, procured from
the ‘ Lignite Bed’ of Cromer, being an extension of the Hap-
pisburgh deposit. In this fragment, which is of a young adult,
‘there were four premolars and three true molars. Of the
latter, two are in place, and the last emerging ; of the former,
the alveoli of the first remain, the two next are in place, and
the fourth or last is embedded in the jaw under the last milk
molar, which had not yet been shed. A portion of the wall of
the jaw has been excised, and the milk tooth is seen super-
imposed to its successor. The pre-antepenultimate premolar
in this case had dropt out, but the fang-pits prove beyond
question that it had been there. Professor Owen has selected
this specimen as a standard example of Rhin. tichorhinus, for
comparison with a corresponding jaw of his Rhin. leptorhinus,
and he has given measurements of the two in contrast. I
have seen the specimen in question, in Mr. R. Fitch’s col-
lection in Norwich, and both the form of the jaw and the
relative proportions of the teeth conveyed to my mind the
impression that it belonged neither to Rhin. tichorhinus nor
to Rhin. hemiteechus, but to the same species as the specimens
above described, of Messrs. Gunn and Layton, i.e. to the
‘Rhinoceros & narines non-cloisonées’ of Cuvier, from Tus-
cany. Not having the fragment now before me, I am
desirous of expressing this opinion with diffidence and reserve.
Professor Owen was probably influenced, in arriving at
the above identification, by the beliefthat he had established
the fact that the first premolar is present in the lower jaw
of Rhin. tichorhinus, although Cuvier had asserted the con-
trary. In the ‘British Fossil Mammalia’ he has given a
representation, natural size (fig. 137, p. 363), of the two
anterior teeth of a young fossil jaw from Lawford, preserved
in the Oxford Museum. These teeth he considers to be pre-

348 RHINOCEROS.

molars, and he contrasts the second with an antepenultimate
premolar (fig. 136, op. cit.), also natural size, of the Clacton
species. The difference both in size and form between the
two is assuredly very great, and if the comparison were well
founded and sound, it would furnish a strongly marked dis-
tinctive character of the species; but it appears to me, that
in this case this eminent paleontologist has fallen into the
error of comparing the milk tooth of one jaw with the corre-
sponding permanent premolar of another. The Clacton
tooth is unquestionably a permanent premolar of the second
set; but the Lawford jaw (figs. 128 and 137, op. cit.) contains
four teeth, presenting as it seems to me the characters of
milk molars. Without going, on the present occasion, into the
details of the evidence for this conclusion, I may state4hat I
have compared the figure of the pre-antepenultimate (p. 1 of
Cut 137, above referred to) with the pre-antepenultimate milk
molar of a very young jaw of Rhin. hemitechus, in Col.
Wood’s collection from ‘Minchin Hole,’ and found them
agree in size and form, to the most minute particulars.
Brandt, with access to the rich collections in the Russian
Museums, distinctly states, in his monograph on the Siberian
Rhinoceros, that he had never seen an adult lower jaw
of this species showing more than six molars, thus con-
firming the early inference of Cuvier. The definite settle-
ment of this point, when well ascertained, will be of much
greater importance than merely determining the precise
number of inferior molars in an extinct species. Hence
the reference to it now. ‘The presence of seven lower
molars in the lower jaw from Cromer furnishes of itself,
independently of the other evidence, strong grounds, to my
mind, in favour of the specimen being referable to the
‘Rhinoceros a narines non-cloisonées,’ and not to Rhin.
tichorhinus. It will be a remarkable fact in Geology if it is
proved that the latter species was a contemporary of the
Sub-Apennine Hlephas meridionalis, as well as of the Glacial
Mammoth.

‘The above was written in 1860. | the Norfolk Coast’ was subsequently
The Rhinoceros of the Val d’Arno | designated by Dr. Falconer Rhin.
and of the ‘Submarine Forest-bed of | Etruscus. See pp. 310 and 355.—[Eb.]

RHINOCEROS HEMITGICHUS. 349

APPENDIX TO MEMOIR ON RHINOCEROS HEMITCCHUS.
Extracts from Dr. Falconer's Note-books.

I,—Note on Lower Jaw, Ricur Sipe, or Rarioceros Hemracuvs,
' From Bacon’s Hore, in Swansea Museum. (See p. 340.)

27th April, 1858.

Compared the original of Spence Bate’s drawing (Plate XXI. fig. 1)
with Mr. Gunn’s specimen from the Norfolk coast. They are very dif-
ferent. Spence Bate’s drawing is not in exact profile. In the original,
the collateral last premolars attain nearly the same height, and are worn
exactly as a single tooth, the outer one a little lower. They are not
milk and true premolars, but double premolars of the second set. The
contour is not well shown in the drawing, particularly of the anterior
end, the jaw not having been placed vertically, but sloped outwards, to
show the crowns. The enamel of the teeth is smooth. There are two
very large mentary foramina, the one under the front of antepenulti-
mate premolar, the other under the back of penultimate. The front
one round and very large.

The most remarkable difference is in the contour line of the lower
jaw, which is curved in the are of a circle very much as in the African
two-horned rhinoceros; whereas, Gunn’s specimen (Pl. XXII. fig. 1)
is nearly wedge-shaped, without any considerable curvature. It is
certainly not &. leptorhinus. Gunn’s specimen is also thicker; the
inner longitudinal channel more marked, and the posterior one also; the
teeth are shorter and thicker in Gunn’s (relatively). The antepenul-
timate true molar in Mr. Gunn’s specimen is also very much thicker in
proportion to the length.

Gunn's Swansea
Extreme length of fragment at base. . ; : 11-4 12°
Length of three last teeth 6 : é : 3 5°3 6:0
Length of last worn crown : : 3 : 0 1°75 18
Ditto near base : $ - > 18 2°18
Width of ditto behind. 1:13 1-3
Length of penultimate . 18 1:95
Width of ditto, behind : 115 1:25
Ditto, front, base of crown x 3 : 1:2 1:
Length of antepenultimate . . ; g 5 1°55 ue /
Width of ditto behind. - . ; : - 1:15 1:2
Width of ditto in front . : . : 11 1:08
Height of jaw behind last tooth, inside’ 39 4:
Ditto in front of antepenultimate 3°35 35
Extreme thickness under last tooth . : C 2°3 2:25
Length of space occupied by five last teeth “ 815 8:8

Ii.—Nore on Mortars or Raurnoceros Hemiracuus, From Durpuam
Down, 1n Bristot Museum.

4th May, 1858.

But the most interesting of all are a set of upper molar teeth of
Rhinoceros, identical with the Rhinoceros of Bacon Hole! Of these,
four belong to the left side and fit in pairs, of which two are worn

350 RHINOCEROS.

premolars, and the two others the antepenultimate and penultimate
true molars.

The antepenultimate true molar is worn very low down, and the
anterior barrel is broken across diagonally from the outer anterior angle
inwards, so that it cannot be fitted to the premolar preceding it. The
posterior notched valley is ground down into an isolated pit, with a
shelving inner wall (not vertical, as in Rhin. tichorhinus). The trans-
verse valley terminates in a very round sweep, without any combing
processes thrown into it. The enamel edge is thin, and the surface of
enamel very smooth, with an enormous coat of cement.

The penultimate agrees exactly in measurement with the Swansea
tooth, but it is more worn. The posterior valley is spacious and angu-
lar, and not yet isolated; the transverse valley is divided into two di-
visions by a bold projecting curved crochet, given off from the posterior
barrel; the posterior division of the valley is roundish lengthwise, but
no combing processes; has a distinct basal tubercle.

The coat of cement is enormous, and very much like that of the
Swansea specimen. ;

The two other teeth which fit are also of the left side; and probably
the penultimate premolar and antepenultimate, both well worn. The
posterior tooth has the posterior valley reduced to an oval fossette, iso-
lated. The transverse valley is also isolated, with three comb-shaped
processes from the posterior barrel, but none from the outer wall.

Dim nsions.—Length, along outer edge, 1°7 in. Length of inner ditto, 1:5 in.
Width in front, 2°05 in.; width behind, 1°65 in., approximative.

These agree very closely with the Swansea measurements.

The antepenultimate premolar is still more worn; the posterior fos-
sette smaller, less oblong (rounder), and more isolated; the trans-
verse valley has three processes thrown into it from the posterior barrel,
but none on the outer side. The tooth has distinctly two barrels, and
is too large for the antepenultimate.

Diinensions.—Length, outer side, 1:5 in. Length, inner sido, 1°35 in. Width
of crown in front, 1:9 in. Width of crown behind, 1°6 in.

IIJ.—Menmoranpum or Sxuty or Rutnoceros Hemitacnus, In THE CoL-
LECTION OF Mason Woop; From Mincuin Howe. (See p. 325,)

The specimen is a superb fragment, comprising the whole of the
cerebral part of the skull, but vertically broken through about two
inches in front of the posterior termination of the temporal fossa. It is
clear from the recent condition of the fracture that the facial part of the
skull was broken and destroyed during extrication. The following
parts are present. The sphenoidal region quite entire, also the two
condyles with the foramen, and nearly the whole of the occiput up to
the niche of the occipital crest ; the lateral margins quite entire The
two auditory foramina quite entire, also the left mastoid, but the sty-
loidal process on both sides broken off. The zygomatic arches both
broken, but the base present on the left side; and on both sides, but
more especially on the right, the greater part of the articulating sur-
face for the lower jaw is present, broad, and somewhat of a cordate
pattern, with the sinus directed backwards. (See Plates XXIII. and
XXIV.)

RHINOCEROS HEMITQCHUS. 501

TV.—ComPARISON OF THE GoWER CAVE RHINOCEROS, WITH SPECIMENS IN
British Museum.
30th September, 1858.

Spent a long day with Mr. Waterhouse upon the Fossil Rhinoceros.
Took with me all Major Wood’s specimens from Minchin Hole and
the Swansea Museum—specimens of upper and lower jaw, and the
Minchin skull. Compared the Minchin skull with the two crania, the
Clacton one figured by Owen and the other from Northampton, and
found them to agree exactly in the form of the occiput, little amount of
backward extension and vaulting of occipital crest, and in the form of
the occipital plane, z.e. contracting upwards, and not a parallelogram,
as in 2h. tichorhinus. Thus inferred that the Rh. leptorhinus of Owen's
cranial figures is the same as our 2h. priscus! (R. hemitechus) of the

Gower Caves. (See Plates XXIII. and XXIV.)

V.—Nore on tue NortHampron and Ciacron SKuLLs or Ruinoceros
HEmMIT@cHUs.
1st October, 1858.

The Northampton Rhinoceros skull in the British Museum, No. 2,
Zt. leptorhinus, Owen, and labelled 20,013, is entered in the book as
having been purchased in 1846 from Miss Baker of Northampton, sister
of Baker the historian. The exact locality is not mentioned, but other
specimens of the same lot are referred to Blisworth, Kilsby Tunnel,
Bugbrook, Northampton, &c., all in Northamptonshire. This specimen
comprises the occiput and condyles, quite entire, and the whole of the
frontal on to the naso-frontal suture, which is also quite entire, as are
also the base of the right zygoma, the right articulating surface, and the
right styliform process; the left zygoma is less perfect. The animal
was very young, although large as compared with that from Minchin.
There is no evidence as to the age of Brown’s Clacton skull (B. M. 182,
183); there is no sign of any of the sutures being open; but the upper
part of the occiput is not broader than in the young Northampton
specimen. Of the three molars which Brown gave with the skull, the
last molar is implanted in the maxillary, with part of the palatine bone
present. The tooth is of the left side, and is in the middle stage of
wear, and is precisely like the pair of Minchin molars. If this fragment
belongs to the skull it would prove the animal to have been adult.
The antero-post. length inside of the tooth is 2:15 in. in Clacton, and
2:1 in. in Minchin. Like the Minchin specimen, the Clacton last molar
has a basal lobe behind and an intercolumnar tubercle, but beth are
wrapt up in an enormous mass of cement. The complexity of pattern
is equally great in both. In the Clacton skull there is no distinct mark
of a frontal horn. ‘The base is not quite smooth, but it is not rugous
enough. The frontal of the Northampton skull is absolutely smooth,
but the animal was young. (See Plates XV., XXIII, and XXIV.)

Dimensions of Clacton Skull—Length from tip of nasals to summit of occipital
crest, measured along the curve, about 29°in. Length from tip of nasals to sun:mit
of occipital crest, stretched, 28-5 in. Width of inter-temporal plateau of sineiput
where narrow, 1-4in. Length of nasal sinus (septum), 9°5in. Length of nasal sinus
(septum) in skull of &. tichorhinus, 7: in. Length of base of partial septum, about
50 in. Length of unossified part, about 4:in. Width of nasals in a line across with

base of sinus, 58 in. Width of nasals at commencement of septum (posterior end),
47 in, Length from anterior side of styliform to nasal sinus, 13:0 in,

! Rh. priscus was the name first given by Dr. Falconer to 2. hemitachus.— [ Ep. |

352 RHINOCEROS.

Comparison of Minchin Skull with young Northampton Skull.

Minchin Young
Skull Northampton

Across the condyles, outer angles 2 ‘ 3 6: 53
Length, lower surface, left condyle . . 2°65
Depth from lower edge, ven pete to occipital

crest, right side. “ ‘ - 85 78
Width of occipital plane near apex : 6-
Width behind orbitary gets ?) foramen, and

a little above 3 . . 3 : 9°3 84

VI.—NotTEe on Younc Lower Jaws or Ruinoceros Hemitracuus.
College of Surgeons, 15th October, 1858.

Examined a very beautiful young lower jaw, left ramus; the greater
part of symphysis present with whole of horizontal ramus; the posterior
angle wanting, but a part of the ascending ramus present; what there
is of it reclines, but of the posterior lower part the whole is restored in
plaster. What remains of the ramus agrees with the next specimen.
It is from Minchin Hole, and has two foramina near symphysis, like the
other specimens. It contains the deciduous dentition quite perfect, and
all emerged, namely, Ist, 2nd, 3rd, and 4th milk molars; all more or
less worn, except the first, which is perfectly entire. In form it resem-
bles exceedingly the figure in Owen’s ‘ British Fossil Mammalia,’ Cuts
128 and 137 of the young jaw, from Lawford, confirming the impression
that the latter is also a milk specimen. Strange to say, the first tooth
is unworn. (See Pl. XXV. fig. 1.)

Dimensions.— Length of the four m.m. 5:2 in.; of 1m.m., 0°8in.; of 2 m.m., 11
in.; of 8 m.m., 1°6 in.; of 4 m.m., 1°7 in.

Another specimen, also from Minchin Hole, is the right ramus of the
same animal. It is less- perfect, and contains the 2nd, érd, and 4th
milk molars, and the alveolus of a 1st,—a single pit of perhaps two con-
fluent fangs. The symphysis has been partly restored, and does not fit
to the left ramus.

Further, compared the first milk molar of the left jaw with Owen’s
figure (Cut 137, p. 363); the latter is of the inside apparently; they
agree to the minutest particulars. Can the real Rugby specimen really
be of Rhin. priscus? (R. hemitechus. See antea, p. 348.)

Compared the third specimen of lower jaw from Minchin; a little
older. It is superb; comprising the whole of the horizontal ramus and
symphysis, with the ascending ramus and the greater part of the condyle;
surface eroded, and the coronoid broken off obliquely down in a line
with sigmoid notch; the ascending ramus reclines exactly as in R.
tichorhinus, but the contour of the lower jaw is decidedly different.
There is no abrupt step of ascent, as FR. tichorhinus. There is more
convexity below, but the curve is gradual in front, as in R. bicornis,
the ramus of which also reclines much.

VII.—Nores on Mitx Denrition or R. Hemitecuus.
College of Surgeons, August 6, 1859.

Examined two milk molars (second and third), fitting together, of
R. hemitechus, from Colonel Wood (Minchin Hole). They are very
fine, though well worn. There is also a detached shell in germ of the
second mill molar, right side, quite intact, and with the enamel only

DESCRIPTION OF PLATE XXIII.
RHINOCEROS HEMIT@CHUS.

Figs. 1 and 2. Represent the basal view of two crania, the one found
in ‘ Minchin Hole,’ the other from Northampton, showing their
similarity. The figures are one-fourth of the natural size. (See
pages 351 & 509.)

Fig. 1. Represents the basal view of a portion of cranium found in
‘Minchin Hole,’ taken from a drawing executed for Dr. Falconer
by Mr. Dinkel.

Fig. 2. Represents the basal view of the ‘ Northampton ‘Skull’ in the
British Museum (Cat. No. 20,013), also taken from a drawing
executed for Dr. Falconer by Mr. Dinkel.

VOL. II.

VoLIL Plate 23.

owner

W.West imp.

hus.
. Northampton.

mitoce e

ceros he

0
mn Hole.

J.Dinkel del. at tith.

ote ee

2

Mmch

at

DESCRIPTION OF PLATE XXIV.
RHINOCEROS HEMITECHUS.

The figures in this Plate have been reproduced from draw-
ings by Mr. Dinkel of the original specimens. (See pages
351 & 509.)

Fig. 1. Is a lateral view of the ‘ Northampton Skull’ in the British
Museum (Cat. No. 20,013), right side.

Fig. 2. Is a lateral view of skull found in ‘ Minchin Hole,’ left side.

«Fig. 3. Is a view of upper surface of skull found in ‘ Minchin Hole.’

VOL. I.

Vol. IL Plate 24.

Dime del «Lith Rhinoceros hemitoechus. W.West imp.
lip | Northampton. 2,3.Minchin Hole.

val RHINOCEROS HEMITC&CHUS. 353

ossified. A third milk molar in wear is very like Cesell’s tooth from
Rome. (See Pl. XXV. figs. 2, 3, and 4.)

Examined also a right maxillary with milk dentition. (See Pl. XXI.
figs. 2 and 3.) The first, second, and third deciduous teeth are beauti-
fully seen in place. The teeth are worn, and part of the alveolus of the
fourth milk tooth is also seen. The second tooth has three fossettes
besides the entrance of the valley. The specimen is exquisitely fine.
There is no matrix on it, but it is probably from Minchin Hole.

Leneth of three teeth, 3-8 in. Length of 3rd milk molar, outer side, 17 in.
Greatest width of ditto in front, at base, 16 in. Length of 2nd milk molar, 1-4
in. Length of 1st milk molar, 0-9 in.

[ References to other bones of the skeleton of the Rhinoceros hemitechus
from the Gower Caves are to be found in Dr. Falconer’s Note-books.
The femur was compared with the femur of Rhinoceros tichorhinus of
Mr. Lucas from Port Inon, referred to by Dr. Buckland. It was found
to differ remarkably ‘in its much shorter proportions, and in the very
bold curve intercepted between the third trochanter and the outer
condyle. The bone itself is absolutely much shorter and smaller, and
the species must have stood on proportionally shorter legs.’ The fol-
lowing reference to a tibia is also important:—‘'The bone is short
and squat, as compared with the corresponding bone of Rhinoceros
tichorhinus, and the fibula is ossified with the tibia along a much greater
extent of surface. This specimen is of great importance in giving the
characters of the species.’ The bones of the cranium are also referred
to in the author’s essay on ‘ the Ossiferous Caves of Gower.’ In a list
of Rhinoceros remains from Bacen Hole, in the Swansea Museum, men-
tion is made of the lower half of right humerus, upper half of radius
with articulating surface of ulna, pelvis, cervical and dorsal vertebra,
a thick and short metatarsal bone, &c.—Eb. |

VIII.—Norte on Rutwoceros Hemitacuus rrom FOLKESTONE.
27th September, 1858.

Tn Mr. Mackie’s collection of fossils from excavations made at Folke-
stone there is a specimen (labelled ‘ Battery’) of the last upper molar,
left side, of R. hemitechus. 'The shell is nearly entire, but the fangs
are wanting. The grinding surface is a little damaged by minute
chips, but there is no sign of wear. The crown, however, is very
perfect, and presents the characters of the species well marked—namely,
the last barrel compressed, and emitting from the middle forwards a
large crochet plate. The valleys have a thick coat of cement, but the
outside is denuded. This is an important specimen, and ought to be
figured. It entirely agrees with Colonel Wood's specimens from Bacon
Hole Cave.

IX.—Nore on Rurnoceros Hemiracuus From Oreston.
College af Surgeons, 10th August, 1859.
To-day compared the Rhinoceros teeth from Oreston, described by

_ Whidbey in the ‘ Phil. Trans.’ for 1817, -21, and -23, and referred to by

Owen in Brit. Fos. Mam. as belonging to 2. tichorhinus. There are

only three upper molars, Nos. 877, 878, and 879. ‘The first is the

right upper antepenultimate, and the second the left do. of probably the
VOL. II. AA

354 RHINOCEROS.

same individual. Both are broken, but conversely, ¢.e. the anterior
end of 877 and the posterior of 878, so that jointly they give the com-
plete form of one tooth. They agree in both showing the crochet of
the posterior barrel stretching across to joi the anterior barrel, as in
Cuvier’s drawing.! They are quite unlike 2. tichorhinus, and I believe
that they agree with R. hemitachus.

X.—Note on Ruroceros HEmitacuus FROM CRAWLEY Rocks.
Oxford, 11th August, 1863.

The Crawley Rocks Rhinoceros tooth in the Oxford Museum is a very
fine penultimate or last premolar of R. hemitechus, upper jaw, right
side, with crochet in two combing plates. Length of crown outside,
174 in.; do., mner side, 1°25in. ‘The tooth is beautifully marked, and
ought to be figured. The valley is very deep. In the Kirkdale series,
besides the large worn molar there are two premolars, both germs, the
one exactly corresponding in size and form with the Crawley Rock
premolar, but intact, and has only one developed combing plate; the
second is also an intact germ of the antepenultimate premolar, left side,
of the same species; the entrance of the valley here also being vertical.
Both these specimens profess to be from Kirkdale, but they differ in
mineral appearance from the other. They bear no label, and they agree
in condition exactly with the Crawley Rocks specimen. Can there be
amistake? Are they from Gower?

Oxford Museum, 5th July, 1860.

Saw one premolar of Rhinoceros hemitechus, well marked, in a drawer,
and labelled ‘ Crawley Rocks.’

Il. NOTES ON RHINOCEROS ETRUSCUS. (Fatec.)
(Extracted from Dr. Falconer’s Note-books.)

I.—Nore on Rutnoceros Erruscus 1x Oxrorp Museum.?
6th May, 1858.

In Buckland’s collection there is a left upper maxillary and half
palate of a Rhinoceros labelled ‘ Rhinoceros leptorhinus from Venice,’
in a hard ferruginous matrix of gritty sandstone. It contains four
molars in situ, namely, p.m. 3 and 4, and t.m. 1 and 2, and also the
broken-off discs of p.m. 2 and t.m. 8. The two premolars are of
the second set and half worn. ‘The first true molar is much worn; the
penultimate is half worn. The enamel is very smooth, and the teeth are
smaller than in the Kirkdale specimen. There is a considerable basal
bourrelet at the anterior end of the last premolar and of the penulti-
mate true molar. There are no combing processes whatever projecting
into the transverse valley, and no appearance of cement. It reminds
me of Ansted’s specimens from Malaga. (See p. 360.) The outer
surface of the two true molars from the termination of the valley is
gone, but it shows the transverse valley well. The first true molar has
its anterior outer corner broken, and the third and fourth p.m. have their

1 See antea, p. 887.—[Ep.] 2 See p. 848, xote.—[ Ep. ]

DESCRIPTION OF PLATE XXV.
RHINOCEROS HEMITECHUS AND RHINOCEROS HTRUSCUS.

Fig. 1. Outer surface of left ramus of young lower jaw of LR. hemi-
tachus, with greater part of symphysis and whole of horizontal
ramus, and containing the first four milk molars. The figure is
one-half of the natural size, and has been copied from a drawing
of the original specimen executed for Dr. Falconer by Mr. Dinkel.
The specimen is from ‘Minchin Hole,’ and is described at

page 352.

Figs. 2, 3, and 4. Represent upper milk molars of R. hemitechus, from
‘Minchin Hole,’ of the natural size, copied from drawings of the
original specimens executed for Dr. Falconer by Mr. Dinkel.
(See page 352.) Fig. 2 shows the second and third milk molars.
Fig. 3 is a germ of the second milk molar. Fig. 4 is a detached

third milk molar.

Figs. 5, 6, and 7. Represent three upper molars of R. Htruscus. The
drawings have been made by Mr. Dinkel from three casts
presented to Dr. Falconer by Professor Meneghini, of Pisa, and
now in the British Museum. They are of the natural size.
Fig. 5 shows the crown of the last (t. m. 3) upper molar of the
left side. Fig. 6 is the last upper premolar (p. m. 4), right
side. Fig. 7 is the penultimate upper molar (t. m. 2), right
side, mutilated at posterior outer augle.

VOL. II. d

Vol. IL Plate 25.

by
;

oDickel jel vith 1,2,3,4. Rimoceros henutoechus, Gower Caves.
5,6,7. Rhmoceros Etruscus trom Pisa.

WW. West ono

RHINOCEROS ETRUSCUS. 355

outer surface as to the valley broken off. There is a little mammilla
between the barrels of the first and second true molars. In the third
and fourth p.m. the end of the valley is only a very slight cleft ; in the
true molars it is an open flexuous fissure.

Dimensions.—Length of 5 teeth (2nd p.m. to end of 2nd t.m.), 7-5 in. Length
of 2nd t.m. at middle, 1°85 in. Width in front, 2:2 in.

Can this really be from the Sub-Apennines ?

TI.—Comparison oF Ruroceros oF Norwich LAcusTRINES WITH
‘Venice’ Upper JAw IN Oxrorp Museum.

7th May, 1858.

Compared the Rev. Mr. Gunn’s detached upper molar (Pl. XXII.
fig. 5) from the Norwich lacustrines with the upper jaw labelled ‘ Rh.
leptorhinus from Venice’ in Buckland’s collection, and found the most
important agreement. Gunn’s also belongs to the left side. In form
Gunn’s would agree best with the last premolar from the smaller size
of the posterior “barrel, but unluckily the fracture of the outer surface
of the Venice fossil prevents a rigid comparison. They agree in the
following important points :—1. ‘Exact similarity of smooth enamel
surface. 2. Decided anterior basal bourrelet, worn down in Gunn’s.
3. Like thinness of enamel. 4. Sweep antero-posteriorly of termina-
tion of large valley, and its nearly isolated form. 5. Openness of
gorge of transverse valley.

Dimensions.

Gunn’s Venice
specimen second true
molar
Length of outer side at constriction : 6 5 1'8
Length of inner side ; = : : 16
Breadth near middle, anterior barrel - : 3 about 2°2
Breadth behind, at base of crown . C 3 : DD
Tirst true
molar
Length of outer side kerpaiert) 2:0 1°75
Length at constriction . : ; : 1-75 16
Length of inner side ; a 1:85 17
Breadth of middle, anterior eaerolee : ri 2:2 2°2 nearly
Breadth behind at base . ‘ A 5 ee i] 2°15
Height of enamel crown, posteriorly A ; ep 12 1:

Norwich, July, 1863.

Examined the Rhinoceros jaw in Fitch’s collection. It belongs to
ft. Etruscus. M. Lartet detected in it the remains of the large men-
tary foramina. ‘ Got at Anderson’s the fisherman’s a portion without
ends of a femur of an old R&. Htruscus, very characteristic.’

IJI.—DescripTion oF CrANIA oF R. Erruscus In THE GRAND Duca
Museum At FLorence (PLATEs XXVI. anp XXVII.).

18th May, 1859.

In the Museum at Florence is*preserved a superb skull of Rhinoceros
Etruscus from the Val d’Arno, nearly entire; two-horned, and very
old. There are six molars on either side, of which even ‘the last is
worn to the base. The skull is very little crushed, and there are very
few restorations. The nasals are perfect to their very tips on one side,

AA 2

356 RHINOCEROS.

and are slightly emarginate and arched at the side, very much as in
R. tichorhinus. They send down a vertical bony partition, which is
deepest in front; the posterior part is broken, but does not appear to
have been ever complete behind (only partial); what remains occupies
one half of the nasal echanerure. The incisive bones are broken off,
but on the right side a considerable portion of the diasteme remains.
The arch of the nasals is higher than in Ff. tichorhinus; and the greatest
height of the septum is in front—the septum being lower behind, which
is the very reverse of what is observed in R. tichorhinus. The broken
part of the incisives has been badly restored in coloured gypsum, but
the join is easily recognizable. Compared with the Lyons skull of
R. megarhinus (Plate XXXI. fig. 3), the Florence head is consider-
ably smaller in all its dimensions, and the lower jaw and teeth are in
keeping. Viewed from the top, the skull in contour resembles more
that of the R. tichorhinus (Cuv., ‘ Oss. Foss.,’ Pl. 160, fig. 5, and Gervais
of the Montpellier skull, ‘ Trans. Academ. Montp.’ tom. xi. Pl. E. fig. 2)
than any of the others.. Length from about outer margin to occipital
crest, 14: in., and from ditto to tip of nasals about 12°5 in., or as 7: 6.

The nasal horn rugosity is enormous, projecting greatly at its central
nucleus; then there is a smooth interval of about three inches, and then
an indistinct and not much raised rugosity for a second horn. This
frontal horn was probably small; and there is here nothing like the
enormous confluent rugosity of P. tichorhinus. The right orbit with rim
is nearly entire, but the tubercles are broken off; they are smoothly
restored on left side. The maxillary bone on right side is a little
crushed below the infra-orbitary foramen. The zygomatic arches are
quite entire, thin and high, and but little crushed. The articular
surfaces are also entire on both sides. There is only a slight rise for
the frontal horn between the orbits. The frontal and sincipital surfaces
are smooth, with a tablet showing about the same width as in Gervais,
Tab. 11, fig. 2; the two bounding ridges are visible but indistinct.
(There is some restoration between the temporal arches on both sides.)
There is hardly any sincipital pyramid, but the occiput is slightly
crushed on the left side. The occipital plane rises nearly vertically,
but is overarched at the sides by the projecting occipito-parietal crest,
and an easy echancrure in the middle. This part of the skull is formed
very much after Gervais’ figure above quoted. The occipital plane is
wide, and very low as compared with width. (Some little plaster
restoration on right side.)

Florence, 19th May, 1859.

The skull of Rhinoceros Etruscus in the Florence Museum has the
following characters (see Plates XX VI. and XXVIL.) :—

1. It is smaller and more slender than the horned rhinoceros of
Sumatra (Cuv. Pl. IX. Rhin.).

2. The cerebral portion is very elongated and shelving behind over
a vertical occiput; it is but little elevated behind.

3. The skull is very flat from the occipital crest forwards; there is
no pyramid properly so called (vide ‘ Dimensions’).

4. The posterior surface of the occiput (when the skull is placed
upon the plane of the teeth) is inclined forwards, and is overarched by
the shelving occipital crest (Plate XXVI. fig. 1).

5. The nasal bones are more elongated than in the Cape species;

DESCRIPTION OF PLATE XXVI.
Rumoceros HTRUscus.

Three different views of cranium in the Florence Museum,
one-fifth of the natural size. Fig. 1. Upper surface. Fig 2.
Profile view, showing well the incomplete nasal septum.
Fig. 3. Lower surface, showing palate and series of six molars
.on either side well worn. These figures’ have been copied
by Mr. Dinkel from drawings executed for Dr. Falconer by

Vincenzo Stanghi, artist at Florence. (See page 356.)

VOL. Ii.

Vol II. Plate 26.

‘W.West imp.

Rhinoceros Etruscus.

DESCRIPTION OF PLATE XXVII.
Rurnoceros Erruscus.

Views of cranium, lower jaw, and teeth in the Florence
Museum. The figures have been copied by Mr. Dinkel from
drawings executed for Dr. Falconer by Vincenzo Stanghi,
artist at Florence. (See page 356.)

Fig. 1. Posterior view of cranium represented in Plate XXVI., showing

occiput, zygomatic arches, occipital condyles, and foramen mag-
num, one-fourth of the natural size.

Fig. 2. Profile view of lawer jaw, outer surface, one-fourth of the
natural size.

Fig. 3. Same lower jaw, viewed from abeve, showing crowns of molars
far advanced in wear, one-fourth of the natural size.

Fi

i

g. 4. Symphysial portion of same lower jaw, viewed from below,
one-fourth of the natural size.

KR

my

g. 5. Four molars of upper jaw, left side, smaller and less advanced
in wear than those in skull represented in Plate XXVL, fig. 3.
Three-fourths of the natural size. (The dimensions almost
correspond to those given in page 359.)

VOL. Il

; | WMC ‘ep sues
Ear asa\'M snosnyy sorsoouryyy

RHINOCEROS ETRUSCUS. 357

they are vaulted forwards, but not uniformly, as in R. tichorhinus ;
they are bifid at the apex and then throw down a septum which
terminates below in a thick knob (Plate XXVI. fig. 1), and is incom-
plete behind (vide ‘ Dimensions’). The nasal horn is very rough and
overlaps the sides of the nasals with an excessively rugous conical
raised nucleus; there are no ramures, asin R. tichorhinus and R. megar-
hinus ; the edges of the nasals are thin and arched; the nasal echancrure
is narrow at the bottom, and then arches high forwards, followed below
by a rim on either side of the septum.

6. The zygomatic arches in front are nearly horizontal; then the
posterior part rises upwards in the arch to the glenoid surface, but not
nearly so much as in &. megarhinus. (In the detached maxillary and
orbitary fragment there is a distinct post-orbital tuberosity defining
the orbit behind.)

7. The temporal fosse are very much as in Cuvier’s fig. of R. tiehor-
hinus, fig. 5, Pl. IX. Rhin.; and in the two-horned Sumatra Rhinoceros,
fig. 3.

8. The incisive bones join on to the septum, but are broken. (In
the right maxillary specimen, 2°2 inches of diasteme remain.) 'There
are no upper incisors apparent, as certainly there are none in the lower
jaw.

9. The orbit is placed mostly above the seventh molar, but its an-
terior border advances as far as the middle of the sixth or penultimate
molar in the large skull. (In the right maxillary fragment it advances
only to the rear part of the sixth molar; the same remark applies to
the skull in two pieces.)

10. The suborbitary foramen is situated between the third and fourth
premolars in the large skull. Inthe maxillary fragment of the head in
two pieces it is over the fourth premolar, close to the nasal echancrure
between third and fourth premolars.

11. The auditory foramen is large and in a line with the upper edge
of the zygomatic arches.

Viewed above, the skull is very like that of R. tichorhinus, but it is
not so wide and the nasals are more elongated. The interval also
between the orbits is narrower, and the cerebral portion longer. The
temporal fosse are of considerable extent; their bounding edges being
less defined than in R. tichorhinus; they are nearly parallel in the
middle, but diverge into the occipital crest behind, and into the orbits
in front, as in R. tichorhinus. The frontal tableau is longer and less
pronounced ; it is less broad than in AR. tichorhinus, but wider than in
R. Indicus. There is no hole with ramures to the nasal horn. The
occiput is inclined in front with two diverging ridges and a deep de-
pression; but is shelved over by the projecting crest.

Measurements of the Rhinoceros Skull and Lower Jaw, at Florence—Sxv1t.—
Length of 6 last molars, right side, 88 in. Length of 3 last (true) molars, right
much worn, 5:0 in. Length of 3 premolars, 4° in, Total length of skull from oceipital
lateral crest, measured along chord to overhanging tip of nasal, 25°25 in. Total
length of ditto from posterior surface of occipital condyle to tip of nasals (vertical
plane), 25:in. Total length from nasal echancrure left side to tip of nasals (by
eallipers), 7°7 in. Total length from nasal echancrure (left) to anterior border
of orbit (exactly), 4°5in. Total length from anterior border of right orbit to occi-
pital crest (lateral), 140 in. Total length from anterior border occipital foramen to
palatine echancrure, 12° in. From palatine echancrure to tip of nasals, 12° in.
Greatest width across zygomatic arches in line with articular surface, 12°76 in.

358 RHINOCEROS.

Extreme length of right temporal cavity taken at base of skull, 5:2 in. Greatest
width of ditto between pterygoid and inside of zygoma, 4:4 in. Greatest constric-
tion of skull between zygomatic arches, 4: in. Length from posterior surface of
occipital condyle, to apex of pterygoid alar process, 9°4 in. From ditto to posterior
boundary temporal fossa below (edge of articular), 6-4 in. Length of diasteme
remaining, right side, 1:5 in. Interval of palate between p.m. 2,1°5in. Interval
between outer surfaces (posterior end) of p.m. 2, 4:7 in. Interval between anterior
barrels of last molars, 2°5 in. Interval between outer surfaces of ditto, 6-6 in.
Transverse extent of articular surface of glenoid, 3:9 in. Stretch across condyles
to outer border, 5-2 in. Height of occipital crest, right side, from lower surface of
condyle, 6°5 in. Height of right styloid (left a little broken), about 2:1in. In-
terval between ditto, inside, at apex, 3°8 in. Length from palatine echancrure to
posterior edge pterygoid alz at base, 5-8 in. Length from posterior surface condyle
to posterior surface of last molar, 11°6 in. Constriction of skull below auditory
foramina, 7:1 in. From anterior border, right orbit, to tip of nasals, about 12° in.
Length of zygomatic arch from posterior fang of 6th molar or penultimate, in a line
with anterior margin of orbit, to border of auditory foramen, approximatively,
10: in. Antero-posterior extent remaining of septum, upper margin, 4:7 in.
Antero-posterior extent remaining at middle, about 4:2 in. Width of brow between
orbits (right half, 4:5), 9°0 in. Interval between sincipital ridges in line with
ear, 2°5 in. Width of nasals in middle of anterior horn at base, 4°45in. Width of
nasals in line with echancrure, 4°25 in. Height from diasteme to edge of nasal
arch, 3°9 in. Length from posterior angle (tuberosity) of right orbit to oecipital
erest, 11'4in. Height of skull from right condyle to right occipital crest, 6-5 in.
Width of occiput near the apex, 6°3 in. Vertical height, right orbit, 2°1 in.
Diameter of ditto from post-orbitary process to anterior border (obliquely), 2°7 in.
Height of septum from upper surface of incisives to nasal arch, at one inch from
premolar, 2°5 in. From tips of nasals to suborbitary (posterior orbit) apophysis,
about 15: in. Interval between inner borders of glenoid surfaces, 6° in. Width
of zygomatic arches outside, in line with anterior boundary of temporal fossa, left,
(end of last molar), 10°4 in. Width of ditto at middle, 115 in. Greatest width
in line of glenoid surface, 12°2 in. Height of frontal chord at middle of frontal
horn (chord stretches over apex of horn), 1°5 in. Height of frontal chord behind
ditto, 2: in. Height between-horns in middle, 1°8 in. Height in line with posterior
boundary of temporal fossa, 1:1 in. Height of chord from middle of occipital crest
to smooth surface at posterior boundary of front horn, at middle, 0°55in. Height
of chord from ditto to between horns, -45 in. Height from ditto to behind the
horn depressed (broken?) 1:3 in. Width of maxillary over last premolar, 6°7 in.
Width of ditto at commencement of zygomatic arch, 9°7 in. Greatest width of
zygomatic arches, 13-2 in. Greatest thickness of nasals to salient point of disc knob,
2°9in. Medium thickness of ditto to base of conical knob, 2°15 in. Height
of septum from tuberosity in front and below to edge of nasals, near tips, 3°3 in.

Lower Jaw (see Plate XXVII).—Entire length of jaw, from posterior margin
of ascending ramus to symphysis, 19°25 in. Height of ascending ramus to top
of coronoid, 10° in. Breadth of ascending ramus, 5'4 in. Length of line of
molars (six last), 8°56 in. Length of three last molars, 4:9 in. Length of three
premolars, 3°5 in. Length of last true molar, 1°55 in. Length of penultimate
ditto, 16 in. Length of antepenultimate, 1:5 in.

Florence, 20th May, 1859.

The Florence Museum also contains a palate specimen of a young
Rhinoceros Etruscus, showing on the right side the four milk molars
emerged, of which the first three are very slightly affected by wear,
the fourth is hardly emerged from the gum, and is in a state of germ.
The second and third have each a small intercolumnar tubercle, but no
basal c’ngulum sweeping round the inside of the barrels. On the left
side there are only the first and second milk molars, with the anterior
part of the third.

Dimensions.—-Length of the four teeth, 5:7 in. Length of first, 1-in. Width
of ditto, -08 in. Length of second, 1°5 in. Greatest width of ditto, 1°3 in.
Lergth of third, 1:8 in. Width of ditto, 1:6 in. Length of last, 1:9 in.

—.-

RHINOCEROS ETRUSCUS. 359

Another fine palate specimen in the same Museum is a little more
advanced in age, showing on the left side the four milk molars, in
place, and all more or less worn, together with the germ of the first
true molar not out of the gum. On the right side there are only the
last four of these teeth. The three anterior milk molars are worn
nearly in the same degree ; the first, being the least worn, shows three
distinct fossettes ; the second also shows three fossettes, the middle one
of which is caused by the confluence of the ‘ crochet’ with the outer
combing plate. Both these teeth show an intercolumnar tubercle, and
the crochet forms a very open angle with the hind barrel; the same is
the case with the last milk molar, which shows no intercolumnar
tubercle. None of these milk molars have any internal basal cingulum ;
the intercolumnar tubercle is most pronounced in the antepenultimate
or second.

Dimensions.—Length of four milk molars, 5°8 in. Length of first, 1:1 in.
Length of second, 1:5 in. Lengthof third, 1:7 in. Width of ditto in front, 1-7 in.
Length of fourth, 1:9 in. Width of ditto im front, 1-7 in. Length of first true
molar, 2° in.

All these specimens are labelled ‘Rinoceronte a parete internasale, ou
Rhinoceros tichorhinus, Cuvier.’ !

TV.—Memoranpum oF Remarys or Ruinoceros ETRuscus, ETC., IN
Museum art Pisa. (See also Plate XXV. figs. 5, 6, and 7.)
22nd May, 1859.

The cast of the skull of the Rhinoceros with the partial septum is not
of R. hemitechus,? but of the Val d’Arno species (22. Etruscus). The
origmal, which has since been much mutilated, is still preserved in the
Florentine Museum. The cast is wonderfully perfect in what concerns
the septum, which is distinctly limited to the anterior half, and termi-
nates in a thickened portion united to the incisive bone. (See Pl.
XXVIII. fig. 1.)

The posterior part of the skull is wanting. On one side there are no
teeth, but on the other the premolars and one molar remain. The teeth
are worn low, but in the remaining molar the crochet is thick, and at
somewhat of an acute angle. There is both a nasal and a frontal horn,
and the nasal dise is very rugous. Saw also several lower jaws of Rhi-
noceros, some of them evidently of the‘ R. Valdarnensis.’* Another,
much larger, and said by Prof. Meneghini to be from the Val d’Arno,
is certainly of another species, and probably of 2. megarhinus.

Pisa, 1st June, 1859.

Examined a very fine specimen of the right ramus of lower jaw of
Rhinoceros. The six last molars are in place, and the posterior five are
entire; the crown of the anterior molar is broken off. The ascending
ramus is broken vertically through the sigmoid echancrure, so that the
condyle and angle are missing, but the coronoid is perfect to the very
apex, and compares beautifully in its greater dimensions, especially in
breadth, with that of Rhinoceros Etruscus. The coronoid rises very
vertically.4 The teeth are all emerged and are very perfect; the cres-

1 See antea, p. 314.—[ Ep.]

? As stated in a previous note, and at page 332—[Ep.] *° R. Htruscus.—[Ep.]

4 Dr. F. seemed to infer that this was the lower jaw of 2. megarhinus. See
page 356, line 13; and page 369, line 6.—[Eb. |

360 RHINOCEROS.

cents of the first true molar are still distinct; those of the last are but
slightly affected by wear. The specimen was found in the Collines of
St. Regolo.

Dimensions.—Total length of specimen, 15° in. Length of line of 6 molars, 9°6
in. Length of ditto of 3 premolars, 4:1in. Length of 3 last molars, 5:7 in.
Height of jaw under penultimate premolar, inner side, 3-2 in. Height of ditto
under penultimate molar, inner side, 3°9 in. Height to apex of coronoid, 10° in.
Width of apex of ditto, at sigmoid, 1-7 in.

V.—NOoTE ON A SPECIMEN OF RuINOcEROS ETRUSCUS, BELONGING TO THE
MarcuHEsE CarLo StRozzI.

Leghorn, 2nd June, 1859.

This is a magnificent specimen of a symphysial portion of a lower
jaw with part of the two rami. The rami are broken obliquely, so that
only the fangs of two molars are seen in the section. The incisive
border is obtusely bifid, with a very pronounced sinus above and behind
each of the lobes. There is a narrow alveolar pit, as for an incisor that
has dropped out. The symphysial portion is very carinate below, and
is completely drilled by large mentary holes, nine on right side and
seven on left. Seven of the nine holes on the right side are close to-
gether. This is an invaluable specimen.

Further Note on same Specimen—1860.

Mr. Dinkel’s drawing is good (See Pl. XXVIII. figs. 2, 3, and 4).
It shows on the right side the fangs of the anterior premolar, and of the
next adjoining tooth. Mr. Dew’s cast! is chiefly defective in the great
size he has given to the incisive pits, especially on the left side, both in
length and in antero-posterior diameter; the cast also makes them un-
symmetrical, which they are not. Dinkel’s drawing represents the pits
accurately. They are evidently the pits of a small shed incisor.

Dimensions.—Extreme length of fragment, left side, 7-3in. Length of diasteme,
right side, 2°65 in. Length of symphysis, at middle, 4°3 in. Width of symphysis at
middle of diasteme, 1°75 in. Greatest width of ditto at protuberances below, 1:85
in. Width of ditto at incisive pits, 1:4 in.

ViI.—Descrietion or Upper Jaw or Ruinoceros Erruscus, FROM
MataGa.

The specimen consists of the greater part of a right upper maxillary
bone, comprising in situ the second and third premolars, and the three
true molars. The last premolar (p.m. 4) is wanting. The specimen
has been fictitiously repaired with cement, placing all these teeth in
series, without allowance for the missing premolar, and it is in conse-
quence deceptive at first sight. The outer border of the crown is more
or less damaged in most of the teeth. Together with Mr. Waterhouse,
to whom I referred the fossil, I was at first led to believe that it be-
longed to the miocene Aceratherium incisivum of Kaup, from its close
general resemblance to the specimen figured by De Blainville in the
‘ Ostéographie’ (hinoc. Pl. XII.), under the name of Rhinoceros inci-
sivus d'Auvergne. But I have since arrived at the conclusion, after a
fresh examination of the Tuscan collections, that the Malaga Rhinoceros
is the Rhinoceros Htruscus, so named by me from its prevalence in the
Pliocene deposits of the Upper Val d’Arno. This form has hitherto
been confounded, on the one hand with Rhinoceros tichorhinus, and on
the other with . leptorhinus of Cuvier. It had a bony nasal septum,

1 Now in British Museum.—[ Eb. ]

i

DESCRIPTION OF PLATE XXVIII.
RuINOocEROS ErrRuscus.

Fig. 1. Is a profile view of a cast of a skull of the Val d’Arno Rhino-
ceros in the Museum at Pisa, showing the septum distinctly
limited to the anterior half of the nasal bones and terminating
in a thickened portion united to the incisive bone. The figure
is one-fourth of the natural size, and has been copied from a

drawing executed for Dr. Falconer by Pierucci, artist at Pisa.
(See page 359.)

Figs. 3, 4, and 5. Symphysial portion of the lower jaw, with part of
the two rami belonging to the Marchese Carlo Strozzi, and
described at page 360. The figures are one-half of the natural
size, and have been reproduced from drawings by Mr. Dinkel.
Fig. 2. Upper surface. Fig. 3. Under surface. Fig. 4. Lateral

view.

VOL. If.

Vol. II. Plate 28.

J.Dinkel & Pierucci del. Rhmoceros fitruscus. W.West imp.

RHINOCEROS ETRUSCUS. 361

as in the Clacton form, described in the ‘ British Fossil Mammalia,’
under the designation of Rhinoceros leptorhinus, from which, however,
it is essentially distinct in every detail throughout the construction of
the skeleton.

The true Rhinoceros leptorhinus of Cuvier, founded upon the Cortesi
cranium, had no ossified nasal septum, and is distinct alike from the
species here called Rhinoceros Etruscus, and from the fossil Rhinoceros
of Clacton. I have ascertained that the character of an ossified nasal
septum was common to three European fossil species of Rhinoceros,
of the Pliocene and newer Pliocene periods; and that there is only
one known species of this category in which it was wanting. The
characters of these species, and their distribution over the European
area, will be described in detail in a separate essay.—H. F., Oct. 1859.

[The above description appeared as an appendix to a paper by Professor Ansted
in the ‘Quarterly Journ. Geol. Soc.,’ for Feb. 1860. The maxilla with portions
of yertebrz were found a few miles from Malaga in white marl, overlying
Pliocene blue clay, abounding with shells. The following details of a comparison
of the specimen with others in the British Museum is extracted from Dr. Fal-
coner’s Note-books,—Ep. |

British Museum, 16th August, 1859.

Brought with me to-day Ansted’s specimen from Malaga, and com-
pared it again with:—1. Kaup’s Acerath. incisiv., a cast of the old
palate figured in the ‘Oss. Foss. de Darmstadt;’ 2. Kaup’s cast of
entire cranium of ditto; 3. De Blainville’s Rhinoc. incisiv. of Auvergne,
cast figured in ‘ Ostéogr.,’ Pl. XII; 4. Lartet’s Rhinoc. Simorrensis ;
5. Duvernoy’s Rhin. pleuroceros, cast; and 6. Lartet’s Rhinoc. bra-
chypus, Acerath. Goldfussi—all Aceratheria.

Observed the following constant characters:—1. In Acerath. Gold-
Jussi, the last molars even have a basal bourrelet all round, most
strongly marked in the penultimate.

2. In all the Aceratheria, the base of the crown outside presents
an angular bulge, a rudiment of what is seen in Palwotherium. This
is very strongly marked in a beautiful specimen of Lartet’s Rhin.
Simorrensis, a skull with the palate and teeth on both sides (7 on left,
only 6 on right); it is also very strongly marked in Lartet’s Acerath.
brachypus, the British Museum specimen of which is made up of teeth
of different individuals. It is also well marked in the cast of Duver-
noy’s Fhinoc. incisivus of Auvergne, and very marked in the penul-
timate of Kaup’s old palate specimen and in the skull cast.

3. The anterior outer vertical angle and groove are very boldly
defined in all the Aceratheria, and the angular projection is very broad ;
but from that forwards the surface is nearly smooth, and without the
undulated swelling seen in 2h. megarhinus and the Rh. tichorhinus, &e.

4. In Lartet’s Rhin. Simorrensis, which is of an adult with all the
teeth worn except the last, and is in the best stage of wear, besides
the projection of the crochet from the back barrel, there is a constriction
of the anterior barrel, which when worn forms a well-marked emargi-
nation, so that a lobe of the anterior barrel projects into the valley like
a kind of anterior crochet; but overlapped by the true crochet, i.e.
nearer the inside. The same thing is observed in the penultimate and
antepenultimate of Kaup’s cranium of Aceratherium, in his old palate
specimen figured in the ‘Oss. Foss. de Darmstadt,’ and in De Blain-
ville’s Rhinoceros of Auvergne—.e. in the last premolar and penul-

362 RHINOCEROS.

timate true molar. This anterior crochet survives when the true crochet
is worn out; thisis seen in De Blainville’s drawing of the penultimate,
which shows a kind of trefoil to the anterior disc.

5. In Acerath. Goldfussi, the posterior tubercle forms a long crenu-
lated ‘ gradus,’ most salient at the outer end ; the same is seen in Lartet’s
R. Simorrensis and in Kaup’s Aceratherium. The ridge is confluent
inside, free and high outside (like the bourrelet in the Mastodons.)

6. The mouth of the valley of the last molar is very open, and will
admit the forefinger easily.

Compared the Malaga specimen, after making these observations,
and remarked the following peculiarities :—

1. The last true molar behind has only a moderate tubercle, as in the
‘Tuscan specimens, and has no ‘ gradus’ ridge at base behind.

2. The mouth of the valley is comparatively narrow; in the last
molar it will not admit the finger as in De Blainville’s Auvergne
specimen ; the anterior barrel is broad and has a crochet constriction.

3. Unfortunately the apex of the outer ridge-summit of the crown is
broken in the three last molars, but what remains of the low crown
presents an undulated surface.

4. There is no true constriction of the anterior barrel, which in the
antepenultimate is very broad.

5. There is a duck’s bill pattern to the termination of the posterior
valley, with an accessory plate forming a reniform outline, as in Acera-
therium, but no subdivision of the crochet into plates in any of the teeth.

6. The most important and marked difference is that the second
premolar (p.m. 2) has no disc of pressure in front—no p.m. 1! p.m. 8
has two fossettes and the anterior inner cone (barrel) is isolated all
round by a deep fissure and gives a narrow ovate disc.

7. There is a basal bourrelet to p.m. 2 and 3, but not very marked.

8. The basal bourrelet to the premolars of the Auvergne specimen
forms actually a sharp raised rim; the bourrelet is very little pro-
nounced in comparison in the Malaga specimen, in which it does little
more than make a bridge between the barrels, while in the Auvergne
specimen it sweeps round the anterior barrel, rising obliquely in the
posterior.

I infer the specimen to be of Rhinoceros Etruscus.

Dimensions.
Ansted’s De Blain-
specimen ville’s
‘om Auvergne

Malaga palate
Joint length of three Jast molars 5 é 5 55 5°55
Joint length of 2nd and 3rd p.m. ‘ . 2°9 2°8
Length, outer edge of p.m. 2 : < 5 - 1-4 1°35
Width of ditto behind : c - : 1°55 16
Length of p.m. 3 F - is : 16 15
Width of ditto, outer . : F 2: 2°05
Length of t.m. 1 in middle . : : 5 5 18 1°55
Greatest width in front . : : Reina 2-1 2°18
Length of t.m. 2, outer . 2°15 PRI
Length of ditto, middle 19 1°85
Width in front at base 3 2°3 2°25
Length of t.m. 3, from tubercle to outer bourrelet 1-9 1-9

heat \

a” ce ell ee ee le

RHINOCEROS ETRUSCUS. 363

VIl.—Descriprion or Cranium witH TrertH, Humerus, Tipia, aAnp
Fisuua, IN THE Museo pi Srorta Naturate DELLA R. Uni-
VERSITA, AT BOLOGNA.

13th May, 1861.

‘Modello in gesso dell’ intera regione palatina delle ossa mascellari,
eolla doppia serie dei molari quasi interi di un grande Rinoceronte
fossile piuttosto giovine, e probabilmente della specie denominata dal
Cuvier Rhinoceros leptorhinus. L’originale dal quale si & cavato questo
modello fu trovato a poca distanza da Barberino del Mugello in quella
stessa localits dove furono rinvenuti gli altri denti e mandibule di
Rinoceronte che si conservano nel Gabinetto sotto i no. 2,381, 3,450,
3,758, regulati dal veterinario di quel paese Signor Onorio Da Bar-
berino. Vedi per il pezzo ora descritto la di lui lettera che si conser-
va nel museo sotto questo numero. La forma onde ottenere questo
modello é stato levata dall’ originale con tutta diligenza dal modellatore
dei Gabinetti Anatomici dell’ Universit’ Signor Giuseppe Astorri.’ '

Description of the original specimen in the Bologna Museum, to which
the above memorandum applies.—Rhinoceros Etruscus, Pl. XXX.

This specimen (represented in Pl. XXIX.) consists of the maxil-
laries on both sides, with part of the zygomatic arch of the left
side, the palate, the palatine echancrure, with the entire series of
molars on either side in the finest state of preservation. The cranial
portion is broken off behind the palatine bones, and all of the facial
part of the chaffron is broken on both sides in a line a little above the
upper margin of the zygomatic arches; the lower boundary of the nasal
echancrure to the bottom is perfect on the left side, and nearly so on
the right. The left suborbitary foramen is distinctly shown ; that on
the right side is broken and concealed by an attached portion of dis-
tinct bone, enveloped in (Sansino) matrix. There remains in the
front of the series of molars about 24 inches in length of the diastemal
beak; but no indication of the descending portion of the nasal septum,
the position of which is occupied by (Sansino) matrix.

The dentition, as regards the age of the molar teeth, is in the most
perfect state to give the dental characters of the species; the ante-
penultimate true molar being but slightly worn, the penultimate less so,
and the last true molar but very slightly affected by Wear. Some of
the crowns are more or less damaged, but what is wanting from this
cause on one side is happily supplied on the other. The teeth belonged
to an animal that was perfectly adult, but not aged; the three last
premolars are beautifully seen on the left side; on the right there is
most happily preserved the alveolus (triple) of the pre-antepenultimate
premolar, which had dropped out, and the antepenultimate at its
front edge shows distinctly the disc of pressure of the fallen tooth. It
is therefore clear that there were seven-molars in the adult state, viz. 4
premolars and 3 true molars. The following are the principal dimen-
sions :—

1 Tl pezzo originale é stato poscia ac- | questo stesso numero nel museo, dove fu
quistato pel Gabinetto dal lodato Signor | depositato in Marzo del 1847, con altre
Da Barberino pel tenue prezzo di ro- | ossa fossili scavate nella stessa localita.
mani scudi quattro, e si conserva sotto

364 RHINOCEROS.

Extreme length of the line of 6 molars on the left side, measured from the base
outside of the penultimate p.m. to the posterior boundary of the rudimentary pit at
base of Jast molar, 9°1 in. Length of ditto on right side from anterior margin of
alveolus of dropped first premolar to posterior boundary of last true molar, 9° in.
Length of last three premolars, left side at top of crown, outside, 4-3 in. Length
of three true molars to posterior boundary of last, left side, 5-4 in. Length from
the antepenultimate p.m., right side, to the posterior margin penultimate true
molar (to correspond with the Pisa cast), 7°7 in. Length of antepenultimate p.m.
left side, top of crown outside, 1°35 in. Extreme width at base of ditto, behind,
16 in. Length of penultimate ditto ditto, left side, 1°55 im. Greatest width of
ditto in front at base, 2° in. Length of last premolar in front at left side, 1°6 in.
Greatest width of ditto in front at base, 2:1 in. Length of crown ‘of penultimate
true molar (1°95, right side), 2° in. Greatest width of ditto at base in front, right,
2-2 in. Length of crown of ditto at base inside, 1-5 in. Length of crown of
penultimate true molar, right, outside, 2° in. Length of crown of ditto, inner side,
at base, 15 in. Greatest width at base in front of ditto, 2-2 in. Antero-posterior
diameter last true molar, left side, from anterior bourrelet, in front, to posterior
boundary of basal valley behind, 1-8 in. Transverse diameter of ditto, in front,
at base, 2:1 in. Interval between the anterior barrels of antepenultimate pre-
molars, 2°05 in. Interval between anterior barrels of last premolar at base, 2°8
in. Interval between anterior barrels, first true molar, 3-in. Interval between
anterior barrels of penultimate true molar, 2°1 in. Interval between anterior barrels
of last molars, 2°85 in.

Mzmo.—The above dimensions give the width of the palate.

Length of diasteme in front of pre-antepenultimate premolar, right side, 1-75 in.
Interval between the diastemal ridges in front of first premolar, 21 in. (These
comprise the principal dimensions of the teeth.) Height of zygomatic arch, left
side, 1‘8in. Width of zygomatic fossa, left side, 3°5 in.

Description of the Teeth on right side-—There were 4 premolars.
This is distinctly shown on the right side by the triple fang-pits of the
pre-antepenultimate or p.m. 1, viz. one in front and two separate ones
behind: they are more or less filled up.

P.m. 2, the antepenultimate premolar, is quite entire on both sides,
and in nearly the same stage of wear. The discs of the two imner
barrels are distinct, and nearly of the same size; the anterior barrel does
not form an isolated compressed cusp-shaped cone, as in Gervais’ draw-
ings of R. leptorhinus. 'The disc forms a very compressed oval, which
is not confluent with the outer longitudinal disc. The disc of the
posterior barrel is wider, and it is connected by an isthmus with the
disc of the outer ridge, forming a kind of gourd-shaped outline. The
disc of the outer longitudinal ridge is not much advanced in wear, being
where broadest but 04. The posterior valley is nearly quadrangular
in form and well defined, the posterior boundary being quite intact.
The great middle valley forms a large triangular fissure, into which
crochet processes are intruded from behind forwards. There is a
distinct cingulum to the base at the inside, but not in strong relief,
not so much so as the anterior talon. The outer surface of the crown
is convex, antero-posteriorly.

P.m. 3, the penultimate premolar, right side. This tooth resembles
in form the antepenultimate, but is larger and more advanced in wear.
The dises of both barrels being confluent with the outer disc, it is much
broader in front than behind. The anterior outer vertical furrow is
well marked, the posterior valley is very much as in p.m. 2. The
great middle valley forms a large fissure which is divided into two
portions by the crochet processes, and an outer accessory plate is in-
truded from the longitudinal ridge; one little ring of enamel is isolated
on the base of the crochet.

Piet sy

i i i le

RHINOCEROS ETRUSCUS. 365

The penultimate premolar is equally perfect on both sides, and in
the same stage of wear. They both show the basal bourrelet round
the inner barrels, but not very pronounced.

P.m. 4, or the last premolar, has the anterior outer angle of the crown
broken on the right side; it is beautifully perfect on the left, which
shows the crown but very slightly advanced in wear ; the discs of both
barrels are confluent with the outer disc. The posterior valley is well
defined and intact behind; the anterior transverse valley has intruded
into it a large crochet process, and two large accessory plates (or
combing processes), proceeding parallel to each other from the outer
ridge, and converging towards the crochet. A distinct ring of enamel
isolating a pit is situated on the base of the crochet, the whole causing
a complex pattern to the convolutions of the transverse valley. Fine
parallel and wavy grooved lines of enamel are beautifully shown on
the inner surface of the enamel. This tooth, like the others, shows a
distinct basal cingulum ; it is more triangular in form than the two
which precede it.

T.m. 1, the first true molar, is quite perfect on the right side; on the
left side the posterior barrel is broken on its inner surface. The crown
is more advanced in wear than any of the others, but still not very
much so, being not yet half worn. The posterior valley is quite intact
behind, but is narrower and more vertical than in the premolars. The
transverse valley is divided into two nearly distinct portions by a very
thick crochet, protruded from the posterior barrel; the outer division
has no accessory plates intruded into the fissure from the outer longitu-
dinal ridge; the inner division forms a narrow triangular fissure. The
crochet is emitted at a very open angle from the posterior barrel, more
open even than in &. leptorhinus, and totally different from that seen
in &. hemitechus. There is a little basal mammilla between the barrels
at the inside, but not a trace of an anterior basal bourrelet to the
anterior barrel. The teeth are very much alike on both sides. The
central termination of the middle valley does not exhibit the duck’s
head pattern, figured by Gervais and De Christol in the teeth of ZR.
megarhinus. There is a little tendency to the peculiar twist of the
posterior barrel near the apex of the crown ; the anterior outer vertical
groove is broad, but shallow; the angle boldly overlaps the last pre-
molar.

T.m. 2, or the penultimate, on the right side, is nearly perfect, but the
outer anterior angle is broken off vertically on the left side. The tooth
in general form resembles very much the antepenultimate just des-
cribed, but is less advanced in wear; the crochet is also, as in it, emitted
at an open angle. ‘The transverse valley is divided in two by the
crochet, the inner division being triangular, without any accessory
plates or complication whatever. The summit of the posterior barrel
has the peculiar compressed contortion well marked. The crochet
advances nearly into contact with the anterior barrel; the discs form
narrow bands of wear, which are confluent throughout. There is not
a trace of a basal cingulum on either side.

T.m. 3, or the last true molar, is broken partly on both sides, but in
different directions, so that what is wanting in the one is supplied by.
the other. The crown is but very slightly affected by wear on the right
side. It is of a distinct triangular form, all the parts converging to a

366 RHINOCEROS.

contracted summit. The anterior barrel has a distinct basal bourrelet,
which is wanting in the posterior. The transverse valley is divided
into two parts by a crochet, advancing on the right side to meet an
accessory plate emitted from the anterior barrel. On the left side,
these two plates overlap. On the right side, an accessory plate is also
given off from the outer ridge, converging towards the crochet. The
most striking character about the tooth is, that, asin R. hemitechus,
there is a distant rudiment of a posterior valley restricted to the base,
but not forming a well-defined cup with a distinct rim, as in that
species. This rudiment is distinctly shown on both sides, bounded
posteriorly by a basal cingulum. The basal bourrelet behind the
posterior barrel of the last true molar has barely emerged above the
alveolar margin.

The enamel is smooth in all these teeth, and marked by beautiful,
fine, wavy, horizontal lines. There is not a trace of general superficial
rugosity, and not the slightest indication of a layer of cement.

The outer surface of the enamel is traversed in a dendritic fashion,
by fine channels, like those which are attributed to the work of Marine
Sponges, but the formation out of which this specimen came is fresh
water,

The bottom of the palatine echancrure comes in a line with the
middle of the posterior barrel of the penultimate true molar, and the
suborbitary foramen is immediately over the line of junction between
the penultimate and last premolars. The disc of the crochet of the
penultimate true molar is nearly as broad as that of the posterior
barrel, the crochet being very thick and simple. The anterior outer
vertical furrow is well pronounced in all the molars, from the penulti-
mate premolar to the last true molar inclusive. It is wide and shallow,
but the other vertical hollows are but very slightly pronounced by an
undulation of the surface. There is not the slightest indication of an
outer basal bourrelet, as seen in the Aceratherium incisivum of Kaup,
the outer surface being smooth, and nearly vertical throughout.

Besides the above, there are casts in the Bologna Museum of several
of the principal specimens of Rhinoceros figured by Nesti, and a dupli-
cate cast of the Targioni Tazzetti cranium of the Florence Museum,
made by Savi for Pisa, and of which I got drawings. There are also
casts of the following bones:—A humerus, left side, very closely re-
sembling the figs. 1 and 2 of Cuvier’s Pl. X. Rhin., but more perfect, of
which the following are the dimensions :—

Extreme length from top of tuberosity to tuberosity of outer condyle, 16:25 in.
From articulating head to middle of inner condyle, 14: in. Width of articular
surface of condyles, 3-4 in. Greatest width at inferior end, 5-2 in. Antero-
posterior diameter of inner condyle, 4:1 in. Greatest width of shaft at middle of
median tuberosity, 5-1 in. From sinus, at lower margin of middle tuberosity, to
top of great tuberosity, 7-2 in. Greatest constriction of shaft, below middle tube-
rosity, 2.2 in. Antero-posterior diameter of articular head and tuberosity, 4: in.
Transverse diameter of articular head, about 8-7 in.

Specimen of tibia and fibula, figured by Cuvier, Pl. XI., Rhin., fig. 15.

Extreme length of tibia at middle, about 14: in. Transverse diameter of upper
articulating surface, 4°4 in. Antero-posterior ditto to inner margin of inner
articular surface, 4°8 in. Transverse diameter, lower articular head, ineluding
fibula, 4:3 in. Antero-posterior diameter, inner articular cup, lower end of tibia,
2-8 in. Extreme length of fibula, 12:25 in, Transverse diameter of shaft of tibia,
at middle, 2°3 in.

DESCRIPTION OF PLATE XXIX.
RaINOcEROS EtRuscus.

This Plate represents the palate view of the cranium in
the University Museum of Natural History at Bologna, de-
scribed at page 363. The drawing is one-half of the natural
size, and has been copied from one which Dr. Faleoner had
executed at Bologna, and on which he had inscribed ‘ Rhino-
ceros Etruscus, Museum, Bologna.’ <A cast of the specimen
which Dr. Falconer also. brought from Bologna has been

deposited in the British Museum.

WOlse pie

2 Vol. IT. Plate 29.

‘
P|
¥
'

Rhinoceros Etruscus.
(Bologna)

W.West imp,

Ee SE Ne

a ee Be Tay

RHINOCEROS ETRUSCUS. 367

Cast of right femur, figured by Cuvier, Pl. XI. Rhin., fig. 19, of
which the upper articulating head is wanting

Extreme length taken at the middle, 16°50 in. Antero-posterior diameter of
inner condyle and pulley, the latter partly broken, 6-in. Length of pulley in
middle, 2°5 in. Transverse diameter of ditto, 2.5 in. Least transverse diameter
of shaft below the middle trochanter, 2°5 in. Vertical height of neck of middle
trochanter, 2:1 in. Transverse diameter of shaft, including middle trochanter, at
middle of ditto, 5-2 in. Width of bone at middle of sinus above middle tro-
chanter, 471 in.

All these bones belong to Rhinoceros EHtruscus, and there are still
preserved in the Bologna Museum, the originals of the specimens
represented by Cuvier, figs. 5 to 10, inclusive of Pl. X., Rhin., of ‘ Os.
Fossil.’ These are the upper and lower extremities of a humerus of
the same species (2. Htruscus), stated to have been procured by the
Ab. Ranzani in France.

VIIL—Descriprion or Specimens oF Rurnoceros Etrruscus at Le Puy.

Le Puy, 15th September, 1863.

In the Museum of Le Puy there is a magnificent series of remains of
the skeleton, consisting of three feet, with all the bones en suite to the
terminal phalanges—the tibia, fibula, astragalus, and articular head of
the femur. The shaft of the femur with the third trochanter is exactly
as in Pentland’s specimen in the British Museum. There are also two
detached calcanea, both of the left side, and one astragalus. All are
from Solilhac.

In the same Museum there is also a series of the molars of R.
Etruscus, six right and left, but detached and separate, with the last
molar just coming into use, and in the finest condition for figuring.
They are from Vialette. In addition there is a superb specimen of the
left ramus of the lower jaw of &. Etruscus, having the four last molars
en suite, all a little worn, and the hind portion of the penultimate pre-
molar; the teeth are in a beautiful condition to be drawn.

There is also the muzzle of the lower jaw of R. Htruscus, perforated
below exactly like the specimen of Carlo Strozzi (see page 360), and
with the empty pits of two small median incisors more round, more pro-
nounced and less angular than in Strozzi’s. There is no keel below,
as in the Florence specimen. The specimen is red, heavy and fer-
ruginous, and in the same mineral condition as the Vialette lower jaw.

Dimensions of Lower Jaw, left, from Vialette——Length of fragment, 11°5 in.
Length of series of four molars, 6°75 in. Length of last true molar, 1°75 in.
Length of penultimate, 1:65 in. Length of antepenultimate, 1°62 in. Length of
last premolar, 1:50 in. Height of jaw at posterior edge, last molar, 3°35 in.
Height at anterior edge of first true molar, 2°9 in. Height at anterior edge of last
premolar, 2°65 in. .

The inferior border is perfectly straight along the three true molars;
there is no curve, but there is a strong longitudinal channel along the
middle of the ner side; and to each of the anterior barrels of the four
molars there is an oblique descending bourrelet, strongly marked. The
jaw is truncated along the ascending ramus and in front. The specimen
ought to be figured.

15th September, 1863.

Masel, near Le Puy, with Messieurs Pichot, Robert, and Lartct. Jaw

of Rhinoceros Etruscus found by M. Pichot at Sainxelle, near St. Aune.

368 RHINOCEROS.

There is also a magnificent head, very well preserved, of Rh. Htruscus,
with the series of molars (six) of the two sides present. The anterior
portion is entire, and also the bony wall of the nasal partition. The
two jaws are slightly broken, and likewise the orbit of the left side.
The occipital portion, as well as the condyle, is wanting. Theage of the
dentition is that which best shows all the characters, the last true molar -
being very little worn. The three premolars are much affected by
wear. The antepenultimate has three fossettes; the echancrure of the
first anterior ridge is still apparent, as in the drawing of the Bologna
skull (P1l. XXIX.). The penultimate is less worn and has two fossettes,
the middle one being divided into two parts; and the crochet is serrated,
as in the Bologna jaw. The last premolar of the left side is well worn,
and shows three very distinct fossettes, and the crochet is but little
denticulated. The first true molar is half worn, the crochet is simple
and at right angles, without a combing plate; the median hollow is
quite open on the inner side. The penultimate true molar has nearly
the same form, but on the left side the crochet is confluent with the
anterior ridge, so as to isolate one part of the median hollow which is
situated behind, as in the tooth of Crozes; but on the right side the
crochet isdetached. The last molar is very little worn, with the crochet
free, and a plate projecting from the anterior ridge. In form and size
it perfectly resembles the cast that I have brought from the Museum
at Pisa (Pl. XXV. fig. 5), and the molars (pre- and true-molars) have
a basal crown on the inner side. The length of the series of six molars
is nearly the same as that in the drawing of the Bologna skull (9°8 in.).
The osseous partition and the nasal bones exactly resemble the draw-
ings of the specimens in the Florence Museum, but it seems to both
M. Lartet and myself that the osseous partition is less complete.

The jaw is embedded on the left side in tufaceous greenish grey
alluvium—the ‘ Alluv. inter-volcanique’ de M. Pichot.

Ill. NOTES ON RHINOCEROS LEPTORHINUS (Cov. pro. Parte),
R. MEGARHINUS (Curisrtor).

I.—Descrirtion oF Remains oF Rurnoceros Leproruinus (R.#MEGAR-
HINUS) IN THE Museum aT MONTPELLIER.

18th November, 1858.

Examined the original of the fine lower jaw of FR. megarhinus figured
by Gervais, and also another lower jaw of the same species more per-
fect at the muzzle, but mutilated behind. The former is double, and on
the right side comprises the whole of the ramus from the tip of the
incisive margin on to the condyle and coronoid, the apex of the coro-
noid being alone wanting. On comparing it with Dinkel’s drawings of
Rk. hemitachus, observed the following points of difference (See Pl.
XXX.) :—

1. The lower edge of the horizontal ramus is nearly a straight line
from the angle on to the anterior edge of the first true molar.

2. The low elevation and great thickness of the body of the ramus.

3. The horizontal line (still slightly concave) of the plane of dentri-
tion (very concave in &. hemitechus).

DESCRIPTION OF PLATE XXX.
RHINOCEROS LEPTORHINUS (R. MEGARHINUS).

Three different views of lower jaw, one-fourth of the
natural size. Fig. 1. Inner surface. Fig. 2. Shows crowns
of molars and symphysial spout. Fig. 3. Outer surface.
These drawings have been executed by Mr. Dinkel from a
cast brought by Dr. Falconer from Montpellier, labelled
‘Rhinoceros des Sables de Montpellier,’ and now deposited
in the British Museum. (See page 368.)

VOL. Il.

Biss.

Vol. I. Plate 30.

aA ae eS ee ee

—
LN
al

te San alee >

J -Dinkel del et ith Rhinoceros leptorhinus (Rmegarhinus) ‘W.West imp.

eS

RHINOCEROS LEPTORHINUS. 369

4, The great lencth of the diasteme; the distance from the middle
of the incisive border to the anterior edge of the antepenultimate pre-
molar being exactly equal.

5. The absence of reclination in the anterior border of the coronoid.
Tt makes an open curve below with the ramus, but the superior part is
nearly vertical.

6. The posterior angle projects behind the neck of the condyle, and
is puckered. (The figs. in 5 and 6 of De Christol’s thesis very good ;
his 2. tichorhinus.)

7. Very long diasteme with sharp raised edges and great constriction
of the gutter between them, and then a spathulate expansion in front
towards the incisive border; the anterior portion is curved, and throws
out a step, but the form is very different from FR. hemitechus, and there
is nothing of the scaphoid character in the general contour below.

Dimensions.
R. megar. Col. Wood's
No. 2 R. hemitcechus
Length of 6 Jast teeth . 6 : ; 10°5 10°
Length of 3 (last) true molars. : : 6° 6°25
Length of 3 premolars . c 45 38
Length of diasteme to incisive border 4-4

In both of Gervais’ specimens the teeth are adult: 7.e. the last are
partly worn and the antepenultimate true molar is ground down to a
disc. In jaw No. 2, the larger, there are only six teeth (less perfect in
the other). In No. 2 there is also a very distinct outer included incisor
on the right side, with the alveoli of two middle ones nearly filled up.!

SkuLL or Rutnoceros LEPTORHINUS IN THE
Museum at Lyons.

IJ.—MEASUREMENTS OF

[On his way to Italy, in 1858, Dr. Falconer was presented by Prof.
Jourdan with the cast and an unpublished lithographic engraving of a
skull of Rhinoceros in the Nat. History Museum at Lyons, with the fol-
lowing inscription: ‘Téte de Rhinoceros megarhinus des sables d’eau
douce trouvée par M. Jourdan le 12me Févr. 1856, 4 Lens Létang, prés
Moras, Drome.’ This cast was subsequently compared with the Rhino-
ceros remains in the various museums of Italy, in the descriptions of
which it is frequently referred to. There is no description of the skull
in Dr. Falconer’s Note-books, but the skull and molar series have been
reproduced by Mr. Dinkel in Pl. XX XI. figs. 2 and 3, and I am indebted
to Mr. W. H. Flower, F.R.S., for his assistance in taking the following
measurements of the cast, which is now deposited in the British
Museum.—Ep. |

1, Extreme length of skull from summit of occipital crest to point of pre-
maxillary bones, 25°5 in. 2. Extreme length of ditto from posterior plane of
occipital condyles to broken edge of left diasteme, 23°3 in. 5, Extreme length of
ditto to anterior edge of alveolus of 2nd premolar, 224 in. 4. Length from
anterior border of right orbit to outer edge of oecipital plane, left side, 15-2 in.

© 5s ie kins

' Dr. Falconer’s Note-book also con-
tains a description of a mutilated skull
of R. megarhinus (sic) in the Bishop's
_ Palace at Montpellier, examined by him
on Noy. 21st, 1858. Nearly the whole of
the nasal sinus was filled with pebbles

VOL, Il.

and gravel, so that it was impossible to
be certain whether or not there was a
septum, but Dr. F. was inclined to agree
with De Christol and Gervais in think-
ing there was none.—[Ep.]

BB

270 RHINOCEROS.

5. Length from posterior plane of occipital condyles to posterior margin of last
true molar, 12°5 in. 6, Diameter between outer margins of occipital condyles,
45in. 7. Transverse diameter of left condyle, taken near middle, 1:5 in. 8.
Vertical height of ditto, 2-9 in. 9. Diagonal diameter of ditto (greatest), 3° in.
10. Greatest width of occipital foramen, 1°7 in. 11. Height of occipital plane to
lower surface of occipital condyles, 9°5 in. 12. Greatest width of occipital plane,
just above the condyles, 6: in. 13. Greatest width of ditto about middle, 5-2 in.
14. Length of zygomatic fossa, left side, 7-4 in. 15. Length from posterior
boundary of zygomatic fossa to posterior surface of left occipital condyle, 7: in.
16. Length from anterior margin of auditory foramen to anterior margin of the orbit,
10-5 in. 17. Extreme length from anterior margin of 2nd premolar to posterior edge
of last true molar, left side, 10°56 in. 18. Length of last 3 premolars, left side, 4°7
in. 19. Length of 3 true molars, left side, 6-2 in. 20. Extreme length of 1st
and 2nd true molars, left side, 4°25 im. 21. Length of 2nd premolar, left side,
1:55 in. 22. Transverse diameter of ditto near base, behind, 1:7 in, 23. Antero-
posterior diameter of 3rd premolar, left side, 1:6 in. 24. Transverse diameter of
ditto at base, anterior barrel, 2° in. 25, Antero-posterior diameter of last pre-
molar, outer surface, 1°7 in. 26. Transverse diameter of ditto at base, anterior
barrel, 2°25 in. 27. Length of crown of first true molar, outer surface, left, 2° in.
28. Transverse diam. of ditto at base, anterior barrel, 2°6 in. 29. Antero-posterior
diameter of penultimate molar, anterior surface, 2°3 in. 30. Transverse diameter
of ditto anteriorly, 2.5 in. 31. Antero-posterior diameter of last true molar
(greatest), 2:1 in, 82. Transverse diameter of ditto anteriorly, 2-4in. 33. In-
terval between diastemal ridges at 2nd premolar (inner surface), 1-2 in,

IIJ.— Note on Ruarnoceros Lerroruinus From ILForD.

British Museum, 13th August, 1859.

Compared the cast from Montpellier of the last upper true molar
with a specimen labelled ‘Tooth of Rhinoceros from Ilford, Essex’
(No. 40,482). They are both nearly of the same size and stage of
wear and exhibit exactly the same pattern. ‘The Ilford tooth shows
still a kind of vertical cleft for the posterior valley, and a very thick
layer of cement in the valley.

TV.—WNorte on Mortars or Rurtvoceros Lertorninus? From THE BoNnE
Breccia oF NICE, FILLING A CAVERN IN THE JURA LIMESTONE.

Nice Museum, 11th December, 1858.

Of the Rhinoceros the finest specimen is a sixth or penultimate
upper molar of the left} side (PI. XXXII. fig. 3), but very slightly
advanced in wear; unluckily the anterior outer angle is broken off, as
far as the middle of the great valley ; but the tooth shows in section
the step of the anterior external vertical groove very pronounced, the
whole of the great middle valley, the anterior basal bourrelet, the pos-
terior valley sheeted over with a very thick layer of cement, the ante-
rior and posterior barrels inside entire to the apex, and the crochet
quite entire. The enamel is rugous on the outer surface, with vertical
striz, but hardly so much so as ordinarily seen in &. tichorhinus, and
the enamel is not so thick. Both the anterior and posterior barrels are
very much compressed at the apex, as shown in the drawing, and the
crochet is also much compressed, and given off forwards at a very open
angle with the crown of the posterior barrel. The direction of the
crochet deviates but little from a straight line; but the crochet does
net join on to the anterior barrel as in #. tichorhinus, a point of great

1 Left, not right, as identified by Gastaldi,

Fig. 1.

DESCRIPTION OF PLATE XXXII.
RHINOCEROS LEPTORHINUS (R. MEGARHINUS).

Series of six molars of upper jaw, right side, described at page
395. The figure is one-half of the natural size, and has been
reproduced from a drawing found in Dr. Falconer’s collection,
and on which he had inscribed, ‘ Rhinoceros leptorhinus, R.
megarhinus, Christol, from specimen in Municipal Museum of
Imola. Scarabelli.’

. Series of six molars of upper jaw, left side, one-half of the

natural size, copied from a lithograph found in Dr. Falconer’s
collection, and on which he had written: ‘ Unpublished litho-
graph of skull of fossil Rhinoceros belonging to the Lyons
Museum, for a work by Professor Jourdan of Lyons. Ahino-
ceros leptorhinus, Cuv., pro parte, R. megarhinus, Christol.’ The
artist has improved on the original drawing by the assistance of
a cast of the same skull presented to Dr. Falconer by Professor
Jourdan, and which is now deposited in the British Museum.
(See page 369.)

. Represents the cranium of R. leptorhinus, referred to under

fig. 2, one-seventh of the natural size. The drawing has been
executed from the same materials as fig. 2. As in the case of
the Cortesi cranium, the specimen is somewhat distorted from
crushing. (See pages 369 & 381.)

VOL. IT.

Vol. IL Plate 31.
\ Fig. 2.

Xe

JDinkel déL etlith. W.West imp.

Rhinoceros leptorhinus (Cuv,) R.megarhinus (Christol)
(Limola; 2,3.Iyons)

my Pee “ow

RHINOCEROS LEPTORHINUS. 37]

importance. The termination of the middle valley is angular (as in
the sketch) and there is a very pronounced combing process, emitted
from the outer side, and projected across so as to terminate not far from
the ‘end of the crochet. The projection of this plate is much more con-
siderable than is shown in the drawing, and agrees very much with
that of R. megarhinus in Jourdan’s big drawing (Pl. XXXII. fig. 2),
with the allowance for the latter being more advanced in wear. The
posterior valley is very deep, down to the very bottom of the crown. It
is large and lined with a very thick coat of cement. The general contour
of the crown of the tooth is not prismatic, as in &. tichorhinus. The
opposite sides (inner and outer) converge towards each other quite as
much as in R. hemitechus or R. megarhinus, but the crochet forms a
much larger plate than in the latter. The anterior outer vertical groove
is very angular and pronounced, forming a well-marked narrow step,
where seen in the section. No basilar mammilla on the inside between
the barrels. The posterior barrel is narrowed inside, into a kind of
oblique vertical keel, not round and bulging as in the drawing. Besides
the large combing process, the posterior termination of the transverse
valley throws up from below a pillar, which is not laterally attached. It
is represented by the posterior vallicular mammilla in the figure. The
anterior bourrelet is very pronounced and gaping, 7.e. the interspace is
well marked.

The most peculiar character probably consists in the form of the
‘barrels.’ The posterior barrel is constricted about half way up.

Dimensions.—Length of outer surface, 2°3 in. Length at middle along crochet
to outer edge of posterior valley, 2° in. Length, inner side, near base, 1-6 in.
Greatest transverse diameter at base, 2°5 in. Greatest transverse diameter near
top where broken, 1:4 in. Greatest transverse diameter of posterior division
(base), 1°9 in. Greatest transverse diameter of ditto at top, 8 in. The two last
measurements show the amount of convergence.

[In Dr. Falconer’s Note-book a description is given of two other
molars of Rhinoceros in the Nice Museum :—No. 39, a third premolar,
and No. 40, a fragment of the first true molar, left side. Respecting the
latter it is stated ‘there is no indication that the crochet was joined on
to the anterior barrel, so as to form during wear a third pit or well, as
in &. tichorhinus. This confirms the indication of the sixth molar des-
cribed above. Further, there is not a vestige of a basal bourrelet,
although the barrels are not ground so low down as to have caused its
disappearance.’—Ep. ]

V.—Descriprion or Remains or R. Lerroruinus in THE MUSEUM aT
Rome.

April and May, 1859.
A. In Professor Ponzi's Collection from the Gravel Beds of Ponte Molle.

The Rhinoceros remains are much rarer than those of Elephant.
The only specimens are :—

1. Two last true molars, upper jaw, left side. (Plate XXXII. fig. 6.)

2. A penultimate upper premolar, left side, much worn.

3. A finely preserved left upper milk molar. (Plate XXXII. fig. 7.)

BB2

372 RHINOCEROS.

4. A fragment of a left lower jaw with the penultimate and ante-
penultimate true molars.

Of the last true molars, the best preserved, A, (P]. XXXII. fig. 6),
resembles very much in form, size, and amount of wear, the specimen
of R. megarhinus, which I got from Gervais, from the marine sands of
Montpellier. It is deficient only at the outer angle, where the grooved
portion is broken off by a fracture sloping from the inside outwards.
The crown presents the characteristic V-shaped outline. The posterior
barrel is somewhat compressed, and at the posterior inner angle (where
the rudimentary valley of 2. hemitechus is seen), there is a broad ad-
pressed basal tubercle with an obscurely crenated edge, very considerably
larger than the tubercle on the Montpellier cast (Pl. XX XI. fig. 2). This
tubercle leaves a neck between it and the enamel of the crown, filled
up with cement; but there is nothing resembling the pit in R. hemitw-
chus; and there is no decurrent groove ascending from it. The prin-
cipal differences between it and the Montpellier specimen are: Ist,
that the valley is more open, with a wider fissure, from more advanced
wear, and that there are no remains of the crochet process intruding
into the valley (only a sinuous line), nor of a combing plate from the
outer angle causing the ‘ duck’s head’ pattern of the Montpellier crown,
which is seen also in Gervais’ drawings. The termination of the valley
forms a large oval fossa, which contracts into the fissure, that opens
between the barrels. 2nd, that there is a basal tubercle between the
barrels, appended to the posterior barrel. 3rd, that there is a well-
marked layer of cement, which is abundant in the valley near the
intercolumnar tubercle.

The second specimen, B, is nearly in the same stage of wear, but it is
mutilated by a fracture, which has removed a portion of the inner side
of the last barrel, while the outer angle, mutilated in A., is entire. It
corresponds very exactly in form with the other. The posterior barrel
shows the same kind of tubercle, adpressed and near the base, but the
greater part of it is removed by the fracture. The transverse valley is
more contracted in consequence of part of the crochet remaining. The
termination is triangular, and a good deal like Gervais’ fig. 6,! but is
more triangular and has less of the ‘duck’s bill’ pattern. This spe-
cimen very fortunately presents the outer angle entire. It resembles
the Montpellier cast exactly in form, ¢.e. it is broad and salient, with a
well pronounced vertical groove, and the lobes of the emargination sub-
equal, being very different in form from the 2. hemitechus molar ;
where the angle is narrow, the groove shallow, the lobes unequal, the
anterior being much the higher. There is no intercolumnar tubercle
in B., but the barrel is broken at that point.

From the annexed comparative measurements, I am satisfied that
these Roman specimens, including Lyell’s Ponte Molle specimen, are of
the same species, as the Montpellier one, 7c. &. megarhinus of
Christol. Lyell’s specimen shows much cement at the mouth of the
valley, but the outer angle of the chevron is mutilated and rolled, and
the posterior barrel is rolled below where the posterior basal tubercle is
placed. The enamel in all is thin and smooth.

1 Paléontologie Francaise, Pl. iit] Ep. ]

Phi i. 2X

RHINOCEROS LEPTORHINUS. 3790

Comparative Dimensions.

Cast of
Mont- poute Ponzi’s, Ponzi’s
Fas olle, ,
pellier Lyell’s * No, A. No. B.
specimen. y
Antero-posterior diameter of barrels, In. In. In. In.
inner side, at base. ‘ A : 2:15 a1 21 2
Transyerse ditto from base, outer angle,
to anterior border . A 5 : 2-4 2-4 2°3 22,
From ditto to posterior barrel . .| 2°38 2°6 2°6 2°3
(adding nearly,
enamel) but
broken
Width of outer angle at apex 2 5 °65 | broken | broken 6
nearly

* Lyell’s specimen partly mutilated.

The next specimen is a third upper premolar, left side, the crown
worn very low down, so that the basal bourrelet is removed. ‘The
posterior valley is reduced to a ‘round pit, and the central valley to
an isolated fissure, somewhat uniform in outline.

Dimensions.—Width of crown, 1:8 in. Length outer side, 1°45 in. Length
inner ditto, 1:2 in.

N.B. In general form this specimen resembles a good deal (fig 1 of
Pl. LI.!) of the ‘ Ossemens Fossiles,’ which is of the opposite side. In
the latter, which is of R. tichorhinus, and much worn, the posterior fos-
sette is much larger in proportion to the transverse fossette, and is less
round.

The milk molar (Pl. XXXII. fig. 7) is in the finest state of preserva-
tion; it has no fangs; the crown is in the best stage of wear to show
all the characters. The angle formed by the crochet is very open, in
consequence of the obliquity of the disc of the posterior barrel; there
is no basal bourrelet, but a rudimentary intercolumnar tubercle. The
groove of the external angle is deeply marked and broad, and the pos-
terior niche is also broad and well marked, overlapping the next tooth.
The crochet is undivided, and the outer ridge throws off a large soli-
tary combing plate, which is directed parallel to the anterior end of the
crown, and at right angles to the crochet, which it nearly touches.
The barrels are as fully developed as in a true molar, and but for the
small size the tooth would be fixed to be a fourth or last milk molar.
The outer vertical groove is very deep, and its posterior bounding ridge
very high and strongly developed. The posterior valley forms a gaping
triangular fissure, with shelving sides. It is of large size, the posterior
edge intact, and emarginate, so as to form a bi-lobed edge like the
carnassier tooth of a tiger. The transverse valley is very open at its
mouth, forming a triangular fissure. It is then bent gently forwards,
to terminate in the cul de sac. There is a small intercolumnar tubercle,
but not a trace of a basal bourrelet inside. The thick ‘combing plate’
is almost in contact with the point of the crochet. If they had run
together a third fossette would have been formed, as in the premolar.
The opening of the transverse valley is gaping, and the posterior barrel

1 PI, xiii. fig. 1 of hin. 3rd edit.?—[Ep. |

374 RHINOCEROS.

is compressed with the peculiar torsion of R. megarhinus. It is deeply
grooved at the anterior side. The disc of wear points diagonally back-
wards. (This specimen was named R&. leptorhinus by Prof. Owen.)
Dimensions.—Length of crown, outer surface near top, 1‘8in. Length of ditto
in middle, 1°55in. Length of ditto, inner side near base of barrels, 1-1in. Length

of outer surface near base, 1-5in. Transverse diam. anter. end, 1‘7in. Trans-
yerse diam. posterior end, 1°6 in.

The lower jaw specimen is a mutilated fragment of the alveolar por-
tion of the left ramus broken off at one-third of its height. It com-
prises the antepenultimate and penultimate true molar in place, and
the anterior half of the socket of the last true molar behind. The ante-
penultimate is worn in front down to the line of commencement of the
outer anterior oblique bourrelet, and has the discs confluent, but the
posterior crescent is not much worn, less so than in the penultimate of
the Montpellier cast. The anterior end bears a disc of pressure. The
penultimate has the apices in the first stage of wear; the disc of the
posterior crescent is distinct from, and at a lower level than, the an-
terior. The posterior end bears a well-marked pressure surface.

Dimensions.

Montpellier Cast Roman Specimens
Antepenlt.| Penult. |Antepenlt.| Penult.
Length of crown at top ; : . 1:7 1°85 1:9 2:
Width in front . ; : Gi 1:05 11 14 1:45
Greatest ditto behind . , : , 12 1.15 1:45 15
United length of the two crowns . 35 4

Compared with the R. megarhinus cast, they agree in form and pro-
portions, but are much smaller. They have each the same oblique
bourrelet ridge to each barrel (one in front of the tooth, the other
behind) on the outer surface. There is a very thick coat of cement
below. They differ entirely from the square form and thick enamel of
Rk. tichorhinus, with two detached molars of which from Kent’s Hole I
compared them. I was not able to extend the comparison to 2. hemite-
chus, but I believe them to be of R. megarhinus. The specimen is
from Ponte Molle (Professor Ponzi’s ‘ voleanic sand’).

B. In the Sapienza Museum, from Monte Sacro.

Examined a very fine penultimate true molar, upper jaw, left side
(No. 111) of a fossil Rhinoceros from Monte Sacro, showing all the cha-_
racters in the best condition of wear; both the collines and crochet are
worn, but the posterior edge of the hind valley is intact (Pl. XXXII.
fig. 4). The tooth is of very large size, larger than the detached
tooth of R. hemitechus, with which it'was compared.

RHINOCEROS LEPTORHINUS. 375

Dimensions.

Roman R. hemitcechus

Inches Inches
Extreme length of crown, outer surface : : co ORG 2°35
Ditto, inner side. A 4 A ; b ; . 185 2:
Transverse diameter in front, at base of enamel . nl OU 2°35
Ditto ditto, behind ; Z : ’ 22, BD:
Height of crown anteriorly, outer surface se, 17
Ditto ditto, inner “4 : : - . ae lis 0-9

From these dimensions it is seen to be larger than the R. hemitwchus
tooth, and less advanced in wear. In this respect it closely resembles
De Christol’s figure of &. megarhinus.

Compared with R. hemitachus there are the following differences :—
The crochet forms a very open angle with the posterior colline; it is
longer and narrows towards the point. The termination of the valley
throws out from the outer ridge a thick combing plate, which is di-
rected at right angles to the crochet, and divides the cul de sac into
two compartments, as in the milk molar in Ponzi’s collection. The
posterior colline is directed backwards, and has the peculiar torsion
shown in De Christol’s figure. Further, the anterior overlapping sinus is
much more pronounced, and the anterior outer vertical groove is wider
and deeper. The enamel is smooth, with some cement on the outer
side, but the greater part is removed. There are some denticles in the
bottom of the valley, but these have been broken in picking out the
matrix of gravel.

This tooth agrees in every respect with R. megarhinus, and proves
that species to exist in Italy. It has no inside bourrelet, nor inter-
columnar tubercle.

There is another specimen (No. 112) also from Monte Sacro, in the
Mineralogical Gallery of the Museum, of the last successional premolar,
left side, beautifully preserved (Pl. XXXII. fig. 5), and very little
worn (like No. 111). It shows a double crochet plate projected in
front and. downwards, as in De Christol’s drawing (fig. 25), but is per-
fectly free from the combing plate of the outer ridge shown in that
fizure. It is very nearly in the same stage of wear, and has plenty of
cement on the outer surface; the crown is high. There is a well-
marked basal bourrelet on the inner side, proving it to be a premolar,
and a layer of cement above the bourrelet. The second and lower
crochet plate is much smaller than the upper ; the two are very unequal,
and it is also much less worn. There is very little obliquity in the
dise of the posterior barrel, which is parallel to the front barrel; the
posterior valley is very large. Ponzi’s worn specimen is proved by
this to be the third.

Dimensions.—Length of crown, outer side, 1:75 in. Length of crown, inner side,
at base, 1:55in. Length of ditto at middle, 1°55in. Transverse diameter at base
anteriorly, 22 in. Transverse ditto posteriorly, 2°15 in.

Lastly, No. 113, also from Monte Sacro, is a first true molar, upper
jaw, left side. The crown is nearly in the same stage of wear as
Ponzi’s milk molar (Pl. XXXIL. fig. 7), but is a good deal rolled. The
posterior colline is broken off.

1 Reproduced in Pl. xviii. fig. 1 of this work. See antea, p. 328, note.—[Ep.]

376 RHINOCEROS.

C. In Signor Ceselli’s Collections, from Torre di Quinto and Ponte
Mammolo.

In Signor Ceselli’s collections from Torre di Quinto there is an ante-
penultimate true molar, right side, of Rhinoceros. It is in nearly the
same stage of wear as the penultimate in the Sapienza Museum (p. 374),
though of the opposite side, and is very much smaller in all the dimen-
sions, but the height of the crown proves it not to be a milk molar.
The crochet forms a very open angle, and a combing plate is emitted
from the middle of the outer ridge converging a little towards the point
of the crochet. There is also ashort combing process emitted from the
anterior colline overlapping the tip of the crochet, and a little above
it. Further, deep down in the valley, are additional denticular com-
plications, forming a ring or a loop, one leg joined to the outer ridge,
one to the outer colline: very complex.

The disc of the posterior colline is directed backwards, with torsion
of the apex. There is a small intercolumnar tubercle and a good deal
of cement. In all the characters the tooth agrees with the Montpellier
L. megarhinus, and in general plan it is very like the drawing of the
milk molar.

Dimensions.—Extreme length of crown, outer side, 2°2 in. Extreme ditto, inner
side, near base,1°5in. Transverse diameter in front near base, 2‘1 in. Transverse
ditto behind, at enamel edge below (very oblique), 2-1lin. Greatest height, outer
surface, 2°1 in.

The specimen is encrusted below with volcanic gravel. It has no
fangs, and is rolled below. There is some cement at the mouth of the
transverse valley, and an abundant layer of it on the posterior valley
lining the surface.

Signor Ceselli’s collection (from Ponte Mammolo) also contains a very
perfectly preserved second premolar, upper jaw, right side, of Rhino-
ceros, slightly worn (7.e. a little less than the milk molar of Ponzi, Pl.
XXXII. fig. 7), and in the best state to show its characters. The summit
of the crown shows distinctly three fossettes, 7.e. one formed by the
anterior transverse valley, one by the posterior valley, and the third an
oval pit included between the termination of the crochet and the comb-
ing plate, emitted from the middle of the outer ridge nearly in front of
the dorsal vertical ridge. The two are fused into a confluent wall, of
which the combing plate is the thickest. The posterior valley has in-
truded into it, from the posterior outer vertical groove (which resembles
in form that of a horse), a very thick blunt plate dividing the end of the
valley into two branches. ‘The termination of the anterior valley (exclu-
sive of the third fossette) is somewhat reniform, concavely parallel to the
posterior end, and free from any minor plates. The anterior disc forms
a narrow strip, little worn ; the posterior disc is nearly the same, and
has not much obliquity. There is no torsion, no posterior colline at
the apex, and no intercolumnar mammilla, but a well-marked basal bour-
relet to the inner side. The outer surface resembles the molar of a horse.

Dimensions.—Length of crown, outside, 1°45 in. Length of ditto, inner side, at
basal bourrelet, 1:-lin. Length of ditto, in middle, to ditto, 1:35in. Transverse
ditto, in front, base, 1‘6in. Transverse ditto, behind, 16in. Height of crown,
outside, 1°7 in.

The specimen has volcanic sand matrix, and is rolled below; the
fangs are entirely gone. There is no cement remaining. The enamel
is smooth. It is from Ponte Mammolo (Monte Sacro).

—————— eee rrr err eer ele

RHINOCEROS LEPTORHINUS. 377

In the same collection from Torre di Quinto there is also a penulti-
mate, or antepenultimate true molar upper jaw, left side, considerably
more advanced in wear than the others, and differing in some degree
from them in the general form of the crown (Pl. SKIL fig. 8). It
is more advanced in wear even than the detached R. hemitechus tooth
brought for comparison, and the worn summit is much flatter than any
of the others, except Ceselli’s very old compressed tooth. The general
contour of the crown is more square, and with less inequality between the
front and posterior diameters, approaching somewhat in this respect 2.
hemitachus. There are two valleys, the posterior of which is triangular
and ground low, and the inner slopes in a more shelving manner than in
R. hemitechus. The middle valley opens into a triangular fissure; it is
then bent nearly at right angles, by the intrusion of the crochet, and ter-
minates in a complex cul de sac, which is three-lobed, or trefoil-shaped.
The termination of the middle valley is not unlike a more advanced degree
of the large penultimate in Pl. XXXII. fig. 4. A thick short plate is pro-
jected baekwards from the anterior colline overlapping the direction of
the crochet, and pointing parallel to it from the opposite side; the ordi-
nary combing plate from the outer ridge is projected inwards at right
angles to the apex of the crochet, but more as a deep-seated denticle,
the apex of which is still partly free. The crochet makes the third
division. ‘The crochet differs very much in direction from the other
specimens. It is thrown forwards at a right angle, but with none of
the boot-shaped thickening of R. hemitechus. 'The length of its inner
border is fully equal to the width of the posterior collie disc. There
is a small intercolumnar tubercle at the mouth of the valley. The an-
terior colline presents a sausage-shaped broad disc ; the posterior barrel
has somewhat of a horse-shoe pattern (from the posterior valley), but
the disc is very wide. . The anterior outer vertical groove is wide and
deep; but the outer edge of the crown is less angular in its outline
than usual, the points having been probably abraded by rolling. The
anterior overlapping sinus is much more pronounced than in R.
hemitechus.

The crown differs in its general pattern a good deal from the others.
The crochet is at right angles, but it is not the crochet of R. hemitechus.
It certainly is not of R. tichorhinus. On the whole, I regard it as an
unusual form of 2. megarhinus. The greater width at the inner side
and the abrasion of the outer edge give the peculiar appearance.

Dimensions.—Length of crown (antero-post.), outside, 2°25in. Length of ditto
in middle, 2°1in. Length of ditto at inner side, about 1:95 in.

The fangs are wanting and replaced by volcanic sand matrix. The
fangs had been rolled. The enamel is smooth and rather thin; the ce-
ment is entirely gone.

In the next place there is an antepenultimate (penult.?) true molar,
upper jaw, right side, very far advanced in wear, of large proportional
size, but very much compressed (Pl. XXXII. fig. 9). It retains the fangs,
perfect to their points. An oblique fracture (at a) has damaged a small
portion of the posterior barrel, and another recent (at 0) has removed the
anterior outer angle and the layer of enamel. The surface of the ivory
here shows some very beautiful bluish black dendritic crystallization
penetrating into the ivory. The crown is oblong across (the dispropor-

tional width to length being much more than is shown in the figure), and

378 RHINOCEROS.

is ground down very low. There are two fossettes; that of the posterior
valley is a small round hole. The transverse valley is a very contracted
fissure at the commencement, terminating in a ‘ duck’s head’ kind of
cul de sac. The crochet is short and thick, and given off at a very open
angle; at the ends and inner side of the tooth there is a very thick
layer of cement. The crown in amount of wear and pattern (except in
the direction of the combing plate) is not unlike the antepenultimate
R. hemitechus (m. 1), in Dinkel’s drawing with the six molars.
(See Plate XVI. fig. 1.)

Dimensions.—Greatest width of crown near ant. fract. at top, 2:in. Greatest
ditto behind, at posterior barrel, 1-7 in. Greatest ditto of crown at base, front,
2°5 in. Groatest ditto behind, 25 in. Antero-posterior diameter of crown at top,
outer, 1°8in. Antero-posterior diameter of crown at top, inner, 1°6in. Antero-
posterior diameter of crown in middle, 1-7 in. Antero-posterior diameter of crown,
greatest, 1°8 in.

The enamel is smooth, the cement is thick, and there is volcanic
sand below.

Lastly, in Signor Ceselli’s collection, from Torre di Quinto, there is
a detached penultimate right molar, lower jaw, having a disc of pres-
sure in front and behind. It agrees exactly in form with the penulti-
mate described of Ponzi’s jaw fragment (p. 874), but is a trifle larger.
It is nearly in same stage of wear.

Dimensions.—Extreme length, 2°1 in. Width of front barrel, 1‘ in. Width of
rear barrel, 1:2 in.

It has a very thick coat of cement between the barrels, which has
been rubbed off elsewhere; this is as thick as in R. hemitachus. 'The
enamel is smooth, but rather thick. The fangs are present.

In Signor Ceselli’s collection from Ponte Molle there is also a third
premolar, lower jaw, right side, well worn.

Dimensions.—Length of crown, 1*4in. Width in front, 0-9in. Width behind,
1-1in.

D. In the Museum of the Jesuits’ ‘ Collegio Romano.’

Examined a very remarkable fragment of the transverse half posterior
portion of a last true molar, upper jaw, left side, of a Rhinoceros, in
different mineral condition from all the other Roman specimens. It
shows a tubercle with four crenatures, attached to the base as in the
Montpellier specimen, and the addition to the valley of a combing
plate, thick, and pointing at right angles to the crochet; there is also a
very distinct intercolumnar tubercle. The dise of the compressed
posterior barrel is very well preserved. The enamel is of a bluish grey
or lead colour, thin and smooth; there is some cement outside. The
ivory is chestnut-coloured, like the Pignano Elephant ivory ;! the
matrix is seen to bea blue clay. The specimen is certainly not from
the quaternary volcanic sands of Rome ; its origin is not known.

The tooth is very much smaller than Ponzi’s last molars.

‘ c In Ponzi's A. (p.372)
Dimensions. Tchad Inches

Transverse diameter near base : . a2 2°6
But for the small size, I would have nalts this specimen to 2. mega-
rhinus, notwithstanding the combing plate, in consequence of resem-
blance of general form, the exact resemblance of the adpressed basal
1’ See antea, p. 187.—[Ep.]

ae og a ee, er aes

Figs. 1

Fig. 3.

Figs. 4 to 9. Represent six molars in the collections at Rome.

DESCRIPTION OF PLATE XXXII.
RHINOCEROS LEPTORHINUS (R. MEGARHINUS).

and 2. Represent the penultimate and last upper molars of PR.
leptorhinus, about three-fourths of the natural size, and are taken
from two of the casts mentioned at page 398, as having been
obtained by Dr. Falconer at Stuttgart. The original teeth are
those upon which Jiiger founded his Rhinoceros Merckii. The
casts are now in the British Museum.

Represents a sixth or penultimate upper molar, left side, in the
Nice Museum, about three-fourths of the natural size. The
drawing is copied from one brought by Dr. Falconer from Nice.
(See page 370.)

figures have been copied by Mr. Dinkel from drawings brought
by Dr. Falconer from Rome.

. Isa penultimate upper molar (t. m. 2), left side, from Monte

Sacro, in the Sapienza Museum, three-fourths of the natural
size. (See page 374.)

Fig. 5. Represents the last upper premolar (p. m. 4), left side, three-
fourths of the natural size, also from Monte Sacro in the
Sapienza Museum. (See page 375.)

Fig. 6. Is a last true molar, upper jaw, left side, three-fourths of the
natural size. The specimen is in Professor Ponzi’s collection,
and is from the Gravel-beds of Ponte Molle. (See page 572.)

Fig. 7. Is an upper milk molar, left side, three-fourths of the natural
size, also in Professor Ponzi’s collection, from the Gravel-beds
of Ponte Molle. (See page 373.)

Fig. 8. Is a penultimate or antepenultimate true molar, upper jaw, left
side, three-fourths of the natural size, in Signor Ceselli’s col-
lection, from Torre di Quinto. (See page 377.)

Fig. 9. Is an antepenultimate true molar, upper jaw, right side, very far
advanced in wear, about three-fourths of the natural size, also
in Signor Ceselli’s collection. (See page 377.)

VOL, II.

The

VoLIL. Plate 32,.

RHINOCEROS LEPTORHINUS. 379

tubercle behind, and the form of the crochet (open angle), together
with that of the disc of the posterior barrel and the width and broad
deep furrow of the outer angle.

The upper molars of 2. leptorhinus, which I have examined at Rome,
are: Sof tm. 3; 1 oft.m.2; 4o0ftm.1; lofpm.4; lof pm. 3;
1 of p.m. 2; and 1 of mm.

I have not seen a trace of an indigenous tooth of R. tichorhinus in any
of the Roman collections. The teeth in the Kircher Museum are evi-
dently of foreign origin. They consist of one upper molar and of two
lower molars, all detached and worn, with the yellow ochre matrix of
the Devon and Somerset caves.

VI.—Nore on R. Leptoruinus rrom Montienoso, NEAR LEGHORN.
Florence, May 20, 1859.

No. 1.—Is a fragment of the anterior part of the right maxillary,
showing the antepenultimate and penultimate premolars much worn.
Compared them with Jourdan’s casts and drawings from Montpellier
(p. 869), and found them to agree exactly.

No. 2.—Is a penultimate or antepenultimate true molar, upper jaw,
left side, exactly like the Montpellier specimens.

No. 3.—Consists of a penultimate and antepenultimate of upper jaw,
right side, detached and well worn, agreeing closely in form with the
Montpellier specimens. In the penultimate, a very thick layer of
cement lines the posterior valley and both the outer anterior angles;
the groove is broad and deep.

No. 4.—Is a specimen of the last premolar, upper jaw, right side, in
beautiful preservation and showing the characters very perfectly. The
posterior barrel throws forward two crochet processes nearly of the
same size, of considerable thickness, and well separated; the outer ridge
throws off converging ‘combing plates,’ nearly of the same size, so that
the sinuosities of the transverse valley are very complicated.

No. 5.—Is a fragment of the lower jaw, left side, containing the
penultimate true molar, partly worn, but having the crescents still
separated.

These specimens are of great interest in proving the extension of the
Thinoceros megarhinus into the ‘Val d’Arno inferiore.’ They were
found along with remains of Hlephas antiquus.

VII.—Norte on Rurnoceros From VAL DI CHIANA.
Arezzo, May, 1859.

Examined a lower jaw, right side, of Rhinoceros in the same mineral
condition (7.e. white and adhering to the tongue), as the large Hlephant
femur and bovine heads from the Val di Chiana in the Florence Museum.
Only part of the symphysis is present. The anterior margin of the
right ascending ramus is present, but the posterior angle is wanting,
The jaw contains the five last molars zm stu. The antepenultimate
premolar has dropped out, but its two fangs are seen. The molars are
well worn; the crown of the first true molar is worn out; and in the
last the discs of the crescents are united. The teeth show marks of a
thick layer of cement dislaminated ; there is an oblique bourrelet on the
outside of the barrels as in R. megarhinus. The enamel is thickish and

380 RHINOCEROS.

smooth. It certainly does not belong to R. tichorhinus, and the teeth
are too large and too broad for the Rhinoceros of the Val d’Arno. It
is probably, #. megarhinus, and is an important specimen. The lower
jaw is very low for the thickness and size of the teeth.

Dimensions.—Length of line of six molars, 9°5 in. Length of crowns of the five,
8lin. Length of three last molars, 56in. Approximate length of three pre-
molars, 4°in. Lerigth of two last premolars, 2°5in. Length of crown of penulti-
mate true molar, 1°85in. Greatest width of ditto, 1-2in. Length of last molar,
Qlin. Total length of fragment, 14°5in. Height of jaw inside at penultimate
premolar, 2°5in. Height of ditto at last molar, 38 in, Greatest thickness of jaw
below, 2°1 in.

Another specimen of left side of lower jaw is very like the above, but
all the teeth are wanting.

VI1ll.—DescripTion oF Rematns or R. Leprornrnus ry Museum OF
Nat. History at Turin.

April, 1861.

A very beautiful specimen of a right ramus of the lower jaw of a
fossil Rhinoceros, marked ‘ Foss. nei sedimenti fluvio-lacustri pliocenici
tra Dusino e S. Paolo (dono dell’ Ingegnere Commend. Barbavara),’ in a
hard mineral condition, weathered grey, containing the whole of the
molar series en suite, and part of the symphysis, but the diastemal edge
entirely gone; the horizontal ramus quite entire from the first premolar
backwards, but the angle broken off; part of the anterior and basal
portion of the ascending ramus present, but the fracture rounded by
abrasion.

The teeth in form, and amount of wear relatively, is nearly a perfect
reproduction of Gervais’ Montpellier lower jaw (PI. II. fig. 8 of ‘Zoolog.
Frangaise’), from the first to the last, so much so that the one might be
taken for the other. But unfortunately, the Turin specimen wants all
the incisive and diastemal portion. The horizontal ramus is very low,
as compared with the double lower jaw got in the Mastodon deposit.
The posterior boundary of the symphysis is in a line with the posterior
third of the penultimate premolar. The lower edge of the ramus is
very horizontal from behind on to the premolars.

The characters of the molars in this specimen are a good deal as in
Mr. Gunn’s lower jaw. One character of great importance is to be
noticed, that on the outside of the penultimate molar, in the furrow
between the crescents, there is a very thick layer of cement; it is only
partially present, and probably is dislaminated elsewhere.

Dimensions —Length of six molars, 9°5in. Length of three true molars, 5-4in.
Length of three premolars, 4:1in. Length of first premolar, 1°2in. Length of
second ditto, 1°3in. Length of last ditto, 15 in. Length of antepenultimate true
molar (middle of crown), 1:6 in. Length of penultimate ditto, 1-7 in. Length of
last ditto, 1:95in. Greatest height of jaw to alveolar edge of last molar, inner side,
3-3in. Height of ditto at first true molar, 2°75 in. Greatest height at antepenul-
timate premolar, 2'4in. Greatest thickness of Jaw, 1'85in. Height behind last
molar, inner side, 3°4 in.

In the same Museum there is also a penultimate or antepenultimate
truemolar, upper jaw, left side, of Rhinoceros leptorhinus (megarhinus).
The anterior outer angle is a little broken, but the crochet has the same
character as in the Grays Thurrock variety. It is certainly not 2.

RHINOCEROS LEPTORHINUS. 381

Etruscus. But the matrix is exactly like the Sansino of the Val d’Arno.
It is from the railway cutting near Dusino.

There are also two upper premolars showing the bourrelet very strong,
but the crown still covered with matrix, of the same species and from
the same locality, with a yellowish ferruginous matrix.

ITX.—DeEscripTion OF THE Cortes! RatNoceros Cranium.!
Natural History Museum, Milan, April 24, 1861.

The cranium, upon the whole, is in a very remarkable state of pre-
servation, and is now very much in the condition as described by
Cortesi in the ‘Saggi Geologici.’ It is entire, from the tips of the
nasal bones to the occiput; the left side of the occipital crest being
the part chiefly damaged. The skull, like Jourdan’s Lyons Museum
Cranium (Pl. XX XI. fig. 3), had undergone lateral pressure, so as to have
been slightly crushed. This is well shown on the basal aspect ; when
an axial line is drawn along the base of the sphenoid through the
palate, the palatal portion is seen to be deflected towards the right side ;
and the spheno-palatine bones are crushed. The whole of the right
zygomatic arch is present, but partly crushed in upon the zygomatic
fossa. The crush has acted upon the palate, so as to elevate consider-
ably the series of molars upon the left side, above the plane of those of
the right; the former being pressed a little outwards, the latter inwards,
upon the plane of the palate. The left zygomatic arch is partly wanting,
but the basal portions at either end are present, and the posterior stump
shows that a portion at least has been lost by a comparatively recent
fracture (since found). The mastoid process on the left side is broken
off, while the greater portion of the styliform process behind it is
present; vice versd, on the right side the greater portion of the mastoid
process is present, while the styliform is broken off. The two (occi-
pital) condyles are present, and nearly undisturbed, although somewhat
damaged; the occipital part of the cranium has fortunately escaped
pressure; the lateral margins and crest on the left side are nearly
entire; the upper third of the right side is broken off. The right occi-
pital condyle is traversed by two fissures; the left has lost a portion
towards the occipital plane. On making a further search among the
fragments in the case, I found the missing portion of the left mastoid,
which is now seen perfectly entire, and probably a further search
would lead to the discovery of some portion of the missing part of the
left zygomatic arch. I found the specimen taken off its stand, and laid
upon a pad of straw with the palatine surface uppermost, preparatory
to being drawn. Seen in this aspect, it bears a very close and remark-
able general resemblance to Jourdan’s Lyons skull (Pl. XXXI. fig. 3,
p- 869), which is also somewhat crushed, but in the reverse direction
(i.e. according to the lithograph), viz. from right to left (Cortesi’s skull
being from left to right). The occipital condyles in the Lyons cranium
are obliquely displaced, while in Cortesi’s skull they are in their natural
position. The bony part of the cranium is a good deal cracked and
shivered, so as to break off into minute pieces when the matrix is
detached ; but it is highly injected with ferruginous infiltration, and
completely mineralized. The matrix consists of a greyish yellow. com-
pact clay (marna azzurra), which is hard and mottled with ferruginous

" Dr. Faleoner was unfortunately not permitted to take any drawings or casts
of this eranium.—[ Ep. }

382 RHINOCEROS.

blotches. A portion of the diastemal and incisive prolongation, which
is wanting in the Lyons skull, is fortunately present m the Milan
specimen. The series of molars is present on both sides, but the
summits of most of the crowns are more or less involved or concealed
by matrix, which has been left very nearly as when found by Cortesi.
The following teeth are present or indicated :—

A. Right Side—1. Immediately behind the commencement of the
diasteme, ou the right side, the empty socket is distinctly marked of a
single-fanged premolar, being the normal pre-antepenultimate (p.m. 1).
The alveolus is oval in the transverse direction, and about eight lines
in diameter. I have picked some of the matrix out of it, so as clearly
to define the pit, and am quite certain of the accuracy of the observa-
tion. The corresponding socket of the same tooth is present on the
left side, but has not been picked out to the same extent.

2. The antepenultimate premolar or (p.m. 2).—The shell of the
crown of this tooth has been broken off, but it is fortunately preserved
on the left side, and will be noticed in the sequel.

3. The penultimate premolar or (p.m. 8).—The shell of enamel
nearly all round the circumference of this tooth is preserved, but the
central convolutions and the anterior inner barrel have been ground
down and destroyed by a recent crush. The salient point of the outer
shell of enamel is very high, and it almost looks at the posterior point
as if it had not been subjected to wear; but this is very doubtful,
there being no ivory attached to determine the point. The correspond-
ing tooth of the opposite side is present nearly entire; but the outer
half of the crown is covered by matrix which conceals the convolutions.

4. The last premolar (p.m. 4) is present, and beautifully perfect. It
had evidently come but very lately into place, as the edges only of the
anterior and posterior barrels are slightly worn into narrow crescentic
discs, and the outer edge is also but slightly affected by wear. The
level of the crown is depressed below the level of both the tooth which
precedes it, and of that which is behind it; it is in a considerably less
advanced state of wear than the next succeeding teeth (7.e. t.m. 1 and
t.m. 2), and therefore had come into place more recently than either
of them. ‘The basal bourrelet is distinctly marked at the anterior and
posterior ends, but only very obscurely around the base of the inner
side, as a slightly crenulated inequality. The anterior and posterior
barrel ridges are nearly transverse and parallel, the posterior fossette is
very large and intact behind. The crochet plates are very complex,
forming four pectinate lamin, which are directed forwards so as to
meet an accessory plate thrown off inwards from the outer ridge, which
divides the central termination of the middle valley into two nearly
equal divisions. In this respect, the tooth resembles very closely fig. 25
of De Christol’s plate (see Pl. XVIII. fig. 1), but with this difference,
that the crochet plates in the latter are only two, while in the Milan
tooth they are closely approximated and are four in number. But De
Christol’s tooth is more advanced in wear, and the crochet plates would
be reduced in number in the Milan tooth by further abrasion, One of
these plates advances so as nearly to meet the accessory outer combing
plate, and thus isolate a distinct fossette. There is a contortion of the
apex of the posterior barrel (as in Pl. XVIII. fig. 1).

On the opposite side the corresponding tooth is apparently wanting,
its place being occupied by matrix. This would indicate either that

a en =

RHINOCEROS LEPTORHINUS. 383

the last premolar, left side, had never emerged, or that it had dropped
out after emergence. (On subsequently removing the outer alveolar
wall it was visible. )

The basal bourrelet, as a general rule, is but indistinctly exhibited in

‘all the premolar teeth of this specimen.

5. Lhe antepenultimate true molar (or t.m. 1).—This tooth is present
in a more advanced state of wear than either p.m. 2, 3, or 4, the stage
‘of detrition being about the same as in De Christol’s fig. 18 (copied
in Pl. XVIII. fig. 3), but even a little more advanced. The crown of the
tooth had originally been quite entire in this specimen, but it had got
crushed and shivered; the pieces have been replaced in position with
glue. The posterior fossette and the whole of the middle valley are
enveloped by matrix, so that the offset of the crochet is entirely con-
cealed, as is the greater portion of the inner side of the two barrels.
The anterior basal talon bourrelet is very pronounced, with a crenated
margin. The corresponding tooth of the opposite side (left) is also
present, and still more perfectly conserved ; but the crown is nearl
entirely enveloped by matrix, so that the characters yielded by the

crochet are not visible. :

6. The penultimate (or t.m. 2), right side——This tooth is present
and quite perfect, but is pressed slightly inwards upon the palate. The
outer shell of enamel is seen to be quite perfect, and the outer ridge
but very slightly abraded, the boundary of the posterior fossette being
quite entire. The anterior barrel has its edge but slightly abraded, a
little more in degree than p.m.4; the whole of the central valley and
of the inner sides of the barrels are enveloped by matrix, so that the
form and offset of the crochet and the anterior basal talon are com-
pletely concealed. This concealment of the most characteristic part of
the crown is much to be regretted for my present purpose. The corre-
sponding tooth of the opposite side is also present, but fractured and
repaired ; it is slightly dislocated outwards (like the whole of the series
of the left side), exposing completely the inner side of both barrels,
down to their base. There is not the slightest trace of an internal basal
bourrelet, and the summits of the barrels, more especially the hind one,
show very markedly the peculiar twist seen in fig. 18 of De Christol’s
plate (as copied in Pl. XVIII. fig. 3). This character is equally seen
on the corresponding parts of p.m. 1, left side.

7. The last true molar (t.m. 3) of the right side had not emerged,
and there is not a trace even of its presence, the corresponding alveolar
part of the maxillary bone being crushed in and covered by matrix;
but, as will be seen in the sequel, the germ of this tooth is distinctly
present on the left side.

B. Left Side-—1. The pre-antepeniltimate, or (p.m. 1).—The single-
fanged alveolus of the first premolar is present, as in the opposite side,
distinctly defined, and partly occupied by matrix.

2. The antepenultimate, or (p.m. 2).—The crown of this tooth is
present, quite perfect, and but very slightly affected by wear. In a
general way it resembles very closely fig. 1 of Gervais’ Pl. II. (‘ Paléon-
tologie Frangaise’), with the exception that the basal bourrelet, which
is distinctly present upon the anterior barrel, is less pronounced on the
posterior barrel than seen in that figure. The crown has a similar
sub-triangular form, 7.e. broad externally, and contracting inwards.
The apex of the anterior barrel, which is all but intact, forms an

384 ¢ RHINOCEROS.

isolated flattened conical cusp, as in Gervais’ fig. 3. Pl. II. of tom. ii.
of the ‘Memoirs of the Montpellier Academy of Sciences,’ but of a
larger size than in that figure. The posterior colline is seen to be but
slightly abraded by wear; the whole of the posterior fossette and the
central crochet convolutions are entirely concealed by matrix. This
tooth appears to be nearly in the same stage of abrasion as p.m. 4 of
the opposite side.

3. The penultimate premolar (p.m. 3).—TYhe whole of the shell of
this tooth is present, but the outer half of the crown is completely en-
veloped by matrix. The anterior and posterior barrels are seen to be but
slightly abraded, 7.e. to about the same extent as t.m. 2; the breadth of
the tooth across the anterior division is much greater than the length.
There is a crenulated anterior talon, but only a very obscure appearance
of bourrelet at the base of the anterior barrel; none is visible behind,
but this part of the tooth is enveloped by matrix. The vertical furrows
upon the outer surface of the enamel of this and the preceding tooth
are but very indistinctly marked. The same observation applies to
p-m. 4, of the opposite side, in which the anterior vertical furrow is
also indistinctly marked. '

4. The last premolar (p.m. 4) on this side, as already remarked,
appears entirely wanting, and its position is occupied by a block of
matrix; but on reversing the cranium, it is distinctly seen enélosed in
its alveolus, below the mass of matrix.

5. The antepenultimate true molar (t.m. 1).—The crown of this
tooth is nearly perfect, although somewhat shivered. The summit is
almost entirely enveloped by matrix concealing the crochet and other
convolutions. The vertical furrow of the anterior outer angle is broad
and shallow, but well pronounced—broader than in De Christol’s fig. 18
(see Pl. XVIII. fig. 3). The summits of the barrels are in the same stage
of wear as described of the tooth of the opposite side. The outer surface
of the posterior division is slightly concave and flattish.

6. The penultimate true molar (t.m. 2).—This has its crown more
exposed than on the opposite side, but it has been fractured, and the
pieces have been imperfectly replaced. The peculiar twist of the apices
of the barrels has been already noticed. The anterior vertical furrow seen
at the outer angle of the tooth of the opposite side is also here well marked.

7. The last true molar (t.m. 3), on the left side, is distinctly seen
in the state of germ, hardly emerged above the alveolar level, and em-
bedded in the jaw. About an inch in height of the posterior ridge is
exposed by the removal of the alveolar wall. The edges are quite in-
tact, and about an inch and a half below the level of the next preceding
tooth. The principal valley is completely filled up by matrix, but it is
visible that the crown had the §ub-triangular form, which is charac-
teristic of the same tooth in the existing bi-corned African Rhinoceros.

Obs. 1.—The enamel surface in all these teeth is tinged of a bluish
grey, which Cortesi compares to an incipient tint of turquoise. The
enamel is perfectly smooth, 7.e. entirely free from any superficial ru-
gosity, as in the tichorhine Rhinoceros, and I could detect upon none of
the teeth any indications of a coat of cement. There is certainly nothing
approaching the enormous coat of cement seen on the teeth of Khi-
noceros hemitechus; the enamel is not so thick as in that species, nor
so rugous on the surface. The ivory-core of all the teeth is highly in-
filtrated with iron, showing a dark amber colour; the general colour of

RHINOCEROS LEPTORHINUSt® 385

the teeth resembles in its pearly aspect that of the molars of Rhinoceros
Etruscus, in the Museum at Florence. It is important to remark in
reference to the measurements, that on the right side the penultimate
does not overlap the first true molar, there being three-tenths of an
inch interval between. There is nearly the same interval between the
autepenultimate true molar and the last premolar, and also between the
third and fourth premolar, showing that these molars have been dis-
placed, and giving undue length to the measurement of the entire series
on the right side. They are in their natural state of apposition on the
left side. The length of the series, from the anterior end of the second
- premolar to the posterior margin of the second true molar, which in-
eludes five teeth, amounts exactly to 10:2 inches, and from the anterior
border of the first premolar to the same pau behind, to 10-9 inches
(or nearly 11 inches).

Obs. 2.—Cortesi’s figure in the ‘Saggi Geologici’ (Pl. VIL.) is appa-
rently of the left side (the nasals and symphysis pointing to the left, the
occiput to the right); but the figure is exhibited reversed, and in reality
it represents the right side. The same remark applies to fig. 7, Pl.
IX. Rhin. of the ‘Ossemens Fossiles,’ professing to be on the scale of one-
sixth of the natural size. The lower jaw, which is placed in relative
position below the cranium in both these cited figures, is also figured
reversed. Cuvier asserts that his engraving was made after drawings
sent by Adolphe Brongniart, and these have hitherto been assumed to
have been originals; but it is clearly manifest that Brongniart’s is
merely a copy of Cortesi’s figure. The uncouth lower jaw is fore-
shortened precisely alike in both, so as to show the line of molars on
both sides, both coronoid processes, both sigmoid notches, and both
condyles. In fact the figures are so much alike that it is impossible to
doubt that the one was copied from the other. There is the same nick
to the broken edge of the left coronoid process, and to the broken end
of the incisive bone. The principal differences are, that the mastoid
shown in Cortesi’s figure is omitted by Brongniart; that the rim of the
orbit and the outline of the zygomatic arch, together with the shading
of the orbital cavity and zygomatic fossa, are better defined by Brong-
niart than in Cortesi’s figure. The uncouth occipital pyramid rising
into a conical peak, and evidently exaggerated in Cortesi’s figure, is less
salient and more naturally represented by Brongniart. Asi regards the
lower jaw, Cortesi’s figure represents a salient 1 mass of matrix on the
lower margin of the jaw, below the penultimate figured tooth (ie.
t.m. 1), all of which is omitted by Brongniart, who gives a clear
outline to the lower margin. But this mass is still undisturbed with
the rest of the matrix, as “when left by Cortesi.

Obs. 3—De Christol, in his remarks upon Brongniart’s figure of the
lower jaw, passes some severe strictures upon the low height and little
projection of the coronoid process above the alveolar ~ margin, &e.
But these are all explained away by the fact that a great deal of matrix
is still left enveloping the j jaw, and that a part only of the crowns of the
two last molars that are in sitw emerge above the cake of matrix.
When the natural object is compared with fig. 5 of De Christol’s draw-
ing (profile) it is manifest that there is a great general agreement of
form between the Montpellier and Milan specimens, and even an inex-
perienced observer would at once remark the similarity of the sym-
physial expansion in both.

VOL. Il. cc

386 RHINOCEROS.

Obs. 4.—The antepenultimate premolar (p.m. 2) of the left side
in the Cortesi specimen resembles in the closest manner the corre-
sponding tooth represented by De Christol in fig. 27 of his memoir.!
The anterior cusp forms in both an isolated compressed cone, the apex
of which is just beginning to be abraded, and the posterior barrel has its
edge ground down into a narrow crescentic band, which is alike in both.
The principal difference observed is, that the basal bourrelet is more
strongly represented in De Christol’s figure than it is seen in Cortesi’s.

Obs. 5.—The lower jaw of Cortesi’s specimen is seen to be in the most
fragile state of disintegration. On detaching a slab of the matrix, mea-
suring 35 inches by 3 beneath the second and third premolars (p.m. 2
and 3) on the left side, it was seen that the fibrous roots of herbaceous
plants had insinuated themselves between the matrix and the surface of
the jaw, forming a web, and that the bony mass of the latter was cracked
and fissured in every direction, penetrated by roots, and in a state of the
most rotten decay. The lower jaw was evidently discovered uncrushed.
A great mass of matrix is interposed between the rami from the sym-
physis on to behind the last molar, yielding the dimensions given in the
table (viz. Nos. 17 to 21). The details of the teeth in the lower jaw
are as follows:—There is not a trace of incisives, the beak being partly
damaged at its edge, where they might be looked for, and the diaste-
mal ridges being also abraded. The lower contour of the beak expan-
sion is disguised by a cake of matrix, which has vitiated both Cortesi’s
and Brongniart’s drawings; otherwise it would be like Christol’s fig. 6.
There is no trace on either side of an alveolus for the pre-antepenulti -
mate, but it is by no means certain that it may not have been there to

correspond with the tooth in the upper jaw. ‘The antepenultimate .

premolar present upon the left side consists of two crescents, both of
which are only in the slightest degree affected by wear. The lower
half of the anterior end bears a smooth surface, which appears to be
the dise of pressure against a pre-antepenultimate, which had dropped
out. This disc of pressure for the pre-antepenultimate of lower jaw,
left, occupies nearly half the height of the crown and is sagittate in
form, like a Celtic arrowhead of flint. From the broad surface of the
anterior end of the antepenultimate, and the appearance of a disc of
pressure at its base, I am convinced that there must have been a
pre-antepenultimate, corresponding with the upper one. The third
premolar is present upon both sides, and both the crescents are slightly
affected by wear, showing a narrow band of enamel all round. The
anterior crescent in each is elevated about half an inch above the plane
of the posterior crescent. The last premolar is wanting on either side,
its site being occupied by a mass of matrix; the last milk molars had
probably just fallen out, and their successors may be embedded in the
jaw as germs. The two stumps of the fangs of the antepenultimate
premolar are seen on the right side, the crown being broken off.

The first and second true molars are present on either side, both of
them being but slightly affected by wear. The anterior division of
each yields a horse-shoe pattern, of which the front limb is much
shorter than the hind one. The posterior division yields a crescent with
but a very slight curve. The last true molar on either side is wanting.

The condition of the dentition in both the jaw and cranium shows
that they must have belonged to the same individual.

1 Ann. des Se. Nat. 2™e Sér. tom. iv. 1835. PI. iii. fig. 12. See antea, p. 328,
note.—| Ep. |

RHINOCEROS LEPTORHINUS. 387

Obs. 6.—De Christol’s drawing, fig. 11, although stated by him to
represent the left side, is in reality of the right. The ragged black
shaded wall A of his figure represents pretty fairly the existing condi-
tion of the left wall of the nasal cavity, inner side; and although mis-
taken by him for a nasal septum, is exactly what Cornalia states it to
be in his note to Duvernoy.! The light shaded portion included within
dotted lines is not, as De Christol supposed, a fracture where a large
piece was wanting, but in reality it represents a layer of argillaceous
cement, which has been spread over the fossil from the orbit to the
incisive termination, either to strengthen the specimen or to disguise
fractures. A depression is left in the cement, indicating the position of
the suborbitary foramen. It is exactly situated as in Cortesi’s drawing,
but the clay envelope deprives me of the means of deciding whether
it really is the suborbitary hole or not. The fractured slab of the
frontal between the orbits, indicated in Christol’s drawing by the letter
C is a mistake ; the whole plateau of the frontal at this point, although
cracked and broken into minute pieces, is entirely present. The angles
of the lozenge on either side are broken over the orbits, and the
drawing of the fracture on the right side has misled De Christol. On
removing the cement, I find that the suborbitary foramen is present
on the right side, and situated exactly over the line of junction be-
tween the third and fourth premolars; its posterior rim being in a line
with the anterior third of the last premolar, and yielding the following
dimensions :—From anterior rim of orbit to posterior margin of subor-
bitary foramen, 4:2 inches; from the same point, é.e. rim of orbit to
bottom of nasal echancrure, 4°8 mches; from bottom of nasal echan-
crure to tips of the nasals, 8-4 inches; apparent entire length of nasal
bones, measured along curve, from the naso-frontal suture to tips in
the middle, 12: inches.

The uncouth representation in profile of the molar teeth in Cortesi’s,
Cuvier’s, and De Christol’s figures is owing to the fact already stated,
that they are pressed inwards upon the palate, more especially the two
last, and their most salient points therefore appear fore-shortened; the
representation of the opposite side would be much more natural. The
orbit is immediately over the penultimate and last molars, its anterior
margin on both sides falling in the line between the antepenultimate
and penultimate true molars. The outline of the naso-maxillary sinus
is well pronounced, as in Cuvier’s figure, and the present height, which
is partly concealed, is approximately 4°2 inches, taken about the middie.
Strictly speaking, the orbit is situated immediately over the penulti-
mate true molar.

Continuation of Description of the Cranium.

Most of the details in the anterior part of these remarks were taken
when the Cortesi cranium was lying with the palate upwards; it has
since been turned and mounted on a tripod stand, admitting the profile
and upper surface to be compared.

Projfile.-—This bears, as stated by previous describers, a close general
resemblance to that of the Sumatra bicorned Rhinoceros. The skull
has been exposed to lateral pressure, which has crushed in the right

zygomatic arch and the maxillary wall of the face, in front of the right

1 See antea, p. 314.—[ Ep, |
cc 2

388 RHINOCEROS.

orbit, under the chaffron. The occipital part is not nearly so perfect
now as in Cortesi’s time, the left side of the occipital crest being broken
off, together with the posterior and upper part of the parietals, to an
extent of five inches in length by four in width. In consequence, the
posterior termination of the sincipital echancrure and the posterior ex-
tension of the occipital crest behind the occipital plane are no longer
seen. The diploe cells are exposed where the upper plate of the parietal
has been removed, giving rise to the tessellated ragged lines of De
Christol’s figure, but less marked, and not extending so far forward as
he shows. The right zygomatic fossa is covered over by a cake of
matrix, about a quarter of an inch thick, as high as the fracture of the
parietals; the left zygomatic fossa is covered by a thinner cake of the
same. The lower three-fourths of the occipital are entire, more espe-
cially on the left side, and the lower half on either side, downwards
towards the styloid process, is covered by a thick mass of matrix, all
the central portion being bare. A great amount of hard matrix covers
the whole of the facial portion from the orbit forwards, as far as the
anterior third of the nasal arch, concealing entirely and blocking up the
left side of the nasal fossa. This is the mass represented by the dark
shade (A) in De Christol’s figure.

The cranium, as a general character, looks more elongated, more
slender, and much less massive than in the Rhinoceros tichorhinus ; the
cerebral portion is less elongated than in the latter, and the lateral
edges of the occiput less projected backwards. The anterior slope of
the cerebral pyramid makes a very considerable angle with the plane
of the frontal, more perhaps than is seen in Cuvier’s figure, but con-
siderably less than is shown by Cortesi’s, where the pyramid is exagge-
rated. The posterior face of the occiput inclines a little forwards as it
ascends from the occiput upwards, more so even than represented by
Cuvier’s figure, and is then over-arched by the projecting sides of the
occipital crest, which are produced backwards. It differs entirely from
the reclined occipital plane seen in R. tichorhinus. The bones of the
nose are clongated and slender in thickness, rather wide, and not much
arched above; they are nearly of uniform width, thinning as they ad-

vance forwards. The nasal suture between them is distinctly marked |

and open ; there is not the slightest indication of a dividing nasal septum ;
and I confirm entirely Cornalia’s remarks upon this point. They are
not so much arched as represented in Cuvier’s figure, resembling more
the outline given by De Christol. There is a slight central boss along
the axis near the tip of the nasals, but I can detect nothing like an
indication either upon the nasal or upon the frontals of the granular
rugous inequalities which indicate the base of horns; the frontal, it is
true, is cracked and fissured, but the nasal surface is entire and smooth.
A strip of about an inch wide of matrix has been left near the tip and
side of the right nasal bone. The absence of horned rugosities may be
owing to the immature age of the animal, which is shown by the teeth
and open sutures to have been not quite adult. De Christol describes
the vault of the nasal bones below to be excavated in a boat-shaped
fashion ; nothing of this kind is seen in Cortesi’s fossil, but their lower
surface is still concealed by matrix. The character of the nasal bones
entirely warrants the designation of leptorhinus, or thin nasal-boned
Rhinoceros, given to this species by Cuvier; these bones are infinitely
less massive than in the African Rhinoceros or the Indian species.

4

j

RHINOCEROS LEPTORHINUS. 389

The zygomatic arches are crushed in on the right side and wanting
on the left; the extreme height of the arch behind on the right side is
about 2° inches. The characters of the temporal fosse are not shown,
in consequence of the state of the zygomatic arches. The form of
the articular or glenoid surface for the lower jaw is also concealed
by matrix. The intermaxillary portion does not appear to have been
complete even in Cortesi’s time; it has now been considerably further
damaged by a fracture, and the missing piece has not been found. The
diastemal edges, as already described, are prominent and well marked,
bounding a gutter which contracts forwards; they are now very much
in the state represented by De Christol’s fig. 12 of the Montpellier form.
The orbits are placed immediately over the sixth tooth or penultimate
true molar. The position of the suborbitary foramen has already been
described. The auditory foramen is well seen on the left side, but
filled up with matrix; it resembles very closely that seen in fig. 12 of
De Christol, running upwards in a gutter on the side of the occipital
crest. In fact, the lateral and posterior part of the parietals and the
lateral outline of the occipital crest towards the base on the left side
very closely resemble the same parts in De Christol’s figure, with this
allowance, that in the latter the occipital condyles are wanting, while
in Cortesi’s they project boldly backwards. The terminal outline of
the molar teeth of the left side resembles very closely, in a general way,
that of Gervais’ fig. 1 of Pl. II. tom, ii. of the ‘Montpellier Transac-
tions.’ The height from the edge of the penultimate molar, left side, to
the frontal plateau, which is crushed, amounts to about 11 inches. On
the right side the same measurement gives 9°7 inches. Unfortunately
the orbital rim is not perfect on either side; it is best seen on the right,
but the suborbital tuberosity is wanting.

Upper View.—When the skull is seen from above it presents the
same elongated slender character as when seen in profile. This is
somewhat exaggerated by the skull having been crushed laterally, and by
the intrusion of the right zygomatic. In consequence of the immature
age of the animal, there is no indication of the sincipital lateral ridges
which define the temporal fosse, so strongly seen in Gervais’ fig. 2 of the
Plate above referred to, and also in De Christol’s fig. 13. Gervais’ figure
looks much wider in consequence of the presence of the zygomatic
arches. De Christol’s fig. 13 shows the nasals more massive proportion-
ally than in Cortesi’s fossil. In both of these the frontal plane is ele-
vated between the orbits to sustain the second horn. ‘This part of the
skull is cracked, fissured, and depressed in Cortesi’s fossil, and the
angles of the trapezium over the orbits are broken on both sides.
Making allowance for this depression, the profile outline of the Cortesi
skull resembles more Gervais’ figure than De Christol’s, as regards the
line of contour of the nasals and frontals. The crush is so considerable
that on the right side the height from the upper rim of the orbit to the
frontal plateau is only 2°4 inches. ‘The sincipital contraction of the
cerebral portion between the temporal fossee is very much as in De
Christol’s figure; but the absence of bounding ridges on either side
leaves no indication of a defined tablette.

Since the preceding remarks were written, I have been further able
to restore the posterior missing portion of the left zygomatic arch and
the greater part of the left articular condyle of the lower jaw.

On the whole, the Cortesi cranium is in a wonderful state of preser-

390 RHINOCEROS.

vation, considering the numerous removals which it has undergone. It
was first deposited at Piacenza in Cortesi’s time, then removed to the
Museum of Mines in the Stradone di Santa Teresa. After remaining
there many years, it was removed with the other Natural History collec-
tions in 1848 to the Palazzo Dugnani, and finally (1849) transferred to
its present locale in the Museo Civico, Contrada della Maddalena al Cer-
chio, near the Piazza Santa Marta, along with the rest of Cortesi’s fossil
collections, which include the Whale skeletons and the palate (and
other bones) of the Elephas meridionalis figured in the ‘ Saggi Geo-
logici.’ The skull, when De Christol’s figure of it was made by Gené,
appears to have been nearly in the same state as it is now.

Cortesi mentions, that along with the skull he found 10 verte-
bre, 14 ribs, 2 scapule, and the 2 fore legs. On looking over the
fragments in the case, parts of most of these remains are to be seen.
Of the vertebre there is an axis, which is entire, with the exception of
the spinal portion of the neural arch. There are also 8 other vertebre ;
the bones of the fore-legs and the scapule are unfortunately very much
broken. There are 2 humeri, one of which is in three pieces, that do
not admit of being joined. The head of another humerus of very large
size is in the same mineral condition as the other. It belongs to the
opposite side from that in Cuvier’s fig. 9 of Pl. XLI. (Rhin. Pl. IIL,
éd. 8me); but as compared with that it yields the following measure-
ments:—From 6 to d as in fig. 9,7: inches; from a to 6 6:1 inches; from
d to a 6° inches; greatest expansion under the neck, 7°2 inches; trans-
verse diameter of head, 4° inches.

The hooked process below the expansion is present in this specimen,
but all the rest of the shatt is broken off. There are some metatarsal
and metacarpal bones, but.of the radius and ulna and scapule there
are only fragments, not sufficiently perfect for description. Cortesi
mentions having discovered in another place the humerus of a Rhino-
ceros, covered with oyster-shells growing upon it. One of these humeri,
nearly entire (the lower articular head being wanting), is still in the
collection, and the transverse expansion, where greatest below the arti-
cular head, measures only 5°6 inches. It is evidently of an adult
animal, as the epiphyses are united; the bone is impregnated with iron,
and in a very different mineral state from the other decomposed
humerus above measured, and it yields dimensions which are so much
less than that of the other above given, that it probably belonged to a
distinct species, and that species Rhinoceros Etruscus. But I have no
time at present to determine that point accurately. This completes all
that I can do about the Cortesi Rhinoceros.

In the same case are seen the remains of the palate of the Hlephas
meridionalis, figured ky Cortesi. The teeth are the last true molar of
either side; that of the right side is entirely exposed, showing twelve
ridges with a talon plate behind, and also a front talon. Of these, the
front five ridges are more or less worn; the enamel-plate is thick; the
discs wide apart and little undulated, with thick ringed digitations.
The tooth measures in extreme length 11 inches, and the greatest width
of the crown is 43 inches. Alongside of it is the fragment of an enormous
ivory tusk, somewhat oval in section, the greatest diameter of which
yields 9} inches. In the same case there is a portion of a most enormous
sacrum, attached to the last lumbar vertebra. Among the Elephants’
teeth, upper and lower, in this case, I could detect no indications of Hle-

RHINOCEROS LEPTORHINUS. 391

phas antiquus. There is a large collection of Elephant bones in another
compartment, some of them exhibiting enormous dimensions.

Memo.—Cornalia has shown me the posterior fragment of an Ele-
phant’s molar, found in the deposit above the lignite of Leffe (Gandino).
It consists of the last three ridges of the last true molar, lower jaw, right,
together with the talon, of undoubted Klephas meridionalis. The ridges
are worn, but the talon intact. It is a characteristic example of .
meridionalis, with very thick enamel, and thick cylindrical digitations.
It is nowise tinted black, and is stated to have been found above the
lignite. Another fragment of molar, found at the same place, appeared
to me to be of Hlephas antiquus; it was in the same white untinted
condition. Besides these, from the lignite of Leffe itself, Cornalia pro-
cured a worn-out fragment of a large lower molar of an Elephant. It
is difficult to say what the species is, the enamel-plates being too thick
for EL. primigenius, and too thin for H. meridionalis. It is probably
either of EL. antiquus or E. Armeniacus; the discs show very little un-
dulation of the enamel-plates, but the crown is especially remarkable in
having the discs separated by a longitudinal fissure (filled up with
cement) like the singular Hlephant’s molar from Durdham Down,
which I observed in the Museum at Bristol. Besides these, some
lower teeth of Rhinoceros were found in the lignite ; one of these is an
entire penultimate true molar, slightly worn, and of the right side,
exactly resembling in every respect the corresponding tooth in Cortesi’s
lower jaw. It is free from cement, and from the surface rugosity,
observable upon the enamel of Rhinoceros tichorhinus and Rhinoceros
hemitechus. It is certainly not of R. téchorhinus, and I believe it to
belong, like the Cortesi cranium, to Rhin. leptorhinus. Cornalia has also
procured molar teeth and fragments of antlers of small Deer, and some
molars with a long intercolumnar pillar and prismatic form, which I
regard as being of a small species of Bos. Lately he has acquired from
the same lignite deposit some molar teeth and casts of incisors, which
he finds it impossible to distinguish, whether by size or pattern of
crown, from the existing Beaver, Castor Europeus. They are not of
Trogontherium.

The Abbate Stoppani regards the deposit as being a late quaternary,
Gandino being a spot below the horizon, to which the moraines of the
southern glaciers of the Alps in Lombardy extended. On the other
hand, the vertebrate remains, exclusive of the Beaver, appear to me to
indicate a Pliocene age. A fragment of a Mastodon’s molar, tinted
black, is supposed to have come from the same deposit ; but there is no
certain record of its origin, and it cannot be relied upon. Nuts of a
walnut of a very elongated form are very abundant in the same lignite ;
and one of them was got along with the Elephant’s tooth. The species
has been named Juglans Berchenensis? or some such name, by Balsamo
Crivelli. The occurrence of the Beaver’s teeth in this case is very
remarkable, and singularly so, should it really prove to be the existing
species.

Dimensions of the Cortesi Rhinoceros Skull. Extreme length of skull from
broken summit of occipital crest to point of the nasal bones, 28°25in. 2. Extreme
ditto from the posterior plane cf occipital condyles to broken edge (anterior) of
left diasteme, 27-25 in. 3. Extreme ditto from ditto, ditto, to anterior edge of
alveolus of first premolar (left side), about 25:in. 4. Extreme length from anterior
margin first premolar to posterior edge of last true molar, left side (last molar
included in alyeolus), 13: in. 65. Length of last three molars, left side, 6-7 in. 6.

392 RHINOCEROS.

Extreme length of first and second true molars, left side, 4-6in. 7. Length of last
three premolars, right side, 5-6 in. 8. Length of four premolars (to anterior
margin of empty alveolus of first ditto, right side), 6-lin. 9. Length of remaining
portion of-diasteme, left side (measured from anterior margin of first alveolus),
22in. 10, Transverse diameter of empty alveolus of first premolar, right side,
*8in. 11. Antero-posterior ditto of ditto, ‘Sin. 12. Length of second premolar,
left side (crown of tooth broken on right side), 1°95in. 13. Transverse diameter
of ditto near base, behind, 1:7 in. 14. Antero-posterior diameter of third premolar
(left side), about 2-lin. 15. Transverse diameter of ditto at bourrelet (base), an-
terior barrel, 2-4in. 16. Antero-posterior diameter of last premolar, right side,
outer surface (corresponding dooth, left side, broken off, and place occupied by
matrix), 1‘8in. 17. Transverse diameter of ditto at base, anterior barrel, 2°25 in.
18. Length of crown of first true molar, outer surface, left side, 2°3in. 19. Trans-
verse diameter of anterior barrel of ditto (left side), near base, partly concealed by
matrix, about 2-4in. 20. Antero-posterior diameter, outer surface, penultimate
molar, right side (crown shivered on left side), 2-4in. 21. Interval between dias-
temal ridges at commencement, near first premolar, 2°85in. 22. Length from
anterior border, right orbit, to outer edge of cast of occipital plane, right side,
about 16°0in. 238. Length from ditto, ditto, to tip remaining of nasals, 13: in.
24. Length from posterior plane of occipital condyles to posterior margin of last
true molar, about 13-in. 25. Diameter between outer margins of occipital con-
dyles, 64in, 26. Transverse diameter, right condyle, taken near the middle,
2-2in. 27. Vertical height of ditto, 26in. 28. Diagonal diameter of ditto
(greatest), 3°2in. 29, Width of occipital foramen (greatest), about 2°5in. 30.
Height of occipital plane to lower surface of occipital condyles, 10-5in. 31.
Greatest width of occipital plane just above the condyles, 9-lin. 32. Greatest
width of ditto about middle, 7-2in. 33. Length of zygomatic fossa, left side, 5 in.
34. Length from the posterior boundary zygomatic fossa to the posterior surface
of the occipital condyle, left side, about 8°6 in. 35. Extreme length from the tips
of the incisive to the broken edge of the occipital crest, left side, measured as a
straight line, 28-75-29 in. 36. Extreme ditto from the anterior margin of the
orbit, right side, to the tip of the nasal, 13-in. 37. Extreme ditto, ditto, ditto,
left side, to the broken edge of the occipital crest near the left summit, 16°75 in.
38. Length (versed sine) of cord stretched from greatest convexity of nasals to
summit of occipital crest where slightly broken, left side, taken on plateau be-
tween the orbits, 2°3in. 39. Length of ditto, taken at constriction of frontals
between the zygomatic arches, 3-in. 40. Length from the posterior surface, occi-
pital condyles, to tip of the nasals (a long curve), 31-in. 41. From tip of the
nasals to lateral margin of occipital ridge, above the left auditory foramen,
26°5 in. 42. Length from anterior margin auditory foramen to anterior margin of
the orbit, 12°in. 43. Thickness of the nasal bones: taken at the middle, 1:4 in.
44. Width of ditto, ditto, 4-in. 45. Greatest contraction of the cranium between
the zygomatic fosse, 55 in. 46. Height of the occiput above the lower plane of
the occipital condyles (occipital crest partly broken), 10°5 in. 47. Height of
jaw from edge of third premolar to convexity of nasals, left side, 10-7 in.

Measurements of Lower Jaw of Cortesi's Rhinoceros—1. Extreme length from
posterior margin of ascending ramus to broken edge of incisive beak, right side,
23°25in, 2. Length of edentulous beak from beginning of diasteme, 3°25in. 3,
Width of symphysis at. contracted portion at commencement of diasteme, 2°7 in.
4. Length of line of molars, left side, as visibly exposed, 9°6in. 4. Length of
ditto, right side, ditto, 9°6in. 5. Antero-posterior length, right side, of ascending
ramus above alveolar level, 63in. 6. Height from posterior angle to middle of
sigmoid notch, 9°7 in. 7. Length of two last molars, left side, 4-3in. 8. Length
of anterior two, ditto, 3-lin. 9. Length of gap between, 2°2in. 10. Length of
last exposed molar, left side, 2°2in. 11. Length of penultimate ditto, ditto, 2-in.
12. Length of anterior molar, ditto, 1:3in. 13. Length of second ditto, ditto,
16in. 14. Height of jaw at contracted part of symphysis, 2-2in. 15. Height
of jaw to alveolar margin between first and second molars, right side, 3-6in. 16.
Interval between the posterior crescents of the last visible molars, 4°4in. 17. In-
terval between the anterior edges of p.m. 2, inside, 2°7in. 18. Interval between
p-m. 8, inside, posterior margin, 4:in. 19. Interval between anteriorends of t.m. 1,
49 in. 20. Ditto between posterior crescents of t.m. 2, inside, 4-4 in. 21. Height
of jaw to margin of alveolus of antepenultimate premolar, right side, 2°5 in.

RHINOCEROS LEPTORHINUS. 393

X.—Drescriprion oF Lower Jaw or Ruatnoceros LEPTORHINUS FIGURED
By Correst.!

London, October 138, 1862.

The description which follows is believed by me to be of the
missing lower jaw of Rhinoceros figured by Cortesi, and which Ca-
pellini tells me was discovered, since my visit, in a box at Parma, by
Strobelli.

Among the marbles and polished stones of the Italian Court in the
London Exhibition of 1862 are two rami of the lower jaw, evidently
right and left of the same individual, of a fossil Rhinoceros, believed to
have been sent by Professor Scacchi of Naples. The left side is entire
from the ascending ramus to the symphysial margin, the condyle alone
being wanting. On the right side the anterior part of the ramus, as far
as the third premolar, has been crushed by a recent injury. The jaw is
evidently that of an adult animal, with six molars in situ, all of them
fully in wear, but the abrasion of the crown of the last true molar is
not very far advanced. ‘There are six molar teeth out, but no appear-
ance of the socket of the pre-antepenultimate or first premolar. 'The
symphysial beak is perfect on both sides, with a very short diasteme,
which shows a doubtful trace of a socket for an incisor.

Dimensions :-—

Length of the line of six molars, 9:25 in. Joint length of three true molars,
51in. Ditto of three premolars, 3°9 in. Length of crown of last molar, 1°7 in.
Greatest width of ditto, 1:1 in. Length of penultimate, summit of crown, 1°7 in.
Greatest width of ditto, 1:2 in. Length of antepenultimate, 16 in. Length of
last premolar, 1:4 in. Ditto of penultimate ditto, 1-3 in. Ditto of antepenulti-
mate ditto, 1:05 in. Ditto from anterior edge of antepenultimate premolar to
incisive border, 1:7 in. Ditto of diastemal ridge, 0°65 in. Height of ramus under
penultimate premolar, 24 in. Ditto at middle of last true molar, 2-9 in. Greatest
thickness of ramus (about), 271 in.

The first premolar is not very far advanced in wear, the anterior
part of its crown being still intact; the penultimate is further ad-
vanced, having both barrels worn so as to have confluent discs. The
last premolar is nearly in the same state of wear, but less advanced.
The first true molar is worn very low into a uniform sinuous depressed
disc. The second is less worn, showing a horse-shoe pattern to the
front division, confluent with a simple cornu to the hind division. The
last molar has the anterior and posterior discs quite distinct and at dif-
ferent levels, the anterior one showing a disc of a form between a sagit-
tate and horse-shoe pattern; the hinder disc forms a narrow band, but
slightly curved into a kind of clavate form and at a much lower level
than the anterior. Regarded from the outer side, the anterior barrel of
the last true molar and of the penultimate shows distinctly the oblique
erenate bourrelet indicated by De Christol in his &. megarhinus. On
the right side the same bourrelet is shown on the premolars still more
distinctly. The enamel surface is comparatively smooth, as in &. me-
garhinus, and perfectly free from the reticular inequalities so boldly
shown in R. tichorhinus. On the inner side it is perfectly smooth and
shows occasionally the parallel lines characteristic of 2. megarhinus

1 This is evidently a different lower jaw from that already described.—[Ep. ]

394 ' RHINOCEROS.

and R. Etruscus. The symphysial part of the jaw and the diasteme, in
their sudden abbreviation and general contour, remind me very closely
of Gervais’ drawing of R. megarhinus. Unluckily the lower surface of
the symphysis is either broken or covered by matrix, so as to conceal the
character there yielded by the foramina.

The left ramus on its outer surface is distinctly covered by sea-shells,
some of which are of a Patella-looking form. The lower border of the
ramus is nearly in a horizontal line from the posterior angle,as far as
the last premolar; it then curves gently forwards to rise suddenly
upwards into the beak, in a line with the anterior edge of the ante-
penultimate premolar. On the whole, I am satisfied that the specimen
belongs to Rk. megarhinus (A. leptorhinus, mihi.)

The outer surface of the ramus is convex, but the inner is flat, with
a broad longitudinal shallow channel. The teeth appear to have been
covered with a considerable coat of cement. On the right side, at the
middle of the diasteme, and about half way into the incisive border,
there is an indistinct appearance of a triangular pit, as if the residuary
socket of a small shed tooth; there is no such evidence on the left side,
in consequence of a layer of matrix.

XI.—Rurnoceros Lrproruinus at Pisa.
May 22, 1859.
The Rhinoceros specimen from the Ardenza bone-breccia, contain-

ing the antepenultimate and penultimate true molars, left side, isnot of
Kt. hemitechus, but of R. megarhinus.

XII.—Descrietion oF Remains or Rutnoceros LEPTORHINUS IN THE
Museum at Imoua. .

May, 1861.

Came on last evening by Faenza from Ravenna, and went out this
morning at 5 A.M., with Signor Scarabelli the Syndic, and Capellini, to
see the locality where the Rhinoceros bones, &c., in the Museum were
found. Drove about due 58. parallel to the Santerno, towards the hills ;
crossed the river, and then entered a small valley, that of the ‘ Rio
dell’ Acque Marine,’ where the proprietor, Signor Cerchiani, a friend of
Scarabelli’s, had collected through the villagers the Rhinoceros and other
bones. The sections are beautifully shown, somewhat as in the Sewalik-
hills.

1. Uppermost yellow quaternary loam or lehm.

2. A thick bed of stratified gravel in a hard sandstone cement,
quaternary.

3. Thick beds of yellow sand, containing Cardium edule, &c., with
occasional seams of gravelly conglomerate.

4. Blue clay, containing walnuts with elongated fruit, the same as
those at Milan (p. 391). Saw nothing exactly corresponding
to the Sansino beds of the Val d’Arno.

Signor Cerchiani had the bones collected for him by the contadini,
who found a superb skull of a fossil Rhinoceros and broke it into bits to
get their separate reward for each piece, a baiocco per fragment! Sca-
rabelli repaired the broken teeth, and has fitted the whole series of either
side very cleverly into separate slabs of plaster of Paris, exactly in their

RHINOCEROS LEPTORHINUS. 395

natural position, including the six molars of each side from the ante-
penultimate premolar (p.m. 2) to the last true molar (m. 3), inclusive.
(See Plate XXXI. fig. 1.)

The molars (see Plate XXXI. fig. 1), on the whole, are admirably
preserved, better even than the Bologna specimen of R. Htruscus (PI.
XXIX.), and in a beautiful state, so far as age goes, to show the dental
characters, t.m. 2 being about half way worn above the basal bourrelet,
and t.m. 3 with its apex only partially worn; p.m. 4 and t.m. 1 of
either side much worn.

The following are the principal dimensions on right side :—

Extreme length of line of six molars from hind tubercle, last molar, to antepe-
nultimate p.m. 10°6in. Length of three true molars outside, 6-2 in. Ditto in
middle, 5°8 in. Ditto of three premolars, 4:9 in. Length of p.m. 2, top, outside,
1:55 in. Width of ditto, greatest, 1:6 in. Length of p.m. 3,1°8in. Width of
ditto greatest (below bourrelet), 2°2 in. Length of p.m. 4 ditto, 1-°9in. Width of
ditto (greatest in front), below ditto, 2°3 in. Length of t.m. 1 (greater on left
side, but restored), about 2° in. Width of ditto in front (bourrelet worn away), 2°40
in. Length of tm. 2, which is very perfect, 2°3 in. Width of ditto in front,
below bourrelet, 2°5 in. Length of t.m. 3 diagonally from anterior angle to basal
tuberele, 2:3 in. Width of ditto at base of front barrel, 2:25 in.

General Remarks.—1. The first point that strikes is, that the three
premolars have a very large basal cingulum, quite as large as that
figured by Christol. It is largest in the third and fourth, and very
oblique in its direction, rising gradually from the base of the anterior
barrel to the top, behind, of the posterior barrel (¢.e. from the anterior
talon to the edge of the hind valley).

2. The true molars have also a very distinct basal cingulum (!). This
is nearly worn away in the antepenultimate, but is shown in very bold
relief upon the anterior barrel of the penultimate, and interruptedly, but
quite clearly, on the posterior barrel. The same cingulum is shown
very boldly on the anterior. barrel of the last true molar, but is not
exhibited on the posterior barrel of this tooth, which is narrow at the
base.

3. In lieu of the rudimentary pit on the hind part of the base of the
last true molar, which is seen in the R&. Ktruscus of the Bologna Mu-
seum, the Imola tooth (t.m. 3) shows a distinct triangular or sagitti-
form lobe or tubercle (like a Celtic arrow-head), adpressed to the pos-
terior barrel, but separated from it at the apex by a very pronounced
notch. This tubercle is somewhat crenated at the apex, but utterly
distinct in form from that cf R. Htruscus or R. hemitechus. There is
not a trace of a posterior valley running up upon the posterior angle of
the last molar.

4. The vertical external furrow of the anterior angle is broad and
very boldly defined by a deep groove in all the true molars, and also in
pm. 4. This is shown also in p.m. 3, but less boldly. In this respect
the teeth are very different from those of &. Etruscus. The other
ridges and furrows of the outer surface are also shown more distinctly
in the Imola specimen than in F. Ktruscus.

5. There is not the least indication of a basal bourrelet outside (as in
Aceratherium).

6. The crochet in t.m. 2 makes an obvious angle with a re-entering
nick in its offset from the posterior barrel; the angle is much more

396 RHINOCEROS.

pronounced than in the nickless very open angle of R. Htruscus, but
does not form the right angle of R. henitechus.

7. Pm. 2 is about half worn, and has its anterior barrel much
smaller than the posterior, like a compressed conical cup as in Gervais’s
figure; there are no accessory plates, but a distinct ring isolated on the
base of the crochet.

8. P.m. 8 is much worn; the accessory plates are ground away,
with only a sinuous outline.

9. P.m. 4 shows the same characters, but is still more worn.

10. T.m. 1 is ground down to the cingulum ; the inner termination
of the transverse valley shows a ‘ duck’s-head pattern,’ as in Gervais’
drawings; the crochet is short and very thick.

11. T.m. 2 is in the finest condition, only about a third worn; the
posterior valley is not touched behind; the crochet is thick and forms a
nick at its offset, but at anopen angle. There is a peculiar twist of the
posterior barrel at the apex. The anterior transverse valley has a wide
triangular fissure at its central termination ; there are no combing plates,
but there is a pillar of enamel rising in the middle of it, evidently given
off from the outer ridge.

12. The last molar, as usual, is triangular, but is little worn; its
anterior barrel is very broad; the posterior is narrow. There is no
rudiment of a posterior valley; the middle valley is triangular, with
one large combing plate converging from the outer ridge towards the
crochet ; there is also a similar accessory plate sent off from the anterior
barrel to overlap the crochet; the three processes forming three distinct
converging intrusions into the outer termination of the transverse valley.

In the Imola Museum, from near the same locality in which the skull
was found, but not exactly from the same deposit, there are two rami of
a jaw, each portion containing the series of molars from the second pre-
molar to the last true molar, beautifully preserved.

Both rami are fractured anteriorly in a line with the fangs of the
second premolar, and they are likewise broken posteriorly in the middle
of the ascending ramus.

The lower margin is perfectly entire, but unfortunately the sym-
physial portion and mentary process are missing.

Dimensions on right side :—

Length of the last six molars, 8:5 in. Length of the last three true molars
measured from the middle of the crowns, 5: in. Ditto of crowns of the three
premolars, 3° in. Ditto of the last molar, 18 in. Ditto of the penultimate, 1°65
in. Ditto of the antepenultimate, 1-4 in. Ditto of the fourth premolar, 1:26 in.
Ditto of the penultimate premolar, 1-1 in. Ditto of the antepenultimate premolar,
1:05 in. Height of the jaws between the antepenultimate and penultimate, up to
the alveolar margin, 2°3 in. Height from the middle of the last molar to the
alveolar border, 2°7 in.

The crowns of all the teeth are somewhat worn, 7.e. the animal was
an adult, but not old. Several of the molars of this specimen show the
small characteristic bourrelet, which has been indicated by De Christol.

There is also a third jaw specimen—a left ramus—very well pre-
served, in which the molars are less worn than in the two preceding.
This specimen is broken vertically in front of the penultimate premolar,
and therefore exhibits only the last five molars.

Length of the last five molars, 8:2 in. Ditto ditto of three last true molars, 5°3
in. Ditto ditto of two premolars, 2°8 in.

RHINOCEROS LEPTORHINUS. 397

This specimen is fractured anteriorly and posteriorly like the other
two; the symphysial portion is missing. The crowns of the molars are
very little worn, and are beautifully preserved ; the transverse bourrelet
of the outer side is well shown at the two extremities of the penul-
timate true molar, and is crenated. The same character is seen in the
anterior portion of the last true molar, less so in the antepenultimate,
and still less in the last premolar. The margin of the ramus in this
specimen is exactly equal to that of the other two fossils; it belongs
like them to the same species, to which the skull must also be referred,
i.e. R. leptorhinus (Cuv. pro parte), R. megarhinus (Christol). Of
the detached molars, of which there is a large number, all exhibit
the characters of &. leptorhinus; not one can be referred to LR.
Etruscus.

There are two specimens of the last true molar, upper jaw, one right,
the other left, both showing the posterior Icbe, instead of the fossette as
in &. Htruscus.

In one of the specimens, that of the right side, the crochet forms
a connecting bridge, extending between the anterior and posterior
portions.

XIUI.—DescrieTion or Remains or R. LEPTORHINUS IN THE SCORTEGAGNA
CoLLECTION AT VICENZA.

May 31, 1861.

Tn this collection there is a lower jaw, right side, of a fossil Rhino-
ceros found in an osseous breccia, which corresponds exactly with the
ordinary breccia of ossiferous caves. The jaw is fractured and covered
with a matrix, crammed with fragments of bone. The six last molars
are seen; in the first of these the crown is wanting, but the two fangs
remain ; the last is displaced. ‘The first true molar exhibits De Christol’s
transverse bourrelet, and from all the characters it appears to me that
the specimen belongs to the &. megarhinus of Montpellier.

Dimensions :—

Length from anterior part of penultimate premolar, to posterior portion of pe-
nultimate true molar, 7-2 in. Ditto of penultimate true molar, 2:(?) in. Ditto
antepenultimate ditto, 1°75 in.

In the same collection there is shown the corresponding ramus
perhaps of the same animal, with four teeth 7m situ, the last of which is
very little worn. There is also a mass of matrix, containing Cyclos-
toma elegans, and several other molars of the same species of Rhinoceros,
but so involved in the matrix that their crowns are not well seen.

The crown of the last true molar is worn to the middle, and has an
artificial outline of wax round the posterior portion, so that all the
characters cannot be seen. From what is exhibited, the specimen
appears identical with R. megarhinus.

There is also a radius of Rhinoceros (leptorhinus?). The lower part
is entire, but the head is wanting, and the bone is broken in several
places, so that the distinctive characters are not recognizable. It is
described as a tibia of Hippopotamus.

398 RHINOCEROS.

XIV.—Note on Mortars oF Rutnoceros Leptroruinus (R. Mercku,
JAGER), IN THE Museum aT STUTTGART.

June 18, 1861.

Got casts of the three molars upon which Jiiger founded his R. Merckit
of Kirchberg. Dr. Fraas told me that the real history of the discovery
of these specimens is involved in obscurity. They were shown to Jiiger
by the Prince of , residing near Kirchberg, and no additional
specimens have turned up from that quarter. The two upper teeth are
the penultimate and last, evidently of the Grays Thurrock species, 2.
leptorhinus (R. megarhinus). The original penultimate is in very fine
preservation. [Figures of two of these casts, executed by Mr. Dinkel,
will be found in Plate XXXII. figs. 1 and 2.—Ep. ]

XV.—Menmo. or Ruroceros LEPTORHINUS FROM THE FOREST-BED.

August 25, 1863.

In Mr. Gunn’s collection there is a very fine specimen of the last pre-
molar, upper, right, of R. leptorhinus (2. megarhinus ), which shows the
characters perfectly and is a certain proof of R. megarhinus from the
Forest-bed. [The characters are described in detail and are shown to
differ from those of 2. Etruscus. In a letter to M. Lartet, dated June
25, 1865, Dr. F. also remarks:—‘ The Rhinoceros leptorhinus of Grays
Thurrock occurs elsewhere in England in a peat-bed, which is below
the loess, along with Elephas primigenius. —E». |

XVI.—Note on Remarns oF Rurnoceros Leprornimus (R. MEGAR-
HINUS), IN Dr. SpurRRELL’s CoLLecTION aT BELVEDERE.

Sept. 30, 1863.

There are four detached upper molars belonging to this species. One
is a last true molar (t.m. 3), right side, in the finest preservation, and
only slightly advanced in wear. In its transverse diameter from the
outer angle to the inner side barrels, it agrees very closely with the
Montpellier cast brought for comparison, but the width is considerably
less; it shows no indication of any rudimentary basal valley behind.
Another specimen of the same species is a penultimate upper left molar,
which agrees in the most surprising manner in form, size, stage of wear,
and hook of the posterior barrel with the Rk. Merckii cast from Stuttgart,
which was brought for comparison with it. Dr. Spurrell and Messrs.
Woodward and Prestwich were struck with the identity. With regard
to mineral character the four teeth of R. megarhinus present a tint
which seems to me to differ a little from that shown by the R. ticho-
rhinus (see page 401), while the latter have besides a rough and rolled
general character which is not so obvious in the former. On the other
hand, Prestwich considers that there are three teeth of the 2. tichorhinus,
which, in mineral character, closely resemble the R. megarhinus, whilst
the slight difference in tint may arise from difference in the facility
with which the different species stain! the matrix being in both cases
alike—sand with green grains of flint pebbles. He admits, however,
that it is a case for inquiry.

Ne eee
, 7

RHINOCEROS ANTIQUITATIS. 399

XVII.—Nore on Remains or Ratoceros Lertoruinus IN THE Museum
or Le Puy, AUVERGNE.
Sept. 15, 1863.

Examined a fine specimen of left side of lower jaw of 2. megarhinus
from Solilhac (said by M. Robert to have been found along with the
bones of the skeleton which I have attributed to R. Htruscus!). It has
the six molars en suite, the last but little worn. The outer side of the
angle has the deep rugosities exhibited by Gervais’ figure. Length of
four last molars, 6°5 in.

In the same Museum there is also a magnificent palate series of R.
megarhinus (Merckii pattern), according to M. Robert, found in ‘ des
fentes 4 ossements éruptives du collet Polignac.’ It contains the six last
molars on both sides, all a little worn. Length of six molars, 11 inches.

The teeth are very finely preserved, and exactly like the large Grays
Thurrock specimen in the British Museum ; they are very fresh and
modern looking.

IV. NOTES ON RHINOCEROS ANTIQUITATIS (Biums.) R.
TICHORHINUS (Fiscu. anp Cvv.).

T.—Rauroceros ANTIQUITATIS FROM Wooxkry Hor, Taunton, AND
UpHILL CAVERN.
Taunton Museum, April 13, 1858, and Bristol, May 1858.

Examined upper and lower molars of R. tichorhinus from Wookey
Hole, a lately discovered cave in the Mendip Hills. From the same
cave there are molars of HL. primigenius, a magnificent canine of the
Cave Lion, remains of Hyena, &c.

In the same Museum there is a skull of a R. tichorhinus, three-fourths
grown, found in digging the foundation of the jail. It contains on either
side the five posterior teeth, the penultimate and last in germ, and the
last not fully emerged from the alveolus. There are also numerous
detached teeth of the same species.

In the Bristol Museum are two lower molars of &. tichorhinus from
Uphill Cavern, very pronounced by their rugosity.!

Il.— Comparison oF Mr. Boryp Dawkins’s SPECIMENS OF RHINOCEROS
Moxrars rrom Wooxkey Hote.
March 25, 1862.”

They consist of two milk molars, probably from the dimensions
penultimates (m.m. 3) of the upper jaw, the one (10 D) of the left
side, the other (10 A) of the right; 10 A is considerably more
advanced in wear than the other. There are three insulated valleys;
first, there is a fissure formed by the great transverse valley, the open-
ing of which is blocked up by a much higher step than in the same
teeth of R. hemitechus, in this respect agreeing with R. tichorhinus.
There is no basal bourrelet at the inside, but in this case a small and

1 Dr. Falconer also identified speci- | Wookey Hole.’—{ Ep. ]
mens of R. hemitechus from Wookey | 7? In the same year Dr. Falconer iden-
Hole. In a letter to Col. Wood, dated | tified remains of R. tichorhinus in col-
July 8, 1862, he wrote: ‘Mr. Dawkins | lections from Kent’s Hole at Torquay.—
intaly got veritable R. hemitechus from | [Ep.]

400 RHINOCEROS.

rather pointed tubercle is appended to the posterior barrel. The second
valley is formed by the confluence of the combing processes; it is very
round and insulated, with vertical walls differing from all Colonel
Wood’s Gower specimens. The posterior valley is also insulated all
round, with rather vertical walls. The vertical furrows upon the outer
surface are well pronounced ; the enamel surface, especially at the ends,
is decidedly rugous; there are three fangs. I have compared it with
the drawings of Colonel Wood’s specimens of R. hemitechus, and with
the small ‘ Long Hole’ milk molar, from all of which it is decidedly
different. The smoothness and thinness of the enamel in the latter is
strongly pronounced. In the form of the fissure, in the roundness of
the small valley, and in the enamel surface, it closely agrees with the
still more worn milk molar of R. tichorhinus from ‘ Long Hole,’ Gower,
and I infer it to be of R. tichorhinus.

10 D. resembles 10 A. very closely in all its characters, but is con-
siderably less worn, and it shows large fangs. The large transverse
valley forms an isolated fissure, with a high step blocking up its open-
ing as in &. tichorhinus, but there is no tubercle. The small middle
vailey is a round ring with vertical walls, but not quite insulated on its
inner side, there being a narrow cleft between the combing processes.
The posterior fissure forms a deep and rather vertical pit, the edge of
which is intact behind. In the characters of enamel surface, and outer
vertical furrows, it agrees entirely with 10 A. The posterior fissure in
the ‘ Long Hole’ (Gower) specimen is much more gaping and triangular
in its marginal outline, and very much more depressed at its hind
border. I believe Mr. Dawkins’ specimens to be of R. tichorhinus and
not of any form of &. leptorhinus. The R. megarhinus has far more
combing plates.

JIJ].—Memoranpum oF FracMent oF Lower JAw oF Rurnoceros
TICHORHINUS, WiTH Mirtx Dentirion, From Wooxkry Hox.

December 7, 1862.

Mr. Dawkins’ specimen is a fragment of the anterior portion of the
left side of the lower jaw, containing the first three milk molars in situ,
with about one inch of the diastemal and symphysial portions; the last
milk molar (m.m. 4) is wanting. With this exception, the Wookey
specimen resembles in the very closest manner the Lawford specimen
figured by Owen in the ‘ Brit. Foss. Mam.’, pp. 338 and 363, Cuts 128
and 137. The m.m. | is all but intact at the apex of the cusp. The
m.m. ° has the middle cusp and posterior crescent slightly abraded,
but the anterior edge is intact; m.m. 3 has both crescents slightly
abraded.

M.m. 1 in form is exactly like p. 1 of Cut 137, and mm, 2 like
p- 2, both of natural size and seen from inner side, the latter showing
the double cusps of the middle more pronounced.

The diastemal portion, which is shown entire for about °6 of an inch,
is very rounded. The enamel is rugous and there is no cement. The
jaw is low and the inferior edge is rounded forwards, and very broad
and flat. There is not the least appearance of incisors or their pits.
There is one large mentary foramen, at about 3 of an inch in front of
the anterior (m.m. 1) tooth, at about the middle of the height of the jaw.

RHINOCEROS ANTIQUITATIS. 401

The following are the principal dimensions: —

Length of fragment, 4:6 in. Joint length of three milk molars, 3°1 in. Length
of mm. 1,07 in. Ditto of mm. 2,1:0in. Ditto of mm. 3,13 in. Height of
jaw under m.m. 1, inside, 1-6 in. ‘Ditto at hinder end of m.m. 8,2:0 in. Greatest
thickness of ramus. below, at section, 1°5 in.

The jaw at hind section is gnawed, but not deeply scored, as if by
Hyena.

This specimen confirms my former doubts,! that the Lawford speci-
men has the milk dentition, and not the permanent, as described in
the ‘ Brit. Foss. Mammalia.’

TV.—MemoranpDum or SkuLts oF Ruinoceros ANTIQUITATIS IN THE
StuTraart Museum.

June 18, 1861.

Saw two skulls of Rhinoceros tichorhinus, found in the Lehm, near
Stuttgart; one of them very large but somewhat crushed. The molars,
lower jaws, and other bones of this species, are very numerous. Looked
over the whole of them,, but saw nothing in the slightest degree resem-
bling either Rhinoceros. hemiteechus, Phinoc. lepton hinus, or Rhinoe.
Etruscus. (See antea, p. 398.)

V.—Mo tars or Rurnoceros ANTIQUITATIS IN THE CoLLECTION oF Dr. F.
SPURRELL, BELVEDERE.

September 10, 1863.

Of Rhinoceros tichorhinus there are fourteen characteristic and well-
marked detached upper molars, including a pair of last (m. 3) of oppo-
site sides. They are all highly characteristic specimens of the species,
7.e. the enamel is thick and rough, and the valleys are three and ver-
tical. They are in a ruder state and appear to have been rolled or
tumbled about much more, than the leptorhine molars in the same col-
lection. Woodward, Prestwich, and myself are agreed upon this.
(See antea, p. 398.)

VI.—Meno. or Ruinoceros ANTIQUITATIS IN Mr. GRANTHAM’S COLLECTION.

September, 1863.

Posterior part of the cranium of Rhinoc. tichorhinus, including nearly
the whole of the occiput with the left condyle quite entire. The occi-
pital crest is perfect, and on the left side the parietal and temporal re-
gions, with the auditory foramen and the styloid process, are nearly
perfect. The skull, in situ, was probably entire. There ‘are also
several fragments of the upper jaw containing teeth. One left maxil-
lary Peniains four molars of an adult animal in situ. The teeth show
distinctly the character of Rhinoc. tichorhinus. There are no lower jaw
specimens, but several detached lower molars. The only remains of
Rhinoc. leptorhinus in this collection is one molar, very far advanced
in wear and very like Dr. Spurrell’s (p. 398).

' | Expressed in Note-books after exa- Oxford Museum in 1858. See also
mination of the Lawford and Wirks- | antea, p. 348.—[En.]
worth Jaws of R. tichorhinus in the

VOL. II. DD

402 ‘ RHINOCEROS. paras:

VIL.—Menmo. or Jaw AND Mo.ars oF RaINoceros ANTIQUITATIS IN THE
Museum or Le Poy.

_ September, 1863.

Examined a lower jaw, left side, broken in’ Gaur and _ behind, of
Rhinoceros tichorhinus, and two detached upper and three lower molars
of same species, labelled Rhinoc. Mesotropus, by Aymard, in his
handwriting. The jaw is youngish’and contains the last five molars 7
situ, the last not quite emerged. There is also a block of plaster of
Paris containing four molars, not consecutive, of upper jaw, right side
(i.e. two true ‘molars and two premolars), named &. Mesotropus in
Aymard’s handwriting. According to Aymard, both R.megarhinus and -
R. tichorhinus belong 1 to his R. Mesotro opus. The specimens are from
‘ Attérissements de Paradis pres Espaly.’

VIII.—Nore on Specimens or R. AnTIQUITATIS IN MarpsToNE Museum.
September 28, 1863.

In this collection I found five upper molars of Phinoe. tichorhinus,
from Stroud ; six upper molars, including two fine last upper, from the
brick-earth at Thornhill, at back of Maidstone Jail, one of which is
very remarkable and ought to be figured; the lower end of a right
humerus from Burham; and the fragment of a tooth, far ddvanced~ in
wear, from the railway cutting near St. Peter’s Church.

NOTES ON DENTITION’ OF LIVING SPECIES OF
RHINOCEROS.

J.—NoTE oN RHINOCEROS KeErtLoa,.
Saffron Walden, October 8, 1861.

The Saffron Walden Museum contains a beautifully perfect skeleton of
an adult Rhinoceros, got at the same time as the Elephant from Algoa
Bay (see antea, p. 265), but the ticket indicating South Africa. It
bears the name of Rhinoceros camus or I. simus of Burchell.

In the upper jaw there are seven molars all protruded, but the last
true molar barely.touched by wear. There are four .premolars and
three true molars. ‘The premolars are surrounded by a distinct basal
cingulum ; but in the progress of wear only two pits have been left, and
the form of the crown is exactly that of the two-horned Rhinoceros of
Sumatra, and totally different from the Tichorhinus pattern. . Unfortu-
nately the two intermaxillary bones have been lost or omitted in
mounting the skeleton, but it is apparent that there was a short dia-
stemal edge in front of the first premolar.

As regards the lower jaw the dentition is quite complete. There
are four premolars and three true molars, all of them affected by wear,
except the last. The first premolar has a flattened crown, with a single
fang, and is of moderate size, immediately in front of which is the
nearly filled up alveolus of an outer incisor which had been shed, and
of which the fang-pit is in progress of filling up. Inside of it there is,
on either side, and immediately contiguous to it, the pit, nearly eradi-

DENTITION OF LIVING ‘SPECIES. 403

cated, of a rudimentary incisor. There were, therefore, four incisors
below, deciduous in the adult animal, and which were in immediate
contiguity with the molar series without the interruption of a diasteme.
On the whole, the dentition of this skeleton reminds me very much of
that of the adult Rhinoceros bicornis of Africa.

There is one peculiarity in the skull deserving notice, viz. that while
the suborbital foramen on the left side is single, on the right side there
are distinctly three foramina disposed in a triangle. The skin of the same
skeleton has been mounted, forming a very fine specimen. It presents
two horns, of which the aoe is “27 inches long, and 184 inches in
girth at the base. The posterior horn is contiguous at the base with the
‘anterior. * It is of large size, measuring about 13 inches in height and
173 inches in girth at the base. On referring to. the excellent figures
in Anderson’s ‘ Lake Ngami’ (p. 386), it would seem that the Sailr on
Walden skeleton is a R. Keitloa, both horns being of considerable
length; in Anderson’s figure they are subequal.

IJ.—Nore on Ruinocenos CAMUS.

In the same Museum there is also a skeleton of a little Rhinoceros
camus. Both jaws show seven molars on each side, the seventh i in the
upper jaw being barely out. The front. of the lower jaw shows the
pits of four incisors which had fallen out, the pits being nearly filled
up; the two outer are large, the two inner small. The alveolus of the
outer incisor is overarched by the first premolar, there being only two
lines of interval.

IM.—Norte on DENTITION OF RutNoceRos BICoRNIS.
™ Leeds, July 17, 1858.

“e the Natural History Museum there is a very fine skull of Rhino-
ceros bicornis in the best possible age for the comparison of the denti-
tion, but it has no lower jaw. All the permanent teeth.are in place;
the premolars are well worn, and the last molar is just worn sufficiently
to show the pattern perfectly. The 3rd-right premolar has three
distinct fossettes, and the 2nd, 3rd, and 4th have a distinct basal
bourrelet on the inner side very salient and marked, and continuous with
the anterior and the posterior bourrelet. There is no inner bourrelet
to the three last molars. The last molar has an interior bourrelet, but
only one or two warty tubercles. posteriorly. The posterior barrel is
bifureate. There is a very minute rudimentary incisor in the imcisive
bone on the right side.

DD 2

404 - HIPPOPOTAMUS...

VI. NOTE ON THE EXISTING HIPPOPOTAMUS LI-
BERIENSIS (MORTON), WITH A SYNOPSIS
OF THE HIPPOPOTAMIDA, FOSSIL AND
RECENT.!

Dr. Morton’s discovery is one of the most interesting and remarkable
made in Zoology during. the present century. Cuvier, in his ‘ Discours
Préliminaire,’ has entered into an elaborate argument against the
probability of any remarkable existing large species of land animal
remaining: to. be discovered, after the search which has been made
through the continents and great islands of the globe. Dr. Morton’s
discovery proves that the inference was premature. ;
Hippopotamus Liberiensis is perfectly distinct from H. amphibius.
It differs more from the latter than H. amphibius does from H. major
and H. paleindicus, the only two fossil species of the same subgenus
of which the crania are known. The distinctive characters are very
strongly marked, and are chiefly the following, viz. the length of the
cranium proper as compared with that of the face; the advanced po-
sition of the orbits, which are nearly in the middle of the head; the
convexity of the forehead, both from back to front and across the orbits ;
the less hooking forwards of the leafy expansion and the greater eleva-
tion of the coronoid process of. the lower jaw, as well as the absence of
any diasteme. . The dental characters are also different. The trefoil
disc of wear in the true molars is exactly as in Hippspotamus amphibius,
but the canine teeth of the upper’ jaw alone, without reference to the
dimensions and other peculiarities, would establish the distinctness of
the species. In H.. amphibius and H. major, the internal vertical
channel is-shallow, while in H. Liberiensis it is so deeply grooved as
to yield a strongly marked reniform outline in the section. This ¢ha-
racter is of especial interest, as it is constant and nearly to the same
amount as in. an Indian fossil species of the subgenus Hexaprotodon,
viz. H. Sivalensis. The lower tusks resemble closely those of H. am-
phibius in form and direction. There are four incisors in the upper
jaw which are slightly curved and nearly vertical. In the lower jaw
there are only two incisors, which are much stronger than those in the
upper jaw and project almost horizontally forwards. It would appear
to be the intermediate incisor which is alone developed on either side.

1 Tn 1847 Dr. Morton transmitted his | been made to the original note from the
specimens to London, to be examined | author’s Note-books. The Synopsis of
by Dr. Falconer, who contributed the | Hippopotamide and the Memorandum
greater part of this note to Dr. Mor-/} of the Hippopotamus skull at Dublin
ton’s essay which was published at | have also been extracted from the Note-
Philadelphia in 1849. Additions have | books.—[En.]

HIPPOPOTAMUS LIBERIENSIS. 405

Measurements of Hip. (Tetraprotodon) Liberiensis——From occipital crest to
anterior margin of incisive bone, 12°8 in. From ditto to tips of nasals, 11°8 in.
Width between posterior borders of orbits, 5° in. Width between anterior ditto
at foramen, 3°6 in. Greatest width zygomatic arches ($ 3°9), 7-8 in. Width of
occiput, 8:lin. Width of head at contraction (orbit. foramen), 2°5 in. Width of
ditto at bulges of canine alveoli, 5:25 in. Width of incisives, 8°in. Height of
head to margin of alveolus at suborbital foramen, 3.5 in. Length from occipital
erest to suborbital foramen, 7°75 in. Length from ditto to posterior border orbit
(posterior orbital foramen), 4°75 in. From ditto to anterior border of ditto, 6:25
in. Antero-posterior diameter orbit, 1:9 in. Vertical ditto, 2°1 in. Interval
between interior surface of zygomatic arches at point of greatest expansion, 7°6 in.
Length of alveolus at margin of molars, 6° in. Interval between anterior false
molar and canine, 0°25 in. From ditto to inner margin alveolus, middle molar,
2-1in. Width of palate between the’ tusks, 2°75in. Width between the outer
incisors, 1:75 in. Width between outer false molars, 2° in. Width between third
false molars, 1:4 in. Length of three posterior true molars, 3:1 in, Width of
forehead between middle of orbits, 3°75 in. Interval between middle incisors, 1°
in. Length of alveolar margin of two incisors (of one side), °9 in. From anterior
margin of orbit to tip of the muzzle, at middle incisor, 7-in. From ditto to crest of
occiput, 65 in. Length of nasal bones, 6:1 in. Depth of nasal opening, 1°75 in.
Transverse diameter of nasal opening, 1:75 in. Interval between tips of posterior
orbitary processes, 1-1 in. Transverse diameter of tusk, 1-lin. Vertical dia-
meter of ditto, *8 in. Depth of reniform channel, 0°-4 in. Diameter of outer
incisor, ‘4 in. From anterior margin of orbit to orbitary foramen, 1°76 in.

Lower Jaw.—Extreme width of alveoli of tusks, 4:5 in. Interval between
canines, 2°8 in, Interval between canine and incisor, ‘5 in. Interval between
incisors, 8 in. Diameter of left lower incisor, ‘55 in. Between tusk and first false
molar, ‘4 in. Length from posterior margin of condyle to tip of muzzle, 10: in.
Length from posterior margin ramus to ditto, 10°25 in. Length of symphysis, 2:8
in. Width of jaw at contraction below third false molar, 3°75 in. Width behind,
in front of niche, 5°3 in. Greatest expansion of leafy dilatation below, estimated,
9-0 in. Extreme height from lower margin of expansion to tip of coronoid, 5:9 in.
From anterior margin coronoid to posterior ditto of condyle, 2-5 in. Height of
leafy expansion to niche in front of condyle, 4°6 in. Height of jaw to margin of
alveolus between second and third false molar, 2°45 in. Height of jaw at last
molar, 1:9 in. Width of condyle across, 1:25 in. Interval between first false
molars, 2°6 in. Interval between third false molars, 2-0 in. Length of three pos-
terior molars, 3°4 in, Length of alveolar border, four false ditto, 3-16 in. Length
of leafy expansion, 3°5 in.

Hippopotamus comprises two Subgenera, Hexaprotodon, with six in-
cisors above and below, and 7'etraprotodon, with reduced incisors, viz.
four above and below. Hip. Liberiensis, although it has but two in-
cisors in the lower jaw, belongs to the latter subgenus; the excessive
reduction is probably only an individual case of variety, but if proved
to be constant the position of the species would not be altered. The
succession of the species in the subjoined synoptical table indicates the
order of their affinities. No. 1, H. major, is the most divergent form,
with short cranium, posterior orbits, great elevation of the sagittal and
occipital crests, and excessive elevation of the upper margin of the
orbits above the plane of the brow. Next follows H. palewindicus, a
true fossil Hippopotamus from India. Then comes H. amphibius, No. 3
in the series, of which the French naturalists make two species, H. Ca-
pensis and H. Senegalensis. Duvernoy (Comptes Rendus, Oct. 1846)
maintains their distinctness, but I regard them as merely varieties. H.
Pentlandi (No. 4) is the fossil species which prevails in Sicily, Malta,
and Candia. JZ. annectens (No. 5) is a fossil species from the Nile

406 HIPPOPOTAMUS.

above the Cataracts, which was brought to Europe by Dr. Riippell in
1827, and which I have examined in the Frankfort collections (the
Senckenberg Museum). I have named it provisionally H. annectens,
from its forming a link in size between H. amphibius and H. Liberiensis.
The cranium is not known, and further investigations may show that it
is identical with H. Pentlandi. Cuvier’s H. medius has proved to bea
species of Dugong (Halitherium). Next follows Cuvier’s H. minor
(No. 6), which is a doubtful Z'etraprotodon. I range H. Liberiensis
(No. 7) last, from its close resemblance to the Indian Hexaprotodons, in
the form of the upper canines.!

Ot Hexaprotodon there are three well-marked Indian species. H.
Travaticus (No. 8) is a size larger than H. Liberiensis. H. Sivalensis
(No. 9) is less than H. amphibius; and H. Namadicus (No. 10), with
other strongly marked characters, is larger than H. amphibius or H.
Sivalensis. There are portions of every part of the skeleton showing
the closest resemblance to H. amphibius throughout, though it is more
slender in its proportions.

Merycopotamus is a most interesting and well-marked genus, con-
necting Hippopotamus with Anthracotherium. The molar teeth, as in
the latter, are constructed on the ruminant plan; while the cranium,
incisors and canines, together with the leafy expansion of the angle of
the lower jaw, connect it with the former. It was nearly of the size of
HI. Liberiensis.

SYNOPSIS OF HIPPOPOTAMIDZ.
Gen. J.—Hippororamus.
Subgenus L—Tetraprotodon.

1. Tet. Major (Cuv.), European fossil, figured in ‘ Fauna Antiq.
Sival,’ Plate LXII.

2. Tet. palzindicus (Fale. and Caut.), Indian fossil, from the Valley
of the Nerbudda, figured in ‘Fauna Antiq. Sival.,’ Plates
LVII., LVIIL., LXII.

1 Note on a remarkable Hippopota- | right side! and only two as usual on the
mus Skull in Mr. Ball's Collection at | lett. The supplementary incisor about
Dublin.—The specimen is the large end | 1 inch projecting, and between a swan-
of the skull of an adult, but not old, | quill and tip of little finger in thickness.

animal. The rear back molar in both | It is projected in-the same plane as the

jaws is developed, and worn down on | middle ones, being about one-third of
the left side into a trefoil; it is intact | the distance from the canine, and two-
on the right. There are six molars on | thirds from the next incisor. The
the left side of the lower jaw; there are | canine of the same side diverges very
five on the right, the anterior being | much outwards, while on the left side
missing. There are seven molars on | it is erect. Further, there is a supple-
the right side, upper jaw, the anterior | mentary slip of a separate tooth placed
being isolated, and placed about one- | like a splint on the inside and angle of
third forward on the diasteme. There | the right canine in the same sheath, and
are six only on the left, the anterior one | projecting about an inch. The upper
being wanting. The lower jaw is very | and lower right canines had not been in
remarkable, the anterior part being un- | contact to wear; there is an interval of
symmetrical, with three incisors on the } two or three inches.

»y

7 ae

CLASSIFICATION . AND’ SYNOPSIS. 407

°3.-Tet amphibius.
Plate LXII.
Var. a, ‘Tet. Capensis.

Africa, living, figured in ‘ Fauna Antiq. Sival.,’

Var. 6, Tet. Senegalensis.

> Tet. Pentlandi.

Tet. minor (Cuv.),
. Tet. Liberiensis,

NOU

Tet. annectens, Nubian fossil.
fossil.
living.

Subgenus Il.—Hewaprotodon.'

8. Hexap. Iravaticus (Fale. and Caut.), Indian fossil from Ava,

ficured in ‘ Fauna Antiq: Sival.,’

Plate LVIL.

Be Hexap. Sivalensis (Fale. and Cant. ), Indian fossil from Sewalik

hills, figured in ‘ Fauna Antiq. Sival.,’
LXV,.,

Pe 1 XI LXIV.,

Plates LIX., LX., UX,
LXVL 2

10. Hexap. Namadicus (Fale. and “Cait: ), Indian fossil : from the
3 Valley of the Nerbudda, figured in ‘ Fauna pea Sival.,’

‘ Plates LVIL, LVIII.

Gen. IJ.—Merycoporamus.

LS Merycopotamus dissimilis (Fale. and_ Caut.), Indian fossil from

Sewalik ‘hills, figured in ‘ ae Antiq.. Sival.,’

-LXVIL., LXVIIL. :
2. Merycop.. nanus (Falc.).4

Be ageier is referred to a paper

-MeLelland,’‘On the genus

b)

by Dr.
en of Falconer. and Cautley.
—Journ. As. Soc. Dee. 1838, vol. vii. p.

1038.—[ Ep. ]
. 2 See also vol. i. Plates xi. and xii.—
[Ep. ]

3 See also vol. i. Plate xiii En.]
4 Jn Plate Lxvil. of the ‘Fauna Ant.
Siv.’ two varieties of Merycopotamus
are figured: ‘ M. major’ and ‘ M. minor,’
and in one manuscript note written
in 1846, by Dr. Valconer, the genus
is divided into two species, viz. M.
dissimilis and M. nanus. This sus-
picion as to two species seems to have

Plates LXIL,

been confirmed by a specimen from
Attock sent to the author in 1857 _by
Dr. Oldham. This was the last upper
molar, and was compared with the cor-
responding tooth of Merycopotamus dis-
similis in the British Museum (No.
1075 a) with the following result :—

Mer. Mer. of
dis. Attock

Length of last molar .° 1:1 0:75
Width of ditto in front .1:2 08

The author adds: ‘The difference in
area is so great that the Merycopotamus
of Attock must haye been a distinct
species. —[ Ep. |

408 PLAGIAULAX.

VII. DESCRIPTION OF TWO SPECIES OF THE
FOSSIL .MAMMALIAN GENUS, PLAGIAULAX
FROM PURBECK:!

Unttn very lately, the only fossil mammifer known to science
from the Upper Oolite beds (Purbeck series) was the Spala-
cotherium tricuspidens of Professor Owen, a small insectivorous
form referred by him, with some reserve, to the placental
series. It was discovered by Mr. W. R. Brodie in one of
the so-called ‘ Dirt-beds’ of Durdlestone Bay, Purbeck. That
meritorious collector continued his researches during the
years 1855-56, and had the good fortune to discover some
other mammalian remains, which were transmitted to London
about the end of last December for description by Professor
Owen. They were all found in what is called the ‘ Dirt-bed’
No. 93, of Austen’s ‘Guide.’ Before these remains had
reached London, Mr. Samuel H. Beckles, so favourably
known from his researches in Sussex and the Isle of Wight,
after free communication with Sir Charles Lyell about the
importance of a close and sustained search for mammalian
remains at Purbeck, proceeded to Swanage for the express
purpose of carrying it out. Before a fortnight had elapsed,
Mr. Beckles, by a series of well-directed excavations, had
discovered several mammalian jaws besides numerous rep-
tilian remains, in the ‘ Dirt-bed’ No. 98. When the first
line of section ceased to be productive, or could no longer be
worked, he opened new ground, under difficulties which
would have damped the ardour of a less earnest inquirer.
The labours of Mr. Beckles have been crowned with the
success which they deserved. He has discovered a large
number of mammal remains, many of which are new, and in
very fine preservation. The united acquisitions of Messrs.
Brodie and Beckles have already attained the important _
figure of about thirty mammalian jaws, more or less com-
plete, the majority of them lower, but two, at least, upper

1 This paper was communicated to | have been reproduced on stone (Plates
the Geological Society of London on | xxxiii. and xxxiv.) from the erigiael
March 11, 1857, and is reprinted from | woodcuts.—[Eb. ]
the ‘ Quarterly Journal’-of the Society ? Quartorly Journal of, the Geological
for August, 1857. The illustrations | Socicty, vol. x. pp. 481 and 432, 1854.

|
!
.
:

P. BECKLESIIZ AND P>MINOR. 409
jaws, with one well-pronounced cranium. The most amportamt
portion of these are the discovery of Mr. Beckles.

The explorations have been conducted under a conjunction
of unusually favourable circumstances. : Sir Charles: Lyell
gave his sage and long-experienced advice with the deep
interest in the case which befits the author of the ‘ Principles
of Geology ;’ Professor Owen aided the good cause by keeping
Mr. Beckles advised of the importance of the additions which
he was making to Paleontology; and, having had the leisure,
from confinement to my rooms by indisposition, to examine
the objects as they were successively discovered and forwarded
to me, I was enabled to communicate to Mr. Beckles, con-
stantly, an approximative opinion as to the nature of ‘each
fresh acquisition, and thus encourage him to persevere.
From his correspondence, apart from ‘the results, I can bear
testimony to the rare zeal, minute care, and admirable vigour
with which Mr. Beckles has followed up the inquiry. So
productive have his labours been latterly, that hardly a -week
passes without its regular instalment of a couple of dis-
patches of mammalian jaws from Purbeck.

It is intended that, when the collection is completed, the
Purbeck Fossils shall be made over to Professor Owen for
description and publication; and, from what is already
manifest, it may safely be stated, that they will furnish
materials for one of the most interesting and important of
the many chapters which our distinguished countryman has
contributed to. the record of Mammalian Paleontology.
Without forestalling Professor Owen’s detailed results, I
may be permitted to state that I have satisfied myself of
there being among the Purbeck fossils at least seven or eight
genera of Mammalia, some of them unquestionably Marsu-
pialia, both predaceous and herbivorous; and others of them
conveying to my mind the impression, so far as the evidence
goes, that they belong to the Placental Insectivora, having
affinities more or less remote to existing types." ‘

ing to Dr. Falconer’s notes, had an

1 Hight genera, comprising at least
twelve new species, received provisional
names from Dr. Falconer, before the
specimens were forwarded to Professor
Owen. One was believed to be a minute
placental insectivore, closely allied to the
insectivorous genus Hviculus peculiar to
Madagascar. Another lower jaw indi-
cated an insectivorous mammal of fair
size, leaning to the placental, and with
affinities to the European Talpide.
One was allied to the type of the mar-
supial Amphitherium of the oolitic slates
of Stonesfield, though generically dis-
tinct. In this species the jaw, accord-

elongated. slender ramus, containing 7
uniform. back molars im situ, and the
empty alveoli of 4 or 5 false molars in
front, together with a. prominent laniari-
form tooth. The dental formula agreed
numerically with that of the Amphi-
therium, but differed from it in the
double-rowed and complex arrangement
of the crown cusps. Two other species
(afterwards designated Triconodon by
Professor Owen) Dr. Falconer inferred
to be carnivorous marsupials. The
smaller one was believed to have been
nearly as large as the common hedge-hog.

410 PLAGIAULAX.

Having undertaken a description of one of the most re-
markable of these Purbeck mammal genera, in compliance
with the expressed wishes of Mr. Beckles, to accompany some
illustrations which will appear in Sir Charles Lyell’s forth-
coming: Supplement to the 5th edition of the ‘Manual of
Elementary Geology,’ I have thought it: desirable to place
the anatomical evidence for the results more in detail than
could be admitted in a brief abridgement in that work.

The. genus ‘ Plagiaulaz,’! which is inferred to have been
herbivorous and marsupial, comprises two well-marked spe-
cies, Pl: Becklesii and Pl. minor. It has been determined
upon two distinct specimens, which were among the earliest
ot Mr. Beckles’ acquisitions, each a right ramus of the lower
jaw. Latterly, two additional specimens? have been received
of the larger form, Pl. Becklesii, supplementing important
points of evidence which were wanting in the first instance.
The-illustrations and descriptions now submitted are derived
chiefly, from the two original specimens. Of these, the one
of Plagiaulax Becklesti (Plate XXXIII. figs. 1 & 4, a, b, and
b, d); in two-pieces on reversed slabs, consists of the lower
jaw, right side, perfect from the tip of the incisor to the
proximal surface of the condyle, including the ascending
ramus and -coronoid, with the exception only of the raised

The lower jaw contained eight molars,
a large and. prominent, canine, and one
broad and thick incisor on each. side.
«The compressed ‘crowns of the inferior
molars-in this Triconodon haye each of
‘them three sub-equal. sharp-pointed
cusps, rising. nearly yertically into the
same longitudinal plane, with basal end
lobules, but. without additional interior
eomplication.. They are so arranged, in
a continuous and compact series, as to
present a uniform ‘serrated edge, like
the teeth of- a saw.’ Similar tricuspid
teeth of larger dimensions indicated the
existence of another species of Tricono-
don of a more-elongated form, and about
one-third larger in size.. Dr. Faleoner
noted the following evidence as to its
marsupial character. - ‘1, The plurality
of true molars. 2. The strong inflected
angular process. 3.'The most signifi-
cant proof is the broad salient everted
rim of the ridge which is decurrent on
the outer side from the condyle along the
inferior margin, exactly as in.the car-
nivorous marsupials. 4. The marked
development of the.mylo-hyoid groove.’
‘The two species of Triconodon from
the cutting character of their teeth, and
their comparatively formidable canines,

together with the form of the ascending
ramus, are more like small ferine ani-
mals than mere insectivorous marsu-
pials. It is more probable that they
fed on prey less minute than insects,’

Two other species had dental charac-
ters closely allied to those of the mar-
supial herbivore Hypsiprymnus, or kan-
garoo rat, of Australia.

Lastly, there were numerous remains
of a reptilian form, which Dr. Falconer
intended describing in a memoir to the
Royal Society, under the designation of
Saurechmodon (the essential part of
the term being derived from oixun, in
reference to the spear-head form of tha
teeth), when the specimens were con-
signed for description to Professor
Owen, who designated it Echinodon
Becklesit. (Fossil Reptiles of Purbeck
Strata. — Paleontograph. Soc. 1860.—
[Ep]

1 An abbreviation for ‘ Plagiaula-
codon,’ from mAdy.os, oblique, and abaaé,
groove, haying reference to the diagonal
grooving of the premolars.

2 A fifth specimen, subsequently ac-
quired, is described in the sequel (see
Pl, xxxiv. fig. 1).

DESCRIPTION OF PLATE XXXIITI.

PLAGIAULAX BECKLESII.

The drawings in this Plate have been reproduced from the
woodcuts which accompanied the memoir as originally pub-
lished in the Quart. Journ. Geol. Soc. for August, 1857.
(See page 410.)

Figs. 1-6. Plagiaulax Becklesti (figs. 1-5), and Hypsiprymnus Gai-

Fig. 1.

Fig. 2.

Fig. 3.

Fig. 4.

mardi (fig. 6). Figs. 1 and 4 show the entire right ramus of the

lower jaw, in two pieces, on reversed slabs of the same piece of

matrix. Magnified two diameters.

[ Figs. 1 and 4 represent the same right ramus of the lower jaw
seen on the opposite surfaces of a split stone, the two taken
together affording data for a complete restoration of the jaw. ]

a, b, e’. Outer side of the anterior portion of the right ramus of
lower jaw; magnified two diameters. a,b, outer side. b, 0’, d’, é,
impression of inner side.

a. Incisor.

b. Line of vertical fracture behind the premolars.

d'. Impression in the matrix of the condyle.

e’. Impression of top of coronoid process.

o'. Broken-off inflected fold of inner margin buried in the

matrix.
m. Place of the two molars.
pm. Three premolars, the third or last divided by a crack.

J. Section of the anterior piece of the jaw at the fracture, fig.
16; x, inner surface; y, outer. The notch at the top is formed
by one of the sockets of the double-fanged true molar.

g-. Section of the hinder piece near @ in fig. 1; x, inner surface ;
y, outer surface.

a’, d. Inner side of the posterior portion of the same lower jaw
on the opposite slab of stone; 6, d, e, inner side; 6, a’, h, cast
and impression of outer side.

a’. Outline of the incisor restored.

b. Line of vertical fracture.

d. Condyle.
. @. Coronoid process.!

' The artist has made the point of the coronoid (e) project too much back-
wards, and the curve of the posterior margin too great; the line being nearly
vertical in the original. The projection of the condyle behind the coronoid margin
is more considerable than is shown by the figure, and the neck longer.—H. F.

VOL. II. ée

Fig. 5.

Fig. 6.

Description oF Prats XX XIII.—continued.

h. Impression of the three premolars on the matrix.
i. Empty sockets of the two true molars.

n. Orifice of dentary canal.
o. Indication of the raised and inflected fold of the posterior

inner margin.
k. Third or largest premolar, showing the seven diagonal grooves ;
magnified 5} diameters.

1. Corresponding premolar in the recent Australian Hypsiprymnus
Gaimardi, showing the seven vertical grooves; magnified 3}
diameters.

Figs. 7, 8, 9, and 10. Plagiaulax Becklesii. Portion of the right ramus

of a lower jaw, and different views of the two true molars.

Fig. 7. A portion of the jaw, with two molars 7m situ, mag-
nified 10 diameters; fig. 8, inner side of the molars, magnified
10 diameters; fig. 9, outer side, 7 diameters; fig. 10, summits
of the crowns of the molars, 7 diameters.

. a. Anterior margin of the coronoid process.

6. Fractured posterior margin.

pm. Impression of the last premolar.
m. 1. First true molar.

m. 2. Second true molar.

Figs. 8, 9, and 10. a. Anterior inner point of penultimate molar.

b. Posterior inner point of the same molar, showing the disc of
wear,

c. Anterior outer point.

d. Posterior outer edge.

ee. Fractured surface of interior edge of the last molar.

Jf. Ground surface of outer edge of the same.

Figs. 11, 12, and 13. Plagiaulaw Becklesii. Fragments consisting of

the anterior portion of the right ramus of the lower jaw,
magnified 2 diameters; fig. 11, outer surface; fig. 12, inner
surface; fig. 13, vertical view seen from above; a, incisor ;
pm, premolars; 6, symphysial harmonia; c, mentary foramen.

Vol. I Blate 32.

del Lens Aldous ith: Harrison & Sons Imp®

Plagiaulax Becklesi.

Pay

P. BECKLESIT AND P. MINOR. 411

posterior lower margin and-inflected angle; it shows three
premolars (p, m) in situ, with the empty sockets of the two
back molars. Another specimen (Pl. XXXIII. fig. 7) fortu-
nately supplies these two back molars in situ. The lower
jaw (also of the right side) of the other species, Plagiaulax
minor (Plate XXXIV. fig. 2), is less perfect. It contains all
the teeth im situ, beautifully preserved, but it is mutilated
vertically behind the alveolar border; the ascending ramus,
with all the proximal portion, being wanting. Besides the
two true molars, it contains four premolars ‘instead of three,
as in the other species. :

Teeth.—Together, these specimens furnish nearly complete
evidence as to the characters of the lower jaw of the genus.
And, first in regard to the teeth, the dental formula is :—

incis. 1—1 ; can. 0O—0; prem. 8—3; mol. 2—2=12 in P. Becklesii:

4—4 =14 in P. minor.

To save unnecessary technical details by reference to a
well-known existing genus, which will constitute in other
respects an important term of comparison, it may be stated
at once that the incisor of Plagiaulax Becklesii (Pl. XX XIII.
fig. 1, a,and Pl. XXXIV. fig. 1), in every particular of form,
namely, edge, point, and section, and in relative amount of
projection, bears a very close general resemblance to the
incisor of the marsupial Hypsiprymnus, or Kangaroo Rat, of
Australia; it differs chiefly in being, for the relative size,
more robust and curved more abruptly upwards, in the fossil
animal. Its line of implantation in the socket is more ver-

.tical, and the alveolar sheath shorter and thicker. The
diasteme is exceedingly abbreviated, not exceeding a line in
length. f

The three premolars, in Pl. Becklesii (Plate XX XIII:
fig. 1, p, m), are in the finest state of preservation, showing
the details of every minute character. They are limited to
three in all the specimens (see Plate XX XIII. figs. 4,11, 12,
and Pl. XXXIV. fig. 1). This is a point of some importance
to establish distinctly, as there are four of these teeth in the
other species. They form a closely adpressed and compact
series of very unequal size, diminishing rapidly in succes-
sion from the last to the most anterior. The last premolar
(Pl. XX XIII. fig. 5) presents a square oblong side, convex
from back to front, and sloping upwards and inwards to the
edge, which is finely serrulated, as in Hypsiprymnus, the ser-
ratures being caused by the terminations of about seven well-
marked parallel grooves, which descend upon the side, not
vertically as in Hypsiprymnus (Plate XX XIII. fig. 6, 1), but

412 PLAGIAULAX.

diagonally downwards and forwards, disappearing about the
middle of the crown-side, upon a smooth and discoidal sur-
face. The enamel below the four last grooves is inequal and
raised into a well-marked crenated step (Pl. XX XIII. fig. 5),
which is exhibited on all the specimens. The interior surface
being adherent to the matrix, the characters of the inner side
of the last premolar are not shown by this specimen. But in
two of the other specimens the inner surface of all the pre-
molars is free and seen to be furrowed by diagonal grooves,
exactly like the outer, the principal difference being that the
inner side is more flattened, and the enamel smoother than
on the outer. The two sides, therefore, slope from the base
upwards to a sharp or thin edge, which is serrulated, agreeing
in every respect, except the inclination of the grooves, with
the corresponding tooth of Hypsiprymnus. The tooth is im-
planted by two distinct fangs. The penultimate premolar is
somewhat spathulate in outline, the lateral surface of the
crown is convex in the longitudinal direction, and slopes
inwards to the apex, which is diagonally grooved like the
last premolar, the grooves being fewer in number. It is
inserted by two inequal fangs, the posterior of which is
barely visible; in size it corresponds with one fang division
of the last premolar. The antepenultimate or anterior pre-
molar is greatly reduced in all its dimensions, being hardly
one-fourth the size of the tooth immediately behind it; in
form it exhibits more the ordinary appearance of an incisor.
In the other species, Plagiaulaw minor (Plate XXXIV.
fig. 2), which is very considerably smaller, the incisor (a) pre-
sents a corresponding general form, but it is more elongated,
less robust, and is not so much curved upwards. <A portion of
the point has been broken off in the specimen, and it is seen
by the impression (a’) that the inner side near the apex was
hollowed out in a longitudinal depression. The premolars,
in number four, are higher in proportion to the depth of the
jaw than in the other species. The last. one is similar in
form and grooving to the corresponding tooth in Plagiaulax
Becklesvi, but exhibits a slight difference in the inequality of
the enamel-surface below the basal terminations of the
grooves. In front of it there are two spathulate premolars,
i.e. the antepenultimate and penultimate, both diagonally
grooved near the apex; and at the base of the antepenulti-
mate, but pressed somewhat inwards, there is a very minute
anterior or first premolar. The basal enamel-surface bulges
out over the fangs in these teeth in a rounded angle which
points downwards. Regarded as a series, they decrease in
size very rapidly from the last to the foremost. The sharp _
edge of the crowns of the three anterior teeth slopes down

P. BECKLESIZ AND P. MINOR. 413

towards the diasteme from the anterior margin of the last
premolar; while that of the latter slopes in a reverse manner
downwards and backwards: towards the true molars, the
anticlinal planes meeting at an obtuse angle.

- The true molars in both species were limited to two, the
sockets of which alone remain in the more perfect jaw
(Plate XX XIII. figs. land 4, m and 7). But they are shown
im situ, in the most perfect preservation, in the jaw of
Plagiaulax minor (Plate XXXIV. fig. 2, m).. It is clearly
apparent from the relation of the second tooth to the muti-
lated base of the anterior margin of the ascending ramus (b)
in the latter, and from the empty sockets of the fallen teeth
in the alveolar rim of the perfect specimen (Plate XX XIII.
fig. 4, 7) of Pl. Becklesti, that the true molars in the lower
jaws of both species did not exceed two, a very unexpected
and reduced number to occur in forms otherwise inferred to
be marsupial, and therefore demanding rigorous determina-
tion, there being no other corresponding instance known
within the whole range of this sub-class, fossil or recent.

; Fortunately the specimen represented by figs. 1 and 4 of
Plate XX XIII. shows the whole of the teeth of Pl. Becklesii
in nearly as perfect preservation as they are in the specimen
of Pl. minor (Pl. XXXIV. fig. 2). It consists of a right lower
jaw in two continuous fragments, presenting the two last
molars of an adult animal, well worn and in situ. The anterior
edge of the ascending ramus is entire, forming a well-defined
boundary to the alveolar border, and so closely contiguous
to the second and last true molar, which it partly overlaps,
that it is manifest there could not have been more than two
of these teeth in the jaw. The true molars are not only
reduced in number in Plagiaulaw minor, but they are also
dwarfed in size, and comparatively insignificant in contrast
with the last premolar, the united length of the two being
less than that of the latter, while the vertical height of their
crown is little more than one-third of that of it.. They are
situated both horizontally and vertically in different planes
from the premolars; a perpendicular from the acute serrated
edge of the latter would coincide with the line of the inner
row of the crown-lobes of the true molars, to be described in
the sequel, as occurs in the recent Hypsiprymnus; and the
outline of the coronal surface of the whole molar series would
be included within a curve, of which the last premolar formed
the most salient part, with a considerable descent on either
side of it.

- The crown of the first true molar in Plagiaulax minor
(Plate XXXIV. figs. 3 and 4), which I take first as the more
perfect, is of a broad oblong form. Its inner or axial margin

414 PLAGIAULAX.

supports two well-raised and bluntly conical points or tuber-
cles, separated by a wide cleft, which is partly continued
down upon the body of the tooth, vertically in a sinus, form-

ing an obsolete mesial constriction. The points are isolated.

and start up in considerable relief. The outer edge of the
crown is not divided correspondingly; it supports mesially
but a single prominent conical tubercle, which is intruded
upon the plateau of the crown, and opposed to the sinus
between the two inner points. It alternates therefore with
the latter. The base of the outer pont is continued on either
side, backwards and forwards, in a well-marked lunate be-
velled rim, which is convex outwards, and rises at either end
into a low terminal lobule. The posterior lobule may be
considered as the homologue of the posterior inner point,
although it forms but an insignificant tubercle. The an-
terior lobule is still less developed, but opposed to the
anterior inner point.» The middle of the crown is occupied
by a sinuous hollow surface, intervening between the outer
and inner rows. This hollow, from the intrusion of the
outer mesial point and the constriction of the inner side, is
divided into an anterior and posterior discoidal depression.
The two rows are separated at both ends of the tooth by a
longitudinal chasm. ‘There is no indication of any low trans-

verse concave ridge connecting the opposed tubercles. From.

the above description it will be seen that the two sides are
unsymmetrical, whether the outer row is considered as con-
sisting of a single point with an accessory tubercle on either
side, or of three unequal tubercles.

The second or last true molar (Pl. XXXIV. figs. 5 and 6},
viewed in profile, is smaller than’ the: first, and the crown-
surface, although of nearly equal extent, is less complex in
its subdivision. It is of a broad oval, with the base applied
to the contiguous anterior tooth. .The coronal eminences
consist of an outer and inner raised marginal and more or
less lobulated edge, separated by a broad central depression.
The outer edge is very narrow and nearly horizontal, rounded
off at either end and presenting no marks of composition be-
yond four or five obscure indications of crenulation, like one
of the rows in the tooth of: Microlestes (see Pl. XXXIV.
fig. 7). _ It is incurved so as to overarch slightly the central
depression. ‘The inner-edge presents at its anterior end a
well-elevated and isolated conical tubercle, resembling in size
and form the outer mesial point of the first true molar. It
is bounded posteriorly by a deep wide cleft from which a
convex edge is continued backwards, which is raised, but not
sufficiently to attaim the importance of a cusp, although
homologous to the rear cusp of the anterior tooth. The

*

P. BECKLESIT AND P. MINOR. 415

centre of the crown is occupied by a hollow—smooth, and,
as it were, scooped out and depressed considerably below the
raised margins. <A well-marked chasm intervenes between
the marginal edges, both in front and behind, the latter be-
ing narrower and less pronounced, so that when the tooth is
viewed endwise, the opposite rows are seen apart somewhat
as in the tooth of Wicrolestes. 'The crenation of the outer
margin into a row of tubercles is more decided in Microlestes'
(PL XXXIV. fig. 7), the crown is narrower and more elon-
gated in proportion, and the opposed rows are more approxi-
mated. None of the raised points of these teeth in Plagiaulaa:
minor show any considerable marks of abrasion, nor exposure
of the ivory ; the white spots in the figures represent adherent
specks of matrix, and not depressed discs of wear; the apices
of the outer crenations alone are a little worn in the last
molar. The animal from which the: fossil was derived is in-
ferred, from the intact condition of the molar teeth, to have
been a young adult, as these teeth have been found well worn
in one specimen of the other species.

In the jaw of Plagiaulax Becklesii (Pl. XXXII. figs. 1
and 4), there are three small pits on. the alveolar border,
behind the last premolar. The anterior two, seen upon the
fragment fig. 1, a, b, are closely approximated, with a thin
plate intervening, indicating that they are the sockets of the
two-fanged first molar. One of them is shown in the niche
at the apex of the vertical sections, Pl. XX XIII. figs. 2 and
3,f/,g- The last pit, fig. 4, 7, 1s larger, square, and undivided,
but shows obsolete marks of a mesial constriction, indicating
that the second or last molar had fangs converging in a com-
mon socket. The rim of this socket is distinctly defined, and
the inner wall raised into a prominent gibbosity which leaves
a deep impression on the matrix. This gibbous point corres-
ponds with what occurs in the lower jaw of the recent Hypsip-
rymnus Gaimarda, with which itwas compared,and is commonly
seen on the alveolar wall of the last molar of other animals.

. In the second specimen of Plag. Becklesii, Pl. XX XIII.
figs. 7-10, already referred to, the two true molars of an
adult, and, judging by their wear, well-aged animal, are seen
am situ. They present the same general characters as the
corresponding teeth of Plag. minor (Pl. XXXIV. fig. 2), but
appear to be proportionally larger, in comparison with the
premolars. The crown of the first molar bore, as in the
other species, three principal points; two to the inner row;

1 Tam indebted to Sir Charles Lyell | veral duplicate casts, he had very careful
for the only available means of institut- | and highly finished drawings made of
ing a comparison of these mammalian | the teeth during his last visit to Stutt-
teeth from Purbeck with those of the | gart, the whole of which have been
Microlestes of Plieninger. Besides se- | placed at my disposal.

416 j PLAGIAULAX.

and to the outer a single tubercle, which is situated more
anteriorly than in Plag. minor. ‘The outer point is worn
down and indistinctly defined, the wear mvolving what ap-
pears to have been the bevelled lunate rim at the posterior
end, as described in the other species. It is not determinable
whether it attained the magnitude of a separate tubercle.
The two inner points form obtusely conical tubercles, which
are in greater relief, and less affected by wear; so that the
worn disc of the outer row looks like a step at their base.
Of these tubercles the anterior bears an accessory lobe which
is continued across in a front talon, giving a bilobed cha-
racter to the point, which is but slightly abraded at the apex:
while the posterior point shows a well-marked depressed dise,
which is continued some way down upon the inner side of
the enamel, as if caused’ by the grinding play of an over-
lapping cusp of an upper tooth in use against it. The lon-
gitudinal furrow between the two rows is distinctly visible.
There is a well defined. basal cingulum to the last molar in
this species. The inside elevation of the last true molar
agrees very closely with a corresponding view of the larger
tooth of Microlestes antiquus, as represented in one of the
drawings belonging to Sir Charles Lyell (Pl. XXXIV. fig. 8).
The second and last true molar presents a nearly square
crown with rounded angles. It is fully equal in size to the
penultimate, if not a little larger. .The crown bears two
marginal edges, as in the other species, with a well-marked
depression between them. ‘The inner edge is broken off and
seen embedded in the matrix of the opposite slab. The outer
edge is entire, but ground down by wear, so that it shows a
marginal band with no remains of crenation. The detrition
of the crown, as a whole, bears some resemblance to that
seen upon the last molar of an aged Bear; the comparison,
be it understood, not being intended to imply the slightest
idea of affinity. The angle formed by the anticlinal planes
of the crowns of the true molars and premolars is more |
acute in the jaw of Plag. Beckles than in that of Plag.
minor. .

Character of the Lower Jaws.—Although the teeth are seen
in greater perfection in the smaller species, the other cha-
racters. of the jaw are best shown by the specimen of Plag.
Becklesii, Pl. XXXILI. figs. 1 and 4, a,b, and b,d. The jaw,
right ramus, is broken vertically through one of the sockets —
of the first true molar; the anterior fragment (a, b) presents
the outer surface of all the teeth in front, from the last pre-
molar, while the posterior fragment (b, d) exhibits the inner —
surface of all behind, on to the condyle. A very distinct im-
pression of the premolars attached to the anterior piece is left

DESCRIPTION OF PLATE XXXIV.

PLAGIAULAX BrecKLEsit. P. MINOR. MICROLESTES ANTIQUUS.
THYLACOLEO CARNIFEX. CHEIROMYS MADAGASCARIENSIS.

The drawings in this Plate have been reproduced from the
woodcuts which accompanied the memoirs as originally pub-
lished in the Quart. Journ. Geol. Soc. for August, 1857, and
November, 1862. (See pages 412 & 436.)

Fig. 1. Plagiaulax Becklesii. The left ramus of the lower jaw, nearly
perfect, showing the outer surface. Magnified 4 diameters. a,
incisor; b, condyle; c, coronoid process, broken at the edges ;
d, inferior boundary ridge; a depression is seen on the inner
surface of the bone at the base of the coronoid; pm., premolars.

Figs. 2 to 6. Plagiaulax minor.

Fig. 2. Outside of the right ramus of the lower jaw, magnified 4 diame-
ters. All the teeth are in place and well preserved. The
hinder part of the jaw-bone, with the ascending ramus and
posterior angle, is broken away.

a. Incisor with point broken off. a’, impression of same,
showing that the inner side near the apex was hollowed
out in a longitudinal direction.

b. Offset of coronoid, the rest of which is wanting.

m. The two true molars.

pm. The four premolars.

e. The length of the jaw, natural size.

Figs. 3 and 4. The first molar, magnified 8 diameters; 3, the crown;
4, side-view.

Figs. 5 and 6. Second molar; the crown and side-view.

Figs. 7 and 8. Teeth of Microlestes antiquus of Plieninger, from the
Upper Trias of Wirtemberg. Magnified.

Fig. 7. Crown of the smaller molar.!
Fig. 8. Crown of larger tooth,? with part of the crown broken off.
1 See Lyell’s Manual Elem. of Geol. dth edit., fig. 441 4, p. 348.

2 Thid. fig. 442.
VOL, Il,

DESCRIPTION OF PLATE XX XIV.—continued.

Figs. 9, 10, 11, and 12. Posterior half of a carnassial tooth (pm. 4)
from the left side of the lower jaw of Thylacoleo carnifer.
Preserved in the Museum of the Royal College of Surgeons.

Fig. 9. Inner side. a, hinder end, showing the undulations of the
enamel-surface on the base of the crown, and the rugosely
reticulate surface below the summit. Drawn for comparison
with figs. 5 and 6 of Plate XXXIII., showing the same teeth
in Plagiaulax and Hypsiprymnus.

Fig. 10. Outer side. a, hinder end.
Fig. 11. Top aspect, showing the undulations. a, hinder end.
Fig. 12. Section, showing the broken edge of the middle of the crown.

Figs. 13 and 14. The right ramus of the lower jaw of the Aye-Aye
(Chetromys Madagascariensis) ; the outer aspect. Natural size.

a. Incisor.

b. Molar teeth.

e. Coronoid process.

d. Condyle, having its articular surface below the grinding-
plane of molars.

e. Angle of jaw. (The letter ¢ has been omitted by the
engraver, but ought to have been placed at posterior angle
of lower margin, where this turns up into neck of condyle.)

f. Conjectural dotted line. .

Fig. 14. End-view of condyle.

Ries. Rey

Vol. IL. Plate 34.

a Harrison & Sons Imp
“| Plagiaulax Becklesii. 2.6 Plagianilax minor.
7.8. Microlestes antiquus. 9.12.Thylacoleo carmfex.

13.14. Cheiromys Madagascariensis.

P. BECKLESII AND P. MINOR. 417

on the matrix of the slab which contains the hind fragment,
so that it has been easy, by combining the figures, to pro-
duce a restored outline of the general contour, which is shown
- by the aid of dotted lines, in both figures, magnified twice
the linear dimensions.' The most striking characters are:
1st, the shortness of the jaw from the posterior edge of the
ascending ramus to the border of the alveolar sheath of the
incisor (a); 2nd, the great vertical height of the horizontal
ramus in relation to the last dimensions; 8rd, the de vressed
position of the condyle (d), and its great horizontal projec-
tion behind the coronoid process. The lower body deviates
little from the horizontal for about four-fifths of its length,
being but slightly convex. The upper margin slopes a little
downwards from behind the premolars on to the incisive
alveolus, so that if the two margins were produced they
would meet at an acute angle. The diasteme is very short.
The incisor (a) is comparatively large and directed suddenly
upwards, the point being elevated above the level of the pre-
molars, with a short bluff sheath-border. The base of the
incisor is impressed with a shallow longitudinal fossa, and
the upper edge is bevelled. The point shows no mark of
abrasion. The mentary foramen is small, indistinct, and
situated mesially in a line nearly with the middle of the
diasteme. The alveolar border rises in large serrate pro-
cesses between the fangs. The dark-shaded depression (fig.
4, i) behind the vertical fracture (at b) upon the matrix
marks the position of the pointed gibbosity on the inner side
of the alveolus of the last true molar. It is considerably
below the line of implantation of the premolars, indicating
that the true molars were placed correspondingly low, as is
seen in the jaw of the other species. The longitudinal me-
dial shade shows a deeply impressed broad channel upon the
outer surface of the ramus, which is most pronounced under
and behind the last premolar. The amount of this depres-
sion is well exhibited by the vertical sections (Pl. XX XIII.
figs. 2 and 3, f and g), where (x) represents the inner surface,
and (y) the outer. It is seen that the inner surface is slightly
concave in the vertical direction, between the upper and
lower margins ; and that the general convexity of the outer
surface is interrupted by a wide and deep fossa, causing a
mesial constriction (y). Although there is as yet no direct
evidence to the point available, in consequence of the outer
surface of the posterior portion being adherent to the matrix,
still it would appear that this channel runs back towards the

4 1 See also fic. 1 of Pl. xxxiy., illns-
‘ trating a perfect right ramus of a
VOL. II. EE

younger individual magnified four times,

418 PLAGIAULAX.

depressed disc of the outer surface of the ascending ramus,
although it may not have been continuous with it.

The posterior fragment (b, d, fig. 4, Pl. XX XIII.) shows
in very perfect preservation the whole of the inner surface
of the ascending ramus and condyle, with the exception of
the fragile edge of the inflected lower margin and angle,
which are broken off and left in part embedded in the matrix-
cast of the opposite slab containing the anterior fragment ;
it comprises also the posterior portion of the body, in which
the last molar is implanted. The coronoid process (e) is tri-
angular. Its anterior border slopes upwards and backwards
to the apex in a gentle curve, at an angle of about 45° with
the alveolar border. The posterior margin descends in a
curve with little deviation from the vertical into the broad
sigmoid notch (d, e), which is but slightly overarched by the
apex.! Its height and width at the base are nearly alike,
and about equal to the depth of the body of the jaw, ina
line with the last molar. The apex of the process is sharp.
In general form the coronoid process in Plagiaulax resembles
more that of the predaceous marsupials, and of the Ursine
Dasyurus especially, than that of the herbivorous families.
It differs very markedly from the elevated strap-shaped coro-
noid of Hypsiprymnus and the other herbivorous marsupials.
Tt is to be remarked, however, that it is less elevated, and
that its surface is of less area, than in the predaceous genera,
whether marsupial or placental.

Fortunately the condyle (d, fig. 4, Pl. XX XIII.) is in
every respect as perfect as that of a recent bone. The
matrix has been removed, so that it stands out in bold relief,
showing the convex articular surface entire. In no part of
the specimen are the peculiar characters of the fossil more
stronely marked than in this process, the remarkable points
being: 1st, its very depressed position, the upper edge being
below the line of the alveolar border, and the lower extending
nearly to the inferior margin of the jaw; 2nd, its prominent
convex surface, great depth in proportion to the width, and
vertical direction; 3rd, the breadth of the sigmoid notch
(d, e), involving a long neck to the condyle, and its projection
much behind the coronoid process. The articular surface is
convex and protuberant, narrow in comparison to the height;
when viewed endwise, the outline is pyriform, the broad end
being uppermost. The general direction of the articular
surface is vertical; a chord through the upper and lower
edges of the curve would form a very acute angle with a

1 The artist has represented the pos- | jecting too much backward over the sig-

terior margin of the coronoid with too | moid notch.
crescentiec a curve, and the point pro-

P. BECKLESIT AND P. MINOR. 419

perpendicular immediately behind it. The neck-portion of
the sigmoid notch (d, e) is as long as the breadth of the
coronoid, so that the articular surface is not only depressed
below the apex of the coronoid, but projected a long way
behind it. A line drawn from the condyle to the gibbous
prominence on the wall of the last molar (7) considerably
exceeds the length of the horizontal ramus from the latter
point on to the border of the incisive alveolus. The inner
surface of the ascending ramus behind the orifice of the
dentary canal (mn) is smooth, and near its inferior border
traversed by a longitudinal channel which bends back to the
under edge of the condyle. The orifice of the canal (n) is
low and directly in a vertical line with the posterior wall of
the last molar, being relatively situated exactly as in Hypsip-
rymnus. The anterior margin of the canal forms a raised
step, the edge of which is continued upward in a low crescentic
ridge to join on with the base of the gibbous prominence of
the last alveolus. The mylo-hyoid groove traverses the ridge
about the middle, but the adhesion of the inner surface of
the front piece (a,b) to the matrix does not at present admit
of its being determined how far the groove advances upon
the horizontal ramus. The inner surface of the coronoid is
convex, but this appearance is partly obscured by the plate
of the process being traversed by numerous fissures, with
partial displacement of the pieces.

The inferior margin of the hind part of the ramus is nearly
horizontal ; it terminates suddenly in the lower edge of the
condyle. <A narrow fractured surface (Pl. XX XIII. fig. 4, 0),
the continuation of which was evidently directed inwards
(i.e. presented to the observer), is seen on the side of the
lower edge, stretching from the anterior boundary of the
dentary foramen (n) on to the condyle. The rest of this in-
flected margin (what remains of it) is seen embedded in the
opposite slab (Pl. XX XIII. fig. 1, b, 0’). It forms a lamina
of great tenuity. The base of its inner bounding r:dge shows
a triangular fracture immediately under the orifice of the
dental canal (at n, fig. 4, Pl. XX XITI.), where it is compara-
tively thick; thence the section becomes gradually attenuated
on to the condyle. There can be little doubt that this is the
characteristic marsupial inflected angle, although feebly de-
veloped. It may have formed a slender elongated apophysis,
with little inward inflexion, as in Acrobata.

The characters of the anterior portion of the inner side of
the horizontal ramus, which are concealed by the contact
with the matrix in the fragment Pl. XXXITI. fig. 1, a, b, are
beautifully shown by a detached piece lately received from
Mr. Beckles, represented by figs. 11-13. It comprises the

EE 2

420 PLAGIAULAX.

anterior third of the ramus, the fracture having passed
vertically through the middle of the last premolar. The
incisor is broken off near the base of the exserted portion ;
the two anterior premolars are quite entire, and show well
the diagonal grooving on both sides. The alveolar sheath of
the incisor is entire all round, and its upper edge presents an
exceedingly abbreviated diasteme. The symphysial portion
is very short, directed upwards, massive, and obtuse. ‘The
dise of the harmonia or junction between the two mandibular
pieces is distinctly defined (fig. 12, b). It is of comparatively
small antero-posterior extent, of a broad elliptical or some-
what reniform outline, with the sinus directed backwards,
and the long axis of the disc with but slight deviation from
the vertical. The surface is perfectly smooth and without
indentation, as is commonly seen in Hypsiprymnus and other
herbivorous marsupials; while in the carnivorous and in-
sectivorous genera of the same subclass, the symphysial
harmonia is narrower and more elongated, with more or less
inequality of surface for the reciprocal firm apposition of the
united pieces. It is of some importance to take due notice
of this character, however trivial it may appear, as every
point is of value that can assist us in determining the affini-
ties of this remarkable fossil genus. On the outer surface of
this fragment the mentary foramen (fig. 11, c) is seen under
the middle of the diasteme. It is round, well defined, and
of good size. This instructive fragment furnishes direct
evidence upon another point of importance, namely, that there
was but one large incisor on either side of the lower jaw.

It would have been of great interest and importance to
have ascertained the character of the outer surface of the
ascending ramus in Plag. Becklesii; that is to say, whether,
in harmony with the marsupial indications as here interpreted,
it presents a depression bounded by a raised ridge sweeping
round on the lower side from the condyle to join on with a
corresponding ridge descending along the anterior margin
of the coronoid ; and whether the depression terminated in a
trumpet-shaped excavation of the horizontal ramus, common
to it and the dentary canal, as occurs in Hypsiprymnus. But
the fossil is so fragile, that I have not attempted to detach it
for fear of injuring what it now exhibits. We may expect,
however, if the excavations are continued at Purbeck with
the same zeal with which they have hitherto been conducted,
that abundant materials will be acquired for clearing up this
single unascertained point connected with the characters of
the lower jaw of Plag. Becklesvi. From the direction which
the fracture has invariably taken along the line of least re-
sistance, in two fragments on reversed slabs, whenever a

tl et

P. BECKLESIT AND P. MINOR. 421

complete jaw has been discovered, and from the circumstance
that the inner surface has been always exposed, and the
outer remained in adhesion with the matrix, I ain led to
surmise that this has been caused by the matrix forming a
plug in the excavation here referred to, thus causing it to
adhere firmly; and I am prepared to expect that the outer
surface of the ascending ramus will be found to agree in a
great measure with that of the recent Hypsiprymnus.

[P.8. Since the preceding remarks were written, a fifth
specimen of Plagiaulax has been discovered by Mr. Beckles,
which supplies the desired information regarding the cha-
racters of the outer surface of the ascending ramus. See
fig. 1 of Pl. XXXIV. It consists of the left ramus, nearly
entire from the incisor to the condyle, showing the whole of
the outer surface exposed. The specimen would seem to
have belonged to a young individual of Pl. Becklesir.

The incisor is vertically inserted, and projects above the
level of the premolars, of which there are three. The true
molars, if present, are concealed behind the flap formed by
the anterior margin of the coronoid process. This part of
the jaw has been slightly crushed. The coronoid process
rises more vertically, and is narrower than in the specimen,
Pl. XXXIII. tig. 4; but a portion of the posterior margin is
probably wanting. The base of the coronoid is occupied by
a deep depression bounded on the lower side by a raised
ridge, which sweeps round from the inferior part of the
condyle, to be continued into the anterior margin of the
coronoid process. The characters are clearly marsupial; but
it is not determinable whether the depression terminates in
an excavation of the ramus common to it and the dentary
canal, as occurs in Hypsiprymnus. So far as can be seen, the
depression would seem to be more limited. The matrix has
been cleared away from the posterior inner margin, and a
portion of the inflected angle is distinctly visible. The speci-
men, magnified four times linear, is represented by fig. 1 of
Pl. XXXIV. It bears out in every respect the marsupial
inferences deduced from the other specimens; and indicates
for Plagiaulax a position between Hypsiprymnus and the
Phalangers.—June 20,1857. H. F.]

The lower jaw of the other species, Plagiaulaa minor, is
represented by Pl. XXXIV. fig. 2, a, b, magnified 4 diameters,
being double the scale of Plag. Becklesti, Pl. XX XIII. figs.
1&4. The outer side is exposed, the inner being adherent
to the matrix. The notable points are—the shortness of the
horizontal ramus from the offset of the coronoid to the border
of the incisive alveolus, its great relative height on a line
with the premolars (p m), and the bold curve of the lower

422 PLAGIAULAX.

margin. The incisor (a) is projected with a less sudden
curve upwards, and its sheath is longer than in Pl. Becklesii.
The premolars are also larger in proportion to the height of
the jaw than in that species. Unluckily the whole of the
ascending ramus is wanting, and with it are lost the signi-
ficant characters yielded by the form of the condyle, coronoid
process, and posterior angle. At the fractured posterior end
(6), a small portion remains of the external oblique ridge
which rises into the anterior border of the coronoid. A well-
marked wide depression is seen on the posterior part of the
horizontal ramus under the true molars, corresponding with
that shown on the jaw of Plagiaulaw Becklesti. In the great
development of the premolars, and the dwarfed size of the
true molars, there is in the fossil an analogy with Acrobata
pygmea, the ‘ Opossum-mouse,’ or ‘ Pigmy Flying Opossum ’
of New South Wales. But the resemblance goes no further,
the principal premolars in Acrobata being much elevated and
pointed in front, leaning to the insectivorous type,! while
they are uniformly compressed, grooved, and serrated in
Plagiaulax minor.

This concludes what I have to offer in the shape of descrip- ~

tive details. I shall now proceed to consider what may be
legitimately inferred from them respecting the nature and
affinities of the fossils. That the genus was mammal admits
of no question; that it was a marsupial is inferred for the
following reasons, which are given in the order of the direct-
ness of the indications :—

1. The compressed hatchet-shaped last premolar with the
serrulated edge and parallel grooving. These characters are
confined, among all known mammals, to the marsupial genus
Hypsiprymnus; the correspondence in grooving is so exact
that the number of furrows is the same in the fossils and in
the recent species with which they were compared, namely,
seven; the difference, that they are diagonal in the former
and vertical in the latter, being trivial and not typical.

2. The agreement in form, relative size, and direction of
the solitary incisor in the fossil rami, with that of the recent
Hypsiprymni.

3. The indication of the raised and inflected fold of the
posterior inner and lower margin of the ramus. F

4, The form and characters of the symphysial suture.

5. The absence of any character in the jaw or teeth incon-
sistent with the marsupial indications.

The presence of only two true molars might seem, at first
sight, at variance with a marsupial determination, since it

1 Waterhouse, Nat. Hist. of Mammalia, vol. i. p. 338.

=

P. BECKLESIZ AND P. MINOR. 423

has been asserted by an able authority, that, with the excep-
tion of the edentate ‘species of marsupials, or those which
are nearly edentate, like the Tarsipes, and also excepting the
Myrmecobius, all Marsupialia possess four true molars.’* But
the character is not absolute, for all the Pigmy-Phalangers
of the subgenus Dromicia, besides Acrobata, are admitted to
have only three true molars.2 In the Purbeck fossils the
premolars are inordinately developed, while the true molars
are dwarfed and rudimentary in proportion. Where such
characters coexist with an exceedingly abbreviated alveolar
border, there is less reason for surprise in seeing two of the
molars suppressed. It is now well known, that there is no
certain distinctive character, whether of placental or marsu-
pial, that can be founded on the number of their teeth.
Among the marsupials, Myrmecobius presents a case in which
they are in excess; while the Purbeck Plagialauaz would
seem to present the opposite condition, where they are below
the normal number, from suppression.

The same reasons are equally strong for referring Plagiaulaw
to the neighbourhood of the existing genus Hypsiprymnus.
_ The affinity indicated by the premolars and incisor is so
manifest and direct, that details upon the differences from
other terms of comparison, placental or marsupial, would be
superfluous. The large grooved premolar is confined among
the Marsupilia to Hypsiprymnus; that genus, comprising
three subgenera, includes about ten species, in all of which
the premolar is solitary, the true molars being constantly
four. In Plagiaulax there are either three or four grooved
premolars, and only two true molars. In Acrobata and some
of the Phalangers, the inferior premolars are as many as
four, the true molars in these instances being reduced to
three, a dentary formula which closely approximates that of
Plagiaulaz; while in other Phalangers the premolars are
single, the true molars attaining the full complement of four.

In regard to the indications of the true molars, which
might, a priori, have been expected to be the most significant,
the tri-tubercular antepenultimate and the longitudinally two-
edged last tooth are without a known analogue among living
forms. They certainly bear no resemblance to any insecti-
vorous species, placental or marsupial. The general form of
the tubercles of the antepenultimate suggests some resem-
blance to the omnivorous pachyderms, but it is not sufficiently
pronounced to counterbalance the strong leaning of the pre-

1 Waterhouse, Nat. Hist. of Mam- | Transactions, vol. ii. p. 383, of the 8th
malia, vol. i. p. 8. This generalization | Jan. 1839.
was previously stated by Professor| ? Waterhouse, op. cit. pp. 307, 337.
Owen, in his memoir in the Zoological

424 PLAGIAULAX.

molars to a herbivorous regimen. The wear of the two true
molars would seem to indicate a grinding, as contradis-
tinguished from a crushing or cutting action of the teeth;
and this is confirmed by the form of the articulating surface
of the condyle.

The characters of the jaw are so peculiar, and in some
respects of so mixed and complex a nature, that they require
to be weighed with caution, in conjunction with the teeth, in
forming any opinion regarding the affinities of Plagiaulaz.
The low position of the condyle is so pronounced, and the
elevation of the coronoid above it so considerable, that re-
garded per se, supposing no teeth had been discovered, they
might have been considered to imply with some degree of
certainty, a predaceous animal. The condyle is even re-
latively lower in Plag. Beckles than in Thylacinus, Dasyurus,
and Didelphys, the most carnivorous among marsupial forms.
A condyle so placed was considered by Cuvier to be a positive
indicator of the ferine type. But in Plag. Becklesii, the force
of the indication is counterbalanced by another character, of
which, so far as I am aware, there is no example among any
of the predaceous genera, either placental or marsupial, recent

or fossil, namely, the long neck and horizontal projection of |

the condyle behind the coronoid, the term ‘ neck’ being used
for convenience to imply the constricted portion of the ramus
between the bottom: of the sigmoid notch and the lower
margin. In all the ferine animals the pivot of motion or
transverse condyle is, for obvious mechanical reasons con-
nected with the functions of the jaw, brought on a short stem
close under the base of the coronoid process. In Plagiaulax
it is carried out upon a long pedicle behind, and, pro tanto,
there is a great deviation from the predaceous type. The
arrangement is equally without a parallel among the herbi-
vorous or omnivorous types, in which the condyle is ordi-
narily elevated above the horizontal plane of the teeth, with
more or less freedom of lateral or longitudinal motion. Fur-
ther, the convex articular surface of the condyle, and its
vertical instead of transverse direction, are at variance with
the locked implantation of the jaw of a ferine animal. The
other leading indications all lean towards a vegetable feeder,
namely, the limited surface and moderate elevation of the
coronoid above the plane of the teeth; the feeble develop-
ment of the inflected margin, and the absence of a thick
angular process; the advanced position of the orifice of the
dentary canal; the offset of the inflected margin above it,
and the form of the symphysial suture. These characters,
taken in conjunction with the marked signification of the
teeth, would seem clearly to place Plag. Becklesii among the

ee

_

:

P. BECKLESIT AND P. MINOR. 425

vegetable feeders. In this view, the exceptional position of
the condyle would be regarded as a special modification,
having reference to the abnormal character of the teeth, and
the adjustment involved thereby, 7.e. the excessive develop-
ment of the premolars, and the suppression of so large a
portion of the true molars, together with the functional
degradation of the two which remain.

Giving due weight to these various considerations, and
with the above-indicated analogy in the dental formula to
guide us, I am led to the conclusion that Plagiaulax may be
regarded in the natural system as a marsupial form of rodent,
constituting a peculiar type of the family to which Hypsip-
rymnus belongs, and as bearing, in respect of number of
teeth, the kind of relation to that genus which Dromicia
bears to the other Phalangers, and Acrobata to Petauwrista.
Mr. Waterhouse includes the Kangaroo rats among the
Macropodide: Plagiaulax could never be classed among the
Kangaroos. But, although inferred to have been allied to
Hypsiprymnus, the fossils were generically widely distinct
from the existing Kangaroo rats. A great many links of
the chain which would place them in connection are unknown
to us, some of which may yet turn up in the fossil state.

The species of Plagiaulaw must have presented a form of
which there is nothing to remind us among living marsupials.
This is indicated by the extreme shortening, compression,
and depth of the lower jaw, together with the sudden upward
curve of the incisor, and still more by the depressed position
and backward projection of the condyle. For aught that we
know to the contrary, they may have had the volant habits
of the Flying Phalangers, and flitted from tree to tree among
the oolite forests by means of parachute-folds of their skin.
As the Kangaroo rats are strictly herbivorous, gnawing
scratched-up roots, it may be inferred of Plagiaulax that the
species were herbivorous or frugivorous. I can see nothing
in the character of their teeth to indicate that they were
either insectivorous or omnivorous.

The larger species, Pl. Becklesii, I have named after Mr.
Beckles, the discoverer, to whose energetic and well-con-
sidered explorations Palzeontology is indebted for so many
and important additions to the Upper Oolite (Purbeck) fauna,
after the efforts of the Geological Survey Department, specially
directed to the same object, under so able a head as the late
Professor E. Forbes, had proved unsuccessful. This species
equalled the size of a squirrel, or nearly that of Petawrus
macrourus, one of the Flying Phalangers. The other species

1 The skeleton so named in the Cat. (Osteol. Mus.) Roy. Coll. Surgs., No. 1849.

426 PLAGIAULAX.

was very much smaller, and, being one-half the linear dimen-
sions, was probably about one-twelfth of the bulk of the
former, or near the size of the ‘Pigmy Flying Opossum,’
Acrobata pygmea.

About the mammalian associates of Plagiaulaw I abstain
from making any remarks beyond the few which are intro-
ductory to this paper, as the fossils will so soon pass, for a
detailed description, into the hands of Professor Owen, who
has already designated one of the largest of the new forms
by the generic name of Triconodon. The Purbeck mam-
malian genera announced up to the present date are, there-
fore, Spalacotherium, Triconodon, and Plagiaulax.

There are, however, some points. of general geological
interest, on which I may be permitted to make a few obser-
vations. ;

The first is the relation of resemblance which the molar
teeth of Plagiaulax minor bear to those of the Triassic Micro-
lestes antiquus of Plieninger. The agreement in general form
is so close, that, had detached molars of both been met with
in beds of the same formation, they might have been taken
for back and front, or upper and lower teeth of the same, or
of nearly allied, species. The essential crown-characters are
the same in both, namely, two opposed longitudinal marginal
ridges, more or less lobed or crenated, and separated by an
intermediate chasm or depressed disc.! A solution, however
approximative, of so ancient and obscure a mammal as Micro-
lestes is not devoid of interest. Plieninger considered it to
be predaceous, hence the name; other naturalists were dis-
posed to regard it as leaning, however remotely, to the
omnivorous Pachyderms, or omnivorous Insectivora; while
Professor Owen, in recognizing at once the mammalian
character of the teeth, admitted them to be distinct from
anything fossil or recent known to him.? Pictet, in his
‘ Paléontologie,’ doubtingly includes Microlestes among mar-
supials, for reasons which he does not state, upon the autho-
rity of some of the German describers, whose memoirs I

have not been able to consult.

1 Judging from the very careful draw-
ings and casts, the two teeth of Micro-
lestes, figured in Lyell’s ‘Manual of
Geology, would appear, as there sur-
mised, to indicate at least distinct
species. The larger tooth (fig. 442, p.
343 of that work) resembles the penul-
timate molar of Plagiaulax Becklesii,
regarded in the side-aspect, inner sur-
face. There is in both an anterior
talon, forming an accessory lobule

Bronn notices Microlestes in

where it joins on with the anterior inner
tuberele. But I can detect nothing in
either like the basal cingulum referred
to by Mr. Waterhouse (/oc. cit.). Fig.
441, representing the first discovered
tooth of Microlestes antiquus, crown
aspect, is the one which bears the closest
resemblance to the last true molar of
Plagiaulax minor.

2 Cited in Lyell’s Manual, 5th edi-
tion, p. 343.

6 et

>

P. BECKLESIZ AND P. MINOR. 427

the ‘Lethzea Geognostica,’ as being probably a predaceous
marsupial (3rd edit. vol. ii. p. 122).

The next point to which I would solicit attention is, that
Plagiaulaw would seem in some respects to furnish a crucial
test of the soundness of certain generalizations which have
been put forward regarding the order of successive appear-
ance of mammalia upon the surface of the earth. It has
been maintained by British paleontologists and comparative
physiologists | of the highest authority, that, while there is
no good proof of a serial progressive development from the
lower to the higher forms, there is evidence of another order
of development or successive passage, namely, from the general
to the special, as we descend from the oldest to the modern
period. It is urged by the advocates of this doctrine, that
the Mammalia of the Hocene period assimilated more to the
general plan of the archetype and to the embryonic con-
dition of the vertebrate organization; while the Mammalia
of modern times successively furnish more and more numer-
ous examples of deviation from the archetype, all tending
towards special adaptation. Among other arguments, they
insist that the earliest Hocene Mammalia, both carnivorous
and herbivorous, possessed, in most cases, the full comple-
ment of teeth; while forms characteristic of later times, such
as the Felide and Ruminantia, are remarkable for special
suppression of these organs. If the generalization were
really of as wide an application as has been claimed for it,
we ought to find evidence of closer adherence to the general
archetypic model the further back we recede in time. But
so far is Plagiaulax, at present the oldest well-ascertained
herbivorous mammal yet discovered, from giving any coun-
tenance to the doctrine, that it actually presents the most
specialized exception, so to speak, from the rule, to be met
with in the whole range of the Marsupialia, fossil or recent.
Tt had the smallest number of true molars of any known
genus in that subclass, six at least of the normal number of
incisors being also suppressed; thus exhibiting, at the most
remote end of the chain, the very characters which, under

1 Carpenter, Principles of Compar.
Physiology, 4th edit. 1854, pp. 107-111.
The doctrine here referred to is deve-
loped in considerable detail by Dr. Car-
penter in the passage above indicated.
In a note (loc. cit. p. 111) he disclaims
it as having originated with himself:
‘ The principle expounded in this para-
graph has been prominently enunciated
and illustrated by Professor Owen in
yaripus parts of his writings. The
remarkable facts here stated with re-

spect to the dentition of mammalia are
contained in his article ‘‘ Teeth,” in the
Cyclopzedia of Anatomy and Physiology,
vol. iy.’ Some interesting illustrations
bearing upon the retention of the typical
formula of dentition in the placental
mammals of the Eocene and Miocene
periods, and upon the departure from it in
modern mammals, are adduced by Pro-
fessor Owen in his memoir ‘ On the Den-
tition of Phacocheerus,’ Phil. Trans. for
1850, p. 495.—H. F., June 20, 1857.

45g PLAGIAULAX.

the generalization in question, we might a priori have ex-
pected to encounter at the near end among existing mar-
supials.

The curious fact, that only lower jaws should have turned
up among the Stonesfield mammalian remains has often
been the subject of speculation or remark. The same, to a
certain extent, has held good with the remains found in the
Purbeck beds. Among the determined fossils, lower jaws
predominate largely. But some upper maxillaries have been
met with, and, while writing, I have received intimation of
the discovery of more. Among the undetermined remains
there is a considerable number of other bones of small ani-
mals, many of them probably mammals, but they are seldom
or ever perfect. In these minute creatures, unless the bone
be complete, and, supposing it to be a long bone, with both
its articular surfaces perfect, it is almost hopeless, or at any
rate very discouraging, to attempt to make out the creature
which yielded it; whereas the smallest fragment of a jaw
with a minute tooth in it, speaks volumes of evidence at the
first glance. This I believe to be one great reason why we
hear so much of jaw-remains, and so little of the other bones.
For, as an inferior maxillary is to the other bones of the
skeleton in the ratio of about 1 to 250, ceteris paribus, a large
number of these should be encountered for every lower jaw
that turns up. No indication has yet been met with at Pur-
beck of the bone of a good-sized terrestrial mammal. But I
do not consider the negative evidence in this case to be de-
cisive of their non-existence. The matrix of the so-called
‘Dirt-bed, No. 93,’ by which most of the mammal remains
have been yielded, is a whitish-grey, fine-grained marl, full
of the exuvie of freshwater animals, hardening into a kind
of stone when the moisture is expelled by desiccation, but
very bibulous, and readily becoming pasty, #fter immersion
in water. It has properly no claim to the designation of a
‘ dirt-bed,’ or ‘ancient vegetable soil,’ as there is rarely a
speck of vegetable matter to be seen in the numerous speci-
mens containing bone-remains which have passed through
my hands.' It appears to me to present more the character
of the deposit near the margin of a patch of fresh water, and
that the probable explanation of the association of so many
small bones of minute mammals and lizards is that they were
the floating objects most readily drifted to the margin by a
surface-ripple from wind, or by a wave-eddy. In India, in

1 I am informed by the Assistant- | in 1854, were of a dark colour and con-
Secretary of the Geological Society, how- | tained vegetable remains, together with

ever, that the hand-specimens of this | freshwater shells. See Quart. Journ.
bed, in which Spalacotheritwm occurred | Geol. Soc. vol. x. p. 423.

P. BECKLESIT AND P. MINOR. 429

the tanks, or wherever running water falls into an artificial
lake, numerous remains may be observed, along the margin,
of the bones of frogs, lizards, mice, and musk rats, forming a
more or less continuous edging, without the admixture of
large bones, which lie in abundance below the deeper water.
The former float and are drifted to the margin by the action
of the wind, and rest there.

M. Lartet pointed out to me, in the rich Falunian deposit
of Seissan, certain parts of the Lacustrine bed where skele-
tons of large terrestrial animals, such as Mastodon and Rhi-
noceros, are more or less abundant; while in other situations
near the margin immense quantities occur of the bones of
small animals, such as frogs, lizards, shrews, and minute ro-
dents, which may be taken up by the handful unmixed with
larger bones. The mammaliferous band, ‘ No. 93,’ where
most productive, does not exceed five inches in thickness. If
the excavations could be carried into a line of section where
the bed is thicker, it does not seem too much to hope that
they might be rewarded by the discovery of larger mammals,
when we consider the numerous acquisitions to Paleontology
which have been made within the last two months alone
from Purbeck, and the improbability that a fauna already
proved to have been so extensive should have been restricted
to small creatures only. Further, where herbivorous mam-
mals are shown to have existed, it would seem in the highest
degree improbable that they should have been limited to a
single genus containing two small species like Plagiaulaw.

430 PLAGIAULAX.

VIII. ON THE DISPUTED AFFINITY OF THE MAM-
MALIAN GENUS PLAGIAULAX, FROM THE
PURBECK BEDS.!

One of the most accurate observers and original thinkers of
our time has discoursed with emphatic eloquence on the im-
perfection of the geological record.? Besides what is yet to
be discovered, so much has been irrecoverably lost that we
may never hope to write more than disconnected pages of
the palzeo-biography of nature. The truth of the assertion
comes home to the conviction of all; but so far from dis-
couraging, it only renders us the more eager to pursue what
we may attain. Hvery now and then, in paleontology, an
unknown form is discovered of so unexpected a character
that our habitual train of ideas is diverted by it into a new
avenue of thought. It may confirm a position which has
before been merely conjectural, or but faintly shadowed out;
or it may shake the foundations of some cherished, but un-
sound, hypothesis. It is hailed with more especial satisfac-
tion if it contribute to fill up any of the great gaps in our
existing knowledge. The form itself is often presented to
the first observer in such a mutilated or imperfect aspect,
that at the best he can effect little beyond an approximative
idea of the outline. From the same cause, or from a balanced
conjunction of unusual characters, he may fail in his first
attempt at the interpretation; but he has no reason to be —
ashamed of the failure, if he has devoted his powers fairly to
the investigation ; for a great part of the solid progress made
in science is mainly effected by the later observer correcting
the errors of those who have preceded him. Reproach can
only be felt when we allow some bias unduly to influence our
interpretation—when we strain facts to countenance a par-
ticular view. If the observer has guarded himself against
this weakness, and with care used the proper means of in-
vestigation, whatever opposition his results may at first en-
counter, generally speaking, he may be at ease, in the

} This paper was communicated to{ been reproduced on stone from the ori-
the Geological Society of London on | ginal woodcuts.—[Ep.]
June 4, 1862, and is reprinted from the * Darwin, ‘On the Origin of Species,’
‘ Quarterly Journal of the Society’ for} p. 287.
November, 1862. The illustrations have

ITS DISPUTED AFFINITY. 431
assurance, that further research and future discovery will
only confirm and extend them. If the conclusions are chal-
lenged, science is invariably benefited by the controversy.
Different modes of analysis and different trains of ideas are
brought into conflict; and landmarks are established — the
warning and euidance of future observers.

Among the mammalian forms brought to light firanen
Mr. Beckles’s important researches in the Purbeck Beds,
there was one which struck me with especial interest. I
found in it a singular combination of characters :—The den-
tition modified by suppression to as great an extent as in
any existing form; strong analogies, in some respects, with
known genera, while in others it diverged from them very
widely. Karly in 1857 I communicated to the Geological
Society an account of the genus Plagiaulax, which appeared
in the 13th volume of the ‘ Quarterly Journal’ (p. 261').
About the same time an abridged description of the form,
illustrated by figures, was brought out in the Supplement to
the 5th edition of Sir Charles Lyell’s ‘ Manual of Geology’
(1857, p.17). On both occasions I arrived at the conclusion
that ‘ Plagiaulaw may be regarded in the natural system as a
Marsupial form of Rodent,’ constituting a peculiar type of
the family to which Hypsiprymnus belongs,’ although widely
distinct from that genus.

The only comment impugning this determination that has
come under my notice, appeared in the Article ‘ Palzeon-
tology,’ by Professor Owen, in the 8th edition of the ‘ Ency-
clopzdia Britannica,*? published in January, 1859, and subse-
quently reproduced as a separate work.* The two accounts
differ in some unimportant particulars. I here cite the later
in date, as presumably conveying the latest views of the
author. The following are extracts :—

‘Two specimens exemplified the shape and proportions of
the entire jaw of this species (Plagiaulax Beckles). The
foremost tooth is a very large one, shaped like a canine, but
implanted by a thick root in the fore part of the jaw, like
the large lower incisor of a Shrew or Wombat. The three
anterior teeth in place have compressed trenchant crowns,
and rapidly augment in size from the first to the third.

1 See antea, p. 408.—[ Ep.]

2 T leave the words as they originally
stood; but my meaning would have
been more accurately conveyed by the
expression ‘ Rodent type of Marsupial,’
—rodent being here used in the large
sense, haying reference to the plan of
dentition, characterized by two collateral
incisors in the lower jaw, as typically

shown in the placental series by the
Rodentia and Cheiromys; and in the
Marsupialia, by Phascolomys, modified
in the Macropodide and the Phalan-
gistide by the opposition, in the upper
jaw, of several incisors. (See Cuvier,
Oss. Foss. 4th edit. tom. v. p. 3.)

3 Vol. xvil. p. 161.

4 Paleontology, 2nd edit. p. 353,

432 PLAGIAULAX

They are followed by sockets of two much smaller teeth,
shown in other specimens to have subtuberculate crowns re-
sembling those of Microlestes. The large front tooth of Pla-
giaulax is formed to pierce, retain, and kill; the succeeding
teeth, like the carnassials of Carnivora, are, like the blades
of shears, adapted to cut and divide soft substances, such as
flesh. As in Carnivora, also, these sectorial teeth are suc-
ceeded by a few small tubercular ones. The jaw conforms to
this character of the dentition. It is short in proportion to
its depth, and consequently robust, sending up a broad and
high coronoid process, for the adequate grasp of a large tem-
poral muscle; and the condyle is placed below the level of
the grinding teeth—a character unknown in any herbivorous
or mixed-feeding Mammal: it is pedunculate, as in the pre-
daceous Marsupialia, whilst the lever of the coronoid process
is made the stronger by the condyle being carried further
back from it than in any known carnivorous or herbivorous
animal. The angle of the jaw makes no projection below the
condyle, but is slightly bent inward, according to the Marsu-
pial type.’

‘In the general shape and proportions of the large pre-
molars and succeeding molars, Plagiaulax most resembles
Thylacoleo (fig. 173, p.m. 1 and 2), a much larger extinct
predaceous Marsupial from tertiary beds in Australia. But
the sectorial teeth in Plagiaulax are more deeply grooved ;
whence its name. The single compressed premolar of the
Kangaroo Rat is also grooved; but it is differently shaped,
and is succeeded by four square-crowned, double-ridged
erinders, adapted for vegetable food; and the position of the
condyle, the slenderness of the coronoid, and other characters
of the lower jaw are in conformity to that regimen. In
Thylacoleo the lower canine or canine-shaped incisor pro-
jected from the fore part of the jaw, close to the symphysis,
and the corresponding tooth in Plagiaulaz more closely re-
sembles it in shape and direction than it does the procumbent
incisor of Hypsiprymnus. From this genus Plagiaulax differs
by the obliquity of the grooves on its premolars; by having
only two true molars in each ramus of the jaw, instead of
four; by the salient angle which the surfaces of the molar
and premolar teeth form, instead of presenting a uniform
level line; by the broader, higher, and more vertical
coronoid; and by the very low position of the articular
condyle.

‘The physiological deductions from the above-described
characteristics of the lower jaw and teeth of Plagiaulaaz are,
that it was a carnivorous Marsupial. It probably found its
orey in the contemporary small insectivorous Mammals and

ITS DISPUTED AFFINITY. 433

Lizards, supposing no herbivorous form, like Stereognathus,
to have co-existed during the Upper Oolitic period.’ !

We have here an opinion, professing to be founded on the
high ground of a connected series of physiological correla-
tions, that Plagiaulax was a carnivorous Marsupial; while
the same materials led me to infer that it was phytophagous.
These diametrically opposed inferences recall, in some degree,
the discussion, famous in its day, respecting the disputed
affinities of Amphitherium. The question then was, whether
the fossil was mammal or reptile; and the foundations of
Palzontology were supposed to be concerned in the issue.
In the present instance the area of the field of difference is
less, but the interests involved are still important. Are the
indications of paleontology, more especially in its great
stronghold in the Mammalia—the teeth and correlated or-
eans—so unstable or so obscure, that of two paleeontolo-
gists, the same dental and mandibular materials shall lead
the one to infer that the fossil form was a vegetable feeder,
and the other that it was a predaceous carnivore? Or does
this conflict of opinion arise from different methods having
been followed by the observers in dealing with the evidence ?

As the Geological Society gave to my original communi-
cation a place in its Journal, I feel bound, in the interest of
science, either to support the opinion which I then advanced,
or frankly to admit the correction, if I am found to be in
error. [am further impelled by my sense of self-respect, as
an observer, to consider whether—apart from the conclusions
—lI have fallen into such errors of observation and descrip-
tion as would necessarily be implied, should Professor Owen’s
manner of viewing the objects prove correct; and if so, to
explain the fallacious train of reasoning which led me astray ;
for I cannot plead the excuse that the account was written
in haste, or without due consideration.

Tf the data, upon which the author of ‘ Paleontology ’
professes to rest his physiological deductions, were sound,
the demonstration would be complete. They are put together
with an exemplary show of harmony, and, with a single ex-
ception, every link in the chain is supplied. But there are,
in the case, considerations of paramount import in an argu-
- ment of this nature, that lead me to question their soundness,
and to dissent from the conclusions.

_ And first, as regards the admitted facts. Professor Owen
_ agrees that the Purbeck remains establish two species of
_ Plagiaulax; and as he has adopted two of the woodcuts
_ given in my original description of these species, it is pre-

1 Paleontology, p. 353. I entertain | the deduction which makes Stercogna-
strong doubts about the soundness of | thus to have been herbivorous.

VOL. IL. FF

434 PLAGIAULAX.

sumed that the correctness of the figures is not questioned.
The marsupial nature of the forms is not disputed, nor is
there any difference of opinion about the number or desig-
nation of the teeth.

In both species there is a solitary incisor on each side of
the lower jaw, in the fore part of the incisive border, closely
followed, without the interposition of a canine, by a series
either of three or of four premolars. The rami converge to
a narrow point in front, so that the tooth occupies the entire
width of the incisive border on each side ; and fig. 13, p. 280,
of my former communication,! representing the symphy-
sial portion endwise, shows (what is confirmed by the other
figures) that the two incisors were approximated and colla-
teral, as in the rodent type, placental or marsupial. In
P. minor (Pl. XXXIV. fig. 2), the tooth is procumbent. In
the other and larger species, P. Becklesii, it is more robust,
with a thicker root, and with a more decided curvature up-
wards, suggesting, at the first sight, some resemblance to
the form of a canine. Jn both species the point is bevelled ;?
and I failed to observe in either any mark of the play of an
opposed upper tooth.

What was the function of these incisors? Professor
Owen’s opinion is expressed thus: ‘The large front tooth of
Plagiaulax is formed to pierce, retain, and kill.’ This con-
clusion arrived at, the other characters are naturally regarded
in unison with it, until the genus is finally presented to us as
a predaceous carnivore. It is therefore necessary to examine
the evidence closely. Now, in solving a question of this kind,
comparative anatomy supplies for our guidance fundamental
principles, which govern the interpretation of mere form.
Let us revert to the known marsupial genera, and see what
light generalized observation upon them throws upon the
question. In all the Carnivorous genera and species, fossil
or recent, of which the dentition has been accurately deter-
mined, there are three or more incisors, followed by a canine,
on each side of the jaw, above and below; and the empiri-
cally observed result is consistent with a rational interpreta-
tion of the arrangement, in reference to their food and the
means of procuring it. On the other hand, in all the exist-
ing strictly phytophagous genera, there is only a solitary
incisor (being that next the axis) on either side of the lower
jaw, and no canine; or if, as among the Phalangers, addi-
tional teeth are developed, the outer incisors and canine are

1 Pl. xxxiii. fig. 13 of this volume.— | surface is flattened near the apex, so as
[ Eb. to yield a slightly bevelled point. (See
2 Not in the sense of being denuded | antea, p. 417.)
of enamel by wear; but the posterior

ITS DISPUTED AFFINITY. 435

alike rudimentary. The pair of developed incisors are ap-
proximated and placed collaterally, as in the placental Ro-
dents; and commonly they are projected forwards with but
a very slight upward inclination. They are unequally op-
posed in the upper jaw by two or more incisors on either
side. Why there should be this plurality of incisors above,
and only two invariably occupying the same position below,
is wholly unknown to us; but the constancy of the structure
makes it certain that there must be a sufficient cause for it
in nature; and we employ the generalization, empirically
arrived at, with as much confidence as we do the law of
necessary correlation.' In many critical cases, where the
evidence is limited or defective, the empirical is even a safer
guide than the rational law, since it is freer from the risk of
errors of interpretation. Applied to the instance before us,
it is manifest that the principle on which the incisors in
Plagiaulae are framed, in regard of number, order of sup-
pression, collateral position, and relation to the premolars,
corresponds exactly with the type of the Marsupial Herbi-
vores, such as Halmaturus, Hypsiprymnus, and Phascolarctus,
and that it is wholly at variance with the Carnivorous type.
Let us now test the opinion in its professed character as a
physiological deduction. Throughout the Mammalia, where
teeth perform the functions of canines ‘to pierce, retain, and
lal,’ they are held well apart through the interposition of a
line of incisors—the end being obvious; the points of pene-
tration are doubled, the grasp is strengthened by widening
the base, and the dilacerating and killing powers are multi-
plied. To arrange them collaterally in the axis would be to
place them at a disadvantage to the end to be attained. But
when a gnawing power is required, the middle incisors are
powerfully developed, and placed collaterally in the axis of
the jaws, one on each side, above and below, as typically ex-
emplified in the placental Rodents and Cheiromys. Doubt-
less a Rat when seized can inflict a smart wound on the
hand; but the power is a secondary attribute, complemen-
tary to the main function. Regarded in this aspect, it is
negatively stamped upon the incisors of Plagiaulax by their
collateral position, that they are not constructed upon the
carnivorous plan of design, nor in rational correlation

thereto.

It is obvious that this position of the teeth in Plagiaulax
was not overlooked by the author of ‘ Paleontology’ ; for, on

_ the first occasion he describes the incisor of P. Becklesii as

_ being ‘very large, shaped like a canine, but implanted by a

1 Cuvier, ‘Discours Préliminaire,’ p. 51.
? ?
FF 2

436 PLAGIAULAX.

thick root in the fore-part of the jaw, like the large lower
incisor of a Kangaroo! or Wombat.’ But the shape of the
tooth prevailed in deciding him to pronounce it carnivorous.
Now, the form differs in the two species; and I ask any
Comparative Anatomist to look at fig. 2 of Pl. XXXIV., and
say whether the tooth there represented is formed to pierce,
retain, and kill—being the attributes with which Professor
Owen invests the incisor of P. Becklesii. It is projected for-
wards with a slight upward inclination, somewhat as in the
vegetable-feeding Koala (Phascolarctus cinereus). The incisor
of P. Becklesii? is undoubtedly curved more decidedly upward;
and, when viewed sidewise, it is not very unlike a canine.
But the same may be said equally of the lower incisor of the
Lemurine Aye-Aye (Pl. XXXIV. figs. 13 and 14). In this re-
markable form, the affinities of which were so keenly disputed
by the great French anatomists, Cuvier and De Blainville,
the solitary incisors are collateral, on the Rodent type ; com-
pressed laterally, and very deep at the base, they sweep up-
wards in a bold curve, being scooped vertically behind, to
terminate in a sharp edge ; so that, regarded sidewise, so far
as vertical direction goes, they are more canine-like than in
either species of Plagiaulax. But the resemblance goes no
further. In the former the incisor, which is only partially
invested with enamél, is continued backwards below the
molars, the pulp-nucleus being persistent, and the chisel-
shaped edge is constantly maintained by use *—conditions
which are wanting in the latter. Should the construction of
the skull and other parts of the skeleton of P. Becklesii be
ever discovered, there is little doubt that modifications will
be detected throughout in conformity with those of its in-
cisors, as in the felicitous instance, cited by Cuvier, of the
secret relation between the upper canine-shaped incisors of
the Camel and the bones of the tarsus; this exceptional
character does not remove the Camel from among the
Ruminants, nor does the form of the incisor of P. Becklesia
appear to me to be of sufficient weight to counterbalance
the clear evidence of a phytophagous and rodent plan of
construction.

1 Encyclop. Brit. 8th edit. vol. xvii.
p- 161. ‘Shrew and Wombat’ are sub-
stifuted in the ‘ Paleontology,’ p. 358.

2 Pl. xxxiii, fig. 1.

3 De Blainville asserts that the in-
cisors of the Aye-Aye are invested all
round with a shell of enamel, and that
the posterior facet is not the result of
wear (Mémoire sur l’Aye-Aye, p. 23);
while Dr. Sandwith, in his interesting

account of the habits of this animal,
affirms that the facet is denuded, as in
the Rodents (Zool. Proc. Feb. 22, 1859,
p. 111). Ina finely preserved cranium,
for the transmission of which to London
I am indebted to the great courtesy of
M. Edouard Verreaux of Paris, it is
distinctly seen that the coat of enamel
is limited to a belt which sheathes only
the anterior half of the incisors.

:
:
:

ITS DISPUTED AFFINITY. 437

Professor Owen draws an argument, in confirmation of his
view, from the dentition of Thylacoleo. The statement is:
— ‘In Thylacoleo the lower canine, or canine-shaped incisor,
projected from the fore-part of the jaw, close to the sym-
physis; and the corresponding tooth in Plagiaulax more
closely resembles it in shape and direction than it does the
procumbent incisor of Hypsiprymnus.’! But on referring to
his detailed description of Thylacoleo, we find that the body
of the tooth, of which the shape and direction are adduced
as terms of comparison, together with the fore part of the
symphysis and incisive border, is wanting ?:—‘ The symphysis
(Pl. XIII. fig. 4, s) begins behind, at a vertical line dropped
from a little in advance of the middle of the sectorial, p 4;
it is of a wide and oval form. To judge from the cast, but
little of the jaw appears to have been broken away from the
fore-part of the symphysis. The upper and fore-part shows
the alveolus and base of a tooth (Pl. XI. fig. 3, c) which has
projected obliquely upward and forward. 4t is separated by
an interspace of three lines from the sectorial, and would seem
to be the sole tooth in advance of it. If the ramus be really
produced at the upper part of the symphysis further than is
indicated by the present cast, it may have contained one or
more incisors, and the broken tooth in question may be the
lower canine. If, however, this be really the foremost tooth
of the jaw, it would appear to be one of a pair of large in-
cisors, according to the Marsupial type exhibited by the
Macropodide and Phalangistide.’* ‘But in the lower jaw
the carnassial is succeeded by two very small tubercular
teeth, as in Plagiaulax; and there is a socket close to the
symphysis of the lower jaw of Thylacoleo, which indicates
that the canine may have terminated the dental series there,
and afforded an additional feature of resemblance to the
Plagiaulax.’ 4

In all this, it will be seen, the argument is within the do-
main of conjecture; the tooth oscillates between canine and
incisor ; and not merely so, but the principles which are fol-
lowed as guides in this walk of investigation are set aside,
to give place to the illusory indications of mutilated external
form. Ifthe tooth represented by a stump or socket proves
to be a canine, the comparison will not hold; but if it be

_ solitary with the position of an incisor, will it even then bear
out Professor Owen’s hypothesis, that Thylacoleo, which he

1 Paleontology, p. 352. | curved upwards.’— Stutchbury, Report
2 < Unfortunately this morceau is | on the Discovery of Gold in Australia,

much mutilated, the incisor being bro- | 1855, p. 53.

ken at its entrance into the alveolus; 3 Phil. Trans., vol. exlix. p. 318.
its form cannot therefore be precisely 4 Paleontology, p. 482.
given; but it is evident that it was

438 PLAGIAULAX.

infers to have been one of ‘the fellest and most destructive
of predatory beasts,’ ! may have had the laniary portion of its
teeth in its lower jaw constructed on the type of the most
meek and defenceless of herbivorous marsupials? Bearing
in mind the sense in which the term ‘type’ is accepted
among naturalists, I must avow that I have some difficulty
in realizing the conception. But should the unusual con-
junction of characters assumed above be hereafter established,
there are theoretical considerations which would prove to
demonstration that the types of construction are still abso-
lutely distinct. For in the supposed case the outermost in-
cisor would be the one developed, the inner ones being
suppressed ; while, conversely, in the Macropodide it is the
iwnermost incisor which is developed, the outer ones being
suppressed. Morphologically, therefore, the types of con-
struction would be radically different. If palzontological
investigations were conducted in this manner, there would
be no limit to conjecture; the landmarks which we profess
to follow would be disregarded, and disorder would face us
everywhere. But, happily, science furnishes unerring princi-
ples which provide the corrective. I need hardly add that
the argument drawn from Thylacoleo has, in my view, no
bearing on the incisors of Plagiaulax, and gives no support to
the carnivorous inference.

Next, as regards the premolars. From their peculiar
characters and remarkable development, they furnish the
most striking features in the dentition of the fossil genus.
In P. Becklesvi there are three, and in P. minor, four of these
teeth, which diminish rapidly in size from the last to the
first.? I here take the last as the most determinate in form,
and in its nature the most constant. I compared it rigor-
ously with the corresponding tooth of Hypsiprymuus Gai-
mardi, and I affirm now, as I did in my original paper, that
these homologous teeth, in the two genera, are identical in
every essential point of form and construction. In proof, I
refer to figures 5 and 6 of the representations in Pl. XXXTIT.,
the former showing the last premolar of Plagiaulaa, the latter
of Hypsiprymnus. The resemblance is so manifest and direct,
that I never contemplated that it could be called in ques-
tion; but, as it has been questioned, it is necessary to de-

scend to particulars. In both, the crown viewed from the ~

side is of a quadrately oblong form, the length exceeding the
height ; in both, it is compressed and trenchant, the sides
sloping uniformly from the base to a thin edge like a wedge;
in both, the basal part of the tooth presents a smooth sur-

1 Phil. Trans., vol. exlix. p. 319.
? See Pl. xxxiil. and xxxiy.—[Ep.]

ITS DISPUTED AFFINITY. 439

face, above which the crown is traversed by.a series of close-
set, uniform, and exquisitely defined parallel grooves, sharply
angular, and bounded by linear ridges ; in both, these grooves
occupy both sides of the tooth; and in both, the channelled
sides meet in a finely serrated edge. Not the least remark-
able point in this striking list of agreements is the curious
numerical coincidence—these grooves being developed seven
in number, alike in the homologous premolars of Pl. Becklesia
and of Hypsiprymnus Gaimardi.

As to the points of difference: in Plagiaulax there are
three or four of these teeth, while in Hypsiprymmus there is
but one; in the former they are presented with the maximum
of development, in the latter with the minimum; in the
former the grooves are diagonal, in the latter vertical. With
this exception, and with some trivial details of difference in
the proportion of the length of crown to its height, and in
the amount of the basal surface free from grooving, the last
premolar in Hypsiprymnus is identical in its characters with
that of Plagiaulax. The two convey to my mind the impres-
sion of being typically alike.

The objects strike Professor Owen in a very different light.
His statement is that, ‘in the general shape and proportions
of the large premolar and succeeding molars, Plagiaulax
most resembles Thylacoleo, a much larger predaceous marsu-
pial, from the tertiary beds in Australia. But the sectorial
teeth in Plagiaulax are more deeply grooved; whence its
name. The single compressed premolar of the Kangaroo Rat
is also grooved; but it is differently shaped,’ &. Now,
apart from the inferences, here is a conflict of description,
which can be settled by an appeal to the original speci-
mens. I have described the large premolar as essentially
alike in form, in the Kangaroo Rat and in Plagiaulax. Pro-
fessor Owen states that it is differently shaped in the two ;
if so, | invite him to show wherein the difference consists (I
have failed to detect, and he as yet to indicate it)—bearing
in mind that here it is not a question of slight difference,
such as a modification in the outline of the same organ in
two nearly allied forms, but a difference of type—or of ordinal
importance.

Next as regards the assertion that in the general shape
the large premolar of Plagiaulax most resembles T'hylacoleo,
For convenience, I separate the two terms of the comparison
in the sentence. Professor Owen has figured and described
the sectorial teeth of this large Marsupial, in his late memoir
on the ‘ Fossil Mammalia of Australia.’! In Thylacoleo the

1 Phil. Trans., vol. exlix. p. 318, Pls. xi. and xiii.

440 PLAGIAULAX.

inferior premolars are reduced to a single, but enormously
large and massive, carnassial, with two small tubercular
teeth behind it. The carnassial (Plate XXXIV. figs. 9—12)
consists of a long blade, high in front and lower behind, so
that, if notched in the middle, the divisions would in some
degree resemble the anterior and posterior lobes of the corre-
sponding tooth in the placental Carnivora ;! and the worn
summit is distinctly concave lengthwise; conversely, in both
species of Plagiaulax the corresponding tooth is convex, and
the outline of the whole series describes a convex curve, of
which the last premolar forms the most salient part. The
base of the carnassial in Thylacoleo is ‘ slightly grooved ver-
tically’ on the inside (fig. 9). These indentations disappear
about half-way up towards the edge, where the surface be-
comes reticulately rugose, being precisely the reverse of
what occurs in the last premolar of Hypsiprymnus and
Plagiaulax. Besides the difference of their position upon
the teeth, the grooves of the carnassial of Thylacoleo present
the appearance of furrows, separating superficial undulations
of the enamel. A transverse section of the basal part of the
crown would yield a faintly crenated outline, wholly different
from the salient and re-entering angles of the close-set
parallel grooves of Plagiaulaw and Hypsiprymnus. These
undulations are exhibited chiefly, if not solely, on the inner
side; their presence on the outer is not mentioned. Further,
if the indentations on the premolar of Thylacoleo are to count
for anything as significant of affinity, it should be with
Hypsiprymmus rather than with Plagiaulax, since the furrows
are vertical in the two former. In fact, in the outline and
proportions of the vertical section, the premolar of Thylacoleo
differs less from Hypsiprymnus than it does from that of
Plagiaulax. Ihave failed to realize the asserted resemblance
between Plagiaulax and Thylacoleo in the form of the last pre-
molars; and in the details of outline, section, curvature of
edge, crenulation, surface-markings, &c., [am more impressed
with the differences than with any one point of agreement.

Let us now consider the inference as to the function of
these teeth. It is expressed thus :—‘ The large front tooth
is formed to pierce, retain, and kill; the succeeding teeth
are like the blades of shears, adapted to cut and divide soft
substances like flesh, &c. Professor Owen has elsewhere
described the premolar of Hypsiprymnus as trenchant,? and I
have shown above that the tooth is essentially alike in

1 «The first molar is lunate, the cusps | and a slight. ridge passing to a small
turning inwards, and the anterior cusp | depressed posterior cusp.’—Stutehbury,
rising at a salient angle; the edge is | loc. cit.

trenchant outwards; the second molar is 2 Odontography, vol. i. p. 389.
friangular with a large anterior cusp, :

ee

ITS DISPUTED AFFINITY. 441

Plagiaulax. If, therefore, the function is to be deduced with
such facile certainty from the mere form, the premolar of
Hypsiprymnus ought also to be carnivorous. But we know
that the genus is so strictly herbivorous that the family to
which it belongs has been regarded as representing in the
Marsupialia the Ruminants of the Placental Mammals. With
this fact before us, is it likely that the premolars of Pla-
giaulax were applied to cut and divide flesh? Does the
serrated edge indicate a flesh-cutting function? The sin-
eular agreement between the two genera in their premolars,
down even to the number of grooves, however trivial and
unimportant the character may appear to be, has, I confess,
weighed greatly with me in forming my opinion. No special
function has, as yet, been connected with the peculiarly
erooved tooth of the living Kangaroo Rat. The agreement
is therefore purely empirical; but as the character, according
to our present knowledge, is confined, among many hundred
genera of Mammalia, to certain species of Hypsiprymnus and
to Plagiaulaz, those who have faith in the constancy of the
manifestations of nature will not lightly believe that it was
common to these two genera alone without implying affinity ;
and when this is coupled with the obviously phytophagous
type of the incisors, the conviction will be confirmed. I
need hardly add that I regard the carnivorous deduction
from the shape to be arbitrary and untenable. _

[William Hunter, a century ago, by a parity of reasoning,
arrived at the conclusion that the Mastodon of North America,
from the trenchant form of the transverse crown-ridges of
its molar teeth, was an extinct, colossal, carnivorous animal,
in short, a kind of predaceous flesh-eating Elephant.! The
error in his case, as in the corresponding’ one of Leibnitz,
was excusable, comparative anatomy having been then in its
infancy. But it is not a little startling to see the same sort of
unsound deduction reproduced, in regard of one of the most
pigmy of Mammals, half a century after Cuvier, by his lumi-
nous demonstrations, had indicated the method by which such
signal mistakes might be avoided in future.—Oct. 15, 1862. ]

Professor Owen perceives another indication of resem-
blance between Thylacoleo and Plagiaulax in the proportions
of the large premolar to the succeeding molars. In both,
there are but two molars, and in so far the agreement is
clear; but no further. In Plagiaulaw there are as many as
four premolars; while in Thylacoleo the enormous develop-
ment of the solitary premolar or carnassial is effected at the
expense of the rest of the premolars, which are suppressed,
and of the tubercular teeth, which are dwarfed. In the

1 Phil. Trans. 1767, vol. lyiii. p. 38.

442 PLAGIAULAX.

former, as pointed out in my earlier description, ‘the pre-
molars are inordinately developed, while the true molars are
dwarfed and rudimentary in proportion.’ The operation of
the well-known law of Anamorphosis or Balancement is visible
in both. But examples of it are everywhere seen throughout
animated nature, in the same organ, without reference to
affinity, as, for instance, among the Mammalia, in the canine
of Machairodus and of the Musk Deer. Thylacoleo and Pla-
giaulaz may be regarded as being as wide apart among the
Marsupials as the two former are among Placental Mammals.
The solitary trenchant premolar in some of the species of
Hypsiprymnus is said to attain a very large development.
We have the authority of Professor Owen for the statement,
that in two Potooroos of New Guinea its antero-posterior
extent nearly equals that of the three succeeding molars.'
Tf the teeth of Thylacoleo and Plagiaulax had been on the
same morphological plan of construction, the agreement in
the number of molars would clearly have carried weight ;
but, as such does not appear to be the case, the coincidence
ought not to overrule the other indications, more especially
as the form of the crowns of the molars in the two genera is
totally different. In Thylacoleo, the first tubercular tooth
has the crown compressed, supporting two cusps on its axis,
the anterior lobe being more or less conical, with a smaller
lobe behind it, both on the usual carnivorous type of con-
struction. The second tubercular is only known through its
socket. In both species of Plagiaulax, the two molars present
oblong crowns, supporting two opposed lines of marginal
eminences, separated by a depression. In my original de-
scription, I referred to the fact that in Dromicia and Acrobata
the molars are reduced from the ordinary number, four, to
three. In Plagiaulax the suppression is carried still further,
two only being developed. The agreement in this respect
between the latter and Thylacoleo does not impress me with
the idea of affinity, although admitting, as I do, that it ought
to be duly weighed.

I have entered in such detail upon the dental characters,
because, by the consent of all observers, they are of paramount
weight in the solution of a question of this nature. If the
type be distinctly indicated by them to be herbivorous or
carnivorous, the other characters, however modified they
may be, will ultimately be found to be in relation to the
teeth. The author of ‘ Paleontology,’ having formed his
opinion on the teeth, then examines the characters of the
lower jaw, and finds them in conformity. He adduces the
shortness of the horizontal ramus in proportion to its depth

1 Odontography, vol. i. p. 389.

ITS DISPUTED AFFINITY. 443

as indicative of robustness; also the broad and high coronoid
process, and the pedunculate condyle placed below the level
of the grinding teeth (above, p. 432). They are all regarded
as proving a carnivorous type. They were not overlooked in
my former communication :—‘ The characters of the jaw are
so peculiar, and in some respects of so mixed and complex a
nature, that they ought to be weighed with caution, in con-
junction with the teeth, in forming any opinion of the
affinities of Plagiaulax. The low position of the condyle is
so pronounced, and the elevation of the coronoid above it so
considerable, that, regarded per se, supposing no teeth had
been discovered, they might have been considered to imply,
with some degree of certainty, a predaceous animal.’! But
there were other characters, which, taken in conjunction with
the jaw, appeared to me to counterbalance these indications
—namely, the moderate extent and low elevation of the
coronoid above the grinding plane of the teeth; the long
neck and horizontal projection of the condyle behind the
coronoid ; the form of the condyle itself; and the absence of
a stout angular process behind it. With one exception, I
shall consider these mandibular characters briefly.

And first, as regards the shortness of the horizontal
ramus in proportion to its depth. I refer my reader to
fig. 13 of Pl. XXXIV., representing the side view of the
lower jaw of the Aye-Aye. A glance will satisfy him
that the horizontal ramus is much deeper in proportion to
the length in this form than it is in P. Beckles. The fact
is so obvious that I do not think it necessary to enter upon
the metrical details. Commonly we connect the idea of
robustness in the lower jaw with the form and section of the
mandible presented by the Hyena and Tiger. If the sections
in my original paper are referred to (Pl. XX XIII. figs. 2 &3),
it will be seen that they are totally different. The jaw of
Plagiaulaz in this respect also closely resembles that of the
Aye-Aye.?

The coronoid process comes next for consideration. For
the details of my description of it I refer my readers to
p- 418 of my former paper. It is there stated that ‘in
general form the coronoid process in Plagiaulax resembles
more that of the predaceous marsupials, and of the Ursine
Dasywrus especially, than that of the herbivorous families.
It differs very markedly from the elevated strap-shaped

1 See antea, p. 424.—[En:] rami of the lower jaw present great

2 In the Koala (Phascolarctus cine- | depth in proportion to the length, with
reus), in which the procumbent incisors, | a_ compressed section. (Waterhouse,
as already observed above (p. 436), are | ‘ Mammalia,’ vol. i. p. 264.) But the
projected with an inclination resembling | ascending ramus, in that genus, is on a
that of Plagiaulax minor, the horizontal | totally different plan of construction.

444 PLAGIAULAX.

coronoid of Hypsiprymnus and the other herbivorous mar-
supials. Itis to be remarked, however, that it is less elevated,
and its surface of less area, than in the predaceous genera,
whether marsupial or placental.’ Here, it will be observed,
the comparison was restricted to marsupial forms, beyond
which I did not then think it necessary to carry it. Hf ex-
tended to the Aye-Aye (Pl. XXXIV. fig. 13), additional
light is thrown upon the character. In both, the anterior
edge reclines at an angle of about 45°; in both, the summit is
not much elevated above the grinding-plane of the teeth.
The appearance of elevation, which is at first sight suggested
by the coronoid of Plagiaulaz, arises from the great depth of
the sigmoid notch and the low position of the condyle. If
fig. 1 of Pl. XX XIIT. be referred to, it will be seen that the
process itself is not raised much above the summit of the
premolars. There is a further agreement between the Aye-
Aye and Plagiaulax in the amount of area occupied by the
surface of the coronoid. This is partly disguised, in the
lower jaw of the former, by the broad neck of the condyle
and the shallowness of the lunate notch between it and the
coronoid; if the notch were deepened, as indicated by the
dotted line (f), the resemblance would be complete. I do not,
therefore, admit the force of Professor Owen’s remarks, as
significant of carnivore affinities, that ‘the lower jaw is short
in proportion to its depth, sending up a broad and high
coronoid process for the adequate grasp of a large temporal
muscle’—seeing that all these characters are combined in an
existing gliriform Lemur, which is not a carnivore. The
descriptive terms applied to the coronoid would be suitable
for that of a Tiger or Stoat, but they seem hardly applicable
to the process of Plagiaulaz.

The author of ‘Paleontology’ lays stress on the low
position of the condyle and its long horizontal neck: ‘The
condyle is placed below the level of the grinding-teeth, a
character unknown in any herbivorous or mixed-feeding
Mammal; it is pedunculate, as in the predaceous Marsu-
pialia ; whilst the lever of the coronoid is made the stronger
by the condyle being carried further back than in any known
carnivorous animal.’ But it is not a little remarkable that
he is silent regarding the form of the condyle itself, the most
important of all the mandibular characters after the teeth;
for the peduncle, on which he lays weight, is, like the fang of
a tooth, but the stalk upon which the organ performing the
function is borne. I think it necessary therefore to call
attention to the remarks on the subject contained in my
former paper. In the true carnivorous type the condyle
shows more or less of a cylindrical or terete surface, haying

1TS DISPUTED AFFINITY. 4145

invariably a transverse direction, by which it is locked in the
elenoid cavity of the upper jaw, thus constituting a pivot
like that of a pair of scissors, which constrains the blades to
a vertical motion. In Plagiaulaz all these conditions are
reversed, the condyle being convex, with its long diameter
disposed subvertically ; regarded endwise, it is narrow in
proportion to the height, and the outline is ovate or pyriform,
the broad end being uppermost. This is a form which is
unknown among the Carnivora, but common in the Placental
Rodents, with the difference, however, that in the latter, the
condyle having to work backwards and forwards in a groove,
its articular surface is disposed longitudinally. In the com-
mon Norway Rat, the articular surface of the condyle is
partly vertical, with the pyriform outline of Plagiaulax, but
more compressed; and in one of the American Marmots
(No. 2,259, Mus. R. Coll. of Surgeons) it still more closely
resembles that of the fossil genus. I cite these instances to
show the undercurrent of Rodent analogy which pervades the
jaw of Plagiaulax throughout. But amore conclusive and
irresistible case of correspondence can be adduced in the
condyle of the Aye-Aye. In the words of the celebrated
French anatomist who first settled the affinities of the genus,
‘ La forme générale de la machoire inférieure de ’Aye-Aye
dénote une partie forte, large, ou mieux haute et trés-com-
primée ; la branche horizontale beaucoup plus longue que la
verticale, qui est presque dans la méme direction. Le condyle
qui termine cette branche verticale, dans les autres animaux,
est droite ici, et presque a l’extrémité postérieure de toute la
machoire,’ &c.! The condyle of the Aye-Aye has the same
ovate form as that of Plagiaulax, but reversed, the narrow end
being uppermost (Pl. XXXIV. fig. 13); the articular surface
is broader and somewhat flatter than in that genus, but the
direction of the greater axis is the same, that is, longitudinal
and subvertical.2 The glenoid surface of the upper jaw is
modified in correspondence—being broad and flat, and placed
on an inclined plane that would intersect the tips of the
nasals and the middle of the occipital foramen. Here, then,
is a signal failure in the chain of physiological deductions
requisite to prove that Plagiaulax was a marsupial carnivore.

Next, as regards the depressed position of the condyle—
below the level of the grinding-teeth. The author of ‘ Pa-
leontology ’ states that it is a ‘character unknown among

1 De Blainyille, ‘ Ostéographie:’ Mé- | tudinal, de maniére & empécher tout
moire sur l’Aye-Aye, p. 19. mouvement horizontal, si ce mest de

2‘Ta machoire inférieure, comme | Varriére a l’avant et vice versd.’ (Sand-
celle des autres rongeurs, se meut éyi- | with, Zoological Proceedings, 1859, p.
demment au moyen d’un condyle longi- 113.)

446 PLAGIAULAX.

any herbivorous or mixed-feeding animal.’ I again refer my
reader to the figure (Pl. XXXIV. fig. 13) of the lower jaw of
the Aye-Aye. In it, the articular surface of the condyle,
although directed subvertically, or at the most diagonally, is
wholly below the grinding plane of the molars. It looks
still more depressed in Plagiaulax Beckles ; but this is, in
part, owing to the inflected margin of the angle being broken
off in the fossil, while it is entire and salient in the recent
form, thus elevating the condyle above the lower plane of the
ramus, and leading to an appearance of a greater amount of
difference than exists in nature.!

For my reasoning as regards the signification of the long
neck or pedicle of the condyle, I refer the reader to my
former communication (antea, pp. 419 and 424). It is there
stated that the low position of the condyle ‘is counter-
balanced by another character, of which, so far as I am
aware, there is no example among any of the predaceous
genera, either placental or marsupial, recent or fossil, namely,
the long neck and horizontal projection of the condyle behind
the coronoid,’ &c.; and further on I added that the ‘ arrange-
ment is equally without a parallel among the herbivorous or
omnivorous tribes.’ This latter remark was premature. I was
then acquainted with the Aye-Aye only through the figures
given by De Blainville,? in which the lower jaw is shown
in apposition with the skull, thus concealing the coronoid,
and its relation to the condyle. But if the accompanying
figure (Pl. XXXIV. fig. 15) of the lower jaw detached be re-
ferred to, it will be seen that the condyle is not only below
the level of the grinding plane, but that it is projected a long
way behind the posterior edge of the coronoid, exactly as in
Plagiaulaz, and on the same plan of construction—the sole
difference being that the sigmoid notch is shallow in the
Aye-Aye, and deeply excavated in Plagiaulax. If the notch
were deepened in the former, by removing the plate of bone
behind and below the posterior edge of the coronoid, in the
manner indicated by the dotted line (f/f), the resemblance
would be complete. In order to place these facts of agree-
ment beyond question, I give the following measurements of
the relative proportions of the lower jaw in the Aye-Aye and
P. Becklesia? :—

1 In some of the families of the 2o-
dentia the condyle is barely elevated
above the grinding plane of the molars.
See De Biainville ‘ Ostéographie: genus
Cavia, Pl. ii. Figs. Cavia Cobaya and
C. Capybara; genus Hystrix, Pl.ii., and
Sciurus maximus, Pl.i., while in others,
e.g. Castor, both condyle and coro-

noid are well raised above the same
plane.

? Ostéographie: genus Lemur, Pl. v.

5 It must be borne in mind that fig. 1 of
my previous communication (Pl. xxxiil.
fig. 1), from which the measurements of
P. Becklesti are taken, is magnified two
diameters ; the dimensions are therefore

_ TO

ITS DISPUTED AFFINITY. 447
Cheiromys Plag.
Madagase. Becklesii.
Inch Inch
Length of jaw from condyle to incisive border. 3 . 23 2:0
From condyle to posterior edge of coronoid . : a 18 3)
Height of jaw to summit of coronoid male) 1-0
Height of ramus in front of first true molar Lael 6
Height of ramus behind the incisor : : 5 . 65 45
Height from condyle to a line dropped vertically behind
last molar : F : : : : 2 . 1:25 1:05
Height from the latter point to posterior edge of incisor
at diasteme : : : 6 : : 5 te ‘75

From these proportions it will be seen that both in Chev-
romys and Plagiaulaz the condyle projects behind the edge
of the coronoid to the excessive extent of about one-fourth of
the entire length of the ramus. Professor Owen meets the
argument in my paper by the assertion that the condyle of
Plagiaulax is ‘ pedunculate as in the predaceous marsupials.’
If so, I invite him to adduce the instance, bearing in mind
that the question here is one of degree. The lower jaw of a
tiger now before me measures 9:2 inches from the condyle to
the incisive border, while the projection of the articular sur-
face behind the fall of the coronoid does not exceed °7 of an
inch, or one-thirteenth of the length of the jaw. In Dasyurus
and Thylacinus! the condyle projects behind the coronoid,
but nothing approaching the extent seen in the Aye-Aye and
Plagiaulax. As regards the functional effect of the condyle
being carried so far back behind the edge of the coronoid, it
is a plain question of animal mechanics, which the author of
the ‘ Paleontology ’ thus interprets :—‘ It is pedunculate, as
in the predaceous Marsupialia, whilst the lever of the coro-
noid process is made stronger by the condyle being carried
further back than in any known carnivorous or herbivorous
animal.’ As I regard it, a necessary effect would be to re-
strict the power of separating the jaws in front, essential to
a predaceous animal having laniary teeth constructed to
pierce, retain, and kill, And we have the direct proof in the
Aye-Aye, that the same arrangement there is not applied to
a carnivorous function.’

doubled. But this does not interfere with
the ratios of proportion. Further, in
the Aye-Aye the posterior margin of
the coronoid is assumed to be continued
down vertically, in order to get corre-
sponding measurements. The dimen-
sions of Cheiromys are of the natural size.

1 In the Ursine Dasyurus (No. 1900,
Mus. R. Coll. of Surgeons) the length
of the lower jaw is 4°2 inches, and the
projection of the articular surface be-
hind the deepest part of the sigmoid
notch °4 inch, or about one-tenth of the

entire length of the jaw. In Thylacinus
(No. 19038a of the same collection) the
projection of the condyle is about one-
eighth of the length of the jaw. But in
both these forms the posterior edge of
the apex of the coronoid overhangs the
condyle; while both in Pl. Bechklesti and
the Aye-Aye the articular surface of the
condyle is removed about one-fourth of
the length of the jaw behind the fall of
the coronoid.

2 In the typical Carnivora the ful-
erum is a fixed point, the form of the

448 PLAGIAULAX.

With reference to the angular process, I have nothing to
add to what is set forth in my former communication. This
process, which is a very constant character of the carnivorous
jaw, is wanting as a salient apophysis in Plagiaulaz, although
well developed in the minute insectivorous Myrmecobius.

I have one remark more to make in reference to the form of
Plagiaulax. Fig. 2 of Pl. XXXIV. gives a representation of
what remains of the lower jaw of P. minor, magnified to a
scale of four diameters. The entire length of the specimen,
including the six molars and premolars, together with the
procumbent incisor (according to the metrical line e), does
not exceed *4 of an inch, of which the six cheek-teeth united
make only about two and a half lines (*25 inch). I ask any
zoologist or comparative anatomist to look at it, and say
whether the dental apparatus of this extremely minute crea-
ture is competent to perform the duties required of a preda-
ceous carnivore. Magnitude in this case is an important in-
eredient, as it necessarily involves measure of force. Could
P. minor have preyed on small Mammals and Lizards? Is
it not more probable that this pigmy form was itself an ob-
ject of prey in the Purbeck Fauna ?

In the preceding observations I have gone seriatim into the
objections raised against the view which I advanced of the
affinities of Plagiaulax. In the work referred to, every detail
of external form was regarded in a light different from that
in which it was viewed by me; every inference was contro-
verted ; and the conclusion drawn from the whole was diame-
trically the converse of that arrived at by me. The verdict of
Comparative Anatomists will decide which is right. I have
reconsidered my first inferences, and tried to test their validity
by the strongly contrasted and extreme view put forward by
Professor Owen; and the result has been to confirm the
opinion that Plagiaulaz did not belong to a carnivorous type
of Marsupials. Regarded morphologically, in the plan of its
dental system; rationally, through its condyle and correlated
characters; and empirically, by comparison with Hypsip-
rymnus and Cheiromys, it has led me, through every aspect,
to this conclusion. Enough has been adduced in the fore-

glenoid cavity preventing protrusion or | of the muscular power is more advanced,

retraction of the lower jaw; and the
muscular power being applied close to
the condyle leaves the free part of the
lever longer, or, in other words, admits
of a wider separation of the jaws in
front, for the canines and cutting-teeth
to act. In the Aye-Aye and Rodents
(e.g. Cavia and Hystrix) the fulerum is
moveable, the condyle playing on a flat
glenoid surface; the point of insertion

leaving a short portion of the lever free,
and thus restricting the aperture of the
jaws. These conditions combined with
the oblique direction of the temporal
muscle, implied by the reclining coro-
noid, conspire to produce the antero-
posterior and lateral motions required
by the regimen of these forms. The
same reasoning applies to Plagiaulax.

re

a

=<

EE

ITS DISPUTED AFFINITY. 449

going pages to show that, to whatever family comparative
anatomy may ultimately consign the genus, it must always
be held to be a singularly modified form. I have directed
attention to the numerous points of analogy between the
lower jaw of Plagiaulax and that of the Aye-Aye, itself one
of the rarest and most aberrant of existing Mammalia. They
agree in the collateral position and upward direction of their
strong incisors; in the depth and shortness of the horizontal
ramus; in the backward continuation of the ascending ramus
in the same horizontal line with the body of the jaw, and in
the terminal position of the condyle—the two latter charac-
ters not being found, so far as is at present known, in any
other Mammalia, fossil or recent. They agree further in the
form and direction of the articular surface, in the reclinate
coronoid, and in the backward projection of the condyle
behind it. The two jaws are on the same plan of construction.
Starting from the deep narrow incisors of the Aye-Aye
carried back below the molars, the great depth of its jaw, and
the other associated characters, can be seen to be in necessary
correlation. In Plagiaulaxz they are all presented in a less
degree of development. The resemblance goes no further.
I doubt if in the fossil genus the lower incisors were opposed
in the upper jaw by only two chisel-shaped teeth as in the
Aye-Aye. In all the other dental characters they are widely
distinct. In Plagiaulax the force of the dental system is
manifested in the great development of the premolars, of
which there are none, at least in the adult state, in Cheiromys,
but a vacant bar instead. In the latter there are three molars ;
in the former, only two. While, therefore, admitting that the
common construction of the jaw involves some trait of habit
common to the two and essential to their existence, it does
not impress me with the idea of affinity. For the reasons
which have led me to regard the nearest relationship of the
fossil genus as being in the direction of Hypsiprymnus, I refer
to my former communication passim, and to the preceding
pages. Both genera appear to be Marsupial: their incisors
are on the same morphological plan, and their premolars are
in the main identical, except in point of number. The Aye-
Aye is a nocturnal animal, which uses its strong incisors as
a nipping-apparatus, for breaking and detaching bark and
wood in pursuit of the larvee upon which, in part, it is said to
feed. One of the live specimens procured by Sonnerat, on
the first discovery of this form, lived in captivity two months
fed on boiled rice.! The species of Hypsiprymnus are strictly
vegetable-feeders.

1 «Tl a yveeu prés de deux mois, | riz cuit; il se servait, pour le manger,
n’ayant pour toute nourriture que du | de ses deux doigts comme les Chinois

VOL. Il. GG

450 PLAGIAULAX.

I shall adduce a celebrated case to show how little we
should be authorized to pronounce with confidence on the
nearest affinities of Plagiaulaw from the small measure of
evidence we now possess. The Aye-Aye (Cheiromys Mada-
gascariensis) was discovered by Sonnerat before 1782. The
elder Geoffroy and Cuvier placed it among the Rodents. In
1816, De Blainville submitted the skull and teeth, together
with the bones of the fore-arm, to a rigorous examination,
and convincingly pronounced the Aye-Aye to be a Lemurine
Quadrumane. Notwithstanding the evidence supplied by the
brain-case, teeth, and bones of the fore-arm, Cuvier persisted
in regarding the animal as a Rodent, and in the ‘ Régne
Animal,’ of 1829,’ he placed it between the Squirrels and
Marmots. If, with such a full measure of evidence before
him, the position of Cheiromys in the natural system was so
long erroneously contested by Cuvier, how little warranted
should we be to pronounce dogmatically upon the food and
habits of Plagiaulax from the slender evidence of the lower
jaw! Supposing that Cheiromys were only known to us
through its mandible, what would now be its inferred position
among the Mammalia? While, therefore, regarding Plagqi-
aulax to have been of a phytophagous type in its affinities,
we should not be justified in affirming that it may not have
been a mixed-feeder; it may have fed on buds or fruits, like
the Phalangers; or on roots, like Hypsiprymnus; or on a mixed
regimen of fruits and insects, like the Aye-Aye.

But I maintain that every argument which has been ad-
duced by the author of ‘ Paleontology’ to prove that Pla-
giaulaz was carnivorous has been met in the preceding pages.
The method by which the opposite conclusions have been
arrived at are as different as the results themselves. Pro-
fessor Owen, in so far as his method is disclosed to us, has
gone direct from the indications of form to the supposed
function; and he claims for the inferences, that they are
physiological deductions. Comparative anatomists will de-
cide how far they are entitled to the name. Mere external
form must be handled with caution as an instrument of re-
search; signal mistakes in Paleontology have been com-
mitted through too confident reliance upon it. On the other
hand, the method which I have attempted to pursue was,
first to ascertain upon what morphological plan the teeth of
de baguettes.’ (Sonnerat, quoted in| it took to with great relish, gnawing
Buffon, Supplément, tom. vil. p. 268.) | the larvee of insects out of the branches
The early account of the French tra- | of trees, and feeding on them when it
veller has been confirmed by the later | had the opportunity. (Sandwith, Zoo-
and excellent observations of Dr. Sand- | logical Proceedings, 1859, p. 113.)

with, who fed his captive Aye-Aye upon |! Op. cit. p. 195.
bananas and dates, the latter of which

ITS DISPUTED AFFINITY. 451

Plagiaulax were constructed, and, having determined this, to
supply the rest empirically by comparison with known forms,
using at the same time rational analysis where it could be
applied, e.g. to the condyle. The case is of sufficient interest
and importance to test the sufficiency of the respective modes
of analysis.

In the general remarks appended to my former communi-
cation, I called attention to the contradictory bearing of the
dental system of Plagiaulax upon the assumption that the

earliest Mammals had the full complement of teeth. To that

may be added the further evidence of specialization, in the
analogy of its mandible with that of the Aye-Aye, one of the
most exceptional of Mammals. If we cast a glance over the
instructive table given in Lyell’s ‘Supplement’ (page 23),
and reflect on the interpretation of the hiatus between the
Upper Oolitic beds and the ‘ Sables de Bracheux,’ how vast the
interval in time by which they are separated, and how modern
in comparison the earliest of Tertiary Mammals! Tf, on the
other hand, Plagiaulax be regarded through the medium of the
view advocated with such power by Darwin, through what a
number of intermediate forms must not the genus have passed
before it attained the specialized condition in which the
fossils come before us! What a variety of Mammals may
we not hope to disentomb from the buried Oolitic fauna,
should Mr. Beckles resume his explorations, or another
Beckles take his place!

The remote antiquity of the fossil as a mammalian genus
must alone invest the discussion of its affinities with an in-
terest which will prevent the question from resting in its
present disputed state. Other paleontologists will examine
the evidence, and give their verdict. Mr. Beckles’s speci-
mens have long since passed out of my hands; and I have
deferred my rejoinder in the expectation that they might ere
now have found their way into some public collection, where
I could have again submitted them to examination and com-
parison ; but, as that has not yet taken place, I have thought
it full time to reply, lest my silence should be construed into
a tacit acquiescence in the carnivorous character attributed
to Plagiaulax, which I do not accept—nor the reasoning on
which it is founded.

Geez

452 SPERMOPHILUS.

IX. NOTE ON THE OCCURRENCE OF SPERMO-
PHILUS IN THE CAVE FAUNA OF ENGLAND.

Tue glacial or northern character of the greater part of the Cave
Fauna of England is so well-known, that no surprise will be excited by
the announcement of the addition to it of a mammal genus, the habitat
of which is now restricted to the Arctic Circle and the northern parts
of Europe, Siberia, and America.

In 1842, MM. Desnoyers and Constant Prévost announced the dis-
covery of abundant remains of Spermophilus (citillus), from the bone-
breccia of the caves and fissures of Montmorency, in the gypsum of the
Paris Basin.

In 1859, while examining the rich collection of fossil bones, made
by the late Rev. Daniel Williams, from the numerous caves of the
Mendip hills, and now preserved in the Museum of the Somersetshire
Literary and Philosophical Society at Taunton, I detected two rami
of the lower jaw of a species of Spermophilus, which, by the kindness of
the Rev. W. Arthur Jones, I was enabled to compare with recent
specimens in the metropolitan collections.

The most perfect remain consists of a right ramus, containing the
incisor and three last molars in stu. The most anterior of the molar
series (last premolar) is wanting, probably a recent loss, as the alveolus
is free from matrix. The jaw is in a singularly good state of preserva-
tion, showing the condyle, coronoid, and posterior angle, quite entire ;
the only damage being a slight abrasion of the lower margin near the
angle, and a small film broken off from the anterior side of the incisive
sheath. Figs. 1-3 of Plate XX XV. represent the fossil, top and side
aspects: fig. 1 showing the crowns of the molars, and figs. 2 and 3
respectively the inner and outer sides of the jaw erect.

The well-known characters of the genus Spermophilus are so clearly
shown by the fossil, that it is not necessary to describe the teeth in
detail. The following are the principal dimensions of the Spermophilus
from Taunton : ?

April 8, 1858.—Length of jaw from posterior surface of condyle to anterior edge
of incisive sheath, 1°35 in. Ditto from posterior angle (lower margin) to ditto,
1:35 in. Length of three posterior molars, 0°4 in. Length of line occupied by
the four molars, 0°52 in. Length of diasteme, 0°32 in. Length of mentary suture,
0°32 in. Length of exserted portion of incisor, 0°25 in. Length from posterior
margin of condyle to anterior edge of coronoid, 0°33 in. Length from bottom of
sinus between condyle and posterior angle to anterior edge of masseteric disc, 0°7
in. Length trom ditto to anterior edge of incisive sheath, 1:2 in. Height of coro-
noid to line of bottom of sigmoid notch, 0°22 in. Width of ditto at the middle,
0-lin. Extreme height of jaw to the apex of the coronoid, 0°65 in. Ditto behind

1 This paper was commenced in 1858, | letter to M. Lartet.—[Ep.]
but was never completed. The defici- ? The remainder of the paper is made
ency has been supplied by extracts from | up of extracts from the author's Note-
the author's Note-books, and from a | books, &c.—[Eb.]

8. ERYTHROGENOIDES. 453

it, at sigmoid notch, 0°45 in. Ditto from condyle to lower margin near angle,
0:52 in. Ditto of ramus at last molar, inner surface, to alveolar border, 0°33 in.
Ditto of ramus at second molar, inner surface, to ditto, 0°33 in. Ditto at outer edge
of first molar to alveolar border, 0°23 in. Ditto of ramus, where constricted, close
behind the suture, 0:2in. Ditto of masseteric disc, under last molar, 0°31 in. Width
of ascending ramus from sinus behind to coronoid margin at alveolar border, 0-45 in.
Length from anterior border of masseteric disc to tip of incisive sheath, 0-46 in.

June 2, 1858.—Compared the Taunton Spermophilus with all the
specimens in the British Museum, viz.: Sp. erythrogenys (Pl. XXXV.
figs. 8 and 9) of Siberia; Sp. musogaricus of the Altai mountains; Sp.
lateralis of N. America; Sp. Douglastii, Sp. Harrisii of N. America ;
Sp. concolor (Souslik) of Moldavia (Pl. XXXV. figs. 4 and 5); Sp.
Parryi of N. America; Sp. Eversmanii of the Altai mountains (PI.
XXXYV. figs. 6 and 7); and Sp. Franklinii. ‘The jaw of Sp. Franklinit
is of the same size, but the body is deeper and the sigmoid notch
shallower. On the whole, the fossil approaches nearest to Sp. erythro-
genys, although the teeth are larger.

November 12, 1859.—Again compared the Taunton Spermophilus
with the skulls in the British Museum, along with Gerrard. The fossil
is certainly a Spermophilus. It differs from the Souslik specimen (Sp.
concolor) in being as long to the tip of the incisive sheath, as the latter
is to the tip of the incisives; three teeth also in the fossil are equal in
length to four in S. concolor. The fossil agrees most nearly in form
and size with the S. erythrogenys of Siberia (P]. XXXYV. figs. 8 and
9). It has the same bold masseteric disc, and the form of the coronoid
is the same, but it is a little larger. The fossil differs from the S.
Eversmanii of the Altai mountains in being a little larger. The form
of the coronoid and the sigmoid notch, however, are exactly alike. In
S. erythrogenys the coronoid is more erect than in the fossil from
Taunton, and the notch differently formed.

July 27, 1860.—Mr. Waterhouse compared the Mendip Spermophilus
with the European and N. American specimens in the Osteological
collection of the British Museum, and found it nearest to S. erythrogenys,
but the teeth are larger, and the fore and aft diameter relatively
greater. The length of the line occupied by four molars in S. erythro-
genys 18°4in.; in the Mendip specimen it is ‘52 in. Mr. Waterhouse
considers the fossil wholly different from all the other species. I will
call it Spermophilus erytkrogenoides.

[On Nov. 28, 1862, Dr. Falconer received from Dr. H. P. Black-
more of Salisbury, a letter enclosing sketches of a Spermophilus found
by him in the Pleistocene clay at Fisherton, near Salisbury. The
remains of as many as twelve individuals had been found. The frontal
bone was flattened and depressed, the superciliary ridges elevated,
forming strong post-orbital processes extending backwards, as in the
Spermophilus superciliosus of Kaup. On Dec. 20, 1862, Dr. Falconer
took the drawings to the British Museum and made the following
note.—Ep. |

‘Compared the drawing from Salisbury (Pl. XXXYV. fig. 10). None
of the lower jaws of the living species have the slender reclinate pedicle
to the condyle as in the fossil, nor the same erect stumpy coronoid (is
it entire in the fossil?); but S. musogaricus from the Altai has got the
brow concave between the frontals, with elevated superciliary ridges, as
in Kaup’s S. superciliosus. The horizontal condyle of the fossil differs

454 SPERMOPHILUS.

much. The post orbital processes are very much hooked backwards
in a Russian skull of S. musogaricus, but they are not so divergent or
spread out as in the fossil. The palatine slits are very large in the
drawing of the fossil.’

Extract from Letter to M. Lartet, June 27, 1864.—‘ With regard to
the Spermophilus: there are three species in the British Museum, the
lower jaws of which are nearly of the same size, and, with certain
differences, very much alike in form. These are: S. concolor, S.
Hversmani, and S. erythrogenys. But unfortunately, the number of
specimens of each is too limited to determine what the range of variety
of form is in each. On comparing the materials, I found the cave
Spermophilus of the Mendip hills to come nearest to S. Eversmanii
and S. erythrogenys ; but most closely to resemble the latter. I there-
fore called it S. erythrogenoides. The lower jaw from M. Desnoyers
(Montmorency) appears to me to resemble the Mendip Cave form very
closely. I believe them to be the same. But the Spermophilus from
the gravel of Salisbury is considerably larger, with a very broad
ascending ramus, a long neck to the condyle, and other differences.
Possibly there are two fossil species in England; and neither of them
the ‘“ Souslik,” ze. S. concolor, The name of S. erythrogenoides is
provisional,’

DESCRIPTION OF PLATE XXXV.
SPERMOPHILUS. FELIS SPELMA.

Figs. 1, 2, 3. Three different views of lower jaw, right side, of Spermo-
philus erythrogenoides, from the Mendip Hills, of the natural size.
Fig. 1, shows crowns of molars; fig. 2, outer surface of jaw;
fig. 3, inner surface. The figures have been reproduced from
drawings executed for Dr. Falconer by Mr. Dinkel, and described
at page 452, where the words ‘ inner’ and ‘ outer,’ as referring
to figures 2 and 3 respectively, ought to be reversed.

Figs. 4 to 9. Represent lower jaws of existing species of Spermophilus,
drawn for Dr. Falconer by Mr. Dinkel, from specimens in the
Osteological Collection of the British Museum. Of the natural
size.

Figs. 4 and 5. Spermophilus concolor.
Figs. 6 and 7. S. Eversmanit.
Figs. 8 and 9. S. erythrogenys.

Fig. 10. Right side of lower jaw of fossil Spermophilus, from Salisbury.
Outer surface ; natural size. (See page 453.)

Fig. 11. Right carnassier of Felis spelea from the Mendip caverns.
Copied from a rough pencil sketch in one of Dr. Falconer’s
Note-Books. (See page 455.)

VOL. Il. af

Voll. Plate 35

1.2.5.Spermophilus erythrogenoides 4,5 5.concolor
6.7. 8.Eversmani 68,9.S.erythrogenys. 10.5. from Salisbury.
Jo® Pmikel del Lens Aldous lth: ll,Felis spelza. Harrison. & Sans Imp?

FELIS SPELASA. 465

X. NOTE ON THE OCCURRENCE OF FELIS SPELHA
IN THE MENDIP CAVERNS AND ELSEWHERE,
AND ON A SPECIES OF FELIS FOUND IN ONE
OF THE GOWER CAVES!

Tue late Reverend D. Williams, Rector of Bleadon, is well known
for the zeal with which he devoted himself, during many years, to the
exploration of the ossiferous caves in the Mendip hills. His large
collection of fossil bones was chiefly derived from the caves of Bleadon
and Hutton ; but although the origin of most of his specimens was
probably readily recognized by himself, he, like many collectors, failed
to mark the remains from different localities by distinctive labels; and
when his collection passed into other hands, the precise history of many
specimens was lost. Such appears to have been the case with the three
carnassial teeth about to be described. They form part of his col-
lection in the Taunton Museum, where they were kept together, care-
fully attached to a card, and marked numerically, but without any note
or memorandum of their exact origin.

The specimens are three detached upper carnassial molars of the
milk set of young animals. Two are of the right side of the jaw, and
one of the left. It is evident that two of the three belonged to opposite
sides of the same individual, agreeing, as they do, exactly in size,
colour, and form, down to the most minute particulars.

The most perfect specimen, being the right carnassial of the pair, is
represented by fig. 11 of Pl. XXXV. It is in the most remarkable
state of preservation, both as regards the lobes of the crown and the
very fragile and attenuated shells of the fangs which are entire to their
base. The crown, as compared with the corresponding milk tooth in the
young Lion or Tiger, is thinner and more compressed. As is normal
in the Felide, the blade is distinctly trilobed. The anterior lobe, which
is low, is deeply bifid, showing two conical and subequal tubercles
(a, 6), separated by a very decided notch. The anterior tubercle
(a) is roundish, terminating in an edge which is directed diagonally
inwards. ‘The second tubercle (4) forms a sort of obtuse three-sided
pyramid, the inner and posterior face of which is elevated by an in-
clined step above the inner surface of the middle cusp. It is thus
intruded inwards beyond the plane of the latter and of the anterior
tubercle, its inner angle forming a sharp curved ridge, and presenting
somewhat the appearance of a surface abraded by wear. The middle
lobe (c) forms a compressed semi-conical entire cusp, flattish on the
inner side and convex on the outer; its anterior margin is but slightly
inclined, straight, and obtuse; the posterior margin is convex in outline,
and thinned off to a trenchant edge. This middle cusp, at its base on

1 This paper, on the carnivorous teeth | Subsequent examination and compari-
from the Mendip caverns, was com-| son, however, as shown by extracts from

menced by the author on the supposition | his Note-books, satisfied him that they
that the teeth belonged to Machairodus. | belonged to Felis spclea.—[En.|

456 FELIS.

the inner side, contracts into a narrow band which slopes inwards,
terminating in a middle fang which diverges inwards. The sloping
basal surface is the obsolete representative of the inner tubercle of the
upper carnassial in the /elide, which is well developed in the Tiger
and Lion, but greatly reduced in the Drepanodons. This rudimentary
tubercle and diverging fang are given off from the middle of the blade
in a line with the apex of the central cusp, proving the tooth to have
belonged to the milk set. In the adult carnassial tooth of the Felide
generally, the inner tubercle is thrown out much further forward, and
opposite the smus between the anterior and middle lobes with which
it alternates.
Extracts from Dr. F.’s Note-books.

British Museum, October 2, 1858.— Received from the Rev. William
Arthur Jones the three carnassier teeth from Taunton. Compared
them at the British Museum with the cast of Bravard’s Felis Megan-
tereon (Machairod.); with an original carnassier of Smilodon, be-
longing to the defective side of De Blainville’s skull; and with the
young Machairodus of the Sewalik hills, noticed in Owen’s ‘ Brit.
Foss. Mam.,’ p. 178, also with a young Felis spelea, having milk
dentition, from Kent’s Hole.

In the Sewalik specimen the carnassier (right upper) is formed with
a very thin blade. The anterior lobe is damaged, but judging from
what remains, it would seem to have been two-lobed, as in the Taunton
specimen. The middle cusp is thinner, but pointed and formed like
the English specimen. But neither the anterior lobe nor the middle
one bears the slightest indication of bearing 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 com-
pressed and sharp-edged. All the points rise.

Length of crown, °75 in.

There is an interval between the carnassier and canine of 0°3 in.,
part of which has been artificially 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 distinct show of the fang—
two, fore, and aft—of a tubercular in a line with the sectorial behind
it. The breadth of the canine at its base is 0:5. It is very compressed.
The posterior concave edge is finely serrulated; the anterior is entire.
Owen says that both edges are distinctly serrated.!

In Bravard’s cast of Megantereon (No. 28,882) there is a trilobed
premolar on the left side. The carnassial has the anterior lobe entire,
as in the true Lelid@, and not bilobed. It is thicker than in the Sewalik
or English form. The middle cusp is rather thick and short, as in the
Cave Lion, and the posterior cusp is horizontal and bilobed externally.
But the principal point to notice is, that the internal tubercle is given
off at 3 lines from the anterior margin, and about 8 lines from the
posterior, or at the anterior fourth nearly of the tooth.

Length of carnassier, 1-1 in. Breadth of canine,*75 in. Ditto in Lovell Phillips’
English specimen,? 1:11 in.

’ See vol. i. p. 550, and Plate xxv. fig. 5—[Ep.]
2 See p. 461.—[ tp. |

ee St FS

FELIS SPELAA. 457

The latter is also young, so that it is of a different species.

In the adult Cave Lion (No. 28,553), from Lundwig Cavern, the
carnassier is thick and massive. The anterior lobe. is simple. The
internal tubercle is very much in front, being in a vertical line with
the commissure of the two front lobes, sloped off and bearing no tuber-
cular or raised edge. The length of the carnassier is 1-5 inch.

But in a specimen of young Cave Lion from Kent's Hole (No. 18,982),
comprising the left upper maxilla with the orbit, milk premolar, milk
carnassier, and the permanent carnassier coming behind it in germ, the
following characters are observed. The length of the crown of the milk
carnassier is 0-9 in. The anterior lobe is distinctly bitubercular, and
the tubercles are not in the same line, the second being placed inwards
alteraately with the front tubercle and the middle cusp, which are in
the same line. The middle cusp is not much elevated, and descending
from the middle of the point, and very near the middle of the tooth
(but slightly in front), is a distinct tubercle, supported on a pronged
fang. ‘This tubercle is distinctly raised like that of a Hyena.

The Taunton specimens are very distinct from all.

The specimen belonging to De Blainville’s Smilodon skull is the
upper carnassier of the right side, with a premolar in front. It is thick
and short, and 4-tuberculate. The central cusp is large. The length of
the carnassier, which is partly broken behind, is 1-7 in. The height of
the middle cusp outside, ‘95 in. Ditto inside to surface of inner tubercle,
1-1in. The tooth is long and massive, and the crown high; it is not
so thick comparatively as in the Tiger. Anteriorly there is a cusp
bearing a small tubercle in front in the same line, and separated by a
distinct notch. The middle cusp is shaped as in the Taunton speci-
mens; but it descends on the inner side as a distinct fang, which is
nearly vertical, There is no mark of an inner tubercle as a distinct
object. ‘The fang divaricates less than in the English tooth. The
posterior lobe is long, but the crown characters are not shown, owing
to the fracture. The posterior fang is very broad, or two-thirds the
length of the tooth. The internal tubercle is situated at one-third from
the anterior margin, much more in advance than in the English
specimen.

On careful comparison, however, with the Felis spelea, the Taunton
teeth appear to be the milk carnassiers of that species.

[In Dr. Falconer’s Note-books there are also descriptions of frag-
ments of the upper and lower jaws of Felis spelea, among the Bielbeck
fossils in the York Natural History Museum; of the upper and lower
canines (probably from same individual), fragments of the lower jaw,
humerus, and phalanges of the same species, in Dr. Spurrell’s collection
at Belvedere ; also of a lower jaw, supposed to be from Erith (but this
was doubted by Dr. F.), of Felis spelea, in the collection of Mr. W. H.
Newsted at Maidstone. The following note also appears to refer to
another species of Felis——Ep. |

Nore on Ferris rrom Norta Hityt Tor.

British Museum, July, 1862.—Compared Colonel Wood’s three spe-
cimens of Felis, from North Hill Tor, with skulls of Lion, Tiger,
Leopard, Panther (Leopardus varius), Jaguar, and Puma; they cor-

458 FELIS.

respond nearest with the skull of a young, but adult, Lion or Lioness
from Africa, No. 112, e.

Two of the canines from N. Hill Tor are a pair from the same
individual, which must have been young, though adult, as the crown
bears no mark of wear, either on the side or the apex. The right
canine is the more perfect in the fang portion, while the left is the most
perfect as regards the crown portion, half the shell of the fang in the
latter being removed longitudinally. The principal points deserving
notice about the fossil are the following :—

1. The posterior vertical cutting edge is very salient, forming a
sharp keel.

2. The anterior mesial vertical ridge is very pronounced, termi-
nating below in a gibbous bulb or prominence, near the termination of
the shell of enamel ; both the ridge and the bulb are more pronounced
than in either the Tiger or Lion.

3. The outer vertical fissure is solitary, both in the fossil and in the
Lion; but in the fossil it has a much longer stretch. In the Lion the
fissure does not occupy one-third of the length of the crown, while in
the fossil it occupies more than one-half (nearly three-fourths); the
form, size, and contour of the crown of the canine of the Lion (112, c)
and of the fossil are very much alike, the proportions, if anything, being
larger in the recent animal.

The third canine from N. Hill Tor is a left lower, evidently of a
very old animal, as the point, anterior surface, and part of the inner
side, are far advanced in wear; in form and proportions it agrees
closely with the young Lion, but it is shorter in the crown and longer
in the fang ; and the convex contour of united fang and crown is dis-
posed in a curve of less convexity in the fossil; the bottom part of the
fang is completely closed up, while in the young Lion it is still open.

The young fossil canines are very much of the same age and size as
those of the young Tiger (114, cc).

DREPANODON. 459

XI. NOTE ON THE REMAINS OF DREPANODON OR
MACHAIRODUS OF REPUTED BRITISH ORIGIN.!

Remains of the great feline carnivore of the Cave period, armed in
the upper jaw with long compressed and falciform canines, are every-
where so rare in Europe, that I have been induced to examine all the
specimens of reputed British origin. The dentition, characters of the
skull, and closely feline affinities of the numerous forms belonging to
the genus variously named Drepanodon, Megantereon, Machairodus,
Stenodon, and Smilodon, by different paleontologists, are now well
known, through the finely preserved materials discovered by Bravard
in Auvergne, of his Felis megantereon, and by Lund in Brazil, of his
Smilodon populator. But of the largest and most formidable of the
European species, only one or two incisors and a few upper canines are
at present known collectively, from a single cave of Kent’s Hole in
England, from Auvergne, and from the Val d’Arno. It is of importance
in instituting a comparison between the Mammalian Faunas of England
and Italy, during the Pliocene and Post-Pliocene periods, to collect all
the materials calculated to throw light on the form or forms called
Drepanodon or Machairodus cultridens and latidens, regarding which
our information is as yet so limited.

Professor Owen founds the distinctness of his Machairodus latidens
from the Italian Machairodus cultridens, on the proportionately greater
breadth of the three English canines found by Mr. McHnery in Kent’s
Hole. The length of the Italian tooth is 8°5 in., and the breadth of the
crown at the base 1‘5 in.; while the corresponding measurements of the
English specimen are 6: and 1:2in. The breadth of the English tooth
ought to be only 1-06 in., were the proportion to the length the same as in
the Italian. Owen says these differences are constant and well marked.
But are they sufficient for a distinction of species, or are the materials
sufficiently abundant to affirm their constancy? I think not. In my
opinion, the English MMachairodus latidens is probably the same as the
Italian WW. cultridens.

The only specimens of Machairodus teeth of reputed British origin
are, one in the Museum of the Geological Society, one in the Wood-
wardian Museum at Oxford, two in the British Museum, and one in
the Museum of the Royal College of Surgeons.

1. The specimen in the Geological Society’s Museum (No 23,413)
is a fine canine of Machairodus latidens, very like the one figured

by Owen in the ‘ British Fossil Mammalia’ (fig. 69, p. 180), but of a

1 The first paragraph in this memo- | but which careful comparison proved to
randum was written as an introduction | belong to Felis spelea (see page 455).
to a description of three carnassier teeth | The remainder is made up of extracts
in the Taunton Museum, which the author | from the author’s Note-books.—[Eb. |
at first believed to belong to Machairodus,

460 DREPANODON.

different individual. . It is obliquely worn at the summit, and closely
serrulated on the concave side; behind, there is a ridge without
serrulation. The tooth is darkish externally, but very white within. ‘The
specimen is in a collection from Kent’s Hole, presented by Mrs. Cazalet,
February 16, 1826. There is a cast of it at Oxford, and in the British
Museum.

During a late visit to Oxford, Professor Phillips told me that he had
sent to York for an old letter in the records of the Natural History
Museum, in which McEnery, as far back as 1826, mentions Ursus
cultridens teeth, from Kent’s Hole, as having been shown to Cuvier and
Buckland.

In Plate F. of an unpublished work by Mr. McEnery, which was to
have illustrated the fossil remains found in Kent’s Hole Cave, Torquay,
several figures are given of the teeth of Ursus cultridens (Machairodus
cultridens). Figs. 1,2, and 3! are serrated on both sides, and apparently
are different views of the same tooth. Figs. 6 and 7 are only serrated
on the concave side; they represent apparently two teeth, although this
is not quite certain. The teeth were found by Mr. McEnery in Kent's
Hole, in January, 1826, and were mixed with teeth and gnawed bones
of rhinoceros, elephant, horse, ox, elk, deer, hyenas, bears, wolves,
foxes, &e.

2. In Professor Buckland’s collection in the Woodwardian Mu-
seum, at Oxford, there is the original ‘ Ursus cultridens’ canine of
McEnery from Kent’s Hole, one of the three found by him there. At
McEnery’s sale, two of these three teeth were purchased by Dr. Lovell
Phillips. One (this one) he presented to Dr. Buckland, and the other
to the British Museum. On May 8, 1858, I had an opportunity of
examing the Oxford tooth, and comparing it with the cast of a specimen
also in Buckland’s collection, and labelled ‘ Ursus cultridens of Issoire.’
‘The tooth substance of the Kent’s Hole specimen is very fresh looking
and yellowish, apparently glazed over and preserved by drying oil.
The apex is worn off and rounded. The wear is chiefly on one side,
as in old Tiger canines, and a dirty vascular core is exposed. The
serrature is very finely exhibited on the concave attenuated edge; there
is none on the convex side, which is partly worn vertically, showing a
denuded furrow through the enamel into the ivory. The convex side
is rather thick. The base of the crown and the fang portion are also
thick. In addition to the terminal loss, the fang is also corroded. The
Oxford canine is evidently of a different age from that figured by Pro-
fessor Owen in the ‘ British Fossil Mammalia’ (fig. 69, p. 180). The
cast above referred to is that of a corresponding, but younger, tooth.
It is quite entire to the extreme tip, which is very sharp.

Kent’s Hol
Cast Res ae
in. in.
Extreme length along convex border - : 7°25 4:0
Length of crown portion . : F : 7 45 2°75
Greatest width of crown . : : 5 - 14 1:15
Greatest thickness of ditto . : : c bd 65

1 Figs. 2 and 3 of McEnery’s Plate have been reproduced in figs. 8 and 7 of
Plate xxy. of vol. i—[Ep.]

i i iS

DREPANODON, 461

In the Oxford Museum there are also six other casts of Machairodus
canines ; but it is not known if they are English or Italian.

3. The other specimen purchased at McEnery’s sale, by Dr. L.
Phillips, was presented by him to the British Museum, where it now is
(No. 14,954). It is an intact germ specimen, broken across at the fang
portion, which is hollow. The tooth is very broad, flat, and pointed
like a shark’s tooth. It is excessively crenulated at both edges. The
mineral condition is very white, as in Kent Hole specimens.

4, There is a specimen in the Museum of the Royal College of
Surgeons (No. 103, Pal. Cat.), presented by Lord Enniskillen, and also
reputed to be from Kent’s Hole, Torquay. [In a letter from Dr.
Lovell Phillips to Dr. Falconer, it is stated that the third Machairodus
tooth at McKnery’s sale was purchased by Dr. Battersby for ‘some
nobleman.’—Ep. |

5. The last specimen is also reputed to be of British origin, although
this is doubtful. It is in the British Museum, No. 15,433, and is
designated Ursus cultridens by Koenig. The blade is narrow, and
there is no crenulation of the edges. It was purchased at McEnery’s
sale, but its mineral conditivn is certainly not that of Kent’s Hole; it
is dark and discoloured, as if from the Val d’Arno.

462 CANIS.

XII. NOTE ON THE REMAINS OF A HYANOID
WOLF FROM SPRITSAIL-TOR CAVE.!

Juty 21, 1862.—Came with Colonel Wood to compare the lower jaw
of the Hyenoid Wolf from ‘ Spritsail-Tor,’ with the collections of recent
species in the Osteological department of the British Museum.

This specimen consists of the greater part of the left ramus entire on
the incisive border of symphysis, but the whole of the coronoid, condyle,
and angular process wanting. There is also a fissure running from
behind forwards into the ramus, below the carnassial, the bone at this
part being wanting; but the inferior contour, nearly as far back as the
angle, is perfect. The specimen contains 7m situ the four anterior
molars quite perfect ; the carnassial is also present, but the two tuber-
culars had dropped out, the fangs of the anterior one being shown; the
canine and incisors had also dropped out. The alveolus of the canine
is of very large size. (See Pl. XXXVI. figs. 1 and 2.)

The following are’the dimensions as compared with a large European
Wolf, with the Canis occidentalis of N. America, and with the Lycaon
venaticus (or Canis pictus of De Blainville) :—

Canis occi-| Canis
ritsai Wolf : :
“oscil, | 168 4° deo) ere
in. in. in. in,
Extreme length of fragment . 0 : 5°70 — -— —
Joint length of 4 premolars and se 3-15 3 3:85 2-37
nassial : :
Joint length of 4 premolars : 2 : 2° 1:90 2°20 1:52
Length of last premolar : - cp hk 0°45 0°65 0°50
Length of carnassial . : 1:25 1-20 1:20 0-92
Greatest width of ditto near middle .| 0°50 0°50 0°50 0°38
Greatest width of last premolar. 0°40 0°35 0°35 0°25
Transverse diameter of canine alveolus 0:60 0-40 0-45 x
somewhat oblique : :
Antero-posterior ditto ~ . 0°50 0°50 — —
Height of jaw in front of first Seni 6 1:05 0:90 1° 0°95
Ditto between carnassialand last premolar; 1-40 115 1°30 0°82
Ditto behind carnassial . , : 1-40 1:30 1:30 0°92

With reference to the above comparative measurements, it is to be
remarked, that in the American Wolf (Canis occidentalis) the premolars
are very loosely set together, while in the three other forms they are

compact. The striking difference of the fossil from the American and —

“uropean Wolves is, that the last premolar has a small cusp in front of
the large middle cusp, with two serratures behind it; this anterior
cusp, or serrature, is entirely wanting both in European and N.
American Wolves, fossil and recent, while it is very pronounced in the

} This note is from an entry in one of the author’s Note-books.—[Ep. |

CANIS. 463

Spritsail-Tor fossil specimen; as compared with the Canis pictus, the
correspondence in this respect is exact, that species having a very pro-
nounced anterior cusplet, which circumstance, together with the decided
acumination of the central cusp, led De Blainville to name it O. Hy-
cenoides. All the premolars have very pointed cusps in the latter species.
As regards the carnassial, the form of the talon in the ‘ Spritsail’ fossil
resembles more the European Wolf in its considerable width and general
form. The talon is narrower in the Canis occidentalis, and in the C.
pictus it takes more after the Hyena in form.

Further, in the Gower fossil, all the premolars (except the first) are,
ina marked degree, thicker and more massive, in proportion, than in the
existing European Wolf; and in this respect it differs still more from
the C. occidentalis.

464 HY NA.

XIII. NOTES ON HYAINA.!
I.—SyNopsis OF NOMENCLATURE OF LIVING Hyanas (1863).
A.
Hyena, Storr. Canis, Linneus.
Suscenus I. Euuyzna. Large tubercular above; lower carnassial

with internal accessory point, either rudimentary or large.
1. H. (Euhyena) striata (Zimmermann).

Syn. H. vulgaris (Desmarest).
Rey me 15 le antiquorum (Temminck).
ee ees orientalis (Tiedemann).
se aL fasciata (Odmann). Stockholm Trs.

» Canis Hyena (Linn.).
» Canis Hyenomelas ? Bruce.
Suscenus II. Crocorra (Kaup). Crocuta (J. E. Gray).
2. H. (Crocotta) Crocuta (Zimmermann).
Syn, H. maculata (Thunberg. Petersburgh Trans. ).
iy ee ek Capensis (Desmarest).
Crocuta maculata (Gray).
3. H. (Crocotta) brunnea, Thunb. Stockholm Trs., 1820.
Syn: HH. fusca, Geoffroy St. Hilaire.
ate villosa, And. Smith,Linn. Tr.,vol. xv. 1829, p.460.
Crocuta brunnea (Gray, ‘ Tiger Wolf’).
Hyene brune, Cuv., Fred. Cuv., ‘ Straand Wolf’ or Shere
Wolf of the Boers.
Hyene var. Cuv., Oss. Foss., tom. iv. 384.
‘ Hyene dont la patrie est inconnue,’ Cuvier.

B.

1. H. (Euhyena) striata, Zimm. Habitat—India, N. W. P. of
Hindostan ; Persia ; Syria; Asia Minor, to shores of Black Sea; Egypt;
Barbary ; Abyssinia? According to Andrew Smith, not found in
South Africa; but H. brunnea (villosa), Smith, when young, resembles
it. Bruce’s Canis Hycenomelas on the Atbara affluent of Nile,
doubtful ; if real, of what species? Said by him to be very large. An
H. (Crocotta) brunnea? I have seen H. striata tamed in India by the
Rev. R. Everest, and kept in my grounds. Hindostanee : ‘ Luckrabugga’
or ‘Bughera.’ Striped Hyena.

2. H. (Crocotta) maculata, Thunberg. Habitat—South Africa
and Abyssinia ‘ Tyger Wolff’ of the Boers.

3. H. (Crocotta) brunnea, Thunb.

villosa, Smith. fusca, Geoffroy ; ‘Straand Wolff,’
‘Shore’ or ‘Strand Wolf’ of the Boers.

Habitat—Cape of Good Hope; Natal; Senegambia; Mozambique.

Frequents seashore ; nearly as large as Hyena spelea. Devours offal

1 Extracted from the Author’s Ncte-books. The memoranda A and B are of
the same date.—| Ep. |

HY ANA. 465

left by the tide. Is a true Crocotta, by the teeth above and below, and
- not a Huhyena. Blainville’s figure, Pl. IIL. is certainly not it, but
Hyena striata. Vide three fine skulls in the British Museum; one
with Mr. Waterhouse, two with Dr..J. E. Gray.!

II.—Nore on Sxutu or Fossrz Hyana mn FLorence Museum.

Florence, May 20, 1859.—Examined skull of Hyena of large size,
but broken vertically through the cerebral portion, about an inch and a
half behind the orbits; the facial part with the palate and both orbits
present. On the right side, one large premolar and the canine in situ ;
on the left, the carnassier with three premolars in situ, but a good deal
broken, also the six incisive teeth. Looks like Hyena spelea, but
differs very remarkably in showing two disc-shaped eminences, above
the post-orbitary processes, forming a sort of step between the facial
and cerebral portions, with a channel between.

Interval between the orbits, 2-9 in. Width of palate behind, 4-4 in. Length of

. palate in middle, about 4:4 in.
Not quite certain, but the disc-shaped appearance may have been
caused partly by a crush (?). The specimen is a late acquisition, and
from the Val d’Arno; it bears a label of interrogation, ‘ Hyena Arver-
nensis 2’

IIJ.—Norte on Hya£na FROM THE Cave oF SAn TrEoporo IN SICILY.

‘ Oxford, July 5, 1860.—Compared with Baron Anca the upper and
lower jaws of Hyena from Sicily with Buckland’s recent specimens.
The Sicilian bones are certainly not of the Indian striped Hyena, but
of the Hyena crocuta, or spotted Hyzna of the Cape.’?

IV.—Notre on Hyayna From Caverna DE Peprara, Proyincta DE
SEGOVIA.

[This note is dated September 11, 1864, and refers to a fragment of
the left ramus of the lower jaw, comprising the whole of the symphysis
and the three premolars. It was transmitted to Dr. Falconer from Don
Casiano de Prado of Madrid, through M. Lartet. On careful comparison
in the British Museum, Dr. Falconer pronounced it to be identical with

the H. spelea of the English caves, and to be entirely different from
Baron Anca’s San Teodoro specimen.—Ep. ]

1 Since Dr. Falconer’s death it hasbeen drunnea. See Proc. Lin. Soc., May 38,
shown by Mr. Busk that he was led into 1866, p. 59.—[Ep.]
error by the three skulls in the British | * Misled by the specimens in the Bri-
Museum, none of which really belonged | tish Museum, Dr. Falconer, subsequently
to H. brunnea, the name they bore, and | in a letter to M. Lartet, dated Sept. 9,
that, in fact, there was then no accessible | 1864, expressed the opinion that this
specimen of a cranium of H. brunnea | identification was wrong and that the
either in the British Museum or in the | San Teodoro Hyzna was H. brunnea.—
College of Surgeons. De Blainville’s | [Ep.]
figure, according to Mr. Busk, is of H. |

VOL. II. HH

466 URSUS.

XIV. NOTHS ON FOSSIL SPECIES OF URSUS.

I.—NoteE on Foss. Remains or Ursus 1x Museum at Nicer.

Nice, December 11, 1858.—Found one large detached canine of
Ursus, very perfect, presented by Dr. Perez; also a fragment of an
upper incisive ; and a maxillary of a youngish animal containing three
imcisives, with the canine broken off short, and the transverse section
of a very small premolar nearly touching the canine; also several de-
tached molars, some upper, some lower. On a rough examination,
without the means of comparison, approximated them to Ursus priscus,
rather than to Ursus speleus. Among the pieces sent to Gastaldi,
there is also an upper incisive containing a canine, crown broken off,
probably of the same species of Bear.

II.—Nore or Lower Jaw or Ursus 1n tHe Norwicu Museum.

Norwich Museum, June 6, 1862.—Ursus ? labelled, ‘Cromer,
from Miss Gurney’s collection.’ This is a left ramus, lower jaw, com-
prising the whole of the horizontal ramus from the incisive border back
to near the angle, the ascending ramus, with the angle and posterior
part of coronoid process, broken off; anterior edge of coronoid process
present. Sockets of three incisors present, without teeth; canine in
situ, beautifully perfect, comparatively small in size, with facet of wear
inside towards upper third.

Four last molars 7m situ beautifully perfect; diastemal edge free for
about an inch and a quarter ; presents a small rudimentary stump-shaped.
premolar in situ, very much resembling that of fig..35, p. 106, of Owen’s
‘Brit. Foss. Mam.’ Dimensions :—

Extreme length from incisive border to fracture, 9° in. From incisive border to
posterior edge, 7-in. Width of ramus at symphysis over sheath of canine, 1°3 in.
Length of series of molars, 3°7 in. Extreme length of crown of last molar, 1-in.
Extreme width of ditto, 0'8 in. Length of penultimate, 1- in. Width of ditto behind,
0-7 in. Width of ditto in front, 0°65in. Length of antepenultimate, l-in. Width
of ditto behind, 0°55 in. Ditto ditto in front, 0-4 in. Length of last premolar,
0°6 in. Height of jaw to alveolar border between penultimate and last molar, 23 in.
Ditto to alveolar border at middle of last molar, inner side, 2°65 in. Ditto
at middle of last premolar, inside, 2°5 in. Height of jaw at middle of diasteme,
2°3 in.

The chief characters to be noted are these. The last molar shows
the crown beautifully preserved, there being only slight marginal abrasion
along the inner side; the pattern of crown surface is very complex ; it
differs from other Bears in the great proportion which the last molar
bears to the penultimate ; the length being the same, and the width much
greater (vide fig. 35, above referred to, A. B. C.). The penultimate has
the inner margin worn down to sublunate discs, the outer tubercles
being nearly intact; form of crown oblong. The antepenultimate is of
about same length as two last, but more compressed, and more ad-
vanced in wear. The pre-antepenultimate shows the disc of one large

—————

URSUS. 467

sub-anterior outer cusp, bounded on the inside by three marginal tuber-
cles of small size, but very distinctly defined. The canine is compara-
tively small; antero-posterior diameter of base of crown where enamel
begins, 0°8 in.; transverse ditto, 0-6 in.; height of remains of crown,
1:02 in ; apex slightly abraded.

Ursus speleus has an enormous canine in proportion.

Iil.—Nore. Upper Maxituary or Ursus in CoLuecrion or Rev.
S. W. Kina.

This specimen is a right maxillary, comprising the last two molars in
situ, with the empty fare-pits of the antepenultimate, of an older animal
than above jaw in the Norwich Museum, as is clearly proved by the
crown of the penultimate being ground down to a uniformly smooth
disc. The base of the zygomatic process and portion of palate, together
with infra-orbitary foramen, are present; all rest broken off. Di-
mensions :—

Joint length of 2 molars, crowns, 2'4in. Length of crown, last molar, 1°5 in.
Greatest width in front, 0°8 in. Ditto behind, 0°65 in. Length of penultimate,
0:95 in. Greatest width of ditto, 0°75 in.

The last molar corresponds very closely in relative dimensions with
those of the lower jaw in the Norwich Museum, and, as in it, the
unworn talon portion exhibits a very complex pattern. The teeth in
both are jet black, with a very bright vitreous polish. They would
seem to have been yielded by two different individuals of the same
species.

The detached canine found along with upper jaw (detached) is of
comparatively small size, agreeing in this respect with canine of above.

IV.—Note on Ursus From Deporan’s Den.

British Museum, July 22, 1862.—Comparison of lower jaw of Bear
from Deborah’s Den, Gower.!

This specimen is a finely preserved and nearly perfect right ramus
of the lower jaw, entire from the incisive border to the condyle, and
defective only in a slight portion of the outer casing of the canine
alveolus. The posterior point of the coronoid is also slightly abraded,
and a small portion is broken off the point of the angular process. The
symphysial suture is shown throughout; the three incisors had dropped
out, but the alveoli are distinctly shown; the canine is also in situ,
slightly worn at the apex, which is somewhat truncated.

Of the molar teeth, only one is seen in situ, viz. the (m. 2) penul-
timate, and it is not a little remarkable, that although the animal is
shown to have been a somewhat aged adult, there is not the slightest
indication that the last tubercular was ever developed. The surface
where it ought to have been found was perfectly smooth, and on drill-
ing down into the dentary canal, not a trace of the tooth was found,
nor in the base of the coronoid. The same absence of the last tooth
sometimes occurs either on one or on both sides of the lower jaw of the

1‘ Deborah’s Den’ was a cave disco- | half a mile westward of Paviland Cave,
vered by Col. Wood, at Gower, in the | and contained numerous remains of Bear,
autumn of 1861. It is situated about | Wolf, &c—[Ep.]

HH2

463 URSUS.

Polar Bear;! but in the instances observed the corresponding upper
molar was constantly present.

Of the other molar teeth, the alveoli only are shown; of these an-
terior teeth there were four. First, asmall premolar separated from the
canine by the interval of about a line, as in the Polar Bear; then a
vacant barre, 0°8 of an inch in extent, followed by another single-
fanged premolar; this tooth was succeeded by a two-fanged last pre-
molar in close apposition; and lastly, there is seen the double fang of
a first molar, followed by the two-fanged alveolus of m. 2.

The penultimate molar in a general way presents the characteristic
form of crown seen in the Bears; but it differs remarkably from all
the specimens with which we have compared it, in the excessive width
in proportion to the length of crown. In the Pelar Bear, Ursus ferozx,
and Ursus Arctos, the crown of the penultimate is commonly a narrow
oblong, while in the fossil it is of a very broad oblong form, as in
U. speleus; the coronal surface is slightly worn, but not to the extent
of confounding the prominences. Ou the inner side (as in the Grisly
Bear, 1157, B.) there are four points, the two anterior of which occu-
pying the front half are obtusely pyramidal, and somewhat as in the
Grisly Bear, 1137, B., but less salient. They are slightly abraded at
the apex. The two posterior inner points are much smaller, nearly
equal in size, and depressed so as to form a step below the anterior
points.2_ The outer side consists, as in the Grisly Bear, of two prin-
cipal points, the anterior of which is ground down into a three-pronged
disc; the posterior point is lower, and slightly abraded. In all these
details of coronal prominences, the tooth of the fossil corresponds very
closely with that of the Grisly Bear, although still maintaining a con-
siderable difference in the proportion of length to width.

In the Cashmeer Ursus [sabellinus, 1010, F., the crown of the pen-
ultimate is broad in proportion to the length, but the surface is exces-
sively warty and complex, resembling more that of a Hog, and some
forms of U. speleus. In Ursus Arctos the tooth is very much smaller,
and narrow in proportion to the length, with finely warty subdivisions
of the surface. In the Polar Bear, 221, B., the wartiness is still finer
and more complex ; this is beautifully shown in the last molar of that
species; on the other hand, in all these details of surface, the crown of
the fossil, as regards simplicity of pattern, approaches nearest the Grisly
Bear, 1137, B.

The muscular impressions of the temporal show the fossil to be oldish.

In form of jaw, lower line of ramus, the fossil comes nearest the Polar
Bear, in which the angular process rests on the ground plane; it does
nearly the same in fossil. But in the Polar Bear there is no crook
process under the middle of the coronoid. There is more or less crook
in all the others. In U. Arctos, the Cashmeer Bear, and U. ferox, the
angular process is well raised above the ground plane, and enormously
so in U. speleus. Coronoid in fossil broader and lower than in any of |
the others: nearest U. ferov; it has acrescentic outline behind: quite
different in U. speleus. Condyle in great depth and shortness, most like

1 Thus, in the Brit. Mus. specimen of | on the left side, but not on the right.
Polar Bear, No. 221 G., there is no indi- 2 In the Grisly Bear the points are
cation of an inferior last molar on| nearly of uniform height from front to
either side, and in 22] B. it is present | back.

URSUS. 469

Cashmeer Bear! and U. ferox; very different from Polar and Brown
Bears. In short diasteme most like the Cashmeer Bear. In proportions
of penultimate molar, most like U. speleus.

The premolars are very inconstant in number in all the Bears. In
Polar Bear, commonly in lower jaw p.m. 1 and 4. They seem to be
most common and constant in the American Black Bear.

A specimen of Ursus speleus,! said to be from Bacton (No. 16,448
Brit. Mus. Cat.), figured in the Brit. Fos. Mam., fig. 34, c, is a beautifully
perfect right ramus of the lower jaw, with the four last molars zn situ.
The penultimate agrees with that of the Bear from Deborah’s Den, in
its great width in proportion to its length, but the crown is more warty.
I doubt much the asserted origin of this specimen. It differs from all
the Bacton specimens that I have seen. It looks as if out of peat.

A set of Bear’s jaws from Kent’s Hole, although not agreeing with
Ursus spelcus, are entirely different from the Deborah’s Den specimen.
Except that the condyle is much shorter, they resemble the Grays
Thurrock specimens. This has the condyle enormously long, but the
penultimate is of the same size and form as in the Deborah’s Den fossil.

The Muggendorf skull of U. priscus has a small jaw, with a short
diasteme, and the condyle is somewhat like that of the Deborah’s Den
fossil.

[Figures of the lower jaw of fossil Ursus, from Deborah’s Den,
are given in Plate XXXVI. The comparative measurements will be
found over the page.—Ep. |

1 Dr. F. had at first great doubts | with M. Lartet, he came to the ccnclu-
whether this Bacton jaw belonged to U. | sion that it did.—[ Ep. |
speleus, but after careful examination

URSUS.

470

= 29-0 69.0 1-0 9-0 9.0 — —
es! ¢8.0 1.0 G6. 0-1 1.0 — —
1-0 9-0 (i) $¢-0 9-0 9-0 ¢9.0 09-0 1.0
CZ-T °6.0 ¢6'0 0-1 0-1 6-0 6-0 cr
LT ¥1 8-1 8-3 8-1 LT 6-1 (i) #1
CL.3 £3 £3 9-3 69.3 0.3 6-3 13
ChE 9-8 LF 6-8 VP GL oF L¥
— 96-1 0-3 CL3 9-3 L3 ats G¢.Z
— 8-0 9-0 00-1 8-0 10 G9. ¢8.0
— Ni 6-1 Wie lec 8-1 42 CLI
s 8-8 + Lt G 8-8 8-7 06-8
= 0-£ 8 6-6 ¥-8 8-3 3-8 03-8
ar Mai 8.1 rar 3G 8-1 6-1 08-1
L@ 9-1 CLT L3 0:3 LT L3 06-1
9-1 9-1 LZ 6-1 ¢9.T 6-1 C0.3

= Le 8-8 LF 3-7 8 L8 01-8
CL.9 ¢.¢ 8-¢ ¢0-9 ¥-9 6-9 3-9 01-9
- £6 (etre | econ 8-01 0.6 moqn | 096

“sn “1 : ‘Sn ‘414 ; ? : ; : ;
ithe [Sabie ett |Saeiace | GES | MS. | SE | moses
TOOULYSVD ‘Iwog Wwojoeg Fe ;

: ; : , D ‘Op JO YIPTA 4soqvaty
: ¢ : : 2 IvOW jSeT JO YJouerT
2 a * (WMO Jo 4809018) ‘Op Jo YIPLAA
: : : : * eyemyqnued jo yyduerTy

ouluRd 0} 7 ‘urd woIZ ooULySICT

' : * ycurd pur ‘Z *[ “ul Jo Tosaye ‘og
{ : : : : ayeurynued Jo puo
TOTI9}SOd 0} OUTIL) WLOIT 00} Jo OUT] Jo YASUO]

: : * -S[PPrar qtogs prowoe.roo FO) UAPL AA,
: : . * ‘Op JO JYSIoy [BoryA0A 4S0yRaL1)
p * dajourrp osteasuetj—o[Aptod Jo WPL
* yooro puryeq xode jo aTpprur 0} op Jo FyS10FT
: : 0 . OSRq JV PIOMOLOD JO YIPI A
: : ° * ajyeuyqnued puryeq ‘og
: 3 0 * — ayetutyqnued jo yuoay ut ‘og
‘ 9 emloyserp Jo ofpprur ye Mel 50 qYSIoFT
: * — e[Apuod Jo soRZMs 0} “op WoIZ YSUET
* ‘Op 0} plouor0d Jo oSpa AoTAo}U WoT, YISUO'T

; * Jopxog aarstout 0} efApuod worE YYSUOT

SPUOMOANSvOTTL OATQV.ABACUO/

DESCRIPTION OF PEATE XxX vas
CANIS AND URsvus.

The figures in this Plate have been drawn by Mr. Dinkel
from the original specimens, formerly in the collection of
Colonel Wood, of Stout Hall, but now deposited in the British
Museum.

Vig. 1. Upper margin, showing teeth of lower jaw of the Hyenoid
Wolf from Spritsail-Tor Cave, described at page 462. Three-
fourths of the natural size.

Fig. 2. Outer surface of same specimen.

a
Q
oo

- Lower jaw of Ursus from ‘Deborah’s Den,’ described at page
467 ; one-half of the natural size. Upper margin, showing the
canine and penultimate molar, and behind the latter the opening
which was drilled in search of the last tooth.

Fig. 4. Inner surface of same specimen.

Figs. 5 and 6. Represent the outer and inner surface of the left canine
of Felis from N. Hill Tor Cave, referred to at page 458. Three-
fourths of the natural size.

VOL. II,

Vol. II. Plate 36.

W.West- mp.

1, 2. Hyzenoid Wolf from Spritsail Tor. 3,4: Ursus from Deborah’s Den.

\

i dhe

mh ey
cathe)
- ie

)

re

CERVUS SEDGWICKITI. 471

XV. NOTES ON FOSSIL SPECIES OF CERVUS, IN-
CLUDING A DESCRIPTION OF A REMARK-
ABLE FOSSIL ANTLER OF A LARGE SPECIES
OF EXTINCT CERVUS IN THE COLLECTION
OF THE REV. JOHN GUNN, IRSTEHAD.'

THaT remains of large extinct species of Ungulata abound
in the ‘ Forest-bed’ and Laminated Clays, which are inter-
posed between the Crag and the ‘ Boulder-clay ’ along the
Norfolk coast from Cromer to Happisburgh, has long been
well known. They are amply represented in the collection
formed by Miss Anna Gurney, Mr. Foulger, and others, now
deposited in the Natural History Museum at Norwich; in
that of the Rev. James Layton, now absorbed in the British
Museum; and in the private collections amassed during
many years by the Rev. John Gunn of Irstead, by Mr. Fitch
of Norwich, and more recently by the Rev. 8S. W. King of
Saxlingham.

But our knowledge of the Mammalian species belonging
to this extinct Fauna, regarded as a whole, is still in a very
unsatisfactory state. The Proboscidea have yielded three
well-marked species, with indications of probably a fourth :
the former being Hlephas (Loxodon) meridionalis, E. (Huelephas)
antiquus, and H. (Huelephas) primigenius. The genus Rhino-
ceros has yielded two well-marked species, namely Rh. Htrus-
cus and Rh. leptorhinus (Rh. megarhinus of Christol). It is
worthy of remark, that although undoubted remains of EH.
primigenius have been yielded by the Forest-bed, its usual
associate in the Quaternary Valley Gravels, Rh. tichorhinus,
has not yet, so far as I am aware, been established upon reli-
able evidence. Two of the Fossil Elephants above-named
and both species of Rhinoceros belong to the Pliocene depo-
sits of the Val d’Arno and of the Valley of the Po, as ex-
hibited in the Astigiana, and in the Sub-Apennines of Pia-
cenza and the Romagna. It might therefore be inferred that
the rest of the fauna would partake of a Pliocene character.
Remains of the Ruminantia abourd in all the collections,
consisting of antlers and bones of large species of Cervus,
and bones of the Bovide. But, commonly they occur in a

1 The description of Mr. Gunn’santler Cervus in his collection. Additions have
was written in 1863, but was never pub- | likewise been made from the author's

lished. Iam indebted to Mr. Gunn for | Note-books.—[Ep.]
* Dr. Falconer’s description of crania of

472 CERVUS.

very fragmentary state. Of the antler, however, which forms
the subject of the following description, Mr. Gunn has suc-
ceeded in uniting the numerous fragments, so as to render
the specimen unusually perfect.

CErvus.—Sub-gen. Hucladoceros (Fale.)!\—Charact.: Horns
pedunculate, branched, without brow- or bez-antlers ; beam
and compound antlers compressed, the latter thrown off for-
wards and terminating without palmation in long tynes.

C. (Hucladoceros) Sedgwickii (Gunn).—Horns very ample,
indicating a large species. Bur prominent, without obliquity.
Beam cylindrical and straight at the base, compressed up-
wards, giving from its anterior margin three sub-equidistant
antlers, which from the summit downwards are successively
bi- tri- and quadrifurcate. Tynes long, straight, and conical,
diverging in the same vertical plane. A single elongated
tyne or fork ? terminating the beam.

Habit.: Fossil in the indurated gravel pan of the ‘ Forest-
bed’ of the Norfolk coast at Bacton, south of Coal Gap. In
the collection of the Rev. John Gunn.

The cervine horn to which the above definition applies is
shown by the outer oblique ridge of the pedicle to be of the
right side. The specimen was found in January, 1863, em-
bedded in the matrix after the scouring action of a heavy
gale, which denuded the ‘ Forest-bed ;’ and although broken
up into a large number of fragments, the most of these were
recovered ; so that, with the exception of the branch or snag
immediately terminating the beam, the outline of the perfect
horn is fairly presented in its principal features. The whole
of the pedicle is present, with the corresponding basal por-
tion of the frontal, but without any determinable part of the
orbit or cerebral cavity.

The pedicle is well developed, and of sufficient length to
remove, in that respect, the species from affinity with the
Reindeer, the existing Elk, and the Irish Elk. The bur is
prominent, with large warty tubercles. The beam is cylin-
drical and straight at the base, for a distance of about four
inches, where the first antler is given off. It is then recli-
nate in an easy curve, with a slight bend outwards, until
above the offset of the median antler, whence it is curved
gently forward as far as opposite the offset of the third or royal
antler, where the commencement of another slight reclina-
tion is visible. The apex of the beam is broken off here on a
level with the upper edge of the royal antler, leaving doubt
as to the manner in which it terminated upwards, whether in
a single elongated snag, or in a bifurcated branch.?

' See vol. i. p. 587.—[ Eb.] ‘The butt of the beam is very terete at
® Extract from Dr. F's Note-book— | the base and boldly channelled, and

CERVUS SEDGWICKII. 473

From the offset of the supra-basilar, the beam and its main
branches are broad and compressed, but nowhere to the
extent of palmation ; the section yielded by the former, being
ovate, with constriction in the three intervals between the
offsets of the successive antlers. The beam and its branches
are boldly and closely channelled, the grooves being con-
tinued out on the flattened sides of the long straight tynes ;
‘but they are alike free from pearly rugosity.'

The two most striking characters about the horn consist:
1st. In the enormous spread of the tynes of the median and
royal antlers, straight outwards from the beam, as compared
with the length of the latter. 2nd. In the symmetrical direc-
tion of the elongated tynes of the median and royal antlers,
which diverge forward in the same vertical plane.

The supra-basilar, or master-antler, being the lowermost
antler, is thrown forwards and outwards, at right angles to
the beam, its upper edge being about 6? inches above the
bur. It is flattened for about 5 inches, and then bifurcates ;
each division again bifurcating, so that there are three forks
and four snags, which are all in nearly the same vertical plane.
The upper branch is directed nearly straight forwards, the
lower inclines downwards. The snags are of unequal length,
the uppermost or fourth being little more than an elongated
tubercle. The tip of the second snag, which is above entire,
stretches out about 17 inches from the beam. The expansion
of the third snag, which is wanting, must have been greater.
The first snag is also wanting, being broken off at its base.?

The second or median antler is thrown off about 6 inches
above the supra-basilar. It is directed straight forwards
from the anterior edge of the beam in a flattened branch,
which, after stretching about 6 or 7 inches, bifurcates, the
upper lobe consisting of a single, straight, flattened, conical,
and slender tyne of great length, which inclines slightly
upwards, an intermediate piece near the tip being wanting ;
while the lower lobe forms a flattened branch, which again
bifurcates at about four inches. The upper tyne is directed
straight forwards at right angles to the beam, and attains

upon the inner side the convexity of the | outer’—[Ep.

section is continued uniformly upwards
a little above the supra-basilar antler,
while upon the outer side it is flattened.
On examining the continuation of the
beam immediately below the terminal
section it is seen to be less flattened than
the royal, the section measuring about
34 by 14 in.’—[Ep.]

1 Extract from Note-book. —‘The
channelling is much more boldly deve-
loped upon the inner side than upon the

2 Extract from Note-book.— The su-
pra-basilar antler, as a whole, is some-
what concave, the concavity being in-
wards. When the horn is laid flat, the
common plane of the supra-basilar is
raised some inches above that common
to the median and royal antlers. The
section of the supra-basilar is much less
flattened than that of the median and
royal, the section forming an oval, convex
outwards and flattened inwards.’—[ Ep. |

474 CERVUS.

an expanse at its tip of about 2} feet. The lower tyne is
broken off at the base, but the fracture indicates that it must
have been nearly equal in size to the other, with a downward
inclination. All three tynes are in the same vertical plane,
forming a three-pronged fork. This plane does not correspond
with that of the divisions of the supra-basilar, which, as
already stated, inclines more outwards.!

The third or royal antler is given off about 10 inches
above the median, like the others, in a flattened branch,
which slopes upwards, and after a stretch of about 5 inches
bifurcates. The lower snag, the tip of which is wanting, is of
great length ; it is directed forwards, with a slight inclination
upwards, so as to be nearly parallel with the uppermost snag
of the median. The second snag of the royal is broken off a
little above the fork; but the remaining portion of its base
indicates that it was simple, and that its point was directed
nearly straight upwards, or in a line with the axis of the
beam. ‘The expanse of the brcken end of the lower snag of
the royal, measured from the posterior edge of the beam,
attains not less than 2 feet 5 inches, and when the terminal
portion was complete it can have been but little short of
3 feet, while the entire length of the beam from the bur to
the terminal fork does not exceed 1 foot 10 inches. The
vertical plane of the royal and its branches, as already stated,
corresponds with that of the median.?

The terminal part of the beam is broken off immediately
above the edge of the royal antler; but a portion of the fork
between them remains. The width of the terminal portion
of the former does not exceed that of the larger snag of
the latter. It is therefore inferred that the beam was
continued up in a single very elongated slender snag;
or there may have been a fork of two snags. Detached
specimens of very elongated, straight, slender snags, com-
pressed at their base, and from the same Forest-bed, exist in
the collections of the Rev. John Gunn and of Mr. Fitch of
Norwich, one in each. These would correspond with the
supposed termination of the beam, in the specimen here
described.

[There is no conclusive indication as to how the beam
terminated, but Mr. Gunn has got in his collection a magni-
ficent fragment of a flattened antler, the channelled surface
and general appearance of which correspond very closely

1 Extract from Note-book.— The me- { the section being 42 by 11 in., and the
dian is very much flattened, the section | successive tynes are Aattened in a similar
of the oval being 33 by 13 in.—[Eb.] manner, aS we go upwards, the lower

* Extract from Note-book.—‘ The | tyne of the royal at the middle being
royal is still flatter than the median, | 3 inches by 1.—[Ep.]

CERVUS SEDGWICKII. 475

with the characters of the large horn. This fragment,
although the terminal portion of the longest tyne is broken
off, measures upwards of 25 inches. The basal portion
stretches along a length of about 11? inches, maintaining
uniformly nearly the same diameter. It is somewhat flattened,
the section near the middle being 24x14. At 10 inches it
gives off a snag or small tyne, the section clearly showing
that it was of no considerable size. The remaining portion of
the tyne, which continues the beam above the fork of the
small tyne, is no less than 143 inches long, although a con-
siderable portion of the point is wanting. When perfect
this tyne could have been little less than 20 inches in length.
It is continued nearly straight upwards, irregularly conical
and but very little compressed.

This fragment is boldly channelled with broad shallow
grooves.—H.F., Oct. 3, 1863.]'

The following are the principal dimensions :

Length of pedicle (a c), 2} in. Diameter of bur (a-a), 3in. Breadth of bur,
Zin. Girth of beam at base of pedicle, 8} in. Ditto ditto above bur, 83 in.
Length of beam from bur to terminal fork of royal antler (a 6), 244 in. Ditto to
inferior edge of supra-basilar (a d), 44 in. Length of beam from bur to fork of
supra-basilar (@ e),6in. Length from upper edge of supra-basilar to upper
ditto of median (e 2), 83 in. Length of beam from median to terminal fork (7 5),
about 14 in. Width of supra-basilar (d e), 24} in. Ditto of ditto at greatest con-
striction, 2} in. Ditto of beam between ditto and median (f g), 8+ in. Widthof
median (% 7), 34in. Width of beam between median and royal (7 /), 33 in.
Width of royal (m 7), 44 in. Breadth from posterior edge of beam to main fork of
supra-basilar (” 0), 10in. Ditto ditto to 2nd upper fork (” p), 123 in. Ditto
ditto to 2nd snag of supra-basilar (x q), 16 in. Ditto ditto to main fork of median
(7 s), 1384 in. Ditto ditto to 2nd fork of ditto (7 ¢),18in. Ditto ditto to 2nd
snag of ditto (7 w), 29 in. Ditto ditto to fork of royal (v w), 1lin. Ditto ditto
to broken point of Ist snag of ditto (v z), 28 in.

The horn above described is evidently of an adult animal,
but not of the maximum size which the species attained.
Besides specimens in Mr. Gunn’s collection, there are in the
Norwich Museum two basal fragments of shed horns, each
including the offset of the supra-basilar antler, presented by
the Rev. Mr. Foulger. They are both of larger size than the
corresponding parts of Mr. Gunn’s specimen. One mea-
sures 7 inches from the bur to the fork at the offset of the
supra-basilar antler, with a girth of 84 inches above the bur;
the other measures 74 inches to the fork, with a girth of
74 inches. In all these instances, one important specific cha-
racter is constant, namely, that the basal portion of the beam
is straight and cylindrical, and without a trace of brow- or
bez-antler. In Mr. Gunn’s collection, besides detached snags,

1 Mr. Gunn believes he has succeeded | ness it is included in the annexed draw-
in fitting this terminal portion of the | ing, Plate xxxvii. « y.—[Eb.]
beam to the antler, and through his kind- }

476 CERVUS.

there are some large fragments of the upper and dilated por-
tion of the beam, which by their curves, constriction and
expansion, and fluted pattern, agree closely with correspond-
ing parts of the large specimen here described. They all
indicate horns of great size and massiveness, and hence a
very large species.

The next point to ascertain is the frontal insertion and
direction of the horns. The well-developed pedicles, straight
butts, and absence of obliquity to the bur, distinguish the
species from all other large forms of Deer, fossil or recent,
in which the horns are sessile as in the Reindeer and Elk,
divergent from the sides cf the skull as in the latter, or re-
clinate with oblique insertion as in the Irish Elk. The butt
of the beam is continued in the same line with the pedicle,
as far as the offset of the lowermost antler. It is therefore
clear that the horns were reclinate as regards the head, and
slightly divergent after the fashion of the Rusa group or
Hippelaphine Deer.

Of the skull, which it so materially modified, in some of
the recent and extinct species of Deer, I know of no spe-
cimens that can with certainty be referred to the form here
described. Mr. Gunn possesses, in his rich collection at Ir-

stead, mutilated skulls destitute of the facial portion of at

least three, if not four, large species of extinct Deer from the
‘Forest-bed, but the large horn cannot be referred with
confidence to any one of the forms in particular.'’ I have

1 Memoranda of Craniaof Cervusin Mr.| as big as the C. Hippelaphus. The

Gunn's collection, made by Dr. Falconer
for Mr. Gunn, Oct. 3,1863.— No. 20.—
The most perfect is a cranial fragment
comprising both frontals, part of both
orbits, the left superciliary foramen,
both horned pedicles, the anterior portion
of both parietals, and the greater part of
the cerebral cavity. The open condition
of the parieto-frontal suture and of the
sagittal suture proves the individual to
have been young. The basal portion of
the horn is present on both sides, but
most perfect upon the right. The horned
pedicles are very short and divergent.
There is a fracture in front at the base
of the right horn immediately above the
bur, which would appear to indicate that
a brow-antler was given off immediately
above the bur. ‘The insertion of the
beam is very oblique, the bur exhibiting
an inclination somewhat like that of the
Trish Elk. There are two shallow de-
pressions on either side of the sagittal
suture (one on each side), bounded
outwards by the secondary superciliary
foramina. The species must have been

horned pedicles are very short, and the
iuterval between the orbits and the pe-
dicle is inconsiderable. There is no
lateral fossa between the bur and the
orbits. The dimensions are :—Width of
frontal above the orbits, 7° in. Vertical
diameter of the bur (right side), 3°20 in.
Interval between the two horns, mea-
sured at bur, 4°30 in. Girth of right
pedicle, 8+ in.

Iam unable to refer this specimen to
any in the British Museum or in France.
It appears to differ from the Deer of
Auvergne and Chartres.

No. 21.—Another specimen in Mr.
Gunn’s collection is very much like the
last, but with a larger portion cf the horn
preserved on both sides. The specimen is
about the size of C. elaphus. The horns
are much less divergent than in the pre-
vious specimen. Each shows a fracture
in front, a little above the bur, indicat-
ing the offset of one, or two, brow-antlers.
It is impossible to say which. The horn
is terete and suddenly reclinate back-
wards. This specimen may correspond

—--

VARIOUS FOSSIL SPECIES. 477
seen also, by the aid of the Rev. 8. W. King, a skull in the
possession of Mr. Sandford, junior, of Cromer, procured from
the ‘ Forest-bed’ of Sidestrand. It consists of the greater
part of the cerebral part of the skull of a very large species
of Cervus, which, from the form and insertion of the pedicles,
I think it highly probable belonged to the same species as
the Bacton horn.'

None of the established sub-genera groups of Cervus,
founded conventionally on the characters yielded by horns,
will admit the fossil species.’

[During the last two years of his life, Dr. Falconer devoted much
attention to the fossil remains of the genus Cervus, and his Note-books
contain numerous references to specimens which he had examined in
the various paleontological collections of England and the Continent.
Although in most instances the descriptions are too fragmentary to
merit publication, it may be interesting to refer to some of the identi-

fications to which the author was led.

follows :—

The most important are as

1. September 20, 1858.—In Col. Wood’s collection at Stouthall from

the Gower Caves.
a. Cervus eurycerus.
antlers.
b. Cervus Tarandus.

(See p. 498.)

with some of the species figured by
Croizet, but I am unable to make it out.
It deserves to be figured as well as the
previous one.

No. 22.—A third specimen consists of
the frontals, cerebral cavity, and both
horned pedicles, but both horns shed, so
that the distinctive characters are want-
ing. The horned pedicles are short and
slightly divergent, and the frontal plane
above the nasal suture shows a promi-
nent mesial ridge, bounded on either
side, first by a depression and then by a
bulge outwards for the superciliary
artery. Laterally, below the horned
pedicle, there is a deep depression.

The form and characters remind me
strongly of some of the St. Prest speci-
mens, which Lartet and I saw at Char-
tres.

The dimensions are :—Width of brow
between orbits and pedicle, 6} in. Girth
of horned pedicle, 8} in. This specimen
ought also to be figured. It appears to
have come out of the iron pan of the
Forest-bed.

No. 23 is a specimen corresponding
with the three last, but terribly wea-
thered and rolled.

On the right side the pedicle bears a
horn, which differs from the others in

Molars, but no determinable fragments of

Crania from Paviland and Spritsail-Tor.

the collection by emitting a brow-antler,
which is directed inwards, but is so rolled
that nothing can be made out of it, ex-
cept that the species differs from the
three others above referred to.

1 Extract from Note-book.— British
Museum, Aug. 5, 1863.—No. 82,497 is
a magnificent skull of a Deer, dark and
heavy, and evidently a dredged specimen.
It shows the whole of the occiput, con-
dyles, and basal portion, and also part of
the frontal and of the two orbits. Itis of
the size of the Irish Elk, but is of a dif-
ferent species. The form of the occiput
and the offset of the horns are different,
and the distance between the base of the
horn and the orbit is much longer. Can
this be the species to which Gunn’s horn
belongs ?’—{ Ep.]

? From entries in Dr. Falconer’s Note-
books it appears that after careful com-
parison with the remains of deer in the
British Museum, and in the Museums of
Paris, Le Puy, and Chartres, he was
unable to identify the fossil with any
species hitherto described. At one timo
he was inclined to think that it might
belong to Cervus martialis (Gervais),
but this opinion he ultimately gave up,
and, as Mr. Gunn informs me, C. marti-
alis had a brow-antler.—[Ep. |

478 CERVUS.

c. Cervus Guettardi (Desmarest). ‘A complete series of eighteen or
nineteen specimens of different ages, all limited to basal portion of
antler. They agree exactly with the specimens originally figured and
described by Guettard, and reproduced by Cuvier in the “ Ossemens
Foss.” Cerfs, Pl. VI. figs. 14 to 17. They differ entirely from figs. 10,
11, and 12 of the same Plate, which Cuvier conjectured might belong
to a different age of the same species. As this is the form of antler that
was originally described by Guettard, it is but right that the specific
name of Cervus Guettardi should be restricted to it.’

d. Cervus priscus ? Two antlers which agree exactly in form with
figs. 11 and 12 of Plate VII. Cer/s, of the ‘Ossemens Foss.’ from Etampes,
found along with C. Guettardi.

e. Cervus Bucklandi? One specimen comprising the basal part of
the horn, bur, and offset of bez-antler.

f. Cervus Elaphus. Antlers.

g. Cervus Capreolus. Several small specimens, some of them bearing
knife marks.

h. Indications of two other species not identified.

2. April, 1859.—In Signor Ceselli’s Museum at Rome, examined
numerous antlers of Cervidw. One fine specimen appears to belong to
C. intermedius, ‘ but saw nothing in the least degree approaching Rein-
deer.’

3. May 14, 1859.—‘ In the Museum at Bologna is a very interesting
skull of Irish Elk from the Val di Chiana, comprising the cranial portion
with part of both horns, showing the cylindrical beam, the stumps of
brow- and median antlers, and the palmate expansion on the left side.’

4. May 27, 1859.—‘ Saw in Massolongo’s collection at Verona the
lower half of the horn, from bur to commencement of palmation, of the
Italian form of the Irish Elk, exactly like the Val di Chiana specimen
in the Bologna Museum, also a portion of the cranium noted in the
Catalogue as being “ Corno giganteo e pezzo di cranio di Cervo fossile
raccolto nei terreni del Veronese.” ’

5.° September 27, 1862.—In Mr. Grantham’s collection at Erith, a
typical specimen of a shed antler of Cervus Elaphus, and a fine speci-
men of an antler nearly to fit Strongyloceros spelaus.

6. September 30, 1862.—In Dr. Spurrell’s collection at Belvedere,
numerous specimens of Cervus Llaphus, one nearly big enough to pass
muster for Strongyloceros speleus: a very doubtful fragment of Rein-
deer antler (probably C. Elaphus) ; but no other indication of Rein-deer,
or of any other species of Cervus.

7. July 20, 1863.—In Mr. Wyatt’s collection at Bedford, shed antlers
of Cervus Elaphus, from Summerhouse Hill; and antlers of Cervus
Tarandus, from Howard’s Field, Bedford, and from the Gravel, Bleston.

8. August 13, 1863.—In Mr. Prestwich’s collection from railway-
cuttings at Bedford, remains of Cervus eurycerus, or Irish Elk ; Cervus
Elaphus; and Cervus Tarandus, or Reindeer.

9. September 8, 1863.—In Brown’s Clacton collection in the British
Museum is a very extensive series of Deer horns, nearly all belonging to
one species. They are all terete, with a single brow-antler given off
very low, as in the Val d’Arno Axis, but a little lower and pointing
more forwards above the brow-antler. There is generally a long reach
of beam with no branch. How the beam terminates is not shown. In

VARIOUS FOSSIL SPECIES. 479

size it is like Mr. King’s Axis from the Crag and Forest-bed, but it differs
in the brow-antler being given off lower, and in not having the same
pronounced double curve. ‘The species is evidently distinct (Cervus
Clactonianus).

10. September 12, 1863.—‘ Jardin des Plantes, Paris. Examined,
with Lartet, Cuvier’s horn of Cervus Somonensis (Oss. Foss. tom. iv.
Pl. VI. fig. 19). A small part of the frontal is present. There is
absolutely no pedicle whatever, and the horn base is very close to the
orbit. The summit is united with plaster and is evidently a misfit,
probably not even of the same species, or certainly turned the wrong
way. Very puzzling to determine the species.’

11. September 15 to 21, 1863.—Dr. F. visited the Museums of Le
Puy and Chartres, in conjunction with M. Lartet, and took numerous
notes and drawings of the fossil remains of Deer found in them, and
particularly of Cervus Solilhacus and C. Polignacus, but he failed to
discover any specimens resembling the large antler in Mr. Gunn's
collection.

12. September, 1863.—Provisional list of Norwich-Crag and Forest-
bed Cervide.:

a. Mr. Gunn’s large Cervus Sedgwickii. (See aniea, p. 472.)

6. Mr. Gunn’s large Strongyloceros (sic). The specimen is of left
side, and consists of the basal portion of a huge horn that had been
shed. The brow-antler is given off about 2 inches above the bur, and
is curved abruptly downwards and outwards like a huge hook; it is
perfectly terete, and the portion remaining shows no appearance of sub-
division. It is very boldly channelled on the convex outer side;
smooth inwards. The beam above the bur is not quite terete, but
oval, with a ridge behind, opposite the brow-antler. The beam then
contracts, and becomes nearly cylindrical, and then expands, giving off
from the anterior outer side a large antler, at about 6-7 inches above
the bur and 44 inches (lower edge) above upper side of brow-antler.
The beam is then somewhat flattened in a direction corresponding with
that of the brow-antler. Only the section of the base of the median
antler seen. A ridge descends from lower edge of median antler, outer
side, to the ridge or tuberosity opposite the brow-antler.

Length of specimen, 13:5 in. Girth above bur, 10°5 in. Girth of brow-antler
at base, 7-5 in. Ditto at tip, 6° in. Length of brow-antler fragment, about 5: in.
Vertical length of section of median, 2°5 in. Transverse diameter of ditto, 2-0 in.
Girth above brow-antler, 8:25in. Ditto above median, 9°5 in.

The brow-antler is given off much higher than I have ever seen it in
the Irish Elk; the beam less cylindrical than in the latter, and more
erect, without the elegant long reclinate reach in the latter. The low
offset of the median antler is also very different. It appears to indicate
a huge Deer as large as Irish Elk, but quite distinct.

There is nothing like it in the British Museum.

e. Mr. Gunn’s large Cervus Polignacus—‘a shed-antler throwing off
a brow-antler, which nearly continues the beam downwards; the beam
deeply channelled, round below and flattened a little above; evidently
the same species as the Val d’Arno figures.’

d. Cervus eurycerus, from Miss Anna Gurney, in Oxford Museum
(1858).

e. The Rev. S. W. King’s specimen of Rusa. This isa double curved

4380 CERVUS.

horn. It not only curves backwards, but also (with a slight tendency
to twist) outwards. The specimen includes part of the frontal with the
cerebral cavity, but no part of the orbit. The bur is very salient and
pronounced as in C. Elaphus. The beam at the base is nearly level,
with a tendency to keel between the bur- and the brow-antler, and
deeply channelled. The beam above the brow-antler is nearly terete,
and there is a well channelled offset of the supra-basilar about 1°3 inch
above the bur. The dimensions are: Ant.-post. diam. of beam at bur,
15in. Transverse ditto, 1-4in. Transverse diam. above bur, 1:5 in.
Ant. post. ditto, 1:25in. Height to top of fork,2°5in. In the Museum
of the Jardin des Plantes there is an antler (No. 6,201) from the Val
@Arno, labelled C. Arvernensis, which is twisted exactly like Mr.
King’s.

f. British Museum specimen of C. Dama (No. 27,876 ?).

g- ” ” Axis.

h. n i an Elaphoid form of Cervus, but not

C. Elaphus (No. 33,471 ?),—Ep. |

© citi Fie My

4
f
:
We
\“
:
i
4
Is TN
i

%
{od ita
;

uit |
he

DESCRIPTION OF PLATE XXXVII.

a ee

Cervus (HUCLADOCEROS) SEDGWICKII.

Fig. 1, Fossil Antler found at Bacton, and now in the collection of the
Rev. John Gunn, of Irstead, described at page 472. The figure
is between one-seventh and one-eighth of the natural size, and

an ee

——

has been drawn by Mr. Dinkel from sketches furnished to Dr.
Falconer by Mr. Gunn. The letters refer to the description in
the text: a, bur; a ¢, pedicle; a 6, beam; d e, supra-basi-
lar; h 7, median; 2m, royal; « y, terminal portion of beam.

——

Figs. 2 and 3. Represent in outline the two basal fragments of shed
i horns presented to the Norwich Museum by the Rev. Mr.
Foulger, and referred to by Dr. Falconer at page 475. The
figures have been copied from rough outline sketches found
among Dr. Falconer’s papers, and are one-fourth of the natural

size.

VOL, Il.

SETS TeS eG

oof “TEPDIMSpeg (sorsoopeponiy) snateo
¢

a=

BOS ETRUSCUS, 481 ,

XVI. NOTE OF AN UNDESCRIBED SPECIES OF
BOS IN THE FLORENCE MUSEUM (BOS
ETRUSCUS?)!

Fiorence, May 20, 1859.—I examined in the collection in the Grand
Ducal Museum the entire skull of a Bovine ruminant, to which the
lower jaw is attached. It is a good deal crushed laterally, and in the
frontal portion between the horns it bears a loose ticket, on which is
inscribed ‘284, Bos bombifrons’ of Nesti; but I suspect that there is
an error or misplacement in the ticket, as there is no appearance of any
protuberance of the brow. ‘The frontal plane extends a considerable
way behind the offset of the horns, showing the two temporal fossae,
which at their termination approach close together, with an interval of
only 24 inches. This posterior extension of the frontal plane is much .
greater than what is seen in Bos priscus, and differs entirely from the
overhanging crest of Bos primigenius. The horn-cores are cylindro-
conical and slender, and of considerable length ; they stretch backwards
and outwards with a gentle curve, nearly in the same plane as that of
the brow, their convexity being outwards, something like in domestic
cattle. They are 224 inches apart at their tips, with well defined
pedicles at their base. The dimensions are as follow :—

Length of right horn-core measured along the outer curve, about 203 inches.
Girth at base, 9} in. Length of skull from occipital erest to tips of incisive bone,
203 in.

The form of the brow, occipital crest, and temporal fossae, the
direction of the horns, and the size of the skull appear to distinguish
this animal very remarkably from Bos primigenius and from Bos
priscus. It is also of very much smaller size, and I suspect constitutes
a distinct undescribed species, for which the designation Bos Htruscus
would be appropriate. The name Bos bombifrons must be a misnomer,
and was probably applied by Nesti to Bos priscus.

In the same collection there are four cranial fragments of Bos priscus,
of very large size. They all consist of the frontal tablet, behind the
orbits, and each of them bears the horn-cores having the usual direction
seen in this species, ze. stretching out horizontally, with the tips
slightly curved forwards. In the largest specimen the constriction of
the brow behind the orbits measures 114 inches, and the interval
between the tips of the horns 36 inches. The collection also contains
three cranial fragments of Bos primigenius, which, as well as those of
Bos priscus, are all labelled as coming from the Val di Chiana. They
are very fresh looking, and are evidently from a more recent deposit
than the old Val d’Arno specimens. The skull marked ‘Bos bombifrons’
is well fossilized, and all the molar teeth are out.

1 Extracted from Dr. Falconer’s Note-book.—[Ep.]
VOL. II. II

482 CROCODILUS CATAPHRACTUS

XVII... NOTE UPON CRANIA OF CROCODILUS CA-
TAPHRACTUS, AND CROC. MARGINATUS,
IN THE BELFAST MUSEUM!

THE existing Crocodiles are still but imperfectly defined, and
there is little agreement among systematic authors regarding
the number and characters of the species. This remark ap-
plies with especial force to the Crocodiles of the Nile and of
the Ganges. Geoffroy assigns five species of true Crocodile
to the Nile, all of which are considered by Cuvier as varieties
of a single species, C. vulgaris. Dumeril and Bibron, in their
‘Erpétologie,’ published in 1836, follow the view taken by
Cuvier, although it would appear from a verbal communication
of M. Bibron, that their opinions have been considerably
altered since Mr. J. E. Gray, in his ‘ Synoptical Catalogue,’
published in 1844, admits two species, C. vulgaris and C.
marginatus. In like manner, the Crocodiles proper of the
Ganges were restricted to a single species by Cuvier, C. bi-
porcatus, in which view, also, he is followed by Dumeril and
Bibron, although C. palustris of Lesson is inserted with
doubt as a variety of C. vulgaris in their systematic work ;
but it would appear from the labels of the specimen in the
Paris Museum, that they now recognize it as a distinct species.
On the other hand, Mr. Gray gives three species to the Ganges,
viz., C. biporcatus, C. palustris, and C. bombifrons. It is of
interest therefore to record the existence of any specimens
bearing upon the disputed or ill-determined species; and

1 This memoir was communicated by
Dr. Falconer to the British Association at |
Southampton in 1846, and is reprinted |
from the Ann. and Mag. Nat. Hist for
Dee. 1846, vol. xvii. p. 361, where it ap-
peared with the following note :—‘ Com-
municated by Mr. W. Thompson, Presi-
dent of the Society to which the Museum |
belongs, with the following remarks :— |
The crania which form the subject of the |
. present notice were presented to the Na- |
tural History and Philosophical Society |
of Belfast by Dr. McCormace, of thattown. |
They were taken in the waters of the |
Sierra Leone river or its tributaries, and
given to that gentleman by his brother,
Mr. John McCormae, of Freetown, Sierra

Leone. My friend Dr. Falconer, on
visiting the Museum with me early in
1845, called my attention to the rarity
of these crania. On leaving home for
London, a few months afterwards, I took
the specimens with me for the purpose —
of comparison with others in the collec-
tion there, and the result is set forth in
the paper. To the kindness of Mr.
Grattan (Treasurer tothe Society already
named) we are indebted for drawings of
the specimens made by means of a
camera-lucida. These, for the sake of
comparison with the figures in Cuyier’s
““Ossemens Fossiles,” have been drawn
of the same size.’ —[ Ep. |

EE ————EE————EEEo=

AND CROC. MARGINATUS. 483

having observed the crania of two rare Crocodiles in the Mu-
seum at Belfast, the following notes regarding them have
been drawn up by me at the request of Mr. W. Thompson.

I. Crocodilus cataphractus, Cuv., Oss. Fossiles, tom. v. p. 58,
Pl. V. figs. land 2; Dum. and Bibron, Erpét., tom. iii. p. 126;
C. leptorhynchus, Bennett, Proc. Zool. Soc. 1835, p. 129;
Mecistops Bennettii and M. cataphractus, Gray’s Catalog.,
pp: 57 and 58.

This species was founded by Cuvier upon an imperfect
specimen of unknown origin in the Museum of the London

College of Surgeons. It was briefly described by Bennett,

first as a distinct species from Fernando Po, in 1835, and af-
terwards as a variety of C. cataphractus, in the ‘ Zoological
Proceedings’ of 1836. Mr. Gray has erected it into a separate
genus under the name of Mecistops, in which he includes along
with it the C. Jowrnei of Bory de Saint-Vincent, and C. (Ga-
vialis) Schlegel of Miller. So far as is known to us, no
representations have yet been given of the cranium divested
of its integuments. Plate XX XVIII. figs. 1, 2, and 3, repre-
sent the Belfast specimen, viewed from the top, base, and side
of the skull. It is evidently identical with Gray’s Mecistops
Bennettii; the head of the stuffed specimen of this nominal
species in the British Museum collection agreeing with it
exactly in form, and very nearly in size. The muzzle is more
attenuated and narrower than in C. acutus, but less so than
in C. Schlegelii, which constitutes the passage from the true
Crocodiles into the Gavials. The cranial tablet is not so wide
asin the Gavial, C. Schlegelii, and the crotaphite foramina
are proportionally smaller. The muzzle does not contract
abruptly in front of the orbits, but is gradually attenuated
from the back part of the cranium forwards. 'The extreme
width at the condyles of the lower jaw is 7 inches, behind the
orbits 43 inches, and in a line with their anterior border 34
inches. At the seventeenth or last tooth of the upper jaw
the width is 52 inches, and 1# in. between the eleventh and
twelfth teeth; there is an expansion to 2 inches opposite the
ninth tooth, which is the largest in the head: thence the
beak contracts gradually to the space between the fourth and
fifth teeth, where the width is only 1 inch; at the extremity
of the muzzle, between the second and third teeth, it expands
to 1? inch. The margins, when viewed in plan, are there-
fore more undulated and less cylindrical than in the Gavial
or CO. Schelegeli, and there is less dilatation on the point of

the beak.

The orbits are much larger than the crotaphite foramina,
which are separated only by a narrow interval; while in the
Gavial they are large and wide apart. The lachrymals form

I1r2

484 CROCODILUS CATAPHRACTUS

narrow slips of bone, which descend upon the nasals a con-
siderable way below the anterior margin of the pre-frontals.
The nasal bones are extremely narrow and attenuated, but,
as in the true Crocodiles, they descend between the maxil-
laries so as to project into a niche between the intermaxillary -
bones. The same holds good in C. Schlegelii; whereas in the
Gavial the nasals terminate a short way in front of the orbits,
and do not enter into the formation of the anterior portion
of the beak. This character is a good diagnostic mark be-
tween the Crocodiles proper and the Gavials; separating C.
Schlegelii from the latter subgenus under which Muller has
ranged it. The nasal opening is smooth, oval in form, and
of moderate size. There are seventeen teeth in the upper
jaw, and fifteen in the lower; the largest teeth in the upper
are the third and ninth ; in the lower, the first, fourth, tenth,
and eleventh. The dimensions are subjoined. _

II. Crocodilus marginatus (2), Geoff. Croc. d’EHgypte, 165,
and Gray’s Catal. Brit. Mus., p. 61; C. vulgaris, var. C., Dumer.
et Bibr. Erpétolog., iii. p. 110; C. vulgaris, Cuv., Annal. du
Mus., tom. x. 40.

The Belfast specimen is doubtfully referred to this species,
there not being sufficient materials in the London museums
to admit of a satisfactory determination. Neither the College
of Surgeons’ collection nor the British Museum is possessed
of an adult cranium of the common Crocodile of the Nile, C.
vulgaris, or of C. marginatus, although there are numerous
stuffed specimens attributed to both species in the British
Museum collection. The comparison of the Belfast specimen
has in consequence been limited to the reduced figure of the
skull of C. vulgaris in the ‘ Ossemens Fossiles.’

The cranium is 19 inches long, and must have belonged to
an adult animal. The principal distinctive character assigned
to O. marginatus, both by Geoffroy and by Dumeril and
Bibron, in addition to the form of the nuchal and dorsal
scutes, is that the borders of the cranial tablet are raised,
while in CO. vulgaris the frontal area is perfectly flat. In the
Belfast cranium these lateral margins are also considerably
elevated, and the following points of difference from C. vul-
garis are besides observable. The facial portion of the head
is less elongated in proportion to the cranial, and more obtuse
than in C. vulgaris; the interval between the orbits is greater ;
the crotaphite foramina are relatively larger ; the lachrymals
are narrower, and descend further upon the nasals ; the muzzle
is considerably blunter, and the niche for the reception of the
fourth tooth of the lower jaw is larger, causing a greater
amount of constriction. The general outline of the muzzle,
instead of being acute and subcuneiform, is obtuse and oblong,

DESCRIPTION OF PLATE XXXVIIL.

CROCODILUS CATAPHRACTUS AND CROCODILUS MARGINATUS.

The figures in this Plate have been reproduced from those
which appeared originally in the Ann. and Mag. of Nat.
Hist. for December, 1846.

Figs. 1, 2, and 3. Represent the cranium of Crocodilus cataphractus in
the Belfast Museum, rather more than one-fifth of the natural
size. Fig. 1, upper surface; fig. 2, palate surface; fig. 3, side

view.

Figs. 4, 5, and 6. Represent the cranium of Crocodilus marginatus in

the Belfast Museum, rather less than one-fifth of the natural

_size. Fig. 4, upper surface; fig. 2, palate surface; fig. 6, side-
view.

VOL. I.

“STASUIBIEUL SNIIPOOOID 9°C'F -emovaydeyeo enyrpooor) Aa

uly 8TLOG TLOSLEIBT]

STL TRS
rE

ae *AQ QvT TTT A

AND CROC. MARGINATUS.

485

somewhat resembling the form of C. palustris of the Ganges.
There is also a marked constriction behind the twelfth tooth,

considerably greater than in C. vulgaris.

The largest teeth

are the third, the fourth, and the tenth, the last “being the

largest of all.
C. vulgaris.

The nasal aperture is more circular than in
There is no lower jaw to the Belfast specimen.

Plate XX XVIII. figs. 4, 5, and 6, represent the cranium,
viewed from the top, palate, and side, as in C. cataphractus.'
The dimensions of the cranium are as follow :—

Dimensions

C, cata-
phractus

C. margi-
natus,

occipital ridge

to condyle of the upper jaw

Length of orbit .
Width of orbit .
Interval between orbits

Width of the muzzle at the last tooth

Width at the tenth tooth
Width at the ninth tooth

Width at contraction behind the fourth tooth

Greatest contraction behind fifth tooth
Dilatation of the point of the muzzle .
Length of the nasal aperture
Width of the nasal aperture

Length of intermaxillaries on the palate

Length of maxillaries on the palate

Antero-posterior diameter of palatine foramen
‘Transverse diameter of palatine foramen

Length of cranium from the point of the muzzle to the
Length of cranium from the point of the muzzle measured
Extreme width of cranium at the condyles ;

Length from occipital ridge to base of nasals
Length from the point of the muzzle to base of nasals .

Antero-posterior diameter of crotaphite foramen | ; : 1:
Transverse diameter of crotaphite foramen

Width of the muzzle at base of the nasals .
Width at contraction behind the twelfth tooth

1 Part of the above description, taken
from Dr. Falconer’s Note-books, was
published in vol. i. (p. 356), before I
was aware of the aera of the above
memoir. In reference to three species
of Crocodiles of the Ganges mentioned
at p. 356 of vol. i., C. bombifrons, ac-
cording to Dr. Falconer, occurs in the

northern branches of the Ganges, 1,000

In. In.
15°5 16:

=
~I
=
Ne}

ple SoS

i
DRAB HHS HH
AAI

Comoe O Or

wo

3
6°

Phorm | |

Ral ii ollo's)

miles from Caleutta; C. biporcatus ap-
pears to be confined to the estuary ; and
C. palustris to range from the estuary
to the central parts of Bengal. The
Gayialis found along with C. bombifrons
in the North, and descends to the region
of C. biporcatus in the estuary. (See
Sir C. Lyell’s ‘Principles of Geology,’
10th ed. 1867, p. 473.)—[Ep.]

486 OSSIFEROUS CAVE OF BRIXHAM.

XVII. ON THE OSSIFEROUS CAVE OF BRIXHAM,
NEAR TORQUAY.

[Tur great and sudden revolution in modern opinion, res-
pecting the probable existence at a former period of man
and many extinct mammalia, has been universally attributed
to the results of the explorations of the Brixham Cave. The
subjoined documents show clearly in whom this revolution
mainly originated. On May 10th, 1858, the annexed letter,
respecting “the then newly discovered cave of Brixham, was
addressed by Dr. Falconer to the Council of the Geological
Society, and in consequence of a recommendatory resolution
passed by the Council, the Royal Society, on May 15th, gave
a grant of 100/. towards the exploration of the cave in the
manner suggested by Dr. Falconer. Miss Burdett Coutts
contributed 501. towards the same object. At Dr. Falconer’s
suggestion, a committee was appointed to carry the design
into effect. The committee consisted of Professor Ramsay,
Mr. Prestwich, Sir Charles Lyell, Professor Owen, Mr.
Beckles, the Rev. R. Everest, and Mr. Godwin-Austen. Dr.
Falconer was entrusted with laying down the plan and
giving the instructions upon which the exploration was to
be conducted, and the works were carried on under the
immediate superintendence of Mr. Pengelly. The fossil
remains were identified by Dr. Falconer. On the 9th Sep-
tember, 1858, a report on the progress of the operations,
drawn up by Professor Ramsay, Mr. Pengelly, and Dr.
Falconer, was submitted to the general committee, and by
them was forwarded to the Royal Society, which, from the
importance of the results already elicited, voted an additional
sum of 100/. to prosecute the inquiry. Almost immediately
afterwards, Dr. Falconer was compelled to proceed to the
south of Europe, on account of his health; but the ex-
plorations were continued by Mr. Pengelly. At a meeting of
the general committee, held after his death, Mr. Prestwich
was requested to draw up a final report embodying the chief
results of the exploration. The first report is given here in
full, and embodies the important facts determined by Dr.
Falconer of the indiscriminate mixture in the Brixham
Cave of human industrial remains with bones of Rh
Hyena, and other extinct animals.—Eb.]

OSSIFEROUS CAVE OF BRIXHAM. 487

IT. Copy oF LETTER FROM Dr. FALCONER TO THE SECRETARY
OF THE GEOLOGICAL SOCIETY.

£31, Sackville Street, W., 10th May, 1858.
‘To the Secretary of the Geological Society.

‘ Sir,—I solicit the favour of your bringing the subject of
this letter under the consideration of the Council.

‘Tt is well known that a great and popular impulse was
given to Geology in this country by the well-directed and
eminently successful researches of the late Dr. Buckland, on
the Ossiferous Caves of England. After the publication of
the “ Reliquiz Diluviane,” the subject in its general bearing
was regarded as pretty well exhausted, so far at least as con-
cerned the uniformyit of character in the fossil remains
found in the caverns, and their being referable to a single
geological period. Since 1823 the interest in the subject
has gradually fallen off, and it is probably not overstating
the fact to say, that there is hardly a general geological
question in which the majority of geologists in this country
take less interest at present than in what relates to the
Ossiferous Caves. The subject has not advanced pari passu
with the progress in the investigation of the Upper-Pliocene
and Post-Pliocene deposits.

‘Tt is understood that Dr. Buckland, before the close of
his valuable life, had intended to bring out a second edition
of the “ Reliquiz Diluviane,” in which some of the question-
able views, so earnestly advocated in the original work,
would have been greatly modified. But unfortunately the
design remained unaccomplished, and the popular opinions
in the cave-districts, where collections were amassed, have
been mainly regulated by the doctrines embodied in the
work as published in 1823.

‘ The consequences have been thus:—The Tunnel Caves like
“ Kirkdale,” which were the haunts of predaceous Carnivora,
and the Fissure Caves like ‘ Oreston,” that were filled from
above, have been popularly regarded as containing the débris
of the same mammalian fauna, and as having been overlaid
with their ochreous loam by the same common agency at the
same period. The contents of the different caverns were
thus considered as being in a great measure duplicates of
one another, and the exceptional presence of certain forms
in one case, and their absence in another, were regarded more
in the light of local accidents than as significant of any
general source of difference. Hence it followed that more
attention was paid to the extrication of the bones, and to

488 OSSIFEROUS CAVE OF BRIXHAM.

securing good specimens, than to a record of their relative
association and the order of succession in which they
occurred. The remains have been, in some instances, hud-
dled together in provincial collections—the contents of five
or six distinct caves, without a discriminative mark to indi-
cate out of which particular cavern they came. Another
consequence has been, that being regarded in the light of
duplicates, the contents of some of the most important and
classical English caverns have been dispersed piecemeal ;
and so far as regards them the evil is now beyond remedy.

‘ My object in this communication is to bring to the notice
of the Council an interesting case of a newly-discovered and
intact cavern, where the mischief done elsewhere may be
partly retrieved, and probably much effected by combined
action, well directed.

‘Within the last month a new and undisturbed cave has
been discovered on Windmill Hill, overhanging Brixham, in
the same tract of limestone in which the caverns of Kent’s
Hole, Anstey’s Cove, Chudleigh, and Berry-Head are found.
I annex a brief notice of the discovery, cut out of the Exeter
** Western Times ” of the 10th ultimo. Mr. Everest and myself
went to see it on the 17th ultimo. Windmill Hill rises im-
mediately above Bolton Street in Brixham. The limestone
strata crop out on the NE. side, where they are very cavern-
ous. A vertical channel, running up the hill, marks the base
of a fault, the walls being separated by a seam of about two
inches of yellowish loam. Near its base, in quarrying out
a foundation for cottages, a concealed cavern was disclosed,
blocked up by loam, rubbish, and breccia, on removing which
an open cavity was seen, low and narrow at the mouth, but
expanding inwards, and presenting the usual characters of
the Plymouth Limestone caves. Water percolates from above
by a copious drift, and the vault and floor are irregularly
coated with stalagmite. On a shelving stalagmite terrace
in the interior we saw from a distance a pair of large cer-
vine horns, horizontally embedded in the stalagmite, and I
distinguished bones of Hyzena, Bear, Bos, Deer, and Horse,
which had been picked out of the breccia. The interior of
the cavern is blocked up by stalagmitic deposit, but from
the hollow sound yielded on percussion it would appear that
there are under-vaultings, as in Kent’s Hole. In another
direction ‘the stalagmitic flooring descends suddenly in a
chasm of undetermined depth. There are two external open-
ings nearly at the same level, a considerable distance apart,
which would seem to communicate with the same interior
hollow ; and it is probable that like Kent’s Hole the Brixham
Cave is of great extent, with irregular ramifications. As in

CO

OSSIFEROUS CAVE OF BRIXHAM. 489

other similar cases, the principal deposit of fossil bones may
be looked for under the stalagmitic floor not yet touched.

‘The strip of hill, in which the openings of the cave are,
belongs to Mr. Philip Dyer, of Brixham, of moderate circum-
stances in the industrial walk of life, who appeared to be
quite open to an offer from any quarter to lease the cavern
for exploration. Taking into account the vast richness of
Kent’s Hole in fossil remains, the dispersion of Mr. McHnery’s
collections, and the grievous fate of the MS. labours of
about twenty years of his life, it is submitted to the Council
whether there is not a prospect of equal wealth in this
promising and adjoining cave of Brixham, and whether the
case is not one deserving of a combined effort among Geologists
to organize operations for having it satisfactorily explored,
before mischief is done by untutored zeal and desultory work.

‘The importance of following up a case of this description
has been forced upon my attention by some of the results of
an examination of the cave-bone collections both in England
and abroad, in connection with the investigation of the dis-
tribution of the extinct Proboscidea in the Huropean upper
tertiary deposits. I have during the last twelve months been
more or less occupied with the conditions under which
Hlephant remains occur in the caves; and having lately
returned from a tour, in company with my friend the Rev.
R. Everest, during which we have made a reconnaissance
survey of the caves in, or cave collections from, the neigh-
bourhood of Bristol, the Mendips, Devonshire, South Wales,
Kirkdale, and Cefn, some of the results have appeared of
sufficient interest to justify my trespassing on the attention
of the Council with this communication.

“Of these I may mention the following :—

‘1st. The detection in considerable abundance in certain
of the caves of the remains of a species of Rhinoceros,
equally distinct from the tichorhine species of Siberia, and of
the Glacial period generally, and from the leptorhine Rhi-
noceros of Cuvier, of the Sub-Apennines, Hlephant-bed, and
Lacustrine deposits of the Norfolk coast.’ I have seen nearly
the entire series of the upper and lower teeth im situ in the
jaws, and from one of the caves a considerable portion of the
skeleton associated with teeth and cranial fragments. The
characters distinctive of this form from the species above
referred to are so pronounced and so constant, and the
materials so abundant, that I have no doubts on the subject?

1 Dr. Falconer had not at this time | tion of the remains I find that the spe-
distinguished between the true R. lep- | cies (Rhinoceros priscus) is equally dis-
torhinus (R. megarhinus) and R. Etrus- | tinct from the existing African species

cus. (See antea, pp. 314, 348.)—[Ep.] | and from Ph. leptorhinus and Rh. tichor-
* Having gone into a detailed examina- | hinws.—H. F., Oct. 20, 1858.

.

490 OSSIFEROUS CAVE OF BRIXHAM.

I have designated the species provisionally Rhinoceros
priscus.! The interest of the case is enhanced by its pre-
sumable relations to some important late investigations of
Monsieur Lartet, to which I shall refer in the sequel.

‘9nd. Abundant evidence in all the cave districts of two
extinct species of Elephant, viz. Elephas primigenius (Mam-
moth), of the Glacial period, and EH. antiquus, of the Sub-
Apennine period (Norwich Crag and the Astesan). The for-
mer commonly associated in the English caves with the
tichorhine Rhinoceros, the latter with Rhinoceros priscus. I
have not observed among the cave bones any indication of
remains of H. (Loxodon) meridionalis, nor undoubted remains
of Elephas (Loxodon) priscus.

‘3rd. In one of the caves where the evidence is tolerably
conclusive that the bones were washed into a fissure about
the same time, the following undoubted association was seen :

Elephas (Huelephas) antiquus,

Hippopotamus major,

Rhinoceros priscus,
without the admixture, so far as the collection went, of other
species of the same genera.

‘4th. In other caves, Hlephas primigenius and the tichorhine
Rhinoceros were observed, without the admixture of Hlephas
antiquus and Rhinoceros priscus.

‘5th. In one of the caverns the most important part of the
skeleton of one Hlephas antiquus was found together, supply-
ing a desideratum of the Huropean collections.

‘6th. In none of the caves were any specimens observed
referable to the Rhinoceros leptorhinus of Cuvier, as I regard
that species to be limited.

‘From what I have seen, I am strongly of the conviction
that with our present advanced knowledge the thorough
investigation of a well-filled virgin cave in England would
materially aid in clearing up the mystery, either of the con-
temporaneity of the Phocene Mammalian Fauna with the
commencement of the Post-Pliocene Fauna, or of the con-
ditions and association under which the former was replaced
by the latter. M. Lartet, in a late communication to the
French Academy, has thrown out a suggestion, the import-
ance of which, if well founded, can hardly be overestimated,
that the mixed Mammalian Fauna of the Glacial period has
been made up of two distinct geographical elements, the one
a northern division pushed southwards from Siberia and the
north of Europe, consisting of the Mammoth, the tichorhine
Rhinoceros, the Irish Elk, Ursus speleus, Bos primigenius, &c.,

1 Subsequently R. hemitechus, sce antea, p. 351.—[ Ep.]

2 ——— ee ee

OSSIFEROUS CAVE OF BRIXHAM. 491

‘the other a southern division projected northwards from
Mauritania, through Spain and France, comprising the
existing African Elephant, the existing two-horned Rhi-
noceros (Rh. bicornis), the Lion, Panther, two existing
Hyznas, Hog, Antelope, Porcupine, &c. M. Lartet affirms
that Elephant remains from the Quaternary deposits of
Spain, which had been examined by him, belong “ indubita-

BS

blement a VEHléphant actuel d’Afrique et au Rhinocéros
bicorne vivant aujourd’hui dans la partie australe de ce
méme continent.” M. Gervais has described Rhinoceros
remains from the Cave of Lunel Viel, under the name of
Rhinoceros Lunellensis, which he affirms are hardly dis-
tinguishable from those of the existing two-horned species,
the agreement of the teeth being almost complete.' M.
Lartet states that certain Rhinoceros molars from Kirkdale
exhibit the same line of resemblance.

“TI have been induced by these circumstances to bring the
case of the new Brixham Cavern to the notice of the Council.

“Your obedient servant,
(Signed) ‘H. Fauconer.’

Il. Report or PROGRESS IN THE BrixHam CAVE.
Sept. 9, 1858.

Having lately made a joint inspection of the ‘ Windmill
Hill Cavern,’ at Brixham, we think it may be of interest to
the London committee to know our opinion of the progress
already made in the excavations, and of the probable pro-
spective results.

We examined the cavern in company with Mr. Pengelly,
F.G.S., under whose zealous superintendence the operations
are conducted, and of Dr. John Percy, F.R.8., who during
his residence at Torquay has taken an active and lively
interest in the exploration. Most of the points to be noticed
in the sequel were freely discussed among us, and there was
but little difference of opinion as to the bearing of the
observations, and the best plan of carrying on the work for
the future.

1. Extent of the Cavern.—The accompanying ground-plan,
by Professor Ramsay (fig. 3), exhibits the extent and ramifi-
cations of the cave already ascertained. When the first
discovery was made in April of the present year, only the
‘Dyer’s Entrance’ (A), and the ‘Reindeer Gallery’ (H 4),
leading in from it in a SS. Hasterly direction, were then

1 See antea, p. 309.—[ Ep.]

492 OSSIFEROUS CAVE OF BRIXHAM.

known; but another blocked-up entrance, about 25 yards
south of the former, and situated at nearly the same level in
the cliff, gave promise of further ramifications. This ‘ Cen-
tral Entrance’ (B), has now been cleared out, and it has
been found to conduct into several passages. The principal
of these is the ‘ Flint-knife Gallery’ (F F), which stretches
inwards with but little tortuosity, from west to east, and
intersects the ‘ Reindeer Gallery’ at the ‘Hbur chasm’ (G),
after extending a distance of about eighteen yards. Another

Pie. 3:

GROUND PLAN OF GALLERIES IN BRIXHAM CAVE.!

RR, Road. A, Dyer’s entrance. B, Central or main entrance. C, South entrance,
D D, Subordinate passages unexplored. E E, Reindeer Gallery. FF, Flint-knife
Gallery. G, Ebur chasm. HH, Pen Gallery. I, Big Chamber. J, Steep Slide
Hole.

passage leads from the same opening, nearly from north to

south, for a distance of about ten yards, forming the ‘ Pen

Gallery’ (HH), and terminating in the ‘ Big Chamber’ (I).

This expansion of the cavern is at present unexplored, but

it appears to be connected with a third external opening,

forming the ‘Southern entrance mic There are therefore
already three ascertained external openings, connected by
intersecting galleries, and all included within the terms of
the lease held on behalf of the committee.

1 This is the ‘reconnaissance sketch’ | prepared by Mr. Bristow, of the Geolo-

which accompanied the Report. Detailed gical Survey.—[ Ep. ]
plans and sections were subsequently |

OSSIFEROUS CAVE OF BRIXHAM. 493

The general structure of the cavern is now distinctly
intelligible. A road (RR) runs along the outcrop of the
strata from NE. to SW., a little below the level of the open-
ings A and B. The ‘Reindeer Gallery’ (HE E) corresponds
with a line of vertical dislocation, perpendicular to the
strike of the strata, and the ‘ Flint-knife Gallery’ (F F) to a
series of intersecting tunnel-shaped passages, joining on to
the former. From the gallery FF a number of unexplored
passages (DD) diverge right and left. This ‘ Flint-knife
Gallery’ has an irregularly vaulted low roof, which along
the greater portion of its extent is perfectly free from any
crust of stalactite, and bears the most distinct marks of
having been hollowed out by the long protracted action of
running water. The floor of this gallery was formed, nearly
throughout, of ochreous cave earth, free from any stalag-
mitic incrustation upon the surface, while the ‘ Reindeer
Gallery’ (HK EK), terminating upwards in an irregular fissured
cleft, abounded both in stalactitic pendicles, and was over-
laid by a floor of stalagmite, of irregular thickness. At the
‘Kbur chasm’ (G), where the two principal galleries inter-
sect, the floor was formed of a thin crust of stalagmite,
stretching across an empty cavity, the bottom of which is
filled with ochreous cave earth. About the middle of the
‘Reindeer Gallery’ there is a passage, ‘Steep Slide Hole’
(J), given off in a north-easterly direction, at a very steep
incline, which has been followed down to about 40 feet. The -
‘Reindeer Gallery’ is continued inwards beyond its inter-
section with the ‘ Flint-knife Gallery’ at (G); but the
passage is crammed to the roof with ochreous cave earth,
and it remains unexplored. This part of the cavern is
expected to yield a rich harvest of fossil bones. Regarded
in a general way, the Brixham Cavern may be considered as
partaking of the tunnel-character of the Kirkdale Cave, in
the ‘Flint-knife Gallery,’ and of the fissure-character of
Kent’s Hole and the Gower Caves, in the ‘ Reindeer Gallery.’
No vertical flues ascending towards the summit of the cliff
have as yet been detectéd in the explored parts, such as
were found in the Oreston Cave and in Paviland.

2. Workings.—The conduct of the excavations was con-
signed by the London committee to Mr. Prestwich and Mr.
Pengelly. The committee, fully impressed with the proba-
bility of remains of different periods being met with at
different levels in the cavern floor, determined from the out-
set on working the upper deposits horizontally inwards, as
far as might be practicable on the same horizon, and then of
working the lower deposits successively in the same manner.
In this way they considered that they would avoid the risk

494 : OSSIFEROUS CAVE OF BRIXHAM.

of confounding the remains of different levels, which is apt
to take place in excavating cave-bottoms vertically down to
the rock-floor, and which has vitiated the results obtained in
many other cave explorations, more especially in regard of
the contested position of human industrial remains. The
excavations were commenced on the 14th July, by Mr.
Keeping, of the Isle of Wight, deputed by Mr. Prestwich for
the purpose ; and they have been conducted with such vigour
and success, that within six weeks the stalagmitic floor of
the ‘ Reindeer Gallery,’ together with the ochreous cave-
earth deposits below it, was entirely broken up, and the floor
of the ‘ Flint-knife Gallery,’ explored from its interior
termination at (G) to the external ‘Central Entrance’ (B).
The ‘ Pen Gallery’ (H H) was also laid open from its com-
mencement on to the ‘ Big Chamber’ (I); and on a rough
estimate, about 1,500 bones were exhumed between the 14th
July and the 25rd of August.

3. Successive Floor-deposits—We are able to enter on
this part of the subject only in a general way, having taken
no precise measurements of the incomplete sections at differ-
ent intervals in the galleries. Where all the deposits are
present, the following section was yielded :

1. Layer of Stalagmite, of irregular thickness.

2. Cave-earth (ochreous) with Limestone Breccia.

3. Ochreous Cave-earth, with comminuted Shale.

4. Rounded Gravel; depth undetermined.
Occasional waterworn pebbles were found mingled with the
organic remains in the upper deposit of Cave-earth No. 2.
The drip from the roof, in wet weather, is copious in all the
fissure galleries (like EH EH), and the floor in these cases is
covered with a cake of stalagmite ; while in the intersecting
tunnel-gallery (FF), there is little or no drip from the roof,
and the surface of the cave-earth is uncovered.

4. Organic Remains.—Mr. Pengelly estimates that about
1,500 bones had been exhumed during the first six weeks of
the workings. A large number of these, however, belong to
skeletons of small animals, like the rabbit and fox, found
near the surface. We consider that the great harvest of
remains will be found in the low-level deposits, which have
not yet been penetrated. Remains of the following animals
were identified by Dr. Falconer.

Rhinoceros tichorhinus. Detached upper and lower molars,
in considerable numbers, of young and old animals, and an
astragalus, bearing distinct marks of superficial gnawing,
dug up in our presence.

Bos. Species undetermined; teeth, jaws, and other
bones.

OSSIFEROUS CAVE OF BRIXHAM. 495

Horse. Species undetermined ; a few remnants.

Cervus Tarandus. The cranial box of the skull, found
near the surface by the owner, on the first discovery of the
cave; and a very fine entire antler, embedded superficially
in the stalagmite, near the intersection of the galleries H H
and FF, over the ‘Ebur chasm.’ Fragments of antlers of
other Deer, species undetermined.

Ursus spelceeus. Lower jaws of young and old individuals,
with numerous detached canines and other teeth, in fine
preservation. A superb specimen of the bones of left hind
leg, comprising the femur, tibia, and fibula, folded together,
with the patella and astragalus im situ. These were found
near the ‘Ebur chasm,’ and the other parts of the skeleton
may be looked for when that portion of the cavern is dug up.!

Hyena spelea. ‘Teeth, fragments of the skull, lower jaws,
and other bones.

The above is nothing like a complete list of the animal
remains found in the cave; but, considering that the work-
ings have hitherto been restricted to the least productive
and superficial deposits, it will suffice to show that the
anticipations formed of the cavern were not too sanguine,
and that the excavations are well worthy of being followed
up vigorously. Some of the results already arrived at are of
great general interest.

5. Human Industrial Remains (?).—Several well-marked
specimens of the objects called ‘ Flint-knives,’ and generally
accepted at the present day as the early products of rude
Keltic or Pre-Keltic industry, have been exhumed from
different parts of the cavern, mixed in the ochreous earth
indiscriminately with remains of Rhinoceros, Hycena, and
other extinct forms. One of these so-called ‘ Flint-knives’
was brought up from the deposit No. 2, from a depth of thirty
inches below the superficial Stalaomite No. 1. We failed in
detecting evidence that these so-called ‘ Flint-knives’ were
of a different age, as regards the period of their introduc-
tion, from the bones of the extinct animals occurring in the .
same stratum of cave-earth, or that they were introduced

1 Tn a letter to Mr. (now Sir John)
Lubbock, dated May 24, 1863, Dr. Fal-
coner writes as follows:—‘ The Report
on the Brixham Cave was drawn up by
me on Sept. 9, 1858. All the cireum-
stances connected with the entire leg of
eave bear—femur, with tibia and fibula
folded together, and ball of astragalus
partly dislocated—and its position in
comminuted shale, below the ochreous
cave earth, and above a well defined

flint implement, were determined by me
at Torquay and Brixham on September
2. Mr. Pengelly gave us the data... .
I identified the remains and the flint,
and drew the inference that the lee must
have been introduced with its ligament
at least. fresh, after the flint manufac-
tured by the hand of man had been in-
troduced into the lower cave deposit.’—

[ Ep. ]

496 OSSIFEROUS CAVE OF BRIXHAM.

into the cavern by different agencies. Schmerling discovered
in the caves of Engis and Chokier, near Liege, well-marked
‘ flint-knives’? and arrow-heads, mingled in the ochreous
mud and gravel with the bones of extinct Mammalia, which,
he inferred, had been washed in by the agency of running
water; these included Mammoth, Rhinoceros, and Hycna.
Delpon and Jouanet have made corresponding observations
as to a similar mixture in the caves of Quercy and Perigord.
Marcel de Serres, De Christol, Tournal, and Dumas have
inferred the same in numerous caves in the south of France.
The attention of Mr. Pengelly has been closely directed to a
careful and minute observation of the circumstances of the
association in the Brixham Cavern. The results of the
exploration of each day are carefully put aside and labelled ;
and it may be anticipated that data will be arrived at for
settling the disputed question of the contemporaneous intro-
duction, or otherwise, of the supposed human industrial
objects into the cavern along with the remains of the extinct
Mammalia.

6. One result of great interest has already been brought
out, namely, the superposition of undoubted remains of the
Reindeer, above the so-called ‘ Flint-knives,’ from which
the inference arises that the Reindeer continued to be an
inhabitant of Britain after the appearance of man in the
island. A fine horn ofa Reindeer, nearly perfect, from the
basal ‘burr’ to the terminal branches of the beam, and
presenting a bez-antler 17 inches long, terminating in pal-
mated snags, was discovered superficially embedded in the
stalagmite, close to the ‘ Hbur chasm’; near the same place
a ‘ flint-knife ’ was brought up from the ochreous earth, thirty
inches below the stalagmite. Professor Owen has noticed
the occurrence of Reindeer remains in the ‘ Ash-hole
Cavern,’ of Berry Head, explored by the Rev. Mr. Lyte.
Dr. Falconer has identified skulls of the same species, found
in the Mendip Caverns; and Major Wood, of Stout Hall, has
discovered skulls of it, in the caves of Spritsail-Tor and
Paviland, in Gower. In these instances there was no indica-
tion of their association with man; but in the Brixham
Cavern the evidence, so far as it goes, clearly tends to show
that the antler in question was one of the latest introductions
into the cavern, before the ‘Dyer’s Entrance’ was blocked
up by rubbish, and long subsequent to the entombment of
the objects called ‘ Flint-knives.’

7. On the whole, we consider the progress made to be
highly satisfactory, and the promise of future result to be so
encouraging as to merit the best efforts of the committee to
provide ‘the means for following up the excavations. The

OSSIFEROUS CAVE OF BRIXHAM. 497

erant from the Royal Society, together with Miss Burdett
Coutts’ liberal donation, however carefully husbanded, will
not cover the very moderate scale of expenditure within
which the operations are at present conducted, beyond the
month of December. A further grant may with some confi-
dence be expected from the Royal Society next summer, but
we invite the earnest attention of the committee to devise
ways and means to meet the expense of the excavations
until then.
(Signed) H. FAuconer.
A. C. Ramsay.
W. PENGELLY.
Sept. 9, 1858.

Ill. NOTE BY EDITOR.

[Dr. Falconer’s Note-books contain a list of nearly 150 numbered
fossils from the Brixham Cave identified by himself, a copy of which
has since his death been handed to the committee. Most of these speci-
mens were teeth or frazments of bones of the species mentioned in the
above report ; but the list also included the remains of Lagomys, Canis
Vulpes, Putorius, Arvicola, Lepus, and the bones of birds. One specimen
was ‘a superb last molar like that of Rhinoceros hemitechus, and in
another it is noted that there was ‘ a detached Hyena incisor along with a
worked flint!’ Inreporting to Dr. Sharpey, the Secretary of the Royal
Society, the progress of the Brixham Cave excavaticns on June 7, 1860,
Dr. Falconer concluded as follows: ‘ The grant of the Royal Society has
already fructified richly, in the impulse which the Brixham Cave ex-
plorations have given to the whole question of the geological indications
of the antiquity of the human race, founded upon industrial objects occur-
ring in.old deposits, upon which communications are pouring forth
daily from various departments in France by observers of authority.’ |

VOL. Il. KK

498 OSSIFEROUS CAVES OF GOWER.

XIX. OBSERVATIONS ON THE OSSIFEROUS CAVES
IN THE PENINSULA OF GOWER, SOUTH
WALES.!

1. InTRODUCTION.—OBJECT OF THE COMMUNICATION TO GIVE A
SUMMARY OF SOME RECENT OBSERVATIONS.

THE object of this communication is to submit a summary of
the results of some researches with which I have been oc-
cupied, during the last three years, in conjunction with my
friend Lieut.-Colonel Wood, F.G.8., of Stout Hall, upon the
Bone Caves of Gower in Glamorganshire. The caverns in
question are numerous, and the details, both paleontological
and stratigraphical, are much too voluminous for a single
communication. I have in consequence thought that it
might be useful to give at once a brief account of the more
novel and important results already arrived at. I have the
less hesitation in adopting this course, as the Gower caves
have already furnished a considerable amount of new in-
formation regarding the mammalian occupants of the country,
during what may be conveniently called the Cave-period ;
and they appear to me to indicate with more precision than
elsewhere in England the date when the cavern strata
emerged above the sea, to become the receptacles of the
organic remains that are now found in them.

2. IMPORTANCE AND Extent oF CoLONEL Woop’s GOWER
COLLECTION.

It is incumbent on me to premise, im limine, that while
solely responsible for the palzontological determinations,
and the conclusions therefrom arising, I am indebted to the
collections of Colonel Wood for nearly the whole of the
fossil materials, bearing on the Gower caves, that form the
groundwork of this communication. Colonel Wood has been
engaged on excavations in most of these caves in succession,
during the last ten or twelve years. He has discovered and

1 This paper was communicated to the | the ‘Quart. Journ. Geol. Soc.’ for No-
Geological Society on May 30 and June | vember, 1860; but the paper itself has
13, 1860. An abstract of it appeared in | not before been published.—{Ep.]

em

OSSIFEROUS CAVES OF GOWER. 499

explored several that were previously unknown; and the
very extensive collections of fossil bones which he has
amassed, together with his notes and observations, have been
placed most unreservedly at my disposal. Therefore, al-
though this memoir has devolved upon me, I am desirous
that it should be regarded as embodying our joint labours,
and that my friend’s long-continued, disinterested, and meri-
torious exertions in the cause of science should be fully
recognized.

3. GEOLOGICAL SKETCH OF THE GOWER PENINSULA.

The peninsula of Gower forms a broad headland, which
projects into the Bristol Channel, bounded on the H. by
Swansea Bay, and on the W. by the Bay of Carmarthen.
The fundamental rock is a thick mass of old red sandstone
conglomerate, which forms the axis and highland of the
promontory, rising into the hills of Llanmadoe and Cefn-
Bryn. This conglomerate is flanked_-N. and S. by strata
of carboniferous limestone, and the base of the peninsula,
where it joins on to the mainland of Glamorganshire, forms
a part of the coal measures of South Wales. The line of
southern coast stretches from the ‘Mumbles’ on the E. to
the ‘Worm’s Head’ on the W., and with the exception of
the indentations, of Port Hynon, Oxwich, and other smaller
bays, it presents an iron-bound wall of bold, lofty, and preci-
pitous or scarped cliffs, occasionally exhibiting features of
the grandest description. The general character of the coast
section is well seen at ‘Shire Coom’ (Sheer Cwm), near the
‘Three Cliffs Bay.’ In addition to the well-known raised beach
at the ‘Mumbles,’ Mr. Prestwich detected in Mewslade Bay,
at the western end of the section, another example of the
same kind boldly defined, and forming a continuous stretch
of fully a mile in length. In ‘ Rhos Sili Bay,’ and on the
summit of Cefn-Bryn, he ascertained the presence of resi-
duary patches of boulder clay. Mr. Prestwich has favoured
me with a memorandum of his observations, which is an-
nexed as an appendix (p. 536).

4. List or NAMES AND Position oF THE BoNnE CAVES.

With the exception of ‘Spritsail-Tor,’ on the west side of
the peninsula, all the known ossiferous caverns of Gower
occur along the southern coast line between the Mumbles
and the Worm’s Head; they are confined to the carboniferous
limestone, and within a limited range of variation they are
at different heights above the sea: the floor of ‘ Bacon Hole ’
being 30 feet above the highest tides, while the lower story

KK 2

500 OSSIFEROUS CAVES OF GOWER.

of ‘Bosco’s Den’ is breached by the waves and washed out
to a depth inwards of 31 feet. The two most eastern, namely
the Mumbles and Caswell Bay caverns, have been entirely
destroyed by the action of the sea, and a very few only of the
organic remains found in them have been preserved. These
are now deposited in the Swansea Museum. The next in
succession westwards are the group of contiguous caverns
which occupy the but slightly indented line of coast, stretch-
ing between ‘ Pwll dhu’ Head and ‘ Sheer Cwm’ in Three
Cliffs Bay. The best known of these are ‘Bacon Hole’ and
‘Minchin Hole,’ which have been especial objects of explora-
tion by Colonel Wood since 1850, and have yielded a large
quantity of mammalian remains of the highest interest and
importance. ‘ Bacon Hole’ and ‘ Minchin Hole’ are several
hundred yards apart, but between them there occur a series
of detached caverns, which like Minchin Hole are not
entered on the Ordnance Survey Map of Gower. They have
only become known through the persevering exploration of
Colonel Wood. The principal of these are the very remark-
able and instructive two-storied cavern recognized by the
quarrymen in the neighbourhood, under the name of the
“Devil’s Hole;’ then ‘ Bosco’s Den’ and ‘Crow Hole,’ and
lastly on the western side of ‘ Minchin Hole’ towards
Three Cliffs Bay, a cavernous fissure impacted with sand,
which Colonel Wood has named ‘ Raven’s Cliff.’ One cavern,
near Langland Bay, called ‘Ram-Tor,’ and presumed to be
ossiferous, still remains unexplored.

Dr. Buckland gives a brief account in the ‘ Reliquize Dilu-
viane’ of a deposit of fossil bones discovered in 1792, in a
cavity in the limestone of the ‘ Crawley Rocks’! m Oxwich
Bay, which is no longer in existence. With this exception,
no bone-caves have as yet been disclosed in Oxwich Bay
or Port Eynon Bay. From the western point of the latter
to the Worm’s Head, the general alignment of the coast is
comparatively straight, with a slight deflexion to the N., and
with few indentations. The limestone cliffs are here lofty,
exceedingly steep and precipitous, and about the middle of
the stretch, on the face of the ‘ Yellow Top’ rock occur the
two contiguous cavernous fissures, commonly known as the
‘Paviland Caves,’ discovered in 1821, and so celebrated
through the researches of Dr. Buckland. On the western
side of the Gower coast the limestone strata have been
denuded in Rhos Sili Bay; but to the north of Llanmadoc
Hill, where they reappear in a narrow belt, a cave buried in
drift sand was casually disclosed, through quarrying opera-
tions in 1839, on the point called ‘ Spritsail-Tor.’ It was

1 See antea, p. 354—[ Ep. |

OSSIFEROUS CAVES OF GOWER. 601

partially examined by Sir H. de la Beche, and thoroughly
explored by Colonel Wood in 1849. Although but of con-
tracted dimensions, it has yielded a very large quantity of
fossil bones, belonging to many genera and species.

5. ‘MewsntaDE Bay.’—RAISED BEACH OBSERVED BY
PRESTWICH.

Before proceeding to notice the contents of these caves, it
will be convenient to refer to the highly valuable observation
made by Mr. Prestwich upon the raised beach of Mewslade
Bay, when he and I visited Gower last autumn. A thin
layer of marine sand, containing shells of common existing
species, was observed by Mr. Starling Benson and Colonel
Wood, at the bottom of the deposits in ‘Bacon Hole’ cavern ;
and marine deposits of sand and gravel are common on the
floors of some of the other Gower caves, as will be shown in
the sequel. It was of importance to place these beds, either
in correlation with, or distinction from, some definite patch
of raised beach in the neighbourhood. Mr. Prestwich dis-
covered in Mewslade Bay, between Paviland and the Worm’s
Head, the well-marked illustration of this nature, repre-
sented by the accompanying section, for which I am indebted
to his kindness (p. 537). Atten or twelve feet above the level of
high tides there is perched upon the weathered outcropping
edges of the limestone a narrow belt of raised beach, a mile
in extent. The section shows a bed 3 feet thick, composed
of fragments and pebbles of limestone, with entire and broken
shells, imperfectly cemented by a calcareous paste. Above
this there is a bed of reddish sand, 4 feet thick; and the
whole is overlaid by an enormous mass, varying from 20 to
30 feet in thickness, of angular fragments of carboniferous
limestone, being the debris of the immediately adjoining
rocks. This bed is developed in vast accumulations, at a
considerably higher level, on the face of the cliffs between
‘Bacon Hole’ and ‘Minchin Hole,’ and it constitutes a well-
marked example of ‘Bréche en place,’ or the bed which
Mr. Godwin-Austen has characterized, in his memoir on the
‘Pleistocene Period,’ under the designation of ‘ Head.’!
Mr. Prestwich’s remarks on the traces of ‘ Boulder Clay’ in
Gower will be found in his notes in the appendix.

6. ‘Bacon Hour’ CAvERN.

Description.—The cavern of ‘ Bacon Hole’ may, with
advantage, be taken first, as it was very carefully explored,
under the constant superintendence of Colonel Wood ; and an

1 Quarterly Journal Geol. Soc., vol. vii, (1851), p. 121.

502 OSSIFEROUS CAVES OF GOWER.

excellent account of the operations, with plans and sections,
has been published by Mr. Starling Benson,! founded on ob-
servations made during the progress of the work, which
relieves me from the necessity of a minute description on the
present occasion. ‘ Bacon Hole’ is situated about the lower
third of the limestone cliffs on the sea coast, about six miles
west of Swansea, and in a line nearly due 8. from Kilvrough.
You descend from a level grassy plateau by a steep and
difficult path along the cliff, till you come upon the brink of
an enormous accumulation of cemented angular breccia,
below which, but a little more to the E., opens the cavern
towards the sea, and directed to the SE. The mouth is
wide and open, with diverging walls admitting the free access
of light to the interior. The cavern is evidently, as has been
observed by Mr. Starling Benson, in the line of one of numer-
ous faults which intersect the limestone of South Wales. A
cleft continuous with that of the cave can be seen stretching
out into the sea, and bending slightly to the east, upon the
rocky platform below which forms the present sea beach.
The walls of the interior lean to and meet so as to form an
anticlinal axis along the centre of the roof, in part of which
the remains of a wide chasm are visible, which was probably
continued upwards ina flue. This chasm is at present in a
great measure blocked up with masses of breccia, sheeted
over by an abundant coating of stalagmite; but, like the
corresponding flue in the roof of Paviland, as figured in
Buckland’s section, it may have formed a channel through
which injecta were precipitated from above into the cavern.
The upper parts of the interior are mostly covered over
with a cake of stalagmite. There is a copious drip from
above, which keeps the surface soil of the floor in a soppy
state. The roof slopes from the outside inwards, and it is
most probable that the cavernous fissure is continued much
further back than is now apparent, but blocked up by breccia
and stalagmite. The extreme height of the interior is about
20 feet. ;
(a.) Section of the Floor Deposits —The floor of the cavern
at the entrance consists of a thick accumulation of cemented
angular limestone breccia, which is well seen in section from
the outside below. This breccia can be traced seaward upon
the diverging walls beyond the existing portion of the cave,
for a distance of 100 feet.2 The floor slopes down from the
entrance inwards. The uppermost layer of stalagmite is,
according to Mr. Starling Benson, 30 feet above the level of

? Account of the Cave Deposit of | tific Society,’ p. 9.
‘Bacon Hole’ ‘Annual Report for * Starling Benson, Op. cit. p. 13.
1852 of the Swansea Literary and Scien- ;

OSSIFEROUS CAVES OF GOWER. 503

high water. Great difficulty was encountered in conducting
the excavations, from the cemented breccia of the floor being
too hard for the pickaxe, and too cavernous to blast. The
following section, taken near the middle, is given by Mr.
Benson as showing the average thickness and succession of
the various deposits :—

1. At the bottom, and lying immediately above the rocky
floor, patches of yellow sand, a few inches thick, abounding
with shells of Litorina rudis and Litorina litoralis, in groups
and in every stage of growth, implying as Mr. Benson
remarks, ‘that when these sea mollusea lived, the floor must
have been below the level of high water.’ Besides these
marine forms, a smaller number of shells of Clausilia nigri-
cans were found in the same sand, together with bones of
birds and of a species of Arvicola. The Clausilice were en-
veloped by—

2. A thin layer of stalagmite, which could not have been
formed until the floor of the cavern had been elevated above
the level of high water.

3. Upon this stalaemite, a bed of blackish or dark-coloured
sand varying in thickness, as I have been informed by
Colonel Wood, from 18 inches to 2 feet, and containing a
mass of bones consisting almost entirely of various parts of
the skeleton of an Elephant, which I have ascertained to be
Hlephas antiquus, including numerous molars. These bones
when first exposed were quite soft and friable, crumbling
under the fingers when handled ; but after a few days’ expo-
sure to dry air they became hard and susceptible of repair.
The tooth of a Badger (Meles Taxus) and of a Putorius were
found at the lower part of the same sand.

4. Immediately above the black sand, and at places almost
intermixing with it, a deposit of ‘red soil’ or ochreous cave
loam, containing remains of the same species of Elephant, and
several lower jaws, with detached upper molars, vertebrze,
and bones of the extremities of a species of Rhinoceros, which
I have named Rhinoceros hemitechus, distinct alike from the
Rhin. tichorhinus of the superficial gravels, and from Rhin.
leptorhinus of the Sub-Apennine Pliocene deposits. Along
with these were found bones of Hyena, Wolf, Ursus, Bos, and

Cervus. This stratum varied from 1 to 2 feet in thickness,
from the intermixture, in some places of loose limestone
breccia, and in others of sand free from bones. The breccia
is represented in Mr. Bengon’s section as a separate bed.
The blackish sand and cave-earth beds may be regarded as
different conditions of the same deposit. Mr. Benson re-
marks that this latter was the only deposit found in the cave
similar to the soil (ochreous loam) usually met with im bone-

504 OSSIFEROUS CAVES OF GOWER.

caverns resorted to by Carnivora. No coprolites of Hyzena
were observed in any of the floor deposits.

5. Above the stratum of red-earth and breccia a bed of
stalagmite of very irregular thickness, under which, and
partly embedded in it, was found the portion of an Elephant’s
tusk, which measured 5} feet in length with a girth of 24
inches. This tusk lay about 4 feet above the bones of Ele-
phant in the lower deposit.

6. Above this bed of stalagmite, another bed of breccia, in
most places cemented throughout by stalagmite, and con-
taining entangled bones, chiefly of Ursus and Bos, and mostly
entire; but which could rarely be extracted unbroken. It
varied from 1 to 2 feet in thickness.

7. The uppermost bed of stalagmite, forming a very
regular deposit, which ranged generally from 9 inches to 1
foot in thickness, but in some places extended to upwards of
2 feet. Colonel Wood informs me that some remains of
Ursus were found occurring in the stalagmite.

8. A superficial deposit of dark-coloured alluvial earth,
about a foot thick, containing the bones of Bos, Cervus,
Canis Vulpes, together with horns of Red Deer and Roebuck,
the latter in one instance bearing distinct marks of having
been wrought by human hands. Recent shells were also
found in this stratum, consisting chiefly of Patella vulgata,
Mytilus edulis, Purpura lapillus, and Intorina litorea. These
appear to have been brought into the cavern by birds as
food.

‘No remains of man were found below the upper stalag-
mite. In the mud above it were pieces of ancient British
Pottery.’! (Starling Benson.)

(b.) Organic Remains.—It is not my intention now to enter
on a detailed identification of the different species of mam-
malia found in ‘ Bacon Hole,’ nor to analyze the conditions
under which they were associated. I shall confine myself to
such as appear to be most significant, as positive indicators of
the age of the deposits. Sir Charles Lyell, so recently as in
the supplement to the last edition of the ‘Manual,’ states
that ‘to what part of the Pliocene period the cave animals of
Great Britain should be chiefly referred is still a vexed
question.’ The large Pachydermata, belonging to the genera’
Blephas, Rhinoceros, and Hippopotamus, are of the most
weight in their indications, and it is of the last importance
to discriminate with precision the species to which the re-
mains occurring in the caves belong. The vagueness which

1 The above summary of the cave de- | Mr. S. Benson’s excellent description.
posits is mainly a condensed abstract of

OSSIFEROUS CAVES OF GOWER. 505

has heretofore attached to the term ‘Mammoth,’ when Ele-
phants have been cited, is now generally admitted; and I
believe that a corresponding looseness has applied to the
species of Rhinoceros.

Colonel Wood, at whose charge the excavations were made,
presented the collection of fossil bones to the Museum of the
Royal Institution of South Wales at Swansea, where they are
now carefully preserved, and I have had several opportunities
of examining them in detail.

Elephant Remains.—These consist of 11 molars or fragments
of molars of the upper and lower jaws of young and adult
animals ; a fragment of an enormous tusk 24 inches in girth,
and 5} feet long when measured in the deposit; 6 vertebree,
chiefly dorsal and lumbar, but of different sizes, indicating
more than one individual; 2 humeri, right and left, appa-
rently of the same animal, comprising the inferior two-thirds
of the shaft: the head of one of them present, but detached ;
1 radius and ulna quite entire; 1 huge femur, corresponding
in size with the humeri, and probably of the same individual,
the shaft and condyles entire, the articular head present,
but detached ; 2 tibias of large size and entire, corresponding
with the femur and humeri; 2 tibias, also entire, but of a
smaller animal; 1 calcaneum of the right side, of large
size, and in fine preservation ; 14 metacarpal and metatarsal
bones.

The molars present all the characteristic marks which dis-
tinguish Elephas antiquus from the true Mammoth, FE. pri-
migenius ; namely, comparatively narrow crowns, with fewer
ridges, the plates of enamel thick and well crimped, and the
discs of wear wide and open, with a tendency to expansion
in the middle. They agree in the closest manner with molars
of the same species, H. antiquus, from the Oyster-bed of the
Norfolk coast, Saffron Walden, Clacton, Grays Thurrock,
the mud-bed of Pagham, and the Sub-Apennine deposits of
the Astesan. One of them consists of a last true molar of
the lower jaw, right side, presenting fourteen principal ridges,
to a length of 11 inches. Another is the corresponding
tooth of the left side; another is a third milk molar of the
upper jaw, right side—not quite entire, but very character-
istic ; another is a germ specimen of an upper antepenulti-
mate true molar. The remaining specimens are more or less
fragmentary, but are alike characteristic of H. antiquus. Not
a vestige indicative of H. primigenius was detected, either
among the teeth or bones. The remains belonged to at least
three individuals, one of large size, another smaller, and a
third which had not quite passed through the last stage of
milk dentition. The bones were free from marks of gnawing,

506 OSSIFEROUS CAVES OF GOWER.

and they bore no appearance of having been rolled or trans-
ported from a distance.

Rhinoceros Remains.—These were found: in abundance in
‘Bacon Hole,’ and in still greater quantity in ‘ Minchin Hole,’
situated a little further to the westward. LHvery tooth and
bone which admitted of identification was found to belong
to R. hemitechus, so named from the presence of a partial
bony septum between the nostrils, but differing in a very
pronounced manner from Rf. tichorhinus, by the configuration
of the cranium, by the form and characters of the upper
molar teeth, by the absence of an edentulous beak-shaped
prolongation of the symphysis of the lower jaw, in advance
of the anterior molars, and in the form and proportions of
the bones, generally throughout the skeleton. It differs
equally from the original type of the R. leptorhinus of Cuvier,
founded upon Cortesi’s cranium found near Piacenza, which
had no bony septum ; and from R. Htruscus of the ‘Sabbione’
and ‘Sansino’ deposits of the Val d’Arno, which species,
however, had likewise a bony nasal septum, but is distin-
guished at once from R. tichorhinus and R. hemitechus by its
smaller size, light form, and comparatively slight limbs, be-
sides special characters in the shape of the skull and teeth.
The R. henvitechus of the Gower caves is identical with the
Clacton form, the skull of which has been described by Pro-
fessor Owen in the ‘ British Fossil Mammalia,’ under the
specific designation of R. leptorhinus of Cuvier; but I have
been led to the conclusion that it is wholly distinct from the
true type of the latter species.

The following remains, referable to R. hemitechus, were
found in ‘ Bacon Hole’ :—Three adult rami of the lower jaw,
comprising the entire series of molar teeth in fine preserva-
tion; two of the rami belonging to one individual. Six de-
tached molars of the upper jaw, three of which are of the right
side and consecutive teeth, probably of the same individual.
Of the others, two are of the left side, and one of them at
least indicates an individual of a different age. A fragment
comprising the lower half of the shaft of the right humerus,
the outer condyle wanting. A radius presenting the superior
half of the bone, with the articular surface. The corresponding
ulna, mutilated, of the olecranon. Two metacarpal or me-
tatarsal bones. Two cervical vertebree, in fine preservation.
One lumbar vertebra. A right half of the pelvis.

The marine sand at the bottom of ‘ Bacon Hole’ was analo-
gous to the thin bed on the rocky floor of the Grotta di San
Ciro, near Palermo, but contained fewer species of Mollusca.
Here the resemblance ceases. In San Ciro, an enormous thick-
ness of breccia, also of marine origin, and crammed with

OSSIFEROUS CAVES OF GOWER. 507

bones, is accumulated above the sand; while in ‘ Bacon Hole’
the alternate deposits of stalagmite, black sand, ochreous
earth and breccia, up to the top, were all of sub-aerial
formation.

7. ‘“Mincutn Hone’ Cavern.

Description.—The bone cavern named ‘ Minchin,’ or ‘ Min-
chin Hole,’ became the next object of Colonel Wood’s ex-
plorations. It is a very spacious cave, of a grand and lofty
character, several hundred yards to the west of ‘ Bacon Hole,’
and separated from it by a bight or indentation of the coast,
the lower terrace of the intervening cliff being surmounted
by an enormous accumulation of cemented limestone-breccia.
Tt is situated in the bottom of a dark and gloomy recess, and
like ‘Bacon Hole,’ the excavation follows the line of a fault
in the limestone strata. The mouth is open and so lofty
that light penetrates a considerable way into the interior. The
following are the principal dimensions :—

Length from the outer arch to the extremity of the cave, 170 ft. Height at the

entrance, about 34 ft. Width at ditto, below the arch, 17 ft. Greatest width of
the interior, 70 ft. Ditto, ditto, at 6 ft. from the end, 36 ft. .

(a.) Section of the Floor Deposits —The cavern is entered
by ascending a steep step of talus, composed chiefly of
ejected materials; this is succeeded by a more level space,
about 35 feet in length, which in turn is succeeded by a
second step of talus, stretching 51 feet back to the end of
the cave. The floor is very unequal, and the excavations
were but partial, having been greatly impeded by enormous
accumulations of stalagmite covering the inferior deposits.
They were carried on more with a view to the exhumation
of more perfect remains than were found in ‘Bacon Hole,’
and less attention was in consequence paid to the vary-
ing thickness and other minute details of the different de-
posits. At the entrance the floor was covered with (1) loose
debris and breccia of limestone to the depth of 3 feet, suc-
ceeded by (2) a stratum of cave-earth about 9 inches thick ;
(3) one foot of sand; (4) next, a deposit of blackish or dark-
coloured loamy sand, resembling in mineral character the
corresponding ossiferous deposit of ‘Bacon Hole,’ and at-
taining a depth of 24 feet :—abundant remains of Rhinoceros,
Elephas, and Bos were yielded by this bed, and shells of
Helix hispida were found attached to the matrix which en-
crusted some of the bones; (5) at the bottom, and resting
upon the rock-floor, lay a bed of greyish-yellow, coarse, and
gritty marine sand, ranging from one foot to four feet in
thickness, which yielded a considerable quantity of bones of

508 OSSIFEROUS CAVES OF GOWER.

R. hemitechus. One of the lower jaw-fragments of this
species, from this deposit, exhibits a rolled Litorima and specks
of comminuted shells embedded in the encrusting gritty
matrix. Shells of Clausilia nigricans with Litorine were
yielded by other specimens.

Working inwards the sand disappears, and the section
yielded: (1) above, a bed of loose strong breccia of varying
thickness; (2) cave earth, 18 inches; and (3) black unc-
tuous sandy loam, 1 foot. Nothing was met with correspond-
ing to the repeated alternation of the beds of stalagmite
with the other deposits, seen in ‘Bacon Hole.’ The stalag-
mite was accumulated in local masses and not spread out in
horizontal sheets. A large and very important series of
Mammalian remains was exhumed from the interior of the
cavern, chiefly from the shelving floor close to the western
wall, near the middle and back part of the cavern. They
occurred alike in the marine sand at the bottom and in the
black sandy loam overlying it. As a general rule, they were
closely analogous to those met with in ‘ Bacon Hole.’ It is not
my intention, on the present occasion, to go into details on
the genera and species. As regards the Pachydermata,
Elephant remains were fewer, and Rhinoceros remains much
more numerous and important as specimens, than in ‘ Bacon
Hole.’

(b.) Organic Remains.—Among the Elephant bones was a
specimen comprising the anterior two-thirds of the left
ramus of the lower jaw of a very young calf showing the
empty alveoli of the three milk molars. This specimen shows
superficial marks of gnawing. Among the molars were two
very perfect and characteristic specimens, both of Elephas
antiquus, the one being an upper true molar, presenting 15
ridges to a length of crown of about 8} inches, and the
other a lower molar of the right side, showing 16 ridges on
a crown 10 inches long. These two teeth probably belonged
to the same individual. The discs of wear presented the
peculiar characters of Hlephas antiquus in a very perfect
manner; namely, the discs of considerable width and the
enamel plates comparatively thick, with well pronounced
undulations. As in ‘Bacon Hole,’ not a specimen of Hle-
phant was here observed referable to Hlephas primigenius.

The remains of Rhinoceros in Minchin Hole were nume-
-rous, and proved the presence of several individuals, of all
ages. Among these were two adult crania, of which one was
exhumed in a very perfect state of integrity, from the occi-
pital condyles to the incisive border. It had belonged to an
old animal; the series of molars was present on either side,
and Iam assured by Colonel Wood, in the most positive

OSSIFEROUS CAVES OF GOWER. 509

manner, that there was a very distinct and well-marked bony
septum, connecting the nasal bones with the incisives. This
important specimen unluckily met with a grievous accident
in a public museum, by which the skull was crushed and the
fragments lost ; but the palate, with the maxillaries and mo-
lars of either side, remain to prove that it belonged to Rhin.
hemitechus. The other cranium was discovered entire, but
fractured in extricating it from the deposit. The posterior
half, represented by the accompanying figures (see Plates
XXIII. and XXIV.) was preserved, and it agreed very closely
in form with the Clacton cranium, figured in the ‘ British
Fossil Mammalia,’ and with another of the same species,
from Northamptonshire, now in the Paleontological collec-
tion of the British Museum.' Among the other specimens of
Rhinoceros were : 4 fragments of upper jaws containing teeth;
2 upper maxillaries of very young animals, containing the
milk teeth ; 8 lower jaws, of which three are of very young
individuals, with milk teeth only; 1 atlas vertebra nearly
entire ; several dorsal and lumbar vertebre; 1 scapula;
1 humerus, showing the inferior half of the shaft and the
condyles; 1 radius nearly entire; 4 femora, two of which
are of very young or fetal animals; 2 astragali, one of
them superficially gnawed; 1 tibia and fibula in fine pre-
servation; 1 calcaneum; 5 ossa innominata, some of them
showing the cup of the acetabulum perfect. The above list
is intended to show the abundance in which the remains of
Rhinoceros hemiteechus occur in ‘Minchin Hole.’ The diag-
nostic characters of this species, as compared with R. ticho-
rhinus and R. leptorhinus, throughout their respective skele-
tons, have not yet been published. It would be out of place
to enter on such details here, in a communication which pro-
fesses to be merely a summary of observations made on a
series of caverns. But it is especially deserving of remark,
that although molars and other bones of R. tichorhinus have
been found in very considerable numbers in some of the other
Gower caves, such as ‘ Spritsail-Tor,’ ‘ Paviland,’ and ‘ Cas-
well,’ not a tooth or other remain referable to that species
was identified, either from ‘ Bacon Hole’ or ‘ Minchin Hole.’
The same negative observation, as I have already stated,
applies to Hlephas primigenius. Bones of Bison priscus were
common; but remains of Cervide very rare.

Of the Carnivora, the nearly entire skull of a Hyzna,
Hyena spelea, with fragments of lower jaws and various
bones of the skeleton, was met with; also remains of Canis
Lupus and of Ursus speleus, but none referable to Felis spelea.

1 See antea, p. 351.—[ Ep.]

510 OSSIFEROUS CAVES OF GOWER.

No coprolites of Hyzena were encountered, nor bone splin-
ters bearing teeth marks, nor detached molars of Equus.
‘Minchin Hole,’ from these negative results, does not appear
to have been much frequented as a Hyzena’s den, although
some of the bones bear distinct marks of gnawing ; these are
very pronounced upon the astragalus and condyles of the
femur of a Rhinoceros ; while the lower jaw of a foetal Ele-
phant presents superficial grooving, as if gnawed by a Wolf
rather than by a Hyena.

The remarkable state of integrity in which the two large
skulls of Rhinoceros hemitechus were discovered is not a little
significant. They bore no marks of having been dragged in
by any predaceous animal; and it is hardly conceivable that
they would have been precipitated from a fissure above, with-
out being fractured into many pieces, considering the great
height of the roof. It is equally improbable that the ani-
mals were at any time alive in the cavern, for from the
difficult nature of the approach it could never have been
accessible to large Pachyderms like Rhinoceros and Hlephas.
The two crania were embedded in marine sand, at the
bottom of the deposits; and under all the circumstances, it
seems most probable that they were drifted in by the waves,
although it is difficult even then to understand how they
could have escaped without injury or marks of rolling.

8. ‘Bosco’s Den’ CAVERN, UPPER STORY.
*)

Inow pass to the consideration of another series of caverns,
or cavernous fissures, of very high interest, arising not so.
much from the abundance and importance of the organic
remains contained in them, as from the nature and succes-
sion of their marine and alluvial deposits, and from the flood
of light which they throw upon the emergence of the caverns
above the sea, the manner of their filling up by transported
materials, and the proofs of subsidence within a compara-
tively modern period, which they unfold. The merit of the
discovery of this series belongs wholly and solely to Colonel
Wood. Some of the localities, indeed, were known to the
quarrymen who resort to the cliffs, by vague and uncertain
names; others bore no designation whatever, and the names
which will be cited in the sequel are, with a single exception,
those which have been either applied to them or adopted by
Colonel Wood. I must premise further, that the value of
any reasoning upon alleged proofs of subsidence must ne-
cessarily depend much upon the reliability of exact hypso-
metrical data, as to the relation of the level of the sinking or
- sunk deposits to the mean level of the adjoining sea. Obser-

a,

OSSIFEROUS CAVES OF GOWER. 511

vations of this description are of great delicacy—they de-
mand considerable time and repeated local inspection. In
short, they are not of the nature that lie conveniently to the
hand of a non-resident paleontologist; and in the remarks
which follow, when the elevation of the deposits is in ques-
tion, the figures will refer to the height above the platform,
at the foot of the cliffs, which is dry between high and low
water; in other words, to the subjacent beach in the ordinary
sense of the term.

(a.) Description.—Two years ago, a small black opening,
looking in the distance not much bigger than a fox-hole, was
visible from the rocky platform below, in the cliff, about mid-
way between ‘ Bacon Hole’ and ‘ Minchin Hole.’ It was quite
inaccessible by ordinary climbing, and was known to the
quarry-men by the name of ‘ Bacon’s Hye.’ The lower terrace
of the cliff, which here faces the sea, rises perpendicularly about
90 feet above the rugged reef which forms the beach. The
limestone strata of the cliff are intersected by vertical chines,
which widen and contract irregularly at intervals, from the
step of table-land above to the base, and are continued out-
wards as clefts on the denuded platform that stretches under
the sea. The angular aperture, called ‘Bacon’s Hye,’ was
situated about 72 feet high, in one of the largest of these
fissures, and it did not exceed 24 feet in diameter. Below it,
and wedged in between the walls of the fissure, there was
seen a great bulging vertical mass of angular fragments of
limestone impacted in ochreous loam, and resting on the
projecting ledge of a thick and solid mass of cemented blocks
and breccia, which subsequently proved to be the dividing
floor of a two-storied cavern. In the ochreous mass, white
fragments of bone could be distinguished by their bleached
colour from below, at a distance of 56 feet, jammed by, and
irregularly intermingled with, the fragments of the breccia.
On clambering up to the base of the ochreous mass, by a
devious ascent apart from the main fissure, we found our
further progress towards the ‘ Hye’ aperture intercepted by a
perpendicular fall of about 20 feet. The remains, seen from
below, proved to be fractured bones and teeth of Bos, bones
and antlers of Deer, and teeth and phalangeal bones of Wolf,
confusedly disposed in every direction, together with land-
shells cemented by calcareous infiltration. Stimulated by .
these appearances, Colonel Wood determined to follow up the
indications; the contracted aperture above was at length
reached by ladders and widened, when it was found to lead
into a fine cavern full of objects of interest. In order to make
it accessible, and carry on the operations within, it became
necessary to remove the external accumulation of loam and

512 OSSIFEROUS CAVES OF GOWER,

breccia in front of and below the mouth. The projecting
ledge upon which it rested was undermined, and the whole
was precipitated in mass into the abyss, where it soon disap-
peared under the action of the waves at high water. The
mass measured about 16 feet in height; the width was 44
feet at the top, 8 feet in the middle, and 7 feet at the base,
where the depth from the fissure, forwards, amounted to 8
feet. The rock on one side extended beyond the opposite wall
of the fissure a distance of about 50 feet seaward. A streak
of the loamy breccia, continuous with the mass in situ, was
traceable along the projecting side for a distance of about 30
feet, indicating that the progressive encroachments of the sea
must have undermined part of the cliff, and swept off large
masses of the superincumbent breccia.

To avoid the risk of confusion betweén ‘ Bacon Hole’ Cavern
and ‘ Bacon’s Hye,’ I suggested for the new cavern the name
of ‘ Bosco’s Den,’ by which it is designated throughout in the
present communication. The mouth opens upon the sea, to-
wards the SSE., and the cavity extends backwards a distance
of 76 feet, with a slight alteration of direction near the
middle ; the outer division measuring along the western wall
30 feet, and the inner, which is deflected a little to the E.,
46 feet. At the entrance, the width of the fissure near the
middle is 74 feet; it soon expands to 14 feet, and retains this
size with considerable uniformity throughout, the greatest
width being about 16 feet, and the minimum, where contracted
near the end, being 9 feet. The same remark applies to the
height, which is mostly about 14 or 15 feet. The walis of
the fissure are also very regular; there are no lateral offsets,
nor cul-de-sac expansions. The western wall throughout is
much more vertical than the eastern, which leans to it, and
near the shaft or flue in front of the angle of the bend it is
nearly perpendicular. The walls meet at the apex so as to
form a kind of Gothic roof, of which a narrow line of fissure
forms the ridge-pole. The gable end near the extremity has
the slopes inclined at a high pitch. The eastern wall, a little
behind the middle, shelves out so as to form a nearly hori-
zontal or but slightly inclined ledge, which is interrupted
near the great flue, and then re-appears on the western side,
in the anterior division of the cavern. The eastern or more
- inclined wall is sheeted over by a thin coat of stalagmite, of
which but little or none is seen upon the western wall. The
fissure of the roof, a little in front of the angle of bend, is
continued upwards in an irregular shaft or flue, which evi-
dently communicated with the surface. It was choked up
with detritus of the same nature as the upper deposits of the
floor, the surface aperture being covered over with a coating

-

OSSIFEROUS CAVES OF GOWER. 513

of turf. The height of the cavern immediately under this
shaft is 40 feet. The inward termination is rounded, but
abrupt, the end being composed of a great mass of undisturbed
stalagmite, which blocks up the continuation of the fissure.
At 9 feet from the end the roof still maintains an elevation of
16 feet.

(6.) Section of Floor Deposits—The mouth of the cavern is
now laid open to a depth of 22 feet, Colonel Wood having ex-
cavated down to the fundamental stratum of blocks and breccia
through 19} feet of deposits, below the ‘Hye’ aperture of 24
feet. When originally opened, the floor was tolerably smooth,
and shelved down gradually from the mouth to the extremity,
causing the section to exhibit a greater depth of deposits
outwards, and gradually decreasing inwards. Colonel Wood,
who daily superintended the progress of the operations, drew
up an account, at my request, which I have incorporated
with my own observations. The following deposits were
observed :—

1. On the surface a bed of sandy peat, stretching all the
way back from the bend to the extremity through 46 feet,
and varying in thickness from 12 to 15 inches. The peaty
substance consisted of bits of sticks and decomposed and
comminuted vegetable matter, freely mixed with coarse
reddish sand, but maintaining a fair amount of cohesion. In
front of the bend, this bed gradually thinned off as it rose on
the slope towards the mouth, where it disappeared entirely.
The greatest accumulation was immediately under the shaft
in the roof, where it formed a flat conical pile like a grain-
heap. The materials were here loosely aggregated, and filled
several barrow loads when wheeled out by the workmen. It
is deserving of notice that, notwithstanding the fissure cha-
racter of the cavern roof, this stratum of peat was free from
obvious moisture. This is explained by the unusual circum-
stance, that the interior in its existing condition is perfectly
dry and free from drip, notwithstanding the great amount of
stalagmitic deposition at the next lower level. In this peaty
stratum there were found, either embedded or lying on the
surface, bones of Bos and Wolf, and bones and fragments of the
antlers of Deer. Upwards of thirty well-marked specimens,
comprising the basal portions, with the bur, and more or less
of the beams of antlers belonging to the species or variety
allied to the Reindeer, and known under the name of Cervus
Guettardi or priscus, were exhumed from the peat, together
with the cerebral part of one cranium of the same form. It
is deserving of notice that they were all shed horns, no por-
tion of the frontals having in any case been found attached
to them. The same remark, as a general rule, but with a

VOL. II. LL

514 OSSIFEROUS CAVES OF GOWER.

percentage of exceptions, applied to the enormous quan-
tity of antlers of the same species, which were met with in
the lower deposits. The bones and antlers occurring in the
peat, as might have been expected, were fresher, and in a
better state of preservation than those found in the loam below.
The source of this peaty stratum is evidently traceable to the
shaft in the roof, with which all the observed conditions were
in direct and obvious relation.

2. Immediately. underlying the peat there occurred a very
regular and uniform bed of stalagmite ranging in thick-
ness generally from 6 to 9 inches, but in some places attain-
ing 1 foot. It formed a slightly undulated sheet, extending
from the mouth to the bottom of the cave. This was the
only layer of stalagmite encountered in the section. I ob-
served one phenomenon of great significance in relation to
this deposit. Close to the eastern wall, near the angle of the
bend, the stalagmite attains a thickness of 12 mches, and
suddenly rises into an obtusely conical boss, measuring 2 feet
3 inches in height by 2 feet 6 inches of length at the base,
and with a width of 1 foot 6 inches. It is free from adhesion
to the wall of rock. This boss is rent through vertically, and
shivered in every direction, with partial dislocation of the
fragments. But the remarkable circumstance is, that the pieces
are loose and uncemented, proving that notwithstanding the
great amount of drip which must have at one time existed to
form the stalagmitic floor, not a single drop of water charged
with lime in solution has fallen upon it, since the shock
which reft the boss into its present shattered condition. About
eight feet further in front there is another boss close to the
eastern wall, which is also shivered, but not rent to the same
degree. Constant Prévost and Desnoyers, when arguing
against the sole agency of Hyzenas and other Carnivora, in
explanation of the common occurrence of bones in caves,
have earnestly insisted upon the great amount of change
which must have taken place, after upheaval and subsidence,
in the course of subterranean rills. The uncemented con-
dition of the shivered ball in ‘Bosco’s Den’ isa striking
illustration of the truth of what they inculcated.

3. Below the stalagmite was a bed of ferruginous sandy
loam, containing dispersed angular fragments of rock, mea-
suring about 1 foot 4 inches in thickness. Colonel Wood
informs me that very few, if any, bone-remains were found
in this stratum.

. 4. Next, a deposit of sand about 2 feet 6 inches in depth,
without bones.

5. Below the sand, a bed of loose angular breccia. Few or
no bone-remains were met with in this bed, which measured
4A feet.

i

OSSIFEROUS CAVES OF GOWER. 515

6. Next, a bed of cave earth, or yellow ochreous loam 6 to
7 feet thick, containing sparsely disseminated angular frag-
ments of limestone, and occasionally a large block. This
extended down to the cemented fundamental stratum of
blocks and breccia, forming the diaphragm between the
upper and lower stories, and upon the ledge of which the
ochreous bone-breccia rested that was precipitated into the
chasm below.

The above may be taken as fairly representing an average
section of the deposits. But they varied a good deal in re-
lative thickness, according to the distance from the mouth
where the section was taken. Near the entrance, and for a
stretch of 26 feet inwards, what remains of the deposit upon
the walls exhibits nearly throughout, from the stalagmite
bed down to the bottom, a rich yellow ochreous loam or cave
earth, interspersed with small fragments of limestone. In-
wards from the bend, towards the extremity, the loam is
intermixed with a gritty incoherent sand ; but the excavations
have not been carried out so deeply here. Near the end, the
section, when I last visited ‘ Bosco’s Den,’ extended only to a
depth of 24 feet below the stalagmite floor, the workmen
having been afraid of the substratum giving way, and pre-
cipitating them into an unknown chasm below.

(c.) Organic Remains — Singular abundance of shed Deers’
Antlers.—Taking into account the close proximity of ‘ Bosco’s
Den’ to ‘Bacon Hole’ on the one side, and ‘ Minchin Hole’
on the other, and the very considerable aggregate formed by
its deposits, the association of animals found in it is not a
little remarkable. Not a single specimen referable to any
species, either of Elephant or Rhinoceros, was ever met with
in ‘ Bosco’s Den’ by Colonel Wood. Teeth and bones of
Hyena spelea and of the Cave Lion were equally wanting ;
so also were teeth of Kquus, which are excessively abundant
in Paviland and ‘ Spritsail-Tor.’ Bones of Bear were fre-
quent; those of Wolf very abundant, together with bones of
Fox, Deer, and Bos. But the most singular circumstance
was the extraordinary abundance of the antlers of Deer. I
passed through my hands, in the collection at Stout Hall, no
fewer than 750 distinct antlers, belonging either to varieties
of Reindeer or to a species (Cervus Guettardi) very closely
allied to it. I afterwards saw in the cave about 50 more,
and Colonel Wood has by subsequent acquisitions raised the
number to somewhat short of eleven hundred. To eguard
against error, no specimens were taken into the reckoning
but such as presented the base of the beam and bur. Asa
general rule, they were of small size, more or less broken,
and the great majority of them belonged to young animals.

LL 2

516 OSSIFEROUS CAVES OF GOWER.

Another fact was observed, which is worthy of notice, when
speculating as to the agency by which they were introduced
into the cavern, viz., that about 95 per cent. of these antlers
were shed horns, and some of them were very much rolled.
I believe that the above reckoning falls very much short of
the actual number of antlers that were present in the fissure ;
and that a large number of them were probably contained in
the mass of ochreous loam, 15 feet high by 8 feet square at
the base, which was precipitated bodily into the sea.

9. ‘Bosco’s Den,’ Lowrr Story.

(d.) Description of lower chamber.—We were well aware of
the existence of some kind of cavity under the projecting
ledge of cemented breccia upon which the ‘ cave-loam ’ rested,
but it was quite inaccessible from above; and no attempt
was made to penetrate it, until the excavations in the upper
story were nearly completed. At length Colonel Wood, Mr.
Thomas Falconer of Usk, and myself, having brought the ne-
cessary appliances to the spot on September 14 last, accom-
plished our first entry into the interior. By ladders we
ascended between the vertical walls of the empty fissure to a
contracted platform, which led back into a washed out cham-
ber, extending 31 feet inwards from the mouth, and imme-
diately under the floor of the upper chamber. The width of
the chasm at the mouth was 8 feet 9 inches, which increased
inwards to 10 feet, and the height of the chamber to the
projecting ledge of cemented breccia which forms the roof
was 12 feet; chinks intervened between the jammed blocks
of limestone, forming the floor; and by dropping a line, its
elevation of floor was found to be 24 feet above the edge of
the beach. ‘The height of the chamber diminished inwards,
by the inclination of the roof; at the extremity, it was reduced
to 94 feet. Here we got a section of the compact bed of
marine sand and gravel, which forms the end wall, breached
by the sea at high tides, and during heavy south-westerly
gales. In the descending order, it is composed of :—

(e.) Deposits of Marine Sand and Gravel
Lee lial

1. A layer of coarse rolled gravel 2 0
la, A thin layer of pebbles 4
2. Fine comminuted gravel 2
3. Sand mixed with fine gravel . 2 4
4. Coarse rolled gravel wee wo

Bi

No shells or other organic remains were observed either
in these beds or in the chamber. But no very close search
was made for them at the time, and the section was taken

OSSIFEROUS CAVES OF GOWER. 517

from the surface of the bank, without attacking it with the
pick-axe. The deposit bore all the characters of a bed of
compact marine sand and gravel.

_ (f.) Partition bed between upper and lower stories.—The
roof of the lower chamber was nearly flat, and consisted of
huge blocks of limestone mixed with smaller angular frag-
ments, the interstices filled up with gravel, and the whole
cemented by stalagmitic infiltration into a strongly coherent
mass. On the upper side ochreous loam replaced the inter-
stitial gravel. The projecting ledge of this bed was about
6 feet thick, and sloped upwards to the mouth of the upper
chamber, where the thickness was estimated to be about 14
feet. The materials corresponded in the general character
with the bed of angular debris observed by Mr. Prestwich,
on the raised beach of Mewslade Bay.

On the whole, ‘Bosco’s Den,’ of all the Gower caves,
furnishes the most complete succession of marine, brecciated,
and alluvial deposits, disposed in a section of not less than
47 feet.

10. ‘Bowrn’s Parnour’ or ‘ Devit’s Hons.’ Generan
Form AnD DimENSIONS.—UPPER AND LOWER CHAMBERS:
SWEPT OUT BY THE SEA.

The cave named ‘ Bowen’s Parlour,’ by Colonel Wood, is
known to the quarrymen of the neighbourhood by the name
of the ‘Devil’s Hole,’ from its singular and striking appear-
ance. I¢ is situated to the east of ‘ Crow Hole,’ and between
‘Crow Hole’ and ‘ Bosco’s Den.’ Geologically regarded, it is
one of the most interesting and instructive of the series,
inasmuch as it is washed out, thus presenting as it were the
skeleton framework of ‘Bosco’s Den’ denuded of its deposits.
Like the latter, it is situated in an angular fissure, narrow at
the top and gradually expanding below to a width of 14 feet.
The cave, which is about 40 feet high, is divided horizontally
by a thick cake of stalagmite, concreting brecciated fragments
of limestone into a solid diaphragm, so as to form an open
chamber of about 20 feet high at the mouth, and extending
back 53 feet, for the upper division; and under it, a lower
chamber of about 14 feet in height. At high tides and with
heavy south-westerly gales, the waves beat freely up into it,
and the lower story is completely washed out; the outer
thick part of the diaphragm remains forming a solid arch,
but the reverberating action of the waves has operated with
such force against the back part, that the central portion of
the partition has tumbled in, leaving a gap in the floor of
the upper chamber 16 feet long. A fine section of the calca-
reous breccia is thus exposed, showing that it was deposited

518 OSSIFEROUS CAVES OF GOWER.

upon a bed of marine sand; thin tabular aggregations of
sandy matrix adhere to the lower surface of the diaphragm,
penetrated by attenuated irregular septiform plates of iron-
tinted calcareous matter, forming a rude cancellar appearance.
The sand is also permeated by calcareous infiltration. The
upper chamber is likewise washed out, back to the extremity,
and the floor is so free from any overlying matrix that it
was compared by one of the quarrymen employed to a swept
parlour floor. The denuding action of the sea is still in
progress, and at no very distant period the segment of the
horizontal partition now remaining will have been swept
away. The rocky floor of the chasm is overlaid by a massive
and stream-like sheet of stalagmite, which dates from the
emptying out of the lower chamber, and is now in progress
of formation. As above described of ‘ Bacon Hole, a con-
tinuation of the cemented breccia can be traced upon the
face of the cliff beyond the mouth, to a distance of many
feet. The same phenomenon is repeated in ‘Crow Hole.’
The breccia corresponds in the character of the materials, as
I have already remarked, with the angular brecciated debris,
which overlies the raised beach, in the ‘ Mewslade Section’
given by Mr. Prestwich. The penetration, so to speak, of
this angular debris into the interior of so many of the Gower
caverns, and its extension upon the face of the cliffs beyond
their mouths are not a little remarkable, implying clearly
something more than a mere local cause for the phenomenon.
So far as I am aware, no such developed beds of breccia are
met with in the caverns in the mountain limestone of the
south coast of Devonshire.

I regard ‘Bowen’s Parlour’ as an emptied repetition of
‘ Bosco’s Den’ or ‘ Raven’s Cliff: ’ namely, that the bottom
of the cave was originally filled with marine sand or gravel ;
that when this deposit was elevated above the reach of high
water, stalagmite was formed upon the surface, enveloping
the angular debris thickly strewed upon the sand ; that upon
the flooring so formed, cave-ochre or some of the other
common alluvial materials, occurring in the contiguous
caves, were injected through the fissure above, into the upper
chamber ; and that at last a converse action, of comparatively
modern date, took place, by which the level of the cave was
depressed, so that it was emptied of its contents both above
and below the partition by the action of the sea. For
reasons which will be adduced in the sequel, it is not pro-
bable that this effect could have been produced by the
gradual encroachment of the sea upon a receding cliff which .
maintained a constant level.

No organic remains of any description were detected in
‘ Bowen’s Parlour.’

OSSIFEROUS CAVES OF GOWER. 519

11. ‘Crow Hour’ Cavern.—DEscrRiIPpTION AND REMAINS:
IMPERFECTLY KNOWN.

This cavernous fissure, named ‘Crow Hole’ by Colonel
Wood, is situated like the two last between the great
caverns of ‘Bacon Hole’ and ‘Minchin Hole,’ and to the
westward of ‘Bosco’s Den.’ You descend from the plateau
of highland surmounting the cliffs, for some distance, by
the track leading to ‘Minchin Hole,’ and then go straight
down by a very precipitous and perilous scramble on the
erags, to about 60 feet above the level of the sea, to be
landed on a rugged floor, something like that seen below the
‘Yellow Top’ cliff at Paviland. This leads to a ravine with
diverging walls, the bottom of which is not roofed over, but
presents a cliff about 43 feet high, composed of limestone
breccia, chiefly small angular fragments, mixed with some
large blocks. Below the breccia there is a confused mass of
stalagmite, then ferruginous sand and gravel, again succeeded
by breccia, resting upon a thick bed of coarse sand, of which
a depth of 4 feet is exposed, free from admixture with stones.
The width of the ravine where the cliff of cemented breccia
terminates is about 27 feet, and a distinct line of the same
materials can be traced upon the rock a considerable way
out, as at ‘ Bosco’s Den’ and ‘ Bowen’s Parlour.’ Colonel
Wood has exhumed from these deposits bones of Ursus,
Badger, Rhinoceros, and some other remains, but the locality
has not yet been sufficiently explored, and my acquaintance
with the details is at present too imperfect to warrant my
giving any further description of it; more especially as
there is much complexity in the intercalation of the beds of
sand and breccia, demanding careful examination. The
angular breccia bed is here developed in great force; the
section of it in the cliff above the ferruginous sand exhibiting
a depth of 18 or 20 feet.

12. ‘Raven’s Curr’? Cavern.—DEeEscriIpPTion: FLOOR-
BrEps, AND REMAINS.

The cavern which he has called ‘ Raven’s Cliff’ has
been the latest field of Colonel Wood’s excavations, and it
gives promise already of being one of the most productive
in species of all the caves in Gower. It is situated in
the cliff facing the sea to the west of ‘Minchin Hole,’ and
between it and ‘Shire Coom.’ I examined the locality
last September, in company with Colonel Wood, when it was
intact. The cliff presented the section of a cavernous fissure
completely crammed with sea sand in the same plain with
the general face of the scarpment, and affording no indica-
tion whatever of an opening. Below and in front of it there

520 OSSIFEROUS CAVES OF GOWER.

was a great talus of sand accumulated upon enormous masses
of concrete breccia lying huddled in great confusion.
The sand was removed ‘in order to ascertain whether it
was possible to penetrate into a cavern.’ The operations
were commenced in January last, and I am indebted to
Colonel Wood’s letter for the details which follow. After
clearing away the bank of sand in front, a thin stratum of
stalagmite was discovered stretching right across the fissure
from wall to wall, and close up to the roof, there not being
more than a foot of space intervening. They were occasion-
ally formed by connecting pipes of stalactite. In this con-
tracted channel two lower jaws of Mustela foina, decidedly
new to the English fossil Fauna,' together with some fish-
bones, were found resting upon the stalagmite; also several
jaws of Foxes with some bird-bones. On pushing the ex-
cavation further in, a greater depth of sand was attained,
amounting to 4 feet, when several very large balls of coprolites
were encountered, indicating the presence of Carnivora. The
following day some fine remains of Felis spelea were dis-
covered: among which were a fragment of the right upper
jaw containing the carnassier and adjoining premolar in situ,
also a very perfect upper canine, 5 inches long, some de-
tached premolars, and two incisors. Proceeding inwards and
still deeper, the lumbar vertebra of a Rhinoceros turned up,
and close by it the vertebra of a fish, both encrusted with
ferruginous sand, and in the same mineral condition. On
digging down through the bed of sand, which attained an
extreme depth of 9 feet, a second floor occurred, composed, as
usual in the Gower caves, of angular fragments of limestone
cemented together by calcareous infiltration, irregularly in-
terspersed with concrete and detached pieces of stalagmite,
and occasionally discoloured by ferruginous sand. At the
bottom of the sand and resting upon this brecciated floor, a
large block of limestone was laid bare, displaying the ex-
posed surface and edges smoothed and polished by friction.
The carboniferous limestone of Gower weathers everywhere
into a rough honeycombed or unequal surface; and Colonel
Wood infers that the polish upon this block was caused by
the rubbing against it of large Carnivora, as in the pedestal
of Zahnloch described by Goldfuss.? About 2 feet further in
and beyond this block, the perfectly entire and anchylosed
radius and ulna of a Rhinoceros were exhumed. The entire
humerus and several vertebre of an Elephant, together with

> These specimens of Mustela were | Putorius, and with Zorilla Capensis and
carefully compared in the British Mu- | Mephitis Americana.—[ Ep.
seum with Mustela martes and Mustela 2 Cited in Buckland’s ‘ Reliquize Di-
zibellina, also with various species of | Inviane,’ p. 132.

OSSIFEROUS CAVES OF GOWER. 521

a fragment of a tusk, and remains of Bos and Deer, were also
met with in the same deposit.

On clearing out the sand, which at the back part of the
cave decreased in thickness from 9 to 3 feet, the deposit
under the floor of concrete breccia was next examined. The
breccia was found to vary from 9 inches to a foot in thick-
ness ; and, as usual in the other Gower caves, the traces of a
continuation of it were seen extending seaward on the face of
the cliff, many feet beyond the mouth of the fissure. The
stratum below consisted of a very hard, dark-grey, gritty
sand, attaining a maximum thickness of about 8 feet.

At present, in its excavated state, ‘ Raven’s Cliff’ Cavern
exhibits a fissure, about 20 feet high at the entrance, and 29
feet in width. The walls gradually converge inwards, and
the roof declines in the same direction, so that the extremity
terminates in an angular contraction about 43 feet from the
entrance, where the aggregate thickness of the deposits
amounts to 184 feet.

13. PaviLAND CAVE.—SKELETON OF ‘RED LADY’ AND
HLEPHANT REMAINS.

The two cavernous fissures on the face of the cliff,
below the very bold and grand escarpment which is called
the ‘ Yellow Top Rock’ are so well known, under the
name of the ‘ Paviland Cave,’ through the description
given by Dr. Buckland, that on the present occasion I
shall confine my remarks to a single point, namely, the
human bones found in the cavern, and their presumable
relation to the Hlephant remains with which they were
associated. An imperfect female skeleton, minus the skull,
vertebre, and extremities of the right side, but with the
remaining parts asserted to have been found lying ‘ extended
in the usual position of burial,’ was discovered beneath a
shallow covering of nearly six inches of earth, on the floor
of the eastern cavern called the ‘Goat’s Hole.’ The bones
were stained red, by a kind of ruddle, ‘composed of red
micaceous oxide of iron,’ in consequence of which the skele-
ton was distinguished in Gower by the name of the ‘Red
Lady of Paviland.’ None of the fossil or other bones found
in the contiguous deposits were stained in a similar manner.
But numerous fragments of slender and smooth cylindrical
ivory rods, and some portions of ivory rings, supposed to
have been bracelets, also stained red, were exhumed in close
contact with the skeleton. They were in a state of complete
decay, brittle, and so tender as to be readily cut by the nail.
In an adjoining bank, at a higher level, two Mammoths’
molars were discovered, which I have examined, in the

522 OSSIFEROUS CAVES OF GOWER.

collection of Miss Talbot, in Penrice Castle. They are very
pronounced and characteristic specimens of the true Mam-
moth, H. primigenius. At a lower level, and at no great
distance from the bones of the ‘ Red Lady,’ a very consider-
able portion of the skull of an Elephant was discovered,
comprising the sockets of the two tusks and a portion of one
tusk two feet long. The tusk, like the stained rods, was in
a state of advanced decay, and very brittle. Dr. Buckland
inferred that the cylindrical rods, rings, &c., were ‘ certainly
made from part of the antediluvian tusks that lay in the
same cave.’ If this assumption were well founded, there
would be no escape from the conclusion, that the ‘ Red
Lady’ and the Mammoth had been contemporaneous or
nearly so, for the ivory must have been fresh and hard when
the rods and rings were shaped out of it. But similar rods
and rings are met with in the barrows; it is known from
history! that the ancient Britons imported them from France,
and it seems more probable that the ivory ornaments of the
‘Red Lady’ were of the imported class, than that they were
fabricated out of the Mammoth tusks that lay in the cave
deposits.

14. Spritsatt-Tor CAvERN.

The double-mouthed cavern named ‘ Spritsail-Tor’ is situ-
ated on the west side of Gower, facing Carmarthen Bay,
and on the northern flank of the axis of Old Red Con-
glomerate. The carboniferous limestone strata have been
denuded here to a narrow band, and the cave occurs in
the projecting point, forming the northern boundary of
‘Broughton Bay,’ overhanging the Whiteford Sands. This
cavern differs from all those on the south coast, in having its
mouth’s opening near the very summit of the cliff, which,
together with a considerable portion of the crowning plateau,
is buried under an immense accumulation of blown sand,
that rises into hillocky downs in the ‘ Whiteford Burrows.’
It was accidentally discovered in 1839, through the cutting
back of a quarry in the carboniferous limestone.? On remoy-
ing the sand a concealed cavern was discovered, which was
partially explored at the time by Sir Henry de la Beche, and
thoroughly opened in 1849 by Colonel Wood, who, during
the progress of the operations, detected the second opening.
The two apertures, which are low and angular, are about 13

1 «They (the Britons) pay but mode- | of Strabo,’ translated by H. C. Hamilton
rate duties on the imports and exports | and W. Falconer (1854), vol. i. p. 298.
from Kettica, which are ivory bracelets 2 De la Beche, ‘Geological Observer,’
and necklaces, amber, vessels of glass, | p. 308.
and small wares, &¢.’—‘ The Geography |

OSSIFEROUS CAVES OF GOWER. 523
feet apart, and connected by a contracted passage parallel to
the face of the cliff. The principal or eastern aperture is
irregularly vaulted, expanded at the mouth, but suddenly
contracting inwards into a shallow chamber of very limited
extent.' There is still a great amount of stalactite drip from
the roof, and the floor deposits were in relation to this con-
dition. On the surface there was a layer of loose blown
sand a foot and upwards in thickness, which lay upon a floor-
ing composed of several distinct strata, throughout which
bones, teeth, and splinters of the bones of various species of
Mammalia were firmly cemented, forming a compact osseous
breccia. Beneath the stalagmite, the irregular fissure of the
rocky floor was impacted with cave earth or yellow ochreous
loam, containing an immense accumulation of embedded
bones. Among these were molar teeth, milk or adult, and
other remains of Elephants, including both Hlephas antiquus
and Li. primigenius; bones and teeth of Rhinoceros tichorhinus,
Equus, Sus, Bos, Cervus, Lepus, Arvicola, Ursus speleus, U.
priscus, Felis spelea, Canis lupus, C. vulpes, Meles Taxus, and
Mustela. Detached carnassiers, premolars, and canine teeth
of Hyena were very numerous, together with fragments of
lower jaws. Coprolites of the same species were met with.
Bones of the extremities of Bos, Equus, and Cervus, broken
up into massive gnawed splinters, exactly like those figured
by Buckland, were found in large quantities. Detached
molars of Hquus and Bos were also excessively abundant, a
sure sign of the den of Hyena spelea. In short, the con-
tents of the cavern bore all the characters of a lone-tenanted
Hyzna’s den. It is the most perfect illustration of the kind
that I have seen, either in Gower or elsewhere among
British caverns.

The contents of the chambers communicating with the
two apertures were very much alike. When the western
opening was laid bare, a fine specimen of the antler of a
Reindeer presented itself, partly protruding through the
sand, but lying free upon the stalagmite floor.’

In the eastern chamber, the entire femur of a young
woman, together with the lower jaw of a child of about seven

1 «We were much struck by the nar- |
rowness of a part of the entrance, where
predaceous animals, apparently Hyznas

an antler of Reindeer in Tor Hole; cir-
cumference of base 5} inches. Beam,
18 inches long—retaining 43 inches of

(A. spelea) seem to have been stopped
with large portions of the carcasses of
Rhinoceros tichorhinus, numbers of the
teeth of which, among the other remains,
were accumulated close outside it..—De
la Beche, loc. citat.

2 Extract from Dr. Falconer’s Note-
book,—‘ Major Wood found portion of

brow-antler and 2 inches of bez-antler.
The base of this horn and brow-antler
were embedded in stalagmite, while the
upper fractured extremity protruded
through the cavern layer of sand, and a
considerable portion of the beam was
exposed. It was in a highly decomposed
state. —[Ep. |

524 OSSIFEROUS CAVES OF GOWER.

years, were discovered embedded in the sand.! The cave
loam yielded a prodigious quantity of the remains of two or
more species of Arvicola, and other minute Mammalia.
Among the rodents in this case I observed that lower jaws
predominated very largely in number over crania and upper
jaws, in the ejected materials which strewed the face of the
talus below the aperture, and which still remained after ten
years of exposure to the weather. The exact elevation of
Spritsail-Tor above highwater mark has not yet been pre-
cisely ascertained. No trace of marine sand or gravel was
detected. The absence of arenaceous deposits below the
stalagmitic floor is not a little remarkable, considering how
liable the chambers are to be filled with sand at the present
day; and the circumstance indicates a very considerable
change in the physical conditions around the cavern since
the date when the stalagmitic flooring was formed. Although
molar teeth of R. tichorhinus are abundant in ‘ Paviland’
and ‘ Spritsail-Tor,’ no remains attributable to R. hemitechus
were identified among the contents of either cave. One very
finely preserved upper molar of Hlephas antiquus was ex-
humed from the cave loam of Spritsail-Tor, together with
some milk molars of the same species. It is worthy of
remark, that the molars of H. primigenius, from the same
cavern, the distinctive characters of which are strongly
pronounced, commonly present a much fresher appearance
with more animal matter retained in them.

15. GENERAL REMARKS ON THE DISTRIBUTION OF THE
MAMMALIA IN THE DIFFERENT CAVERNS.

This concludes the description of the ossiferous caves. I
append a tabular statement showing the genera and species
of Mammalia which have been discovered in them. The list
is remarkable for the great numerical richness of the species.
With the exception of the Drepanodon (Machairodus) of Kent’s
Hole, it includes all the larger-sized carnivorous and her-
bivorous species that have been met with throughout the
caves in England, while two of the species, namely Hlephas
antiquus and Rhinoceros hemiteechus, have not been heretofore
described as occurring in any of them. Of the smaller-sized
genera, it contains representatives of every one, with the ex-
ception of the Lagomys of the South Devonshire caverns,
and of Spermophilus, the presence of which I have lately
ascertained in the collections from the caves in the Mendip
hills. The latter had not been known before as occurring
in the Fossil Fauna of Britain.?

‘ On the authority of Mr. A. Pytts | tified by Professor Owen when sent to

Faleoner and Colonel Wood, who in- | the Royal College of Surgeons.
form me that the bones were thus iden- * See antea, p. 452.

OSSIFEROUS CAVES OF GOWER.

In the following Fable, 1 indicates that fossil was present ;

525

x, that it was abundant ;

and o, that it was absent.

a 8 3 2 = é iC)

ela pae cab me | ais

an | fq A =I a isa}

S 8 5 3 & s bo

S| GEES as aS a

fA 9 a | 0 ia AY D 4

1. Ursus speleus . ; - af 5s I) Pes || Fal 1 ies ae te |
2. U. priseus ? : F : On exon | Ones on|eon in L 0
3. U. Arctos . c : : CH Ort Gl os |) ole ah sas
4. Meles Taxus. ; - By enall OnE Owe alent On pexn eel:
5. Putorius vulgaris : c NON oO | Vol ea | oi) Gj al 1
6. P.ermineus . : . fale. L Oth Wil | Ce @d| Ol)
7. Mustela foina? . : ; oC Orr)’ 2k SR oa cee aa Te ee ott Sea en
8. Lutra vulgaris . - ; PAO SO Ol Ol Ol Oj) *@a} al
9. Canis lupus. : : pal ea ha. ealtans-e fe IE Il" Wi cb ee Meet
10. C. vulpes : : i IU Mabe silos tLe l\ coil lie Seal ois p oa
11. Hyena spelea . ‘ : ol Sa iemees dhe to Wo MiP Sci ge Ile ce
12. Felis spelea . : : Sih all Cr Oo) |p @ | vil 1
13. F.catus . - : : = OM On [ee On rola ml ol @ | il
14. Arvicolaamphibius . : el (ee (nee Sal ewe: S awa eS) ts. I Mb Se
15. Lepus cuniculus : : i SOA OP Cy oy oh ey Gf Il
16. L. timidus . : - Oy] YN oy amo) IP xo |) oy I] 1
17. Elephas antiquus .. : MELO scout LS xc Or (femelens fae
18. E. primigenius . : : Ol) 0) Oi es] “oil seq) SE
19. Rhinoceros hemiteechus . B60 MON Nig. c 1/8 pallln | <a Me ototat lay ave) |! = l
20. R.tichorhinus . , é Om OSI On| eco ln liege Od ax |) exat |i
21. Equus Caballus. . . Oa rl al PP er eee |
22. E. Asinus . c . : OO OU Ose) | Gel) be A
23. Hippopotamus major : Ol Oo Moo Cel On| OM|eo
24. Sus Serofa : cel bel meet ile Bees |) Soe deh
25. Cervus eurycerus (syn. megacers) Ol) Oe ON OI Oo Weal ]
26. C. (Rangifer) Tarandus . 1 2S al Ossie Onl eexan la ]
27. ~—-var.(a@) C. Guettardi . Ong |e Sen ene Olea Online hat!
28 yar. (>) C. priscus ¢ Hee Oe | eal Ops OR | WO peel ] 0)
29 var. (¢c) C. Bucklandi . One lO) OssiearOy | aleeiee lteal aro
30. C. Elaphus : . tlc te |e Om [ecole On| anil alin tee Tea tee
31.C. Capreolus. - : a all TW oO.) Oi) ol Co)
32. C. (Strongyloceros) . é «(eel On ls Guile On lu Os enol ig: lanes
33. Bison priscus . ; : Bo eo ef ree La | eel gs cl ose a

1. Common in most of the caves;
wanting in Raven’s Cliff.

2. A very fine perfect lower jaw, right, wide
Cuy., Pl. 189, fig. 6, from M.H., with two premo-
lars close to canine, and narrow high coronoid.

8. Rare in the Gower caves generally ; found
in Spritsail- Tor (Paviland ?).

4. Found low in the deposit in Bacon Hole ;
very abundant in Spritsail-Tor.

5. In Spritsail-Tor.

6. In Bacon Hole.

7. ‘ Marten’ found in Raven's Cliff ; 2 lower
jaws on the Stalagmite, also in Spritsail- -Tor.

8. Found in * Long Hole,’—[ ED. ]

9. Very common in most of the caves.
p. 462.)

rare or

(See

10. Skulls abundant in the antler deposits in |

Bosco’s Den ; common generally.

11. Very abundant in Spritsail-Tor, Paviland,
and Minchin Hole ; comparatively rare or
wanting in the others.

12. Only found in Spritsail-Tor and Raven’s
Cliff ; Coprolites very abundant in the latter.

13. In Raven's Cliff above the Stalagmite.

14. Arvic. amphibius ; very common in all,
the smaller species not observed.

15, Found in ‘ Long Hole.’—[Ep.]

1 «Long Hole’ not discovered at date of paper.

16. Only 1 lower jaw of Hare found in Sprit-
sail-Tor ; and not looking very old,

17. Not found in Bosco’s Den or Paviland.

18. Common in Payiland and Spritsail-Tor ;
wanting elsewhere.

19. Wanting in Bosco’s Den, Paviland, and
Spritsail-Tor.

20. Very abundant in Paviland and Sprit-
sail-Tor ; wanting in the others.

Pall Very common in Paviland and §.-Tor.

22. Common in Spritsail—-Tor.

23. Very uncommon, and limited to a few
remains in Raven's Cliff.

24, Common in all the caves,

25. No antlers; but molar teeth in Spritsail-

| Tor only, and not abundant. =

26. Very abundant in Bosco’s Den.
27. Ditto ditto
28. Ditto ditto

- 29. In Bosco’s Den, Paviland, and Spritsail-
or.

‘i 30. In Bacon Hole, Paviland, and Spritsail-
or.
31. In Bacon Hole, Bosco’s Den, and Sprit-

sail-Tor.

32. In Spritsail-Tor.
33. Common in all the caves,

See page 538.—[Ep.]

526 OSSIFEROUS CAVES OF GOWER.

When the eye is cast over the contents of the different
columns, some remarkable points of contrast are observable.
Thus, in ‘ Bacon Hole,’ ‘Minchin Hole’ and ‘ Raven’s Cliff,’
Elephas antiquus and Rhinoceros hemiteechus occur in the two
first in great abundance; while in Paviland they are sup-
planted by H. primigenius and Rhinoceros tichorhinus; the
other associated genera and species being, as a general rule,
the same. It is further remarkable, that these localized
species do not ordinarily occur intermixed, the only well-
marked exception being in the case of ‘ Spritsail-Tor,’ where
both E. antiquus and EH. primigenius have been found. But
neither in Paviland nor in ‘ Spritsail-Tor’ was a single molar
or jaw fragment of Rhinoceros hemiteechws encountered, nor of
Rhinoceros tichorhinus in ‘ Bacon Hole’ or ‘ Minchin Hole.’
I passed all the Rhinoceros remains of the Gower collections
under a very close scrutiny, expressly with a view to the de-
termination of this point, and considering the abundance of
fossil teeth of both species, in the caverns in which they
respectively prevail, the contrasted distribution was very
strongly brought out. It should at the same time be remem-
bered, that ‘Spritsail-Tovr’ fulfils the conditions hypothetically
put by Mr. Prestwich in the appendix, in being a cavern,
situated near the summit of the cliff, and probably ata higher
level above the sea, when marine sands were laid upon the
floors of ‘Minchin Hole’ and ‘ Bacon Hole.’ Detached molar
teeth in great abundance and other bones referable to at
least two species of Hquus go in company with remains of
R. tichorhinus. They were very numerous in ‘ Spritsail-Tor,’
but rare in ‘ Bacon Hole.’ Their prevalence in the former
is explicable on the supposition, which all the conditions
confirm, of its having been long used as a Hyzena’s den: the
hard prismatic teeth of the horse having been rejected, when
the jaws that yielded them were crunched and devoured.

The distribution of Hlephas antiquus in England, and over
the greater extent of the European area south of the Rhine,
is now pretty well ascertained. I found it in very consider-
able abundance in the collections from the bone-caves in the
vicinity of Palermo, and also in some of the newer deposits
of the Valley of the Anapus upon the south coast of Sicily
near Syracuse.' In the environs of Rome it occurs in the
marine volcanic Tuffs, or volcanic gravels of Monte Verde,
Monte Mario, Tor di Quinto, Monte Sacro, La Magliana,
Ponte Molle, and other corresponding localities, in great
numbers. I observed in the Museum of La Sapienza, and
in the private collections of Professor Ponzi and Signor
Ceselli, of Rome, various undoubted molars of all ages of the

1 See antea, p. 188.—[Ep.]

OSSIFEROUS CAVES OF GOWER. 527

true H. primigenius, from Ponte Molle, Monte Mario, and
Monte Sacro, near the Ponte Nomentano. They were from
the same tufaceous deposits which yielded the remains of
Hi. antiquus, and the accuracy of the assigned localities was
placed beyond question, by the matrix which covered the
Mammoth molars presenting abundant grains of Pryroxene
and decomposed Amphigene which characterize the volcanic
Tuffs around Rome. H. antiquus occurs also in the Val
d’Arno and in the Valley of the Po. The range of the spe-
cies which I have named Rhinoceros hemitechus, both geogra-
phically and in time, has not yet been so precisely determined.
In England it occurs in the Clacton-beds so perseveringly
examined by the late Mr. John Brown of Stanway, at Folke-
stone, and in some upper tertiary deposits in Northampton-
shire. It is not a little singular that the prevailing fossil
Rhinoceros of Grays Thurrock, of which the remains are
found in very great abundance, belongs to a species which I
regard as being entirely distinct from the Rhinoceros hemite-
chus of Clacton and the Gower caves.

_ Of the large Carnivora, Bears, chiefly Ursus spelous, were
found in all those caverns. Remains of Canis lwpus were also
very abundant throughout; and in ‘ Bacon Hole’ there was
distinct evidence of their occurrence, in the cave earth below
the second stratum of stalagmite, mixed up with bones of
Hyena spelea, Rhinoceros hemitechus, and Hlephas antiquus.
Hyena spelea was most abundant in ‘Spritsail-Tor’ and
‘Paviland,’ but comparatively rare in ‘ Bacon Hole,’ ‘ Min-
chin Hole,’ and ‘ Raven’s Cliff,’ and entirely wanting in
‘ Bosco’s Den.’ In ‘ Raven’s Cliff’ the ulna of a Hyzena was
found covered with large superficial patches of the slender
burrowed canals, described by authors under the name of
Talpina, which has also been observed on fossil teeth from
Malta, and on the enamel of the molars of Dinotheriwm from
Miocene beds in India. Remains of Felis spelcea are rare in
the Gower caves; they turned up only in ‘ Spritsail-Tor ’
and ‘ Raven’s Cliff’ Meles Taxus occurred in‘ Bacon Hole,’
‘Spritsail-Tor,’ ‘Crow Hole,’ and ‘ Raven’s Cliff,’ and the
remains yielded several finely preserved crania and lower
jaws; in ‘ Bacon Hole’ traces of it were found, in the deeper
deposits, along with H. antiquus and R. hemitcchus.

Of Hippopotamus major a few teeth only were found, at a
low level in the sand beds of ‘ Raven’s Cliff,’ the most cha-
racteristic being a third milk molar closely resembling the
Kirkdale specimen figured by Buckland in the ‘ Reliquiz
Diluviane,’ Pl. XIII. fig. 7, one perfect premolar of an adult
and a portion of another, together with some nearly foetal
teeth. The occurrence of this huge extinct species of

528 OSSIFEROUS CAVES OF GOWER.

Hippopotamus within the precincts of Gower is not a little
remarkable, considering that at the present day there is
nothing resembling a lake or the relics of one, in the penin-
sula, and that the largest existing stream, namely, the
‘ Pennard,’ does not exceed the dimensions of a small brook.
There are no grounds to believe that the few remains of
Hippopotamus in the cavern of ‘ Raven’s Cliff’ were trans-
ported there from a distance, more than in the case of any
of the other large herbivorous animals found in it ; and the
difficulty of explaining how they came there is best ex-
pressed by stating, that the confined peninsula of Gower,
destitute either of lakes or of large streams of fresh water, is
one of the last localities where the remains of Hippopotamus
might have been looked for, consistently with what we know
of the habits of the living species, and of H. major, where it has
been found in the greatest abundance, namely, in the vicinity
of extensive lacustrine deposits, like those cf the Val d’Arno.

The most singular point connected with the distribution of
Fossil Mammalia in the Gower caves is the immense quan-
tity of Deers’ antlers referable chiefly to Cervus Guettardi,
which, as already stated, were encountered in ‘ Bosco’s Den.’
The fauna yielded by that cavern is more restricted in spe-
cies than that of almost any of the others, being confined to
species of Bear (Ursus speleus), Wolf, Fox, Arvicola, Bos, and
Deer, the last of which turned up in very great abundance.
Between one thousand and eleven hundred distinct antlers
were collected, the great majority of them shed, more or less
fractured, chiefly of young animals; some of them much rolled,
but hardly any presenting marks of having been gnawed.
Buckland inferred, that the cave of Kiiloch in Germany con-
tained the remains of 2,500 Bears of the species Ursus spe-
leeus, the calculation having been founded upon the quantity
of animal matter found on the floor of a single vault; and
last year I brought before the Geological Society proofs of the
enormous number of Hippopotami, within and outside the
mouths of some of the caves in Sicily. The antlers of ‘ Boseo’s
Den’ belong to the same class of extraordinary numerical ac-
cumulations. To appreciate to its full extent the singularity
of this case, it ought to be remembered that ‘ Bosco’s Den’ is
not an isolated cave, forming the sole receptacle for the re-
mains of a large district, but that it is in close vicinity to a
group of ossiferous caverns, like ‘ Crow Hole,’ ‘ Raven’s Cliff,’
‘Bacon Hole,’ and ‘ Minchin Hole,’ in none of which was
anything approaching the same number of antlers observed.
On the contrary, they were rare.

Of other Ruminants, remains of Bos (Bison) priscus were
abundant in all the caverns. They occurred in ‘ Bacon Hole,’

OSSIFEROUS CAVES OF GOWER. 529

along with Hlephas antiquus, Rhinoceros hemiteechus, Hyena
spelea, &c. in the cave loam at the lower part of the section,
and also at a low level in ‘ Raven’s Cliff.” No bone of
Bos (Ursus) primigenius was satisfactorily identified ; but the
comparison of the large Bovine remains has not yet been
sufficiently carried out to warrant any definite statement on
the point. Molar teeth referable to the Irish Elk (Cervus
eurycerus) were exhumed from ‘ Spritsail-Tor’; and antlers,
together with crania from which the horns had been shed,
of Reindeer, from the same cave and Paviland. Horns of
Cervus Hlaphus and C. Capreolus were encountered in the
superficial earth covering the floor of ‘Bacon Hole.’ Some
of them bore distinct marks of having been chopped rudely,
or partially cut.

Remains of Arvicola amphibius were met with in great
abundance, in ‘Spritsail-Tor,’ ‘ Bosco’s Den,’ and ‘ Bacon
Hole ;’ in the last at the very bottom of the floor deposits.
Detached incisors and molar teeth are seen attached to a
specimen comprising the radius and ulna of Rhinoceros hemi-
techus from the lower deposits of Raven’s Cliff, and encrusted
with the original matrix. Doubts have been raised whether
the presence of Arvicola amphibius along with extinct animals
in the caves might not be explained by the known burrowing
habits of the species. But the specimen in question clearly
indicates that it was a cotemporary of the Rhinoceros.

16. GrotogicaL AGE OF THE GOWER CAVES.

In the spring of 1858, on my first visit to Gower, I was
struck with the significance of the fact, that marine sands
containing common existing shells lay at the bottom of the
ossiferous deposits in ‘Bacon Hole’ and ‘ Minchin Hole,’
tending, so far as the limited evidence went, to assign a very
modern date in the Pliocene period, for their emergence
above the sea and subsequent filling up. On the other hand,
when I came to examine in detail the fossil bones, I was
equally struck with the fact that the remains of Hlephas anti-
quus and Rhinoceros hemiteechus, the latter especially, were
very abundant in both caverns; while not a single bone re-
ferable to HL. primigenius or R. tichorhinus could I detect in
either. The Gower species of Rhinoceros (R. hemiteechus) was
then new to me; but I found it associated with an Elephant,
which I had hitherto regarded as a characteristic form of
the ‘ Hlephant-bed’ that underlies the ‘ Boulder-clay’ on
the Norwich coast, of the Sub-Apennine beds of the Astesan,
of the ‘ Mud-bed ’ of Bracklesham Bay, and of ‘ Grays Thur-
rock.’ Here then were two lines of seemingly antagonistic
evidence: the marine sands and their contained shells, indi-

VOL. II. MM

530 OSSIFEROUS CAVES OF GOWER.

cating a comparatively modern date, while some of the mam-
malia tended to put the ossiferous beds of the caverns back
to a period not far removed from that of the Norwich Crag.
On examining the bone collections from ‘ Paviland’ and
‘Spritsail-Tor,’ the perplexity of the case was increased by
my observing among them conditions in some respects pre-
cisely the converse of what I had seen in ‘Bacon Hole’ and
‘Minchin Hole.’ Remains of R. tichorhinus were compara-
tively abundant both in ‘ Paviland’ and in ‘ Spritsail-Tor,’
while the species occurring in ‘ Bacon Hole’ and ‘ Minchin
Hole’ was wanting. On the other hand, remains both of Hle-
phas antiquus and EL. primigenius were encountered in ‘ Sprit-
sail-Tor,’ while, as a general rule, the associated herbivorous
and carnivorous genera and species were, with certain local
anomalies, the same throughout the caverns. Hence the
question was suggested: ‘Does this limited peninsula of
Gower comprise ossiferous caves referable to distinct geo-
logical periods ? or are all the mammalian remains contained
in them referable to one and the same period ? ”

On paying a second visit to Gower, in the autumn of 1858,
I became acquainted with the fact, that transported blocks
occurred on the highland of Cefn Bryn, and I observed beds
of gravel on the southern slope of the old red sandstone
conglomerate. JI became very desirous that these boulders
and gravels should be examined by some observer of au-
thority, having special familiarity with that walk of investi-
gation, and in discussing with my friend Mr. Prestwich the
various bearings of the Gower case, I impressed upon him
the importance of his paying a visit to the cave district. This
object he accomplished last autumn, and the results are
given in the notes hereto appended, with which he has
favoured me.

But in the interval between my first visit, in 1858, and Mr.
Prestwich’s late reconnaissance of the Gower district, I had
opportunities of examining the cave collections of the Mendips
in the Museum at Taunton and in Mr. Beard’s possession
near Banwell, the remains of the Devonshire caverns at
Torquay and Plymouth, the Durdham Down collection at
Bristol, the Cefn collection formed by the late Mr. Lloyd
and now in the possession of Colonel H. W. Williams Wynne,
and the Kirkdale collection at York, besides gleanings from
all these and other localities in the British Museum and in
the rich collection left by Buckland at Oxford. I had also
seen the indubitable remains of H. antiquus and EH. primi-
genius, associated in the beds of marine volcanic Tuffa, at
Monte Sacro and other localities in the environs of Rome.
The result was to prepare me for a great change of opinion

OSSIFEROUS CAVES OF GOWER. 531

and for the conviction, that notwithstanding apparent ano-
mazlies in the restriction of certain species to certain caves, all
the extinct mammalia of the Gower caverns might belong to
one Fauna, and to the same unbroken geological period.

There cannot be a doubt that the thin beds of marine sand in
‘Bacon Hole’ and ‘Minchin Hole,’ containing the association
of Intorina litorea, L. litoralis, and L. rudis, were deposited on
the floors of these caverns before they rose above the level of
the sea.!_ The ossiferous beds are in immediate and undis-
turbed superposition upon the sands, and it would seem clear
that they belong to the same period. The species of large
Pachydermata, Ruminants, and Carnivora, were living upon
the emerged land of Gower before the floor of the caves rose
above the level of the sea; but the great accumulation of
their bones now seen in the cave deposits took place during
a long lapse of time after the rise had been accomplished.

The raised sea-beach of Mewslade Bay, and the marine
sands and gravels of ‘ Bacon Hole’ and ‘ Bosco’s Den’ are so
nearly in the same line of section, and they differ so little in
level, that there is hardly any room to doubt that they belong
to the same series of deposits, and to the same period of
upheaval. The species of mollusca found in the raised beach
are more numerous than those that were detected in the
caves; but the three species of Litorina were common to
both, and all the forms are of prevailing recent species.

Mr. Prestwich has expressed his opinion in very guarded
terms upon the evidences of the Boulder-clay deposit on the
highland of Cefn Bryn, and in Rhos Sili Bay, and on its
relation to the caves and raised beach. But I am fully pre-
pared to accept the inference, to which that cautious observer
leans, that they would both appear to be of a more recent
date than the Boulder-clay.

Proofs of the comparatively modern epoch of the cave
deposits and raised beach may also be inferred from the
slight amount of inroad which the sea has made upon them
since the period of their deposition. A mile of raised beach
still survives in Mewslade Bay, perched, as Mr. Prestwich
felicitously expressed it, upon the out-cropping edges of the
limestone strata of the old cliff, which is but very little
changed in the shape of its escarpment since the beach was
formed, although still in close proximity to the sea. The line
of coast in which the group of ossiferous caves between
‘Bacon Hole’ and ‘ Minchin’ are situated presents a broadside
to the waves of the Bristol Channel. These caves overhang

1 Mr. Starling Benson distinctly refers | below the cave deposit.’-—Account of the
to the bed in Bacon Hole ‘as the re- | Cave-Deposit of Bacon Hole, p. 15.
mains of an ancient littoral beach found

MM 2

532 OSSIFEROUS CAVES OF GOWER.

the existing beach, and they contain groups of individuals
of species of Litorina, associated upon the cavern floors
apparently, as they were when alive. It is in the highest
degree improbable that these mollusca could have lived in
distant recesses of the caves, accessible to the waves, but far
removed from light. They must have been in close proximity
to the mouths of the caverns then, as they are now, showing
that in this case also sufficient time has not yet lapsed to
admit of any considerable recession of the cliffs by the wast-
ing action of the sea. Denudation is visible in the indenta-
tions of Swansea Bay and Oxwich Bay upon the softer strata
of the coal measures; but of a date long anterior to the
Cave period.

The next point which suggests itself for inquiry is: are
there any indications of a general subsidence of the coast
line, or of local depression in the vicinity of the caves? Mr.
Starling Benson cites, in support of this idea, as an estab-
lished fact, ‘the well-known remains of large trees, now
several feet under high-water mark, in Swansea Bay.’!
Colonel Wood informs me that there are evidences along the
whole line of coast, from Swansea to the eastern point of
Oxwich Bay, of submerged forests; stumps and trunks of
trees being occasionally exposed by the action of the tidal
currents. I have stated above, that the marine deposits in
the caves are not at a uniform level, either in relation to
each other or to the Mewslade raised beach. The floor of
‘Bacon Hole’ is fully 30 feet above high water, and above the
reach of the waves in the heaviest gales. The marine
deposits in ‘ Bosco’s Den’ have been washed out to a depth of
31 feet back from the mouth, the waves having free access to
the interior during high tides. In ‘ Bowen’s Parlour’ both
the chambers, upper and lower, have been completely swept
out, the solid dividing floor of cemented breccia alone re-
maining of all the former deposits. If the sea has not
encroached much, and the cliffs receded, these appearances
are only explicable on the supposition that there has been a
depression affecting at least the cited caves.

That the newest of the cave strata have been subjected to
at least one violent shock is clearly proved by the rent and
shivered condition of the stalagmitic boss of ‘ Bosco’s Den’ ;
and it is equally clear, from the uncemented state of the
fragments, that this shock is of a very modern date, and
subsequent to the commencement of the date of deposition
of the superficial layer of sandy peat, which covers the floor.
The late Mr. Stutchbury observed in ‘ Durdham Down’ a
dislocation in a vein of spar, which led him to the belief that

’ Op. citat. p. 16.

OSSIFEROUS CAVES OF GOWER. 533

a relative movement of the walls of the cavern had taken
place since the accumulation of the bones within it. I ex-
amined, in the Bristol Museum, a very remarkable specimen
of a lower molar of Hlephas antiquus from the same cave, in
which the three anterior plates of the crown were undis-
turbed, while a fault passed diagonally through the next five
plates, involving a difference of level on the opposite sides of
the dislocation, to the extent of an inch. The fissure was
filled with stalagmitic loam. In the same collection I ob-
served a large upper molar of Hippopotamus, presenting a
vertical fissure and fault, running transversely, the opposite
ends overlapping alternately. These appearances are only
explicable in the manner suggested by Mr. Stutchbury, and
they belong to the same class of phenomena as the rent boss
of ‘Bosco’s Den.’ Distinct evidence of modern subsidence
has been found in Torbay, in the submerged peat or forest-
bed in front of the Torre Abbey Ground, near Torquay, and
along other points of the coast in the vicinity, while on the
opposite points of the bay there are patches of raised beach,
with ossiferous caves in the vicinity, repeating the conditions
which are presented on the south coast of Gower. On the
whole, if not proved, it appears at least highly probable,
that some of the Gower caves have undergone a depression
of level.

One of the most interesting geological phenomena among
the newer deposits in Gower is the enormous development
upon the cliffs along the coast of the ‘ Bréche en place,’ or
angular debris, to which Mr. Godwin-Austen has applied the
name of ‘Head.’ I have had constantly, throughout the
descriptive details of the caverns, to refer to its presence,
sometimes in vast accumulations, in the immediate vicinity
of the caves, and to its intrusion, so to speak, into their
interior, as a flooring cemented by stalagmite, overlying the
marine sands and stretching seaward upon the face of the
cliffs. The origin of this angular debris and the rationale of
the accumulations are at present involved in the greatest
obscurity, seeing that some, like Mr. Godwin-Austen, regard
it as being a sub-aérial deposit, while other able observers
view it as an aqueous deposit, the result of tumultuous
transport. It is not my intention to enter upon the question
here, but simply to call attention to the fact, that there is
probably no part of the South Coast of England in which the
phenomena are developed upon a greater scale, nor can be
studied with greater advantage, than in the peninsula of
Gower.

534 OSSIFEROUS CAVES OF GOWER.

17. CoMPARISON OF THE GOWER CAVE FAUNA WITH THAT OF
OTHER CaAvE Districts In ENGLAND.

Assuming, as a position, that the Gower caves are of a
comparatively late date, it may next be inquired: are there
any proofs, in any of the other caves in Britain, of a Mam-
malian Fauna of an older date? The full discussion of
this subject, involving numerous comparisons, would extend
far beyond the limits within which this communication is
restricted, and I must deal with it very briefly. Some of the
species are nearly constant in all the caves, both in England
and Germany, such as Ursus speleus, Hyena spelea, and
Felis speleea ; and these are commonly associated with Hlephas
primigenius, Rhinoceros tichorhinus, Bos primigenius, Bos
(Bison) priscus, and large species of Cervus, such as C.
eurycerus (the Irish Elk). But in some of the caverns
the latter series is supplanted by H. antiquus, Rhinoceros
hemitechus, and Hippopotamus major. The most remark-.
able case running parallel with ‘Bacon Hole’ and ‘ Minchin
‘ Hole,’ with which I am acquainted, is the cavernous fissure
of Durdham Down, described by Mr. Stutchbury. I have
carefully examined the collection preserved in the Bristol
Museum, and distinguished 15 molars of all ages of H. major,
besides an incisor canine and caleaneum. These were asso-
ciated with a series of upper molar teeth of R. hemitechus,
and with molars of EH. antiquus, together with remains of
Ursus speleus, Ursus arctoides (?), and Hyena spelea. Inthe
Mendip caves (Banwell, Hutton, Uphill, Bleadon, Berring-
ton, &c.), H. primigenius and R. tichorhinus are common,
together with Felis spelea, Hyena spelea, &c. EL. antiquus
is rarer, and I did not observe a single specimen referable to
R. hemitechus. I detected in the Rev. D. Williams’ collec-
tion, in the Taunton Museum, two lower jaws of a species of
Spermophilus (tailless Marmot), from some one of the Mendip
caves, new to the fossil Fauna of England.' In the collec-
tions formed by McHnery, from Kent’s Hole, EF. primigenius
and R. tichorhinus are very common, while H. antiquus and
R. hemiteechus appear to be wanting. The most remarkable
peculiarity of the Kent’s Hole series is the presence of a
species of Machairodus, which has nowhere else been found
in England. The Lagomys of Kent’s Hole has lately turned
up in the Brixham Cave. Of the Kirkdale specimen the
young Elephant’s molar, represented in fig. 1 of Pl. VII. of
the ‘Reliquiz Diluviane,’ belongs to H. antiquus, and fig. 3 of
the same plate to R. hemitechus. I have examined the original

1 See antea, p. 452.—[Ep.]

OSSIFEROUS CAVES OF GOWER. 535

specimens in the Clarendon Museum at Oxford. Along with
- these were found some teeth of Hippopotamus.

18. ConctupING REMARKS AND INFERENCES.

The comparative illustrations here given are taken from
nearly all the principal ossiferous caverns of England, and
they furnish no evidence of a Cave Fauna older than that of
the Gower caves.

But there are some weighty objections to be explained.
The Boulder-clay of the Norwich coast is in direct super-
position to the ‘ Elephant’ or ‘ Submerged Forest’ bed at
Happisburgh and Mundesley, which has been cited by all
authorities as yielding remains of Elephants, Rhinoceros,
and Hippopotamus. I found that the Elephants belonged to
two species H. (Low.) meridionalis and HE. antiquus. The
former has nowhere in England been found at a higher level
than the ‘submerged forest,’ while the latter occurs abund-
antly in the Mud-bed of Bracklesham, in Grays Thurrock,
and other localities in the Valley of the Thames. In like
manner, the Rhinoceros which prevails in the ‘ submerged
forest’ belongs, as I have ascertained by recent investiga-
tion, to a species, R. Htruscus, which occurs very abundantly,
in company with H. meridionalis, in the deposits of the Val
d’Arno; but which, in England, like H. meridionalis, is never
seen above the level of the Norfolk ‘Hlephant-bed.’? The
remains of the two Elephants have never, I believe, been
discovered together im situ in this deposit. Although occur-
ring in immense abundance, they are either brought up by
the dredge from the Oyster Bank, or found stranded on the
beach after heavy equinoctial wales. Molars of EL. antiquus
are also dredged up on the Essex coast, off the ‘ West Rocks.’
There were a great many found thus, in the collection of the
late Mr. Brown, of Stanway, but I never saw any molars of
E. meridionalis among them. It is possible, therefore, that
the molars of the two species found under the sea, on the
Norwich coast, may be derived from beds of different ages.
On a review of the general bearing of the evidence, it
appears to me that the following conclusions are consistent
with the existing state of our knowledge :—

1. That the Gower caves have probably been filled up with
their Mammalian remains since the deposition of the Boulder-
clay.

3. That there are no Mammalian remains found elsewhere
in the ossiferous caves of Britain referable to a Fauna of a
more ancient geological date.

3. That Hlephas (Lowodon) meridionalis and Rhinoceros

536 OSSIFEROUS CAVES OF GOWER.

Htruscus, which occur in, and are characteristic of, the ‘ Sub-
marine Forest-bed,’ that immediately underlies the Boulder-
clay, have nowhere been met with in the bone caverns of
Britain.

4. That Elephas antiquus with Rhinoceros hemitechus, and
HE. primigenius with R. tichorhinus, though respectively
characteristic of the earlier and later portions of one period,
were probably contemporary animals, and each of them were
certainly companions of the Cave Bear, Cave Lions, and Cave
Hyenas, &c., and of some at least of the existing Mammalia.

APPENDIX TO MEMOIR ON THE CAVES OF GOWER.

J.—Letrer From Mr. Prestwich oN THE BOULDERS AND GRAVELS OF
THE GowER Cave District, AND ON A RatseD BEACH TO THE

West or GOWER.
10 Kent Terrace, May 17, 1860.

My pear Fatconer,—I have much pleasure in giving you a few
lines respecting the raised beach I met with last autumn, to the west-
ward of Paviland Cave in Gower. I find my notes on the subject are
not very complete, having taken only a first survey, reserving a fuller
examination of the coast until I could obtain access to the caves. You
will remember how I was baffled on the last occasion by the state of
the tide and the weather. Finding it quite impossible to pass round
the foot of the cliff to gain the entrance to Paviland Cave, I proceeded
westward along that iron-bound and magnificent frontage of limestone
cliffs, ending in Worm’s Head, with the intention of examining them at
the accessible points, to see whether I could detect any facts bearing
upon your very important observations on ‘ Bosco’s Den,’ relating to
the connection of marine remains under, and in association with, the
wonderful mass of bone debris you and Colonel Wood had discovered
there. At the distance of about half a mile west of Paviland Cave I
found a gully, by which I got down to the shore. I then found in
hollows in the cliff, and at an elevation of 10 to 12 feet above the
beach, a layer of sand and rolled limestone pebbles, having all the cha-
racters of a beach; but in the absence of shells, and looking at its
small patchy character, no conclusion could be drawn from it alone.
The passage at the foot of the cliffs being still impracticable, I had to
confine myself for the next mile or two to one or two descents, where
I again found traces of what appeared to be a raised beach. Still I
was not prepared for the very fine and remarkable exhibition I
witnessed, after passing Mewslade, at the bottom of the small bay
formed by Thurba Rock and Tears’ Point, about one mile south of Rhos
Sili. There, perched upon the escarped edges of the grey, weathered -
limestone, is an old beach, raised some 10 to 12 feet above high tide
mark. It is composed of pebbles and fragments of limestones, thinly
mixed with a coarse red sand, and in places full of shells and fragments
of shells. There are very few species; the Patella vulgata is common ;

OSSIFEROUS CAVES OF GOWER. 537

the Littorina litorea abounds; there are a few Purpura lapillus, and
fragments of Mytilus; also pebbles of limestone drilled by boring shells.
The whole, which is 3 to 4 feet thick, is agglutinated into a semi-
compact mass, and is overlaid by a remarkable mass of angular debris,
from 20 to 30 feet thick in some places. The beach goes back only a
few feet, as the limestone hill rises immediately behind. Coastways
the raised beach continues almost uninterruptedly, but diminishing in
importance, for half a mile westward, ending before reaching Tears
Point. Its level is persistent throughout. The sketch I gave you
at the time (fig. 4) is, I believe, nearly correct, and at present I do not
think I could improve upon it, not having made sufficiently full notes.
Taking this in connection with the well known raised beach at the
Mumbles, I think it may have an important bearing, in conjunction
with your discoveries in those Bone Caves in Gower, which are
situated on the coast between these two points. ‘They are evidently on

Fig. 4.

RAISED BEACH AND OLD CLIFF NEAR MEWSLADE, GOWER, S. WALES,

a, Unstratified angular limestone debris, in a matrix of red sand and clay, 21 to 27
feet. 6. Red sand and clay without shells, 4 to 5 feet. c, Rolled shingle with
shells in places, 3 to4 feet. d. Limestone rock. ¢, Level of present beach.

about the same level, and you have found in them sand and sea-shells
under all the bone remains. Should it prove, therefore, that the caves
are of this Raised Beach period, and that the Elephant and other remains
have been subsequently introduced, we shall arrive at the interesting
and curious conclusion that this particular group of mammalia lived
after the formation of those beaches—beaches which have always been
considered as of very recent origin, as they contain nothing, so far as
they have been examined, but the commonest shells of our coasts. At
the same time, it is to be observed that they contain but very few
species, and that no complete and thorough investigation of them has
been yet made. With regard to your suggestion in connection with
the two species of Elephant, I must confess that I saw nothing in the
physical features of the scene, during the somewhat hurried and im-
perfect view I had of it, to lead me to suppose that the caves, or rather
their inhabitants, might be referred to two periods. I should hardly
have hazarded this opinion, without a further examination of the
district ; but I give it for what it is worth, and waiting further data.

538 OSSIFEROUS CAVES OF GOWER.

With respect to the point I had particularly in view, viz. the relation
of the Gower Caves to the Boulder-clay, Iam unable as yet to form a
decided opinion. I got the Boulder-clay within a mile of the raised
beach, but on opposite sides of the point of Rhos Sili. It spreads from
the sea-shore to, as you are aware, the top of the hills. In Rhos Sili
Bay I found intercalated in it, at an elevation almost exactly corre-
sponding to the raised beach on the opposite side of the promontory, a
bed of shingle containing several species of recent shells, but not one of
the species occurring in the raised beach. Yet the two would appear
to be synchronous; the difference might arise from the one beimg on
an exposed and open coast, and the other in a sheltered bay. The
subject requires a fuller and more lengthened inquiry. (J. PREstwIcu.)

JI.—Note oN THE OccURRENCE OF WROUGHT FLINTS, IN ASSOCIATION
WITH TWO EXTINCT SpEcIEs OF Rurnoceros, &c., in ‘ Lone Hore’
Cave, Gower.!

On May 30, 1860, I communicated to the Geological Society a memoir
on the Ossiferous Caves of Gower, founded on the joint researches of
Lieut.-Col. Wood and myself. The main object of that communication
was to show first what the Fauna of these caves was, as compared with
the other ossiferous caves of Britain. 2. The presence in it of species of
Elephant and Rhinoceros, which had not previously been recognized as
occurring in the ossiferous caves. 38. The continuity, in the Gower
Fauna, of Upper-Pliocene and Glacial mammals, without an apparent
break ; and lastly, the epoch whence the Gower Fauna dated. Some
of the conclusions arrived at met in discussion with sharp criticism and
opposition from leading authorities on the newer Pliocene and Drift
deposits. Being desirous of concentrating attention on the points
above indicated, I purposely excluded, with a single exception, from
that memoir any reference to the then newly-agitated question, which
attracted at the time, as it now does, so much attention, namely, the
bearing of the Gower Cave evidence upon the antiquity of man. That
exception was the presumable age of the famous human skeleton, so
familiarly known, through the writings of Buckland, as the ‘ Red Lady’
of Paviland, found associated, under questionable circumstances, with
the skull of Elephas primigenius ; and the inference drawn was against
their having been cotemporary individuals. Lieut.-Col. Wood and
myself had found numerous wrought flints and some bone weapons in
Paviland; but the cave deposits there had been so disturbed by pre-
vious excavations of an old date, that none of the instances was free
from the taint of suspicious occurrence. Asa rule, the Paviland flint-
flakes were undistinguishable in form from those of the Grotto of
Maccagnone.

Last autumn (1861) Lieut.-Col. Wood discovered a new ossiferous
cave in Gower, which supplied conclusive evidence respecting the con-

1 This paper was commenced in 1862, | tory,’ for October, 1864, in a brief notice
but was never completed. The defi- | entitled ‘On the Asserted Occurrence of
ciency has been made up from the au- | Flint-knives under a skull of the ex-
thor’s Note-books. One of the main |} tinct Rhinoceros hemitcechus in an ossi-
facts, however, contained in the paper | ferous cave in the peninsula of Gower.’
was published by Dr. Falconer in the | —[Ep.]

‘Annals and Magazine of Natural His-

»

OSSIFEROUS CAVES OF GOWER. 539

temporaneity of the two extinct species of Rhinoceros of Gower, a point
before open to question, and which at the same time appeared to throw
light upon the relation of man to the extinct Fauna. This cavern,
called ‘ Long Hole’ by the neighbouring farmers, is situated in the line
of cliffs called the ‘ Yellow Top Rocks,’ about a mile west of the village
of Port Eynon, and about one mile east from the Paviland Caves. It
occurs near the summit of an interruptedly escarped limestone rock,
facing the sea. Below, a precipice rises vertically, from the reef
forming the shore, to the height of 51 feet. From its summit a steep
slope, patched over with vegetation, stretches upwards and backwards,
along a distance of 258 feet, and with an inclination of about 25°.
Above this scarp another rugged precipice rises to a height of 20 feet,
its summit being continuous with the surface of the country. The cave
forms a low, irregularly arched aperture, in a scooped bight of the
upper precipice, at its base, and is easily accessible from the plateau
above, by a short descent and then doubling round the limestone bluff
which bounds the cave towards the west. From an observation with
an aneroid barometer, the cavern is about 130 feet above ordinary high-
water mark. It occupies the uppermost part of a cleft, which, as in the
case of ‘ Bacon Hole,’ ‘ Bosco’s Den,’ and other of the Gower Caves that
are situated on the sea cliffs, can be traced down upon the littoral reef
which is exposed at low water. The segment below the scarp of this
fissure is filled with a mass of gravelly conglomerate, forming a perpen-
dicular cliff 50 feet high, which readily dismtegrates when washed by
the waves that are driven against its base, during the highest spring
tides, aided by south-westerly gales. The top of the section is overlaid
by a thin covering of limestone breccia, consisting entirely of angular
fragments of local origin, called ‘ Wark’ by the quarrymen, and corre-
sponding with the ‘ Head’ of Austen and other writers. This ‘ Wark’ is
enormously developed in the neighbourhood of Bacon Hole and some
of the other caves. At the base of the gravel section, and perched
about a foot above the mark of the highest spring tides, lies a huge
detached and residuary tabular mass of the old sea beach, observed
west of Mewslade by Mr. Prestwich.

‘Long Hole’ opens towards SSE., and is visible from the sea, which
- circumstance led to its exploration. When discovered the aperture
presented an irregularly arched low fissure, 7 feet 8 inches wide by
41 feet high, which extended inwards about 44 feet.! At a distance
of 84 feet from the entrance, the cave widens to 12 feet; this also is
the highest point of the cave, the height being 7 feet. Passing inwards
the passage contracts, the width varying from 4 to 7 feet, although at
some places it was at first reduced to 25 feet from the projection
inwards of stalagmitic bosses. The floor inclines upwards towards the
extremity of the cave. The roof is horizontal, but very irregular, and
full of hollows, yet no flue has been discovered to connect it with the
surface. The floor of the cave consisted of ferruginous unctuous cave
earth, extending to a depth of about 7 feet, and intermixed with de-
tached angular fragments of the limestone rock 8 or 9 pounds in weight.
On clearing this away it was found to rest on the solid rock, without
any trace of sand or shingle.

' [The succeeding portion of the paper is compiled from the Author’s Note-
books,—Eb. ]

540 OSSIFEROUS CAVES OF GOWER.

The fossil remains found in the deposits in ‘ Long Hole’ Cave were
transmitted to me for identification by Col. Wood, and are as follows :—

Carnivora.—Ursus speleus. Putorius vulgaris. Canis lupus. Canis
vulpes. Hyena spelea. Meles Taxus. Lutra vulgaris. Felis spelza.
Felis catus. Mustela foina ?

Pachydermata.—Rhinoceros hemitechus. Rhinoceros tichorhinus.
Elephas antiquus. Elephas primigenius. Equus Caballus. Equus
Asinus. Sus scrofa.

Ruminantia.— Cervus eurycerus. Cervus Guettardi. Cervus Taran-
dus. Cervus Elaphus. Bison priscus.

Rodentia.—Lepus cuniculus. Lepus timidus. Arvicola amphibius.

Flint implements, unquestionably of human manufacture, were found
along with these fossil remains, and were also sent to me by Col. Wood.
One very fine flint arrow-head was found contiguous to, and at the
same depth as, a detached shell of a milk molar of Rhinoceros hemite-
chus. The part of the cave where they were found was about 6 feet
from the entrance, and at a depth of 44 feet in the cave earth. Other
flint implements were found at a depth of 3 feet below the stalagmite,
associated with specimens of Cervus Guettardi and the peculiar form
of ‘ Marten,’ of which two lower jaws were found in ‘ Rayen’s Cliff’ (p.
520) above the stalagmite. (Mustela foina?)

OSSIFEROUS CAVE OF CEFN. 541

XX. ON THE FOSSIL REMAINS FOUND IN CEFN
CAVE, NEAR BRYN ELWY, N. WALES.!

THE cave opens towards the west in a nearly vertical craggy
cliff of gray compact limestone which forms the steep eastern
boundary of the Valley of the Elwy. The valley runs north
and south or NW. and SE. The rock is very fragmented,
the cliffs consisting of pieces piled on one another like bricks.
There is much yellow cave-earth in the crevices. The cave
is at about the middle of the vertical height of the cliff. The
opening is something like that of Brixham, but much wider
and higher. Near the mouth is a horizontal tunnel-shaped
flue, as if hollowed out by sea
action. The action of the sea
is visible in the cancellar erosion
near the mouth. The body of
the cavern is formed of a si-
nuous fissure, which is more ///
capacious and loftier than that /
of Brixham, being evidently in '
a line of fault. In the annexed /
eround-plan (fig. 5), A indicates
the opening: B is a ‘steep slide’
branch, running a considerable
way upwards to near the face
of the cliff, as shown by fresh
roots of trees at the upper part;
it does not, however, open into WN
daylight, or if so, the opening | . UN
is blocked up: C is another WY) Drm
eee Wiehiy inclined branch — 1 Me \ . WW
rising upwards into daylight ME
with a wide opening; itappears = GROUND-PLAN OF CEFN CAVE.

to me to have been the flue, Copied from a eee in Dr. Falconer’s
through which most of the ma- ages

terials were injected and washed into the cave. Both B and
C have their floor covered with very slippery yellow loam.
The injecta have been washed down chiefly into d, where
bones are found. D is an insular irregular cylinder, around

Fic. 5.

iil) KA Wp
i Ih y Wii) WA Uy

il

y]
|

1 Extracted from notes of a visit by | 1859. The fossils were subsequently
the author to Cefn Cave, on August 27, j identified in London.—[Fp. |

542 OSSIFEROUS CAVE OF CEFN.

which the cave passage turns, forming a complete circular
communication (d, e, f).

The following is a list of the species which I have identi-
fied among the fossils of Cefn Cave :—

Pachydermata.

Elephas antiquus.
Rhinoceros hemitcechus.
as tichorhinus.
Hippopotamus major.
Equus. Teeth and astragalus. Species undetermined.

Ruminantia.

Strongyloceros speleeus.
Cervus Guettardi.
5 eurycerus.
Bos. Molars of species undetermined.

Carnivora.

Felis spelea.
Hyzena spelea.
Ursus speleeus.
Canis lupus.

OSSIFEROUS CAVES OF SICILY. 543

XXI. ON THE OSSIFEROUS GROTTA DI MACCA-
GNONE, NEAR PALERMO.'

Dr. Fatconer first described the physical geography of that
portion of the north coast of Sicily in which the ossiferous
caves abound, namely between Termini on the east and Tra-
pani on the west. The geological structure of the tract has
been ably investigated and mapped by Hoffmann. A great
mass of Hippurite-limestone stretches from Termini to the
eastern side of the Bay of Castellamare, which on the side
towards Termini forms rugged precipitous or scarped cliffs
skirting the sea-shore. From Cape Zaffarana to Capo di
Gallo, a distance of about twenty miles, the coast-line is
deeply indented by the Bay of Palermo; west of Capo di
Gallo there is a smaller indentation, backed by Carini; and
still further to the west there is the deep Gulf of Castella-
mare. At the bottom of these indentations the mountains of
Hippurite-limestone recede from the coast, forming inland
precipitous cliffs or rugged slopes, from the base of which
stretch slightly inclined flats of marine Pliocene deposits,
which disappear under the sea. These latter form nearly
horizontal strata of a calcareo-argillaceous sandy breccia, full
of marine shells and fragments of corals, &c. Philippi iden-
tified 209 species of Mollusca from this deposit in the neigh-
bourhood of Palermo, the great majority being of living
species. The ossiferous caves had been known from remote
antiquity, and notices of them occur in ‘Valguarnera, Il Mon-
gitore, and other Sicilian historians. The botanist Cupani
had figured and identified some of the bones. The author’s
investigations had been directed to the caverns near Palermo
and Carini. At Palermo the littoral Pliocene plain, celebrated
for its richness as the ‘Concha d’Oro,’ or shell of gold, is
from a mile to 14 mile broad, and where it abuts against the
Hippurite-rocks is from 180 to 200 feet above the level of
the sea. The ancient Pliocene sea-margin is very distinctly
seen at this elevation all round the bay, and the ossiferous
caverns chiefly occur at from 30 to 50 feet above this level.

1 This communication was made by | is reprinted from the abstract which

Dr. Faleoner to the Geological Society | appeared in the ‘ Quarterly Journal’ of
on May 4th and June 22nd, 1859, and | the Society for May, 1860.—[Eb. |

544 OSSIFEROUS CAVES OF SICILY.

Some of them, such as the ‘Grotta di Belliemi,’ are at a
higher level. The caves are studded all round the bay. The
Hippurite-limestone hills skirting the coast are here from
1,200 to 1,800 feet above the sea; some of the heights more
inland, such as Monte Griffone and Monte Cuccio, attain a
height of upwards of 3,000 feet.

ac best known of the caves is the ‘Grotta di San Ciro,’

‘Mare Dolce,’ at the foot of Monte Griffone, about two
waileg from Palermo, and 50 feet above the Pliocene terrace.
This cave had been described by the Abbé Scina in a special
report, and after him by Turnbull-Christie and by Hoffmann.
It is about 130 feet long, 50 feet high, and 30 feet wide in
the middle. The cave had been hollowed out into a well-
marked, irregular, basin-shaped depression near the mouth,
where obscurely stratified and other deposits occur to a depth
of 30 feet in the aggregate. On the bottom was found a
thin layer of sand, in which Philippi detected 44 species
of 23 genera of marine Mollusca. Above this there is an
enormous mass of bone-breccia, consisting of closely-crammed
bones, cemented into a hard rock by an argillaceo-calcareous
concrete matrix, and forming a thickness of 20 feet; above
this a stratum of stones and bones, more sparingly mixed
and similarly cemented, to a depth of 2 or 3 feet; then a
layer of ‘ Lastroni,’ or blocks of limestone, to a depth of
6 feet; and above all'a layer of ochreous earth and rock-
splinters to a depth of 1 foot. The bones in this breccia are
mineralized by calcareous infiltration. The interior and back
part of the cavern was covered by a layer of light and inco-
herent argillaceous soil, containing an enormous quantity of
bones, chiefly of Hippopotami, nearly devoid of gelatine, and in
the ordinary friable condition of grave-bones. The relations
of this deposit were never accurately observed, in consequence
of the rubbish of the excavation-operations having been
thrown up in a great mass of talus extending backwards to
near the roof of the cavern.

In 1829 there was a great demand for bones for the manu-
facture of lamp-black for sugar-refining. The superficial
bones of the San Ciro cavern were collected in large quan-
tities and exported to England and Marseilles. Professor
Ferrara states, that within the first six months 400 quintals
were procured from San Ciro. The great majority belonged
to two species of Hippopotamus. In one heap, out of several
shiploads sent to Marseilles, De Christol, an able paleontolo-
gist, had found that in a weight of thirty quintals all the
bones belonged to Hippopotamus, with the exception of six
derived from Bos and Cervus. Dr. Falconer had examined in
detail the San Ciro collection in the University of Palermo,

OSSIFEROUS CAVES OF SICILY. 546

and found, as a general rule, that Hippopotamus bones pre-
ponderated in a similar proportion. De Christol had counted
about 300 astragali alone of this genus; and Abbé Scina had
collected, for the Museum of Palermo, 76 astragali of Hippo-
potamus, 40 of which belonged to the right side and 36 to
the left. The bone-breccia is chiefly composed of bones of
Hippopotamus, and extends on either side outside the cave to
a length of about 85 yards. Assuming the above ratio of
astragali to the other bones as a standard for an approxima-
tive estimate of the number of the skeletons inside and out-
side the cavern, the author showed what a vast number of
individuals it implied. He considered that they were accu-
mulations of a long series of generations. A lively discussion
having arisen in Sicily as to the origin of these bones, in
which Ferrara maintained the opinion that they consisted of
the skeletons of Hlephants captured by Metellus from Has-
drubal 504 years before the Christian era, and of Hippopotami
imported by the Saracen rulers of Sicily during the Middle
Ages, the government undertook an exploration of the cavern.
A deep trench was dug longitudinally into the cavern; and
the bone-breccia was quarried out, along a considerable ex-
tent, down to the floor of the cavern. Some very interesting
phenomena were disclosed. The eastern wall (left, on enter-
ing) was found to be smoothly polished to a height of 18 feet,
the lower 8 feet of which formed a band thickly drilled with
Pholad-borings. The holes were filled with matrix of the
bone-breccia, and they were greatly reduced in depth by the
grinding action which had produced the polished surface.
The opposite wall of the cavern was equally polished to the
same height, but free from borings.

The walls above the polished band and the roof were
rugged and cancellar, with but a very sparing exhibition of
stalagmite on the latter. The author had identified from
San Ciro two species of Hippopotamus,' Elephas antiquus, Sus,
Bos, Cervus, Ursus, Canis, and a large species of Felis. Hle-
phas antiquus elsewhere indicates the newer Pliocene age.

Another cave, hitherto undescribed, called the ‘ Grotta di
Maccagnone,’ about a mile to the westward of Carini, was
lately the special subject of the author’s research. It is
nowhere noticed by the Sicilian historians. Dr. Falconer’s
attention had been directed in that quarter by J. Morrison,
Hsq., a resident merchant of Palermo, who had many years
ago procured fossil bones from the neighbourhood of Carini,

1 Extract from Author's Note-book.— | Pentlandi, but the large inferior canine
‘Naples, Jan. 29, 1859. The Hippopo- is certainly of another species, and ap-
tamus bones from Sicily are of two | parently of H. major. —[Ep.]
species, the great majority being of H.

VOL. Il. NN

546 OSSIFEROUS CAVES OF SICILY.

which are now displayed in the Museum of the College of
Surgeons. The author was under great obligations to the
kind services, scientific aid, and hospitable cares of Baron ,
Francesco Anca (di Mangalaviti) and Professor Angelo Por-
cari, of the University of Palermo, who accompanied him in
all his visits to Carini, and co-operated with him in the exca-
vations carried on in and near the Maccagnone Cave. Their
assistance applied to every walk of the exploration. The
cave is situated on the north-eastern side of Monte Lungo,
near its base, and about a mile and a half from the sea. Like
San Ciro, the Maccagnone Cave is about 50 feet above the
termination of the Pliocene marine terrace where it abuts
against the Hippurite limestone, and at a corresponding ele-
vation above the sea; both caves partaking in many respects
of common physical characters. But in its form, and some
of its deposits, the Maccagnone Cavern differs materially
from San Ciro. It is much broader and more sinuous at the
sides than San Ciro, with several large cul-de-sac expansions,
but not so long; and the roof is much lower, being but 11
feet high at the principal entrance, and about 10 feet in the
middle. There are two entrances, the principal of which is
251 feet wide, and open down to the floor; the other, on the
same side of the hill, is a much smaller, irregular aperture,
in connection with an irregular expansion of the cavern at its
south-eastern corner, into which it descends. The author
gave the principal dimensions, which were accompanied by a
section and ground-plan (figs. 6 and 7). The uppermost
layer of the floor consists throughout of loose, argillaceous,
finely pulverized soil, containing large embedded blocks of
limestone; beneath this, in the section below the mouth,
was a thick deposit of the ochreous loamy earth (called
‘Cave-earth’), containing blocks of limestone; then, in thick
patches, a reddish-grey and mottled spongy loam, cemented
by calcareous infiltration, and very cellular, called from its
appearance, by the peasants who were employed in the ex-
cavation, ‘Ceneri impastate,’ or ‘concrete of ashes ;’ and
below all, stretching on either side of the mouth, as at San
Ciro, a great aggregation of bone-breccia, full of bones of
Hippopotamus, among which the author in four days collected
a very large number of astragali. The whole of the bone-
breccia was strewed over with huge blocks of limestone which
had fallen since its deposition. Nothing is known of the
nature of the inferior deposits down to the floor. The
author thinks it probable that there may be a great accumu-
lation of bone-breccia below, with polished and bored walls,
as in San Ciro; but the excavations requisite to establish
this were too laborious and extensive for the limited time at

OSSIFEROUS CAVES OF SICILY. 547

Fic. 6.

VERTICAL SECTION OF THE GROTTA DI MACCAGNONE,

a. Bone-breccia, with blocks of limestone.

6. Grey cancellar deposit.

e. Yellow ochreous bed, or ‘ caye-earth,’ with blocks.

d. Humatile layer, with blocks of limestone.

e. Stalagmite coating the ceiling of the cave.

f. Roof-breccia cemented to the ceiling. :
g. Roof-breccia in the back of the cavern, coated with stalagmite.
h. Hippurite limestone.

w. Wall of the entrance.

Fic. 7.

GROUND-PLAN OF THE GROTTA D1 MACCAGNONE.
NN 2

548 OSSIFEROUS CAVES OF SICILY.

his disposal. The interior of the cavern is coated over
throughout by a crust of rough, reddish or ochreous stalag-
mite. The surface-layer of the floor had been previously dug
for fossil bones as far back as 1830, and but few remains
were found in it. One—an important specimen—was a mill
molar of Hlephas antiquus. At the side below the southern
wall of the cavern, and about halfway in, a thick layer was
observed of the ‘Ceneri impastate’ immediately below the
superficial earth, and corresponding exactly with the ‘ Ceneri
impastate’ seen in the section outside below the principal
aperture. The attempts at making a section of the floor
were frustrated by the great blocks of limestone, which im-
peded the operations throughout. In the superficial layer,
Elephas antiquus and horns of two extinct species of Cervus
were found, besides other bones of Ruminants, but all in
small quantity: in the ochreous cave-earth below the mouth,
abundant coprolites of Hyena, with fragments of detached
bones of Hippopotamus, and some astragali: in the ‘ Ceneri
impastate’ below the main aperture, metacarpal and meta-
tarsal bones of a species of Felis as large as F’. spelea, but not
yet specifically identified ; some remains of a large Ursus,’ and
numerous remains of small Ruminants, all broken or splin-
tered, but none of them bearing marks of gnawing. In the
“bone-breccia’ below and outside the cavern, the bones of
Hippopotami very-largely predominated. The author dug up
an enormous quantity of these remains within an area of 12
or 14 feet square. In an angular recess in the rock outside
the cave, and near the small opening, a very large quantity of
coprolites of Hyzenas were observed, superficially embedded
in the ‘ochreous earth.’ The quantity collected together
would indicate that this spot had been used as a common
cloaca of Hyzenas.

The author next described some remarkable conditions in
the roof of the cave. About halfway in from the mouth
(fig. 6, f) and at 10 feet above the floor, a large mass of
breccia was observed, denuded partly of the stalagmitic
covering, and composed of a reddish-grey argillaceous matrix
cemented by a calcareous paste, containing fragments of
limestone, finely preserved entire land-shells of large size,
splinters of bone, teeth of Ruminants and of the genus Hquus,
together with comminuted fragments of shells, bits of carbon,
specks of argillaceous matter resembling burnt clay, also
fragments of shaped siliceous objects, of different tints, vary-

1 The specimens of Ursus from Mac- | were believed by Dr. Falconer to be
cagnone were found to agree with those | of the same species. (Note-book.)—
from Grays Thurrock in the British | [Ep.]

Museum, purchased from Mr. Ball, and |

OSSIFEROUS CAVES OF SICILY. 549

ing from the milky or smoky colour of chalcedony to that of
jaspery hornstone. This brecciated matrix was firmly at-
tached to the roof, and for the most part covered over with a
coat of stalagmite. In the SSH. expansion of the cavern
near the small aperture, a considerable quantity of coprolites
of Hyena was found similarly situated, in an ochreo-calcareous
matrix, adhering to the roof, mingled with some bits of car-
bon, but without shells or bone-splinters. In the back part
of the cavern, where the roof shelves towards the floor, thick
masses of reddish calcareous matrix were found attached to
the roof, and completely covered over by a crust of ochreous
stalagmite (fig. 6, g). It contained numerous fragments of
the siliceous objects mixed with bone-splinters and bits of
carbon. In fact, all round the cavern, wherever the stalag-
mitic crust on the roof was broken through, more or less of
the same appearances were presented. In some parts the
matrix closely resembled the character of the ‘ Ceneri impas-
tate’? with a larger admixture of calcareous paste.

With regard to the fragments of the siliceous objects, the
ereat majority of them present definite forms, being long,
narrow, and thin; having invariably a conchoidal smooth
surface below, and above a longitudinal ridge, bevelled off
right and left, or the ridge replaced by a concave facet, in
the latter case presenting three facets on the upper side.
The author is of opinion that they closely resemble, in every
detail of form, obsidian knives from Mexico, and flint knives
from Stonehenge, Arabia, and elsewhere, and that they
appear to have been formed by the dislamination, as films, of
the long angles of prismatic blocks of stone. These frag-
ments occur, intimately intermixed with the bone-splinters,
shells, &c., in the roof-breccia, in very considerable abund-
ance; other amorphous fragments of flint are comparatively
rare, and no pebbles or blocks occur either within or without
the cave; but similar reddish flint or chert is found in the
Hippurite-limestone near Termini.

In regard to the theory of the various conditions observed
in the Maccagnone Cave, the author considers that it has
undergone several changes of level, and that the accumula-
tion of bone-breccia below and outside is referable to a period
when the cave was scarcely above the level of the sea.
Dr. Falconer pointed out the significance of the fact that,
although Hyzena-coprolites were so abundant against the
roof and outside, none, or but very few, of the bones of
Hyznas were observed in the interior; he remarked also on
the absence of the remains of small mammalia, such as
Rodents. He inferred that the cave in its present form, and
with its present floor, had not been tenanted by these animals.

550 OSSIFEROUS CAVES OF SICILY.

The vast number of Hippopotami implied that the physical
condition of the country must have been greatly different, at
no very distant geological period, from what obtains now.
He considered that all deposits above the bone-breccia had
been accumulated up to the roof by materials washed in from
above, through sinuous crevices or flues in the limestone,
and that the uppermost layer, consisting of the breccia of
shells, bone-splinters, siliceous objects, burnt clay, bits of
charcoal, and Hyzna-coprolites, had been cemented to the
roof by stalagmitic infiltration. The entire condition of the
large fragile Helices proved that the effect had been produced
by the tranquil agency of water, as distinct from any tumul-
tuous action. There was nothing to indicate that the dif-
ferent objects in the roof-breccia were other than of contempo-
raneous origin; subsequently a great physical alteration in
the contour, altering the flow of superficial water and of the
subterranean springs, changed all the conditions previously
existing, and emptied out the whole of the loose incoherent
contents, leaving only the portions agglutinated to the roof.
The wreck of these ejecta was visible in the patches of
‘ Ceneri impastate,’ containing fossil bones, below the mouth
of the cavern. That a long period must have operated in
the extinction of the Hyena, Cave Lion, and other fossil
species, is certain; but no index remains for its measure-
ment. The author would call the careful attention of cautious

List of Fossil Shells found in the ‘ Grotta di Maccagnone,’ as deter-
mined by Padre Lebassi, Curator of the Museum of the ‘ Collegio
Massimo dei Gesuiti’ in Palermo, March 18, 1859.

Where found Remarks Remarks by Padre Lebassi
Roof matrix,)| A solitary Vaur canoe
1. Helix Mazulii. back part }| shell, partly fi ae a t on ye
of cave. broken. 0 a
: Very abundant in the
Helix aperta (Born.) : aig ae
2.{ (H. naticoides, Drsp.)) Roof-matrix. — pes like Z.
Very abund- -
3. Helix vermiculata . .|Roof-matrix }|ant: shells in Rit ee mm the
perfect inte- neg eet
grity. ‘ise
(Very common in liy-
4, Helix cellaria (Miller) |Roof-matrix. as | eee we
| Philippi.
5. Trochus fragaroides sal : Ve 2 Ties
(Lam.) Ty common : ving.
3 Very rare in the sea
6. Patella ferruginea(Linn.) _ { 4 sphlary around Sicily: very
P : common fossil.

OSSIFEROUS CAVES OF SICILY. 551

geologists to the inferences :—1. That the Maccagnone Cave
was filled up to the roof within the human period, so that a
thick layer of bone-splinters, teeth, land-shells, Hyzenas’
coprolites, and human objects was agglutinated to the roof
by the infiltration of water holding lime in solution. 2. That
subsequently, and within the human period, such a great
amount of change took place in the physical configuration of
the district as to have caused the cave to be washed out and
emptied of its contents, excepting the patches of material
cemented to the roof and since coated with additional
stalagmite.

Note.—The author has lately received a letter from Baron
Anca di Mangalaviti, dated Palermo, the 12th of March,
intimating that he had followed up the Cave-researches
which form the subject of this communication, with import-
ant results. He had discovered two caves, hitherto unknown
to naturalists—the one in Monte Gallo, which forms the
western boundary of the Bay of Palermo; the other in the
north of Sicily, at the foot of Monte San Fratello, near the
village of Acque Dolci. In both caverns, but more especially
in the latter, Baron Anca found an immense accumulation
of fossil bones, among which there was ‘une prodigieuse
quantité d’os des Carnivores.’ This is the more remarkable,
as in the Caves of San Ciro and Maccagnone Carnivora were
but very sparingly encountered.—H. F., April 5, 1860.

APPENDIX TO MEMOIR ON THE GROTTA DI
MACCAGNONE.

J.— Notice oF THE Discovery oF Two Bone-Caves In NorTHERN SICILY.
By Francois Anca, Baron DE ManGaLavit.!

Since you left Sicily, I have continued my paleontological researches,
and I am happy in having discovered two bone-caves previously un-
known. One of these is at Monte Gallo, at the western extremity of
the Bay of Palermo, and is situated at an elevation of 160 feet above
the sea-level; the other is situated near the village of Acque Dolci, at
the foot of Monte San Fratello, in the north of Sicily, and is 214 feet
above the sea-level.

These caves, especially the last, are very rich; and, what will
astonish you, they contain a prodigious quantity of bones of Carnivora,
including perfect jaws armed with molars and canines. I have collected
also two molars and a tusk of Hlephas, teeth and bones of Hippopotamus
(of the two species, I believe, determined by yourself). Altogether,
remains of the following were met with :—

1 From a letter to Dr. Falconer, dated | lished in the ‘ Quart. Journ. Geol. Soe.’
‘Palermo, March 12, 1860,’ and pub- | for May, 1860.—[Ep.]

552 OSSIFEROUS CAVES OF SICILY.

Hippopotamus ; two species. Cervus (two species ?).

Elephas antiquus.! Canis.
» Africanus.” Ursus.
Sus scrofa ? Hyena.
Equus. Felis.
Bos; two species. Lepus. And others.

Having this large group of genera, we may say that we have re-
covered in this cave an entire fossil Sicilian Fauna.

I have also found in these two caves a large quantity of flint imple-
ments (‘de silex en armes’); and it is remarkable that we do not
generally see them but where there are great deposits of bones of Deer
—never otherwise. Lastly there occur coprolites of Carnivores, and
another kind of coprolite, which, I suppose, belonged to herbivorous
animals.

I have been fortunate also in detecting teeth of Carnivora in the
Cave of Olivella (‘la grotte de l’Olivella’).

The necessity of having means of comparison at hand induces me to
prosecute the study of these cave-bones at Florence, where I shall have
the assistance of M. Meneghini. Afterwards I hope to publish the
results of the exploration of these caves, and to describe them and the
more interesting of the objects obtained.

IJ.—Exrract oF A Letrer rrou Dr. Fatconer To Capt. Spratt, C.B.
Dated July 21, 1860.

A Sicilian friend of mine, Barone Francesco Anca, who accompanied
me in my cave explorations, followed them up after I left, and in the
cave of San Teodoro, near 8. Agata, west of Cape Orlando, on the north
coast, he discovered molar teeth, which prove to be of the existing
African Elephant,? and about twenty jaws, upper and lower, of the
Hyena crocuta, or spotted Hyzna of the Cape of Good Hope.* Here,
then, we have two existing African mammalia occurring in Sicily, and
proving beyond all question that Sicily within the Pleistocene period
was connected by land with Africa. The severance of the island from
the continent must have been quite as modern an event as the separation
of England from France. The continuity of the land evidently lay in
Admiral Smyth’s ‘Adventure Bank’ and ‘Skerki Shoal.’ Smyth’s
section of the bank shows how very shallow the soundings are. There
must have been continuous land from Ras Addar (Capo Bono) through
Sicily on as far as Messina, with a distinct eastern and western basin.
It now becomes of great interest to ascertain whether the Strait of
Messina was then an open channel, or closed up by land like ‘ Adventure
Bank.’ A great many complex problems are involved in the case.
You know what countless Hippopotami have been met with in Sicily—
literally tens of thousands, of two species. Your remains from the
Krendi Cave are identical with one of the Sicilian species, and it is clear

1 The statement at page 250 of the 3 Extract from Letter to M. Lartet,
paper on E£. Colwmbi, written two years | Nov. 1, 1864.—‘ The San Teodoro speci-
after the publication of this letter, ren- | mens are certainly of the African Ele-
ders it doubtful if this was really £. | phant.’—[Eb.]
antiquus.—|{ Ep. } 4 See antea, p. 465.—[Ep.]

2 See antea, p. 283.—[Ep.]

OSSIFEROUS CAVES OF SICILY. 553

that Malta and Sicily must have been continuous. But a great fresh-
water Mediterranean lake hypothesis will not apply to the case. There
are no freshwater deposits on the north coast of Sicily, but abundance
of marine Pliocene strata. Further, the Hippopotamus remains from
Candia are the same as the Sicilian. Candia must at that time have
been connected with the Morea.!

1 Tf Dr. Falconer had lived, he con-| the Continent of Africa, within the
templated writing an essay ‘On the | modern period.’ See also p.596.—[En.]
Proofs of the Severance of Sicily from

354 OSSIFEROUS CAVES OF GIBRALTAR.

XXII. ON THE FOSSIL CONTENTS OF THE
GENISTA CAVE, GIBRALTAR.

BY HUGH FALCONER, M.D., V.P.R.S., FOR. SEC. G.S., AND G. BUSK, ESQ.,
F.R.S., F.G.S.

[In a Letter to His Excellency General Sir W. J. Codrington, K.C.B., &e. &e.
Governor of Gibraltar. |}

THE circumstances which led to our visit to Gibraltar, and
the objects we have had in view, are so well known to your
Excellency that it is unnecessary on our part to do more
than refer to one or two incidents in the early history of the
cave.

When the interesting objects contained in the upper
chambers of the ‘Genista Cave’ on Windmill Hill were
brought to light by Capt. Brome, your Excellency addressed
a letter to the Secretary-at-War, giving a preliminary report
on the results; that communication was forwarded from the
War Office to the President of the Geological Society of
London, with a request for an opinion as to the importance
in the interest of science of following up the exploration,
and for suggestions as to the manner in which it could be
best conducted. The reply led to the sanction of the Secre-
tary-at-War for the further exploration of the cavern by
means of the labour of the military prisoners, under the able
superintendence of Capt. Brome; and, to pass over minor
incidents well known to your Excellency, the objects dis-
covered were forwarded to us in London for identification
and scientific examination.

Having devoted several months to the study of the cave-
collections successively transmitted to us, which were so
carefully classified, by means of distinctive marks, by Capt.
Brome, the Governor of the Military Prison, as to place the
main facts clearly before us, we were so strongly impressed
with their importance that we determined, on your Excel-

1 This letter, written at Gibraltar in | British Association placed 150/. at the
October 1864, was communicated by | disposal of Dr. Falconer and Mr. Busk,
the Secretary of State for War to the | for the purpose of promoting researches
Geological Society after Dr. Falconer’s | in the ossiferous caves of Gibraltar, and
death, and was published in the ‘Quar- | up to the time of his death Dr, Falconer
terly Journal of the Geological Society,’ | was busily engaged in identifying the
for March 1865. In September 1864, the | fossils which were sent home.—[Eb.]

OSSIFEROUS CAVES OF GIBRALTAR. 555

lency’s invitation, to visit Gibraltar and examine the general
condition of the cave on the spot; for the discoveries in the
Windmill Hill Cave have not only yielded unexpected results
regarding the former state and the ancient animal population
of the rock itself, but they further point to a land connection
between the southern part of the Iberian peninsula and the
African continent at no very remote geological epoch.

Capt. Brome’s Report, dated 21st August, 1863, with the
plan and section which accompany it, so clearly explains the
nature of the Windmill Hill Cave, that it is unnecessary for
us to enter on the present occasion into any detailed descrip-
tion of it. The rock abounds in caves, which are of two
classes. Ist. Seaboard caves at various heights above the
level of the sea and horizontally excavated in the ancient
cliffs by the waves. 2nd. Inland caves descending from the
surface and in connection with great vertical fissures, by
which the mass of the rock has been rent at remote epochs
during disturbances caused by violent acts of upheavement,
like the well-known cavern of St. Michael. The ‘ Genista’
Cave of Windmill Hill belongs to the second class; it forms
part of a great perpendicular fissure, which, by the vigorous
measures adopted by Capt. Brome, has either been excavated
or traced downwards to a depth of upwards of 200 feet below
the level of the plateau of Windmill Hill. It was full of the
fossil remains of quadrupeds and birds, of the former of
which some are now wholly extinct, others extinct in Europe
and repelled to distant regions of the African continent,
others either now living on the rock or in the adjoining
Spanish peninsula.

The following is a list of the species which we have at pre-
sent identified :—

Pachydermata.
Rhinoceros Etruscus (?).1. Extinct.
Rhinoceros leptorhinus (syn. R. megarhinus). Extinct;
abundant.
Equus. Young animals only; species undetermined.
Sus priscus (?). Extinct.
Sus scrofa. Living.

Ruminants.

Cervus Hlaphus, var. barbarus. Fossil remains abundant.
Cervus Dama, or a nearly allied form. Abundant.

1 In the Gibraltar cave we have got | specimens —teeth, tibias, astragalus,
an upper half of a femur of a Rhinoceros | atlas, &c., belong to R. leptorhinus (R.
which is small, old, and indistinguish- | megarhinus).—Letter from Dr. Falconer
able in size and form from Rf. Htruscus, | to M. Lartet, Sept. 9, 1864.—[Ep.]
but no teeth as yet. All the other

556

Bos.

OSSIFEROUS CAVES OF GIBRALTAR.

A large form, equalling the Aurochs in size; re-

mains few and imperfect; species undetermined.

Bos taurus.
Capra hircus (?).

Abundant in the upper chamber.
In the upper chamber.

Capra Aigoceros, form A.; Capra Aigoceros, form B.
Two forms of Ibex, probably extinct, but in vast abund-
ance throughout the fissure.!

Rodents.
Lepus timidus.
Lepus cuniculus.
Mus rattus.

Rare.

Carnivora.
Felis Leopardus.
Felis pardina.
Felis serval.
Hyzena brunnea.?
Canis vulpes.

Very abundant at all depths.

Ursus, sp. Not the Cave Bear; form undetermined.

Delphinide.
Phoczena communis.

Birds. Remains numerous ;

genera and species undeter-

mined.
Tortoise. Rare; species undetermined.
Fish. Remains numerous in the upper chamber.

1 Extract of Letters from Dr. Falconer
to M. Lartet, Sept. 9, 1864.—‘I am of
opinion that there are two distinct
species of Ibex among the Gibraltar
fossils ; the one large with a step to
the last lower, and a grooved flange to
the talons of the last upper, molar. We
have a beautiful upper maxillary of this
form which you have not seen. It is
larger than a very large male Siberian

Ibex. The following are the dimensions
of the 6 molars :—
Upper Lower
maxilla maxilla
In. In.
Small Gibraltar Ibex 2°85
Large 4 A 3°25
Siberian Ibex . 2°85 2°82
Pyrenean Ibex, female 2°8 2°95
Nubian Ibex, male 2°8 2:97

‘The larger upper maxillary of Gib-
raltar has the teeth quite as large in
proportion as your large lower jaw from
the “ Gorge d’Enfer.” In my view the
latter agrees with the characters of the
Ibex of the Alps, and differs from the
Siberian or Nubian forms. We have

no jaw so large.’ Sept. 11, 1864.—‘ Our
large upper maxillary of Gibraltar Ibex
compares well with the Ibex of the Alps
(female). From another cave in Gib-
raltar (the Judge’s Cave) we have the
lower jaw of an Ibex, entire to the
tip of the coronoid, which is nearly of
the same age as your “‘ Gorge d’Enfer”
specimen,

Gorge Gibral-
d’Enfer tar (B)
‘ In. In.
Height of four last
molars . ors8 27 2°45
Height of jaw at
penultimate . 17 1:15

‘The lower jaw of the female Ibex of
the Alps has the penultimate and last
lower molars very much as in the Gib-
raltar form (B) and as in the ‘“ Gorge
d’Enfer” specimens, but the horn-core is
nearly round in section and less abruptly
acuminated, and there is no lateral
compression.’—[ ED. |

* Hyena crocuta. See antea, p. 465,
note 2, and Mr. Busk’s paper in the Proe.
Lin. Soe, for May 3, 1866, p. 62.—[ Ep. ]

OSSIFEROUS CAVES OF GIBRALTAR. bo7

Apart from the still immature state of the investigations,
it would be quite beyond the limits within which we are re-
stricted in this communication for us to enter in detail upon
the conclusions to which the data furnished by the fossil re-
mains lead; we shall therefore confine ourselves to a few of
the more important general points.

The rock is now bared of natural forest-trees, and destitute
of wild animals, with the exception of the hare, rabbit, fox,
badger, and a few magot monkeys, the last in all probability
the descendants of introduced animals. The fossil remains of
the ‘ Genista Cave’ establish beyond question that the rock
was formerly either peopled by, or the occasional resort of,
large quadrupeds like the elephant, rhinoceros, aurochs, deer,
ibex, wild horse, boar, &c., which were preyed upon by hyzenas,
leopards, the African lynx, and serval: that the remains
were transported by any violent diluvial agency from a dis-
tance is opposed to all the evidence of the case. The manner
in which they were introduced into the Windmill Hill Cave
we believe to have been thus :—The surface of the rock and
its level in relation to the sea were formerly different from
what we now see. The wild animals above enumerated, dur-
ing a long series of ages, lived and died upon the rock. Their
bones lay scattered about the surface, and in the vast ma-
jority of instances crumbled into dust, and disappeared under
the influence of exposure to the sun and other atmospheric
agencies, as constantly happens under similar circumstances
at the present day. But a certain proportion of them were
strewed in hollows along the lines of natural drainage when
heavy rains fell; the latter, for the time converted into tor-
rents, swept the bones, with mud, shells, and other surface-
materials, into the fissures that intercepted their course ;
there the extraneous objects were arrested by the irregulari-
ties of the passages, and subsequently solidified into a con-
glomerate mass by long-continued calcareous infiltration.
That elephants frequented the rock is proved by a valuable
and perfect specimen of the penultimate upper molar tooth
of an extinct species, which we have ascertained to be Hle-
phas antiquus discovered by Mr. Smith of Jordan Hill, in a
sea-beach on Europa Point, 70 feet above the level of the
sea. That the hyznas were dwellers upon the rock is also
established by the fact that, in addition to numerous bones,
we have discovered a considerable quantity of coprolites of
Hyena brunnea' among the ‘ Genista Cave’ relics. Some of
the species must have peopled the rock in vast numbers. We
infer, upon a rough estimate, that we have passed through
our hands bones derived from at least two or three hundred

1 See antea, p. 586, note 2.—[ Ep. ]

558 OSSIFEROUS CAVES OF GIBRALTAR.

individuals of Ibex swept into the Windmill Hill fissure ; in
no instance have we observed fossil bones attributable to one
complete skeleton of any one of the larger mammalia.

That the rock now so denuded of arboreal vegetation was
then partially clothed with trees and shrubs, as the corres-
ponding limestone mountains on the opposite side of the
straits are at present, is so legitimate an inference as hardl
to be open to rational doubt. It is now a pinch to find suffi-
cient food at the end of the hot season for the flocks of goats
which are reared on the promontory; while it is a matter of
absolute difficulty to find fodder at all for the few cows that
are kept by some of the officers of the garrison. When ele-
phants, rhinoceros, wild oxen, horse, boar, deer, &c., &e.
either peopled or resorted to the rock in considerable num-
bers, there must have been abundant trees and more or less
constant green food for them. Bare exposed masses of rock
get intensely heated by a southern sun, they repel moisture
by being thus heated, and raise the mean temperature of
the locality by radiation ; while, on the contrary, a clothing
of trees and of fruticose vegetation both tempers the heat,
attracts moisture, and greatly increases the fall of rain. We
are aware that your Excellency’s attention has been directed
to planting-operations on the ‘rock.’ Numerous and re-
peated failures must be looked for at the commencement ;
but the facts above mentioned would indicate that suc-
cess may ultimately be attained, with much benefit to the
station.

The next prominent point in the case is the character of
the extinct fauna of Gibraltar regarded as a group. Of the
prevailing fossil forms which occur in England, Germany,
and France, as far south as the northern slope of the Pyre-
nees and the shores of the Mediterranean, such as the Mam-
moth, Rhinoceros tichorhinus, Ursus speleus, Hyena spelea,
&e.,' not a vestige has been detected among the fossil re-
mains of Gibraltar. In the latter the Carnivora are the most
significant. The three species of Felis are of African affini-
ties; and Hyena brumnea,? now for the first time ascertained
to have existed formerly in Europe, is at the present day chiefly
found near the Cape of Good Hope and Natal. That any of
these wild animals could have crossed the straits from Barbary
to Europe is contrary to.all probability. The obvious inference
is that there was a connection by land,’ either circuitous or

1 Letter to M. Lartet, Sept. 29, 1864.| ? This identification has since been
‘ We found not a trace of Cervus Taran- | shown to be an error.—See antea, p.
dus, or of the Lagomys figured by | 556, note 2.—[Ep.]
Cuvier.’ —[Eb. ] 3 See antea, pp. 552-3.—[ Ep. |

OSSIFEROUS CAVES OF GIBRALTAR. 559

direct, between the two continents, at no very remote period,
somewhere within the Mediterranean area. To arrive at any
further evidence bearing upon this very important question,
from the rock of Gibraltar, becomes an object of the highest
general and scientific interest.

Human remains were found in great abundance in the
upper chambers. They appear to have belonged to between
thirty and forty individuals. They were accompanied by
stone implements of the polished-stone period, broken querns,
a large quantity of pottery, marine shells of edible species,
and some other objects enumerated in Capt. Brome’s Report.
No way of access from the surface by which these materials
could have been introduced has been discovered; but, on
carefully examining the ground, we believe, with Capt.
Brome, that the entrance was somewhere under the southern
half of the east wall of the prison-enclosure. Until the
aperture from the surface is discovered, no certain conclusion
can be arrived at. Considering the time and labour which
have been expended on the cavern, it would be a subject of
ereat regret if the exploration were left incomplete on this
important point. We would therefore venture strongly to
recommend that the excavations be continued through the
ground over which the east wall runs, until the external
aperture is detected. We believe that it will be found in the
fissure outside the east wall, which Capt. Brome has so
sagaciously and perseveringly explored.

The human bones are of high interest in consequence of
certain peculiar characters which many of them present.
They appear to belong to widely different epochs, although
none of them perhaps of very high antiquity (i.e. before the
historical period). That the upper chambers of the cave
were ever inhabited by savage man we consider to be highly
improbable. It seems more likely that they were used as
places of deposit for the dead.

As regards the final disposal of the interesting and import-
_ant relics discovered in the ‘ Genista’ Cave, a complete series
ought to be deposited in London, either in the British Museum
or in the Museum of the Royal College of Surgeons. But we
consider it to be of still higher importance that a collection
should be retained for Gibraltar. In the progress of the vast
defensive works which have been carried on during the past
century, in scarping and tunnelling the rock, objects of high
interest, relating either to its natural history or archeology,
have been brought to light; but in the great majority of
cases they have either been disregarded or lost. Instances
might be cited from Col. James’s ‘ History of the Herculean
Straits,’ 1771, and from Major Imrie’s ‘Memoir on the

560 OSSIFEROUS CAVES OF GIBRALTAR.

Mineralogy of the Rock,’ in 1797. In 1844 a laudable effort
was made by the late Archdeacon Burrow to establish a
museum on the rock; but, after languishing some time, it
failed from the want of proper support. The relics of the col-
lection were afterwards exhibited in the Soldiers’ Home; but
when that institution was given up, no place remained either
for displaying or taking proper care of the collection. Some
of the brightest records of the military glory and prowess of
our country are indissolubly connected with Gibraltar. A
great nation like England cannot afford to neglect, or disre-
gard without reproach, whatever bears on the natural history
or archeology of so renowned a possession. That the naval
and military services take the liveliest interest in such objects
is placed beyond doubt by the United Service Museum of
London, founded upon collections contributed by them from
all parts of the world; but it appears to us that the forma-
tion and maintenance of a local museum at Gibraltar, illus-
trative of its products and relics, ought not to fall upon the
garrison, who are only temporary residents, and that it is
more properly an Imperial obligation. The least expensive
and best mode of carrying the object into effect would pro-
bably be to have a room in the Library reserved for the pur-
pose, and under the management of the Library Committee.
The only outlay would be in the construction of the apart-
ment and in the glass cases for the objects ; no establishment
would be required.

In case of any proposal of this nature being entertained,
we would venture to suggest to your Hxcellency that the
collection should be strictly limited to objects of local interest,
having reference to the rock, the bay, the straits, and the
immediate vicinity. Everything from beyond these limits
should be excluded. A museum of reference of this nature
should include :—

1. Herbarium collection of the plants yielded by the rock.

2. A zoological collection of all objects, terrestrial and
marine, produced within the limits.

. A collection of specimens of minerals of the rock.

. A complete collection of the fossil remains yielded by
the ossiferous caves and bone-breccia of Gibraltar.

. An archeological collection of coins, pottery, and other
antique relics occurring within the circuit of the bay.

(yy Sb)

In illustration of the absolute need there is of a local col-
lection of the kind here indicated, we may mention that,
being anxious to fix the age of the pottery yielded in such
abundance by the Windmill Hill Cave, no similar materials
for comparison derived from the ancient ruins of Carteia, or

OSSIFEROUS CAVES OF GIBRALTAR. 561
from points in the Mediterranean resorted to by the Phceni-
cians, were to be found in the British Museum.'! The proof
of the antiquity of the human race is one of the leading ques-
tions that occupy the attention of educated and scientific
men at the present day. That human remains and other
objects bearing upon it are considered of high value is suffi-
ciently proved by the fact that a grant of 10001. was passed
for the purchase of a collection of this kind from the Valley
of the Vézére, in the south of France, during the last session
of Parliament, for the British Museum. One of the human
skulls yielded by the rocks many years since appears to us to
point to a time of very high antiquity. In fact, it is the most
remarkable and perfect example of its kind now extant.? In
the absence of a properly organized museum no record exists
of the precise circumstances under which this interesting
relic was found, and that it has been preserved at all may
be considered a happy accident; it has cost us much labour,
and with but partial success, to endeavour to trace its his-
tory on the spot where it turned up.

Our time has been so fully occupied by the examination of
the cave collections and collateral subjects that we have
only been able to make a cursory examination of the geology
of the rock. We entirely agree with the opinions expressed
in the excellent memoir of Mr. James Smith, of Jordan Hill,
that it bears unmistakeable evidence of having undergone
extraordinary disturbance, both of upheaval and depression,
during the Quaternary or immediately pre-modern period ;
but the data are complex, and in some instances obscure.
Now that a complete topographical survey of the rock has
been completed on a large scale, a geological survey would
be a matter of comparative ease; and we would submit to
your Excellency’s consideration the expediency of an appli-
cation being made for the services of an assistant upon the
Geological Survey of England, to be deputed for the purpose.
The area is so compact and limited that the survey, includ-
ing that of the surrounding bay, need not occupy much
time.

We cannot bring this letter

to a close without expressing

our opinion of the value and importance of Capt. Brome’s

1 Extract of Letter from Dr. Falconer
to M.le Due de Blacas——*‘ Of the pot-
tery, several rude vases are quite entire.
None of them in the slightest degree
resembles the ‘Hut Vase’ to which you
refer, but rather some of the small vases,
cups, &c., which surround the Hut Vase
figured by Visconti (PI. i.), or as figured
in “ Birch’s Ancient Pottery,” yol. i. p.
197, fig. 175.

VOL. Il,

2 Respecting this skull, Dr. Falconer
wrote to a relative as follows: ‘If you
hear any remarks made, you may say
from me, that I do not regard this pris-
can pithecoid man as the “missing link,”
| so to speak. It is a case of a very low
type of humanity—very low and savage,
and of extreme antiquity—but still man,
and not a halfway step between man
| and monkey.’ —[Ep.]

00

562 OSSIFEROUS CAVES OF GIBRALTAR.

exploration of the Windmill Hill Cavern, under the support
and enlightened countenance and encouragement which we
are well aware he has uniformly received from your Excel-
lency during the progress of his operations, and which have
led in a great measure to their successful issue. The only
account of the mineralogy of Gibraltar that has been pub-
lished is in the excellent ‘ Brief Description ’ by Major Imrie,
of the Royal Artillery, which appeared in the ‘ Edinburgh
Philosophical Transactions’ in 1797. In 1844 Mr. Smith,
of Jordan Hill, brought out his valuable memoir on the
Geology of Gibraltar; but the fossil mammalian remains of
the bone-breccia were only very cursorily noticed by both
authors. In the latter half of the last century they attracted
the attention of William and John Hunter, in papers which
are to be found in the ‘ Royal Transactions,’ but without an
attempt at precise identification. Cuvier, in his great work,
the ‘Ossemens Fossiles,’ in 1823, gave a special chapter on
the ossiferous breccias, and devoted much attention to those
of the Mediterranean. From the materials derived from the
rock which passed under his hands, he was able to detect
evidence only of two extinct species, one of which is doubtful.
He concludes his remarks on the Gibraltar remains in the
following terms :—

‘Voila done dans ce petit nombre d’os de Gibraltar que
j'ai pu me procurer, au moins une espéce de liévre et proba-
blement une espéce de cerf, dont les pareils ne sont pas con-
nus en Europe.

‘Que seroit-ce si quelque naturaliste résidant sur les lieux
prenoit la peine de recueillir et de dégager avec soin ceux
qui se découvriroient pendant quelques années, comme je Vai
fait pour les ossemens de nos gypses? D’aprés ce que nous
allons voir dans les articles suivans, on ne peut douter qu’il
n’y fit des récoltes abondantes. et intéressantes.’ (Op. cit.
tome iv. p. 174.)

From that period down to the present day hardly any ad-
dition has been made to our knowledge of the subject, during
a lapse of forty years, until Capt. Brome undertook the ex-
ploration of the ‘ Genista Cave ;’ and the best commentary
upon the preceding citation is furnished by the fact that the
materials collected by him have enabled us to determine up-
wards of twenty species of mammalia above enumerated,
many of them extinct, and all of them bearing importantly
on the ancient condition of Gibraltar. Indeed it is within
the facts of the case to say that, in the important walk of the
mammalian paleontology of Gibraltar, Capt. Brome has done
more than was effected by the united labours of his prede-
cessors since the rock became a British possession. The per-

OSSIFEROUS CAVES OF GIBRALTAR. 553

severing energy and vigour with which he has followed up
the inquiry, and the minute and scrupulous care with which
he has discriminated and arranged the objects, are worthy
of the highest commendation, and more especially so as the
subject was new to him. We are inclined to believe that the
labour of military prisoners was never better directed in the
interest of science.

We have to tender our best acknowledgments to your
Excellency for the very cordial reception which you have
given us, and for the pains you have taken to forward the
objects of our visit in every respect. We beg leave also,
through your Excellency, to offer our thanks to the military,
naval, and civil departments of the service for their hearty
co-operation. Our thanks are more especially due to Major-
General Frome and the officers of the Royal Engineers, and
to Capt. Ommanney, R.N., the senior naval officer of the
station, who have rendered us every assistance,

t

oo 2

564 FOSSILS FROM FOLKESTONE,

XXIII. NOTES ON A COLLECTION OF FOSSIL BONES
DISCOVERED IN A SECTION OF GRAVEL
IN EXCAVATING THE FOLKESTONE BAT-
TERY.!

Tue fossil remains to be noticed in the sequel were disclosed during
the excavation of the Folkestone Battery. A careful section of the
ground was made by Lieut. Robert Vetch of the Royal Engineers, con-
sisting first of the lower greensand and blocks of Kentish rag to a depth
of 10 ft. 8 in.: next, in ascending order, a bed of marl and flint gravel,
which immediately caps the lower greensand, and in which alone the
fossil bones were found; this bed was 18 in. thick: next, a whitish
loam, varying from 4 ft. 6 in. to 7 ft. 6 in., which is used as a brick-
earth: and above all was a superficial layer of made earth 6 in. deep.
At the bottom of the section abundant remains of Perna and 2
were found, these being characteristic greensand fossils. The collection
of fossil bones is of great interest from the unusual association of old
and glacial forms of Mammalia which it discloses, including Hippopo-
tamus major and the Irish Elk (Cervus eurycerus), with remains of
Rhinoceros and other forms. They are all in precisely the same
mineral condition, ana some of the largest bones are in a state of
such perfect integrity that it is clear they cannot be other than of local
origin, and that they cannot have undergone any considerable amount
of rolling.

Hippopotamus major.—The bones of this species are numerous, and
in such integrity of preservation as I believe have nowhere as yet been
met with elsewhere in Great Britain, with the exception of the remains
found in the Valley of the Aire, near Leeds. Of these, the mest
remarkable are the bones of a right fore-leg, consisting of the humerus
(No. 1) and united radius and ulna (No. 2), perfectly entire, of an
adult animal. The articular surfaces of the bones are as smooth and
perfect as if they had been yielded by a recent animal. Dimensions of
humerus :—

Extreme length, outer side, 20° in. Extreme length, inner side, 17-75 in. Girth
of lower articulation, 20° in. Transverse diameter of articular surface, 4°3 in.
Girth of shaft at constriction, 10° in. Girth of ditto at tuberosity, 12-5 in. Girth
of ditto above tuberosity, 12°75 in. Girth of articular head and bicipital tube-
rosity, 21:5 in. Antero-posterior diameter, 9° in. Transverse diameter of arti-
cular head alone, 4°4 in, Antero-posterior ditto, 4°83 in.

The right radius and ulna are completely anchylosed, with the ex-

1 These notes were written in March | contain full descriptions of fossils from
1863, but were never published. The | similar excavations, in My. Mackie’s
collection of bones to which they refer | collection at Folkestone.—[ Eb. ]
is in the Museum of the Geological ? Blank in MS.—[Eb.]

Society. Dr, Faleoner’s Note-books also

FOSSILS FROM FOLKESTONE. 565

ception of a small fissure, proving them to have belonged to an adult
animal. The articular surfaces and olecranon are perfectly entire, and
there is no appearance anywhere of gnawing.

Dimensions.—Extreme length of radius, measured in front, 13: in. Extreme
length of ulna, measured along the curve to olecranon, 17: iv. Extreme length of
ditto measured straight from top of olecranon to outer articular edge, 17:5 in.
Girth of lower articular head, 16: in. Girth of united shafts, 11:5 in. Girth of
united upper articular surfaces, 14°5 in. Transverse diameter of articulation
with humerus, 4:2 in. Vertical height of ditto to hook of olecranon, 2°8 in.

No. 3. Right Scapula.—About half of this bone remains in a state
of integrity, what is wanting having been lost by recent fractures. The
glenoid cavity is perfectly entire, with a large projecting tuberosity ;
the spine and crochet are broken off, and all the basal and leafy portion
near the base.

Dimensions.—Antero-posterior diameter of glenoid cup, 4°8 in. Transverse
ditto, 3:8 in. Transverse diameter of proximal end, including tuberosity, 7:2 in.
Transverse diameter of neck, 4°4 in.

No. 4. Axvis.—A superb specimen of the second vertebra nearly
entire, and with the articular surfaces and the odontoid processes quite
entire. The following are the dimensions :—

Transverse diameter of lower surface of body, 4°6 in. Antero-posterior ditto,
2:9 in. Transverse diameter (chord) of upper articular surface, 68 in. Height
of body to summit of odontoid, 6: in. Antero-posterior diameter of united body
and spinous process, 7°8 in. Transverse diameter of the spinal foramen, 2-2 in.
Antero-posterior ditto, 1°6 in.

The only parts wanting of this fine specimen are the left inferior
oblique process and the upper part of the spine, both of which have
been lost by a recent fracture.

Besides the axis there are three other cervical vertebre (Nos. 5, 6,
and 7), one of which is perfect, with the exception of the spinous
process; another has the body perfect, but is minus the spinal arch ;
and a third shows only the body. The next specimen to be noticed is
a perfect metacarpal or metatarsal bone, probably a metatarsal, but the
exact position of which has not yet been determined. There is no
specimen of a canine or incisor of Hippopotamus, but, strange to say,
one detached true molar (No. 8), which has undergone more rolling
than all the rest of the bones together.

Dimensions.—Length of crown, 2°3 in.; transverse ditto, about 1°5 in.

Cervus eurycerus (Irish Elk).—The determinable specimens of this
species consist of the frontal part of the cranial box, together with the
surfaces left by two shed horns; a considerable portion of the skull-
cavity (for the cerebrum) is also shown, but all the rest of the skull is
wanting (No. 9). The frontal plateau exhibits the sudden fall with the
two depressions below the horn-pedicles characteristic of the species,
and on the left side also a portion of the orbit, together with the
characteristic supraorbitary foramen. Although mutilated, this cranial
fraement leaves no doubt that the species was the Irish Elk, and in
mineral condition it agrees entirely with the Hippopotamus bones, and
what remains of it is in an equally unrolled condition,

566 FOSSILS FROM FOLKESTONE.

No. 10 isa basal fragment off the beam of the right antler of the
Trish Elk,a shed specimen. The cicatrix of the shed surface is smooth
and distinctly marked; the brow-antler had been broken off before
embedding, but the fractured surface obscured by matrix is visible.
The upper extremity of the fragment is rough and unequal, showing
that it also had been fractured before embedding.

Dimensions.-—Extreme length of specimen, about 8: in. Girth at bur, 12° in.
Girth of shaft above brow-antler, 9° in.

The basal surface shows the usual oblique insertion characteristic of .
the species.

No. 11. Rhinoceros.—Fragment comprising the inferior half of the
right humerus, broken off a little above the constriction; unfortunately
the tuberosity above the outer articular surface has been broken off,
preventing a close comparison with a corresponding specimen of
RL. tichorhinus from Lawford, near Rugby, belonging to the series
described by Dr. Buckland in the ‘ Reliquie Diluviane.’

The following are the dimensions compared with that specimen :—

Folke- R. tich.

stone Lawford
Transverse diameter of articulating surface, outer side,

nearmiddle . ; : : : . : 3°8 4°8
Antero-posterior diameter of larger articulating division 3°6 4:2
Girth of shaft at constricted portion . : ; : 875 11°25
Extreme length of fragment . ; : . - : 8-5

The lower articular surface is completely united to the shaft, preving
that the animal was adult, and the size and form of the specimen, so far
as a comparison can be instituted, differs so decidedly from that of
Rhinoceros tichorhinus that it probably belongs to another species,
which may be &. hemitechus, as | found in Mr. Mackie’s collection at
Folkestone, from corresponding excavations there, a singularly perfect
last true upper molar of R. hemitwchus.

No. 12.—The corresponding part of the left humerus subsequently
found, but minus the articular head. It agrees exactly in size with the
other half,

Vo. 13 is a fragment of the shaft of the femur comprising the middle
portion immediately above the condyles, the expanded part of the third
tuberosity, and on the inner side the rough portion immediately below
the articular head; both articulations are unfortunately wanting, and
the hooked terminal part of the third trochanter is also broken off.
There are, therefore, no decisive characters left to determine the species
with certainty, but in all its proportions, and, in general form, it differs
from the corresponding parts of Rhinoceros tichorhinus. The muscular
ridges and depressions are in very bold relief, proving that the animal
must have been very old. The dimensions are :—

Length of fragment, 11-in. Girth at constricted portion of shaft below third
trochanter, 8°75 in.

Nos. 14, 15, 16, 17. Elephas.—Species unknown; three fragments
of a large scapula of an Elephant, very much mutilated. The proximal
fragment exhibits a portion of the glenoid surface ; but there is nothing
to show what the species was, whether EZ. primigenius or E. antiquus,
both of which I found to occur in the Folkestone depcsit by Mr.

FOSSILS FROM FOLKESTONE. 567

Mackie’s collection. These are all the determinable fragments refer-
able to Elephas.

. No. 18. Bison priscus.—A fine specimen of the cranial box, com-
prising the whole of the occipital, the condyles, the base of sphenoid,
and a considerable part of both temporal fosse ; but the whole of
the frontal, and indeed all in front of the occipital crest broken off by a
very irregular fracture ; quite enough, however, remains of the occiput
and temporal fosse to determine the offset of the horns and to prove
that the species is Bison priscus. The sutures are so open as to
show that the animal, although adult, was not very old. The follow-

ing are the principal determinable dimensions :—
Folkestone Museum
E Abe specimen specimen
Transverse diameter of the two occipital condyles : 59 53

Transverse diameter of left condyle 2°8 2:2
. x foramen magnum ‘ - 1°6 ilbré
Vertical ditto : : : 0 about a7) 1:8
Height from summit of occipital crest to inferior border
of foramen magnum : : a : . : 6-2 6-4
Width of occiput between terminations of temporal
fossa). : . : : : e . 5 76 75

No. 19.—F ragment of a horn-core, very much mutilated, and forming
the middle portion, probably of the right side, 10 inches long, and
11°5 inches where thickest.

No. 20.—Left scapula of a large Ruminant, showing the whole of
the glenoid surface,! the greater part of the salient crest and the thin
basal portion alone wanting. The rough muscular impressions prove
distinctly that it must have been yielded by a very old animal. The
following are the dimensions :—

Transverse diameter of glenoid surface, 3°8 in. Antero-posterior ditto, 3°1 in.
Width of neck where constricted, 3°6 in.

This scapula belongs probably to Bos priscus, but this determination
is approximative.

No. 21. Right seapula.—Corresponding very closely in general form
and fracture with the specimen last described, but of smaller size, and
somewhat different in form, being more slender and somewhat differ-
ently formed. It also shows the whole of the glenoid surface and the
greater part of the spine; it is proved by the muscular rugosities to
have been yielded by an extremely old animal.

Dimensions.—Transverse diameter of glenoid cavity, 3°3 in. Antero-posterior
ditto, 2-7 in. Transverse diameter at neck, 3-4 in.

This specimen may be of the frish Elk. The glenoid surface is
singularly perfect, and free from any show of rolling.

No. 22. Large Ruminant.—Fine specimen, comprising the inferior
half of the left humerus of a large Ruminant, most probably Bos
priscus. The only deficiency is in a portion of the outer articular
surface by recent injury.

Dimensions. — Width of articular surface, lower, 4:4 in. Antero-posterior
diameter of outer condyle and tuberosity, 45 in. Girth of shaft, 8-5 in. Width
of fissure behind, for hook of olecranon, 1°3 in.

1 The glenoid surface of this specimen | recent animal. There is no mark of
is as perfect as if it had been that of a | gnawing or rolling,

568 FOSSILS FROM FOLKESTONE.

No. 23.—A corresponding left humerus, but without articular sur-
face.

No. 24.—Radius and ulna united, of a large Ruminant, correspond-
ing exactly in size with the humerus just noticed, but belonging to
the opposite side (right). It is in the most perfect state of entirety,
with the exception of the olecranon, which is broken off. The upper
and lower articular surfaces are quite perfect. The following are the
dimensions :—

Extreme length, 15°5 in. Girth (least), 8° in. Transverse diameter, lower
articular surface, 4°5 in. Antero-posterior ditto, 2°5 in. Transverse diameter
upper head, 4-6 in, Transverse diameter, upper articular surface, 4°3 in. Antero-
posterior ditto, where greatest, 2-2 in.

No. 25.—Mutilated fragment in two pieces of the central part of the
shaft of a metacarpal or metatarsal of a large Ruminant, probably
metatarsal ; indeterminable.

No. 26. Large Ruminant.—Atlas vertebra of a very large Ruminant,
mutilated, the wings of the transverse processes being breken off; it is
probably that of Bison priscus, but not certain.

Height of body from articular edge to articular edge, 4°3 in. Antero-posterior
diameter of condyloid surface, 2° in. Transverse ditto, 2-4 in. Antero-posterior
diameter of spinal canal, 2-3 in. Antero-posterior diameter of body from tube-
rosity to tuberosity, 4:4 in.

The specimen is small for the skull of Bison priscus, although it
probably belongs to that species.

No. 27 is another atlas, more perfect, and probably that of Irish Elk.

0. 28. Cervus Tarandus (Reindeer).—Basal part of the antler com-
prising a short part of the beam, the brow-antler very much depressed
and given off in a line with the bur. There is barely enough to deter-
mine what it is, the specimen appearing in the same mineral condition
as the other remains.

No. 29.—Cervical vertebra of a Ruminant, minus the anterior part of
the body. The rest perfect, probably of Cervus eurycerus.

No. 30.—Another more perfect, but minus the spinal arch and the
processes on one side.

No. 31.—A dorsal vertebra of a large Ruminant of very large size,
and extremely old, probably Bos priscus.

Nos. 32, 33.—Two lower ends of tibia, probably of Bos priscus (right
and left).

No. 34. Sus.—Fragment of right ramus of lower jaw broken off, in
front of the penultimate molar, and behind by a rugged fracture through
the ascending ramus; has a very rolled appearance, and has no teeth,
the alveoli being filled up with matrix.

No. 35.—A corresponding fragment of the lower jaw of a large
Ruminant, either Bos priscus or the Irish EJk.

No. 36.—Body of the vertebra of a Ruminant, probably cervical, but
very much mutilated.

No. 36a.—Detached fragment of a large cervical vertebra, comprising
nearly the superior and inferior oblique processes, probably of Bison
priscus.

Nos. 37, 38, 39.—Thrce splintered fragments of the horn of Cervus
eurycerus.

FOSSILS FROM FOLKESTONE. 569

®

Nos. 40, 41, 42, 45.—Four splintered fragments of leg-bones of a
large Ruminant ; indeterminable, but probably of Bison priscus.

No. 44.—Articular head of the femur of one of the large Ungulata ;
detached from the shaft, and undetermined; greatest diameter 3:8
inches.

No. 45.—Portion of a scapula; intermediate part; probably of
Rhinoceros.

No. 46.—Fragment of an axis vertebra of a Ruminant.

No. 47.—Basi-sphenoid and occipital condyles, with foramen mag-
num of a large Ruminant.

No. 48.—Fragment of a large flat rib, 2°6 inches wide; species un-
determined.

No. 49.—Point of snag of a large Cervine antler; undeterminable.

No. 50.—Basal fragment of a horn-core of a large Ruminant, pro-
bably Bos priscus.

Nos. 51, 52, 55.—Three large bone splinters; undetermined.

No. 54.—A rolled iliae bone; undetermined.

No, 55.—Fragment of a large rib, probably Elephant ; much rolled.

Nos. 56-63.—Hight fragments of the frontal portion of the skull of a
large Ruminant, and probably derived from the cranium of Bison
priscus, No. 18, but undeterminable.

No. 64.—A bone splinter; undetermined.

Nos. 65, 66.—Two bone fragments, to be determined.

No. 67.—A bone splinter, to be determined.

No. 68.—Mutilated portion of a right humerus of Rhinoceros, lower
end, like No. 11, but very much mutilated, and barely determinable.

570 PRIMEVAL MAN,

XXIV. PRIMEVAL MAN, AND HIS COTEMPO-
RARIES.!

Opinionum commenta delet dies, nature judicia confirmat.
Cicero ‘de Natura Deor.’ lib. ii. cap. 2.

; Fabricated and used by a people who had not the use of
metals. . . . The situation in which these (flint) weapons were found
may tempt us to refer them to a very remote period indeed, even
beyond that of the present world.

JOHN Frere, Esq., F.R.S., F.A.S., 22nd June, 1797.
‘ Archeologia,’ 1800, vol. xiii. p. 204.

PreFratory NOTE.

THE AUTHOR feels that he comes before the public at great
disadvantage with the present work. He is late in making
his appearance in a field in which he has been preceded by
a digest of the whole subject, produced by a philosophical
writer of high eminence and authority. But he has con-
fidence that men of science, and educated readers of all
classes, will take an interest in knowing from one of the
practical observers the steps by which the now generally
accepted belief in the remote antiquity of the human race
has been led up to its present advanced condition ; and that
they will not be indifferent to a narrative of the labours of
those who have pursued the inquiry alike through bad and
good repute, and who have contributed to it the greatest
impulse. For in casting a retrospective glance on the history
of any great question concerning the advancement of human
knowledge, with a view of awarding to each inquirer his
due share of desert, we must revert to the state in which the
subject was when they successively struck us. Once fairly
launched and accredited, the evidence with the conviction
founded upon it accumulates, like a snow-ball when set in
motion. The present work is intended mainly pour servir,
in the history of science, as a vindication of the part taken
by the author and his friends, M. Boucher de Perthes and
Mr. Prestwich, in the investigation of the remote antiquity
of the human race upon the paleontological and geological
evidence.

1 The following important historical | proofs of the remote antiquity of the
Essay was written in 1863, and was | human race, and the physical condition
intended as an introduction to a distinct | of the earth’s crust prior to, and at the
work with the above title, the object of | date of, man’s first appearance.—[ En. ]
which was to set forth the physical |

AND HIS COTEMPORARIES. 571

PrimEvVAL Man, anp His CoTEMPORARIES.

INTRODUCTORY REMARKS.

T

The geological proofs of the antiquity of the human race
have recently attracted a large measure of attention both
among men of science and among educated minds of all
classes. The question by common consent is admitted to
be one of the most important which has been raised of late
years; it is further generally admitted that it has made ad-
vanced progress towards a secure basis of demonstration,
and that English observers have given the greatest impulse
to the inquiry. They prefer no claim to have been the first
in the field; but when the inferences arrived at by the
earlier inquirers were either forgotten or discredited, whether
from suspicion of the evidence, from the undue weight of
opposed authority, or from deep-rooted prejudices in favour
of established opinions, they reopened the question upon
fresh observations, and tested the facts with such scrupulous
care, that they speedily carried with them the convictions
of men of science of all countries. Among these leading ob-
servers, Mr. Prestwich, in one walk of the investigation,
beyond all others holds the first rank. Once fairly Jaunched,
in 1858-59, the subject excited lively interest; the field was
soon occupied by a host of labourers ; additional evidence
poured in from various quarters; and one important con-
tribution was exhumed from the published transactions of a
‘ Jearned society, where it had slumbered for sixty years in
undisturbed repose.

My attention has been directed to the subject, more or less
continually, during many years of paleontological research,
and I propose to give a brief narrative of the circumstances
which lead me to take up the question of the antiquity of
the human race, in order to satisfy my readers that I do
not now approach the subject wholly unprepared by ante-
cedent study.

EE,

The vast expanse of the plains of Hindostan consists of a
fundamental stratum of very ancient alluvium, which is
developed in great force. It is longitudinally traversed,
after their escape from the Himalayah Mountains, by the
Ganges and Jumna rivers, which unite at Allahabad. Be-

572 PRIMEVAL MAN,

tween Agra and that station the fall of the country is con-
siderable, and the waters of the Jumna have produced the
usual effects of river-action operations during a long lapse of
ages. The stream had gradually cut its way down through
the ‘Ancient Alluvium’ to a depth of from 100 to 140 feet
below the level of the adjacent plains, thus exposing a very
instructive section of great extent. The ‘ Ancient Alluvium ’
here spoken of consists of beds of sand and clay, containing
embedded in them enormous quantities of the impure cal-
careous aggregations called nodular Kankar, and commonly,
at the bottom of the section, partial beds composed of
tabular masses of Slab-kankar, which are removed for archi-
tectural purposes.!. In scouring the channel down through
the Ancient Alluvium, the course of the river has intersected
these beds of Slab-kankar, which form, here and there, sub-
aqueous reefs obstructing the navigation of the river. The
Government of India, in 1828, undertook a series of ope-
rations, which extended over seven years, for the removal
of these and other obstacles, from the bed of the Jumna.
They were conducted chiefiy by Captain Edward Smith, an
accomplished officer of the Bengal Engineers. Fossil bones
of extinct Mammalia were discovered, in great abundance,
below the tabular extensions of Slab-kankar, which, beyond
question, is a formation of very great antiquity and long
antecedent to the flow of the present channel of the Jumna.
Captain Smith believed that among these osseous remains
he had detected human bones, some of which have been
figured. But after submitting them to a close examina-
tion, to which I shall have occasion to refer in the sequel,
the identifications proved to be untrustworthy or erroneous.
Another observation, however, was made by Captain Smith,
upon which, as a professional expert, he was competent to
give an authoritative opinion; namely, that some of the
fossil bones ‘ were dug from depths of 6 to 18 inches in the
firm shoal, which is composed of substances (sic) kankar
stone, rounded bricks (vitrified clay ?) more or less rolled
and cemented by mud and clay.’* These observations, pub-
lished in 1833, excited lively interest and discussion in the
Indian scientific journals. Mr. James Prinsep referred,

1. The nodular kankar of India, al-
though on a much greater scale of
aggregation, is identical in its nature
and composition with the impure cal-
careous nodules called ‘ Race,’ which are
so abundant in the ‘ Brick-earth’ de-
posits of the Valley of the Thames:
‘ Race, from their resemblance in form
to arhizoma or Race of Ginger.

The |

tabular kankar is not represented in
England by any equivalent deposits ;
but although more impure, it resembles
the travertine of the ‘Mammoth Felder,’
at Cannstadt.

* Journ. Asiat. Soc. of Beng. 1833,
vol. 11. p. 630, Pl. xxy.

3 Loe. cit.

AND HIS COTEMPORARIES, 573

in connection with them, to the then recent inferences ar-
rived at by Marcel de Serres, Christol, and Tournal, regard-
ing the human bones found in caverns in the south of
France, and used the following terms in his notes ‘ On the
Occurrence of the Bones of Man in the Fossil State’: —‘ The
problem being thus resolved, it follows that man must also
be included among the fossil species, or rather that the
sudden transition from one condition of being to another
must be disallowed, and that the same gradual alteration of
species, already so fully developed by M. Deshayes in his
comparison of the fossil shells of the different periods of the
tertiary formations, must be extended to animals, and per-
chance to man himself: in fact, that the barrier of fossil and
non-fossil must henceforth be a distinction of convenience
only, to separate such remains as may be found buried in
the regular geological strata, from those of more modern or
accidental inhumation.’ !

In 1835, while the interest of these Jumna explorations
was still fresh, Capt. (now Colonel Sir Proby) Cautley and
myself, then occupied with the investigation of the Fossil
Fauna of the Sewalik hills, discovered the remains of the
extinct gigantic Tortoise of India. This remarkable form
was briefly referred to in a memoir communicated to the
Geological Society in 1836; but the detailed account of it
did not appear until 1844. The huge chelonian was inferred
to have had a shell twelve feet lone, eight feet in diameter,
and six feet high; and the anterior or “episternal portion of
the plastron exhibited a thickness of 64 inches of solid bone ;
proportions which rendered it a fit object for comparison
with the Elephant. The following remarks bearing upon
its possible relations to the human period are extracted from
the ‘ Proceedings ’ of the Zoological Society in 1844: ?

‘ Colossochelys Atlas.—The first fossil remains of this colossal Tortoise
were discovered by us in 1885 in the tertiary strata of the Sewalik
hills, or Sub-Himalayahs skirting the southern foot of the great Hima-
layah chain. They were found associated with the remains of four
extinct species of Mastodon and Elephant, species of Rhinoceros, Hip-
popotamus, Horse, Anoplotherium,? Camel. Giraffe, Sivatherium, and a
vast number of other Mammalia, including “Syria or five species of Quad-
rumana. 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 akso-
lutely identical with species now living in the rivers of India: we
allude in particular to the Crocodilus longirostris,* from the existing

1 Journ. Asiat. Soc. Bengal, 1833, volt $ Now Chalicotherium Sivalense, being
ii. p. 634. Nestoritherium of Kaup, [See voli. p-

2 See vol. i. p. 365.—[Ep. ] 223.—Ep.|
4 Gavialis Gangeticus.

574 PRIMEVAL MAN,

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
Emys tectum,! now a common species found in all parts of India... .
This is not the place to enter upon the geological question 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
opinion that it may have descended to the human period?” Any
a priori improbability that an animal so hugely disproportionate to
existing species should have lived down to be 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 circum-
stances. We have as yet no direct evidence to the point, from remains
dug out of recent alluvial deposits, nor is there any historical testi-
mony confirming it; but there are traditions connected with the cos-
mogonic 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 a mere creature of
the imagination, or was the idea of it drawn from a reality, like the
Colossochelys?’

Reference is then made to the most remarkable cases in
which the Tortoise figures in mythological conceptions that
are traceable to an Oriental source: first, in the Pytha-
gorean cosmogany, where the infant world is represented as
having been placed on the back of an Elephant which was
sustained on a huge Tortoise ; next, to the Second Avatar of
Vishnoo, in Hindoo mythology, where the god is made to
assume the form of a Tortoise, and to have sustained the
newly-created world on his back, to make it stable; and,
lastly, to the exploits of the bird-demigod Garida, during
one of which he was directed by his father, Kishyiipa, to
appease his hunger at a certain lake where an Hlephant and
Tortoise were fighting. 'The dimensions of both are expressed
in figures of extravagant magnitude. Garida with one claw
seized the Elephant, with the other the Tortoise, and flew
to a mountain, where he regaled himself with the viands
yielded by the two animals. The speculative remarks, sug-
gested by these traditions, viewed in connection with the
discovery of the Colossochelys were expressed in the following
terms :—

‘In these three instances, taken from Pythagoras and the Hindoo
mythology, we have reference to a gigantic form of Tortoise, com-
parable in size with the Elephant. Jlence the question arises, are we
to consider the idea as a mere figment of the imagination, like the
Minotaur and the Chimera, the Griffin, the Dragon, and the Cartazo-

1 See yol. i. p. 882, xote 2.—[ Ep. ]

AND HIS COTEMPORARIES. 575

non, &¢., 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 imagi-
nation; but in the Indian cosmogonic forms we may trace an image of
congruity through the cloud of exaggeration with which they are in-
vested. 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 Gariida, with all his
attributes, we may detect the gigantic Crane of India (Ciconia gigantea)
as supplying the origin. In like manner, the Colossochelys would

Fie. 8.

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,

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 Garida. 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

576 PRIMEVAL MAN,

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
entertained bya 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 epoch of the human period, and become extinct since :—
Ist, from the fact that other Chelonian species and Crocodiles, con-
temporaries 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.’-—Procecd. Zool. Soc. Lond. 1844. Part xii.
p- 85.1

It is not meant to be urged now, after a lapse of more
than twenty years, that any serious claim can be preferred
on the speculation put forward in the passages above cited.
But it will perhaps be admitted that the mind of the observer
from whom it emanated was then occupied with the subject
of the possibility of the remote antiquity of man in India,
on paleontological evidence.” It is true that the expressed
view is that the Colossochelys may have lived down to an
early epoch of the human period; and not that man had
lived back to be a cotemporary of the Tortoise, now proved
to have been Miocene. But the two views are reciprocal ;
and the form of expression selected on the occasion was
that which was least calculated to provoke ridicule, or to
shock the strong prejudices on the subject which were then
dominant among educated men. And so firmly was—not
merely the possibility, but—the probability of the case im-
pressed upon our minds, that Capt. Cautley and myself were
constantly on the look-out for the turning up, in some shape
or other, of evidences of man out of the strata of the Sewalik
hills, partly from considerations of a different order, to
which I shall briefly allude.

The cataclysmic speculations of Cuvier and the Diluvial
theory of Buckland were then exploded. The wide spread
of the plains of India showed no signs of the unstratified
superficial Gravels, Sands, and Clays, which for a long time
were confidently adduced as evidence that a great Diluvial
wave had suddenly passed over Hurope and other continents,
overwhelming terrestrial life, and leaving the marks of its
course and violent action in these enormous deposits of
transported debris. Hvery section along the Gangetic plain
indicated that the superficial strata there were of local
origin, and the result of tranquil sedimentary deposition.
Viewed in the light of a strictly physical inquiry, the chief

1 Reproduced in vol. i. p. 868.—[Ep.] | appended to the memoir on Hmys
2 As additional evidence on this | tecta, written in 1844. See vol. 1. p.
point, the reader is referred to the note | 888.—[Ep.]

AND HIS COTEMPORARIES. 577

rational argument in support of the opinion that the advent
of man upon the earth dates from a very modern epoch, was
first, the negative evidence in the non-occurrence of human
relics, and next the fact, that taking him in conjunction
with the Mammals, with whom he is now associated, they
appeared, as a group, to belong to a new order of things
strikingly different from that of the immediately preceding
period. The Mammoths, the wool-clad Rhinoceros, the
Cave-Lions, and Speleean Hyzenas, the Irish Elk, &c. of the
Kuropean fauna were all extinct, although the carcasses of
some of them had been discovered, under favourable circum-
stances, in the most perfect state of preservation. Facts of
corresponding import were yielded by a glance cast upon
the latest paleontology of the American Continent. There
also the huge extinct HEdentata, the Mammoth, and the
Mastodon indicated a different order of life, especially from
that now existing. But in India the problem presented itself
under another aspect. There, no break was visible in the
tranquil succession of deposits, no interference of a general
oceanic submergence, followed by incoherent beds of sand
and gravel, no intercalation of glacial phenomena to disturb
the previous system. The present physical order of things—
modified only by alterations of level, by upheavements and
depressions—could be traced back, in an unbroken chain, to
the ossiferous strata of the Valley of the Nerbudda and of
the Sewalik hills. Results in harmony with these indica-
tions were yielded by a retrospect cast upon the system of
organized life. The Mastodons, the Stegodons, and the Lowodon
Hlephants were extinct, as were also the Sivatheriwm, the
Chalicotheriwm, the three-toed Hipparion-Horse, the Hexapro-
todon, the Merycopotamus, and other peculiar forms. But
they were found associated in the same Sewalik deposits with
species of true Hquus, of Camel and of Giraffe, the two last
being characteristic cotemporaries of man at the present
time. The Pliocene fauna of the Nerbudda Valley produced,
along with the Miocene Stegodon insignis of the Sewalik
hills, an extinct Elephant, H. Namadicus, the dental system
of which is closely allied to that of the existing Indian
species; a true Hippopotamus, and not to mention others, a
true Taurine Ox, Bos Namadicus, and a huge Buffalo, B.
(Bubalus) Paleindicus, which is nearly approached by the
living ‘ Arnee’ of the forests of Assam, being the stock from
which the domestic Buffalo of Oriental countries is sup-
posed to have sprung. That the actual order of the present
system of life had begun during the Sewalik period, was
indicated by the living Gharial! Crocodile and Emys tecta,

1 See vol. i. p. 350, xo¢e 1.—[ Ep. ]
VOL. Il. PP

578 PRIMEVAL MAN,

referred to in the above extract as being found associated
with the extinct mammalian forms. And of the latter, some,
like Stegodon imsignis, accompanied by a species of Hewxa-
protodon, descended to the Pliocene period of the Nerbudda
fauna, to be associated with a true Taurine Ox and with a
Buffalo which hardly appears to differ more from the living
Arnee than does the ancient Bison priscus from the living
Aurochs. Another fact chimed in with especial force.
Among the four or five species of Sewalik Quadrumana
alluded to above, one was inferred by Sir Proby Cautley and
myself, in 1837, to have been a large Ape, exceeding the size
of the Orang-Outang, but of unknown immediate affinity.
This opinion was founded upon a canine tooth of an old
animal, which is figured and described in the Journal of
the Asiatic Society of Bengal.! Five years afterwards, in
1842, I instituted a close comparison between the fossil
specimen and the corresponding tooth of three skulls of the
Orang-Outang, contained in the Museum of the Asiatic
Society in Calcutta, and found that their agreement was so
close, that I conjectured that the extinct Sewalik form had
been a large Ape allied to Pithecus satyrus.?

A quadrumanous astragalus, derived from the same strata,
approached, in form and proportions, so near to that of the
existing Hoonuman Monkey, Semnopithecus entellus, that the
help of the callipers had to be put in requisition to enable
us, in 1836, to discriminate them, by differences not exceed-
ing millimétres. The distinction between the fossil and the
recent bone is hardly greater than that which might be ex-
pected to occur in any two individuals of the living species.*
Here, then, was clear evidence, physical and organic, that
the present order of things had set in from a very remote
period in India. Every condition was suited to the require-
ments of man. The lower animals which approach him
nearest in physical structure were already numerous. The
wild stocks from which he trains races to bear his yoke in
domesticity were established. Why then, in the light of a
natural inquiry, might not the human race have made its
appearance at that time in the same region? Cuvier, not-
withstanding his strong bias in favour of the modern appear-
ance of the human race, admitted, in language which has
often been overlooked in later discussions, that man may
have lived before the last great revolutions which were the
subject of his disquisition :—‘ Tout porte done a ecroire que
Vespéce humaine n’existait point dans les pays ot se décou-
vrent les os fossiles, a l’époque des révolutions qui ont enfoui

1 Vol. vi. p. 359, Pl. xviii. fig. c. 3 See vol. 1. p. 294.—[ Ep. ]
* See vol. i. p. 304.—[ Ep. |

AND HIS COTEMPORARIES. 579

ces os; car il n’y aurait eu aucune raison pour qu’elle échap-
pat toute entiére a des catastrophes aussi générales, et pour
que ses restes ne se retrouvassent pas aujourd’hui comme
ceux des autres animaux; mais je n’en veux pas conclure
que Phomme n’existait du tout avant cette époque. I] pouvait
habiter quelques contrées peu étendues, d’ou il a repeuplé
la terre aprés ces événements terribles,’ &c. The Valley of the
Ganges seemed to present the exceptional conditions here
demanded ; it was exempt from the protracted submergence
under the ocean, the effects of which on Europe suggested
the idea of cataclysmic revolutions. I dwell upon the sub-
ject now in the hope that, when the paleontological explora-
tion of the Sewalik hills and Nerbudda Valley, or of other
equivalent formations, is resumed, these remarks may attract
attention in India, and that a keen look-out may be kept up
for remains of the large fossil Ape above alluded to, and for
traces of man, in some form of equally remote antiquity.
For it is not under the hard conditions of the Glacial period
in Europe, that the earliest relics of the human race upon
the globe are to be sought. Like the Esquimaux, the
Tchuktshes, and the Samoyedes on the shores of the Icy Sea
at the present day, man must have been then and there an
emigrant, placed under circumstances of vigorous and uncer-
tain existence, unfavourable to the struggle of life and to
the maintenance and spread of the species. It is rather in
the great alluvial valleys of tropical or sub-tropical rivers
like the Ganges, the Irrawaddi, and the Nile, where we may
expect to detect the vestiges of his earliest abode. It is there
where the necessaries of life are produced by nature in the
greatest variety and profusion, and obtained with the smallest
effort ; there, where climate exacts the least protection against
the vicissitudes of the weather; and there, where the lower
animals which approach nearest to man now exist, and where
their fossil remains turn up in the greatest variety and
abundance. ‘The earliest date to which man has as yet been
traced back in Hurope is probably but as yesterday, in com-
parison with the epoch at which he made his appearance in
more favoured regions.

TE.

The next case to which I shall refer occurs in the pre-
fatory remarks of a published catalogue’ of Indian fossil

1 Descriptive Catalogue of the Fossil | Asiatic Society of Bengal, by H. Fal-
Remains of Vertebrata from the Se- | coner, M.D., F.R.S., assisted by H.
walik hills, &c., in the Museum of the | Walker, Esq. 8yo. Calcutta, 1859, p. 7.

PP 2

580 PRIMEVAL MAN,
remains, bearing the date of February 28, 1855, from which
the following observations are extracted :—

*The Jumna Alluvium collection consists of specimens in part pre-
sented by Captain E. Smith and Lieutenant Burt of the Engineers, but
the great number by Sergeant E. Dean.

‘Captain Smith’s specimens were readily discovered, and among them,
besides those enumerated and figured in the Journal, I found the perfect
astragalus of a Hippopotamus distinct from the Sewalik Hill species,
and apparently identical with the Nerbudda Tetraprotodon. But
Sergeant Dean’s series proved to be in such confusion, that under the
pressure of an approaching departure for Europe, I felt that I could not
afford them the time they required, and left them unarranged.... I
regret this the more, as I consider the Jumna fossils to be the most
promising of results bearing upon the human period, and I strongly
recommend them to any one who is able and willing to undertake the
task of investigating them.’ !

LY.

I shall now pass to the more immediate researches on the
same subject with which I have been connected in Europe.

Strange and improbable as the assertion may appear, it is
nevertheless true, that the most superficial deposits of the
earth’s crust, forming the very ground under our feet, have
been in many respects the least satisfactorily investigated
of geological formations. The Paleozoic rocks at the most
distant extremity of the chain of life upon the globe have,
by an admirable series of researches by great observers, been

1 The fossil remains here mentioned
are those to which allusion has been
made (p. 572) as having been yielded
by the Ancient Alluvium of the Jumna.
Much precious material was irretrievably
lost from heedlessness during the earlier
part of the seven years’ operations:
but enough was preserved to be of
high interest. The supposed Human
Bones proved to be either inconclusive

or mistaken determinations. But among |

the other fossils I identified a molar
tooth of the extinct Hlephas Namadicus,
a lower jaw with teeth and a perfect
astragalus of the true Indian Hip-
popotamus, H. (Tetraprotodon) Pale-
indicus, both forms belonging to the
Phocene fauna of the Nerbudda Valley.
The ancient Indian Hippopotamus
differs in a very essential manner from
the existing African species, the large
and small pairs of incisors of the jaws
occupying reversed positions respec-
tively in the two species. A quadruped
so remarkable for its size, form, and
habits, must everywhere have impressed

itself forcibly on the attention of man ;

| and the question arose: could, as in the

instance of the ‘Gigantic Tortoise, any
reflection of its former existence be de-
tected in the ancient traditions or lan-
guages of India? Following up the
inquiry, I was informed by Raja Radha-
kanta Deva, the eminent Indian scholar
and author of the Sanscrit Eneyclopeedia,
that the Hippopotamus of India is re-
ferred to under different names of great
antiquity, significant of Jala-Hasti,
‘Water-Elephant, or ‘Living in the
water.’ This inference is confirmed by

| the opinion of Henry Colebrooke and

H. H. Wilson. No definite result was
arrived at; but referring to the line in
italics contained in the above extract,
there is published evidence that in 1855
I was engaged, in connection with pal-
ontological researches, upon the same
question to which my attention was
directed between 1836 and 1844, namely
the remote existence of the human race
in India.

AND HIS COTEMPORARIES. 581

dressed into exquisite order, each successive formation un-
folding the record of the organized beings which were then
in existence, and to a great extent also the physical condi-
tions under which they lived. The series of rocks called
Secondary and Tertiary, which are the cemeteries of the
Mesozoic and Kainozoic creatures that formed the mid-stages
of world-life, have been subjected to a similar course of
analysis, and have yielded corresponding results. But the
nearer we approach to the existing system, of which man is
the head and chief, the more hazy has the prospect been—
the more faint and obscure the definition of the outline.
The reason is sufficiently obvious. All the great disclosed
problems that were most calculated to fascinate the imagina-
tion and exercise the intellectual powers lay in the remote
past. It was there where the medals were buried, by the
interpretation of which the prodigious antiquity of our planet,
as a field of life, has been established; and there, where the
vast forces which have rent its coast asunder were, by their
effects, most distinctly evinced. There was an irresistible
charm in tracing back the firstlings of life to their earliest
dawn in the Cambrian and Silurian rocks, and in following
the progress upwards from simple to more advanced forms.
The Secondary rocks revealed evidence that the surface of
our island was formerly clothed with tropical forests, and
peopled with Saurian quadrupeds, like the Iguanodon, the
Megalosaurus, and the Pterodactyle, which, by their unex-
pected forms and colossal proportions, startled the imagina-
tion and taxed the synthetic powers of science to restore
them. Ascending to the Tertiary series, the genius of Cu-
vier worthily followed up by his successors had reproduced
the exuberant variety of Mammalian quadrupeds with which
the Hocene and Miocene formations teemed. But in the
Pliocene strata the exciting interest of novelty waned, till
in the newer Quaternary deposits it became nearly evanes-
cent. The gravels, sands, and clays, which compose them,
were invested with an eminently prosaic character, little
promising of food to the imagination, or of profitable exer-
cise to the reasoning faculty ; and the treatment they met
with was in keeping with their uninviting aspect. They
were all summarily disposed of by the convenient and lazy
hypothesis of a transitory Diluvian irruption, which had
swept over the earth’s surface and left them behind. The
keen intellect of Agassiz rent the veil of this delusion, and
fixed the attention of observers upon the phenomena of
glacial action, which for a time brooked no rival, but re-
flected a flood of light upon what was before either mistaken
or obscure. Like most other fashions, it was in its day su-

582 PRIMEVAL MAN,

preme, and every investigation of the superficial deposits was
tinged, more or less, of a glacial complexion. The gravels,
sands, and loams, not directly the products of glacial action,
were in every respect unattractive. True that they were
occasionally loaded with organic remains, the most important
of these being fossil bones ; but the latter were regarded in
most instances as being but repetitions of what was already
well known; the deposits themselves were of local ‘origin,
and in many cases their precise geological age long baffled
all attempts at determination. One signal example to which
frequent reference will be made in the sequel may be ad-
duced. In the Valley of the Thames, close to the head-
quarters of geological observations in England, there is
an abundant development of loam and gravel deposits of
freshwater origin, which are extensively excavated for brick-
earth and other commercial objects. They contain beds
which are charged with fluviatile shells, some being of
species now extinct in England; and they are also rich in
fossil bones of large Mammalia, living and extinct. Fine
sections of these beds have been disclosed at Brentford,
Ilford, Grays Thurrock, Erith, and other localities. The
Brentford section was described fifty years ago in the ‘ Philo-
sophical Transactions,’ and some of the most remarkable
organic remains found in it were at the same time figured.!
The Grays Thurrock beds were closely studied by an able
and experienced observer, Professor Morris, who published
an excellent account of their contents in 1838. The whole
series of these brick-earths, both in the Valley of the Thames
and in other river valleys of the same system, has been the
subject since of constant observation by some one of numerous
geologists of authority, such as Godwin-Austen, Prestwich,
Morris, Joshua Trimmer, Lyell, Edward Forbes, &c., yet
has their true geological age, whether preglacial, 7.e. an-
terior to the submergence of the land under the glacial sea,
or post-glacial, z.e. subsequent te its final emergence, re-
mained up to the present time a subject of the utmost per-
plexity and of great difference of opinion among geologists,
the reason of this being that nowhere in the Valley of the
Thames was a section to be found showing the glacial drift
and fluviatile deposits in contact or in order of superposition.
The general physical phenomena, in the absence of direct
proof, led Mr. Prestwich, Professor Morris, and the late Mr.
Joshua Trimmer, to the belief that they were of a later date
than the Boulder-clay, 7.e. post-glacial, while the palzonto-
logical evidence appeared to be at issue with that inference,
and to indicate that they were preglacial. The polyonomous
1 William Trimmer, Phil. Trans, 1813.

AND HIS COTEMPORARIES. 583

bivalve shell Oyrena fluminalis, long regarded as ‘ pre-emi-
nently preglacial,’ from its presence in the Crag, occurs in
‘abundance at Hrith, Ilford, Grays Thurrock, and Stutton ; and
from Grays Thurrock, I had determined undoubted remains of
Hlephas antiquus, Fale., and Rhinoceros leptorhinus, Cuvier (R.
megarhinus, Christol), found along with Hippopotamus major,
free from association with Hlephas primigenius, or Rhinoceros
tichorhinus, the two characteristic forms of the glacial de-
posits of Northern Europe and Siberia. The three fossil
mammals of Grays above-named were identified by me as
occurring in the Sub-Apennine Pliocene deposits of the
Astesan, and in the preglacial lacustrine deposits or ‘ forest-
beds’ of the Norfolk coast. The weight of the paleontological
evidence, molluscous and mammalian, appeared to turn the
scale in favour of the preglacial view. It was only when
Mr. Prestwich had discovered in sections near Holderness
and Hull beds of clay, sand, and gravel, analogous to the
Thames Valley deposits, containing Cyrena fluminalis, and
clearly superimposed on the Boulder-clay, that the problem
was by him satisfactorily solved, stratigraphically, in 1860.
Mr. Prestwich before that time had his early convictions
confirmed by sections in the Valley of the Stour, at Sudbury,
and by a still more important case, in the Valley of the
Ouse, near Bedford, where a bank of gravel yielded remains
of Elephant, Rhinoceros, Hippopotamus, Ox, Horse, and
Deer. Commencing with the labours of Mr. W. Trimmer,
fifty years had lapsed before the point was finally settled ;
such a vast amount of competent observation does the solu-
tion of a complex geological question of this nature involve.
The same kind of mystery in reference to their precise geo-
logical age enveloped the ossiferous caves.

Tt had been supposed, prematurely, by some of those who
had a share in the great achievements of the early days of
the science, that the general field of geology was swept
clean of all the grand questions which could command in-
terest, and that it would be left to those who came after
merely to fill in the details of the outlines which had been
carved by the great masters who had preceded them. Vain,
the complacent idea; for even in those neglected superficial
sands and gravels lay buried the evidence, upon which
problems that now fix the attention of mankind were to
be worked out. Boucher de Perthes, in France, was per-
severingly ransacking the gravel heaps and gravel sections
of the Valley of the Somme near Abbeville, for the wrought
flint-weapons which are now so intimately connected with
his name. They had escaped the discrimination of M.
Baillon, one of the most active correspondents of Cuvier, and

584 PRIMEVAL MAN,

a contributor to the ‘Ossemens Fossiles’; but the keen
elance of Boucher speedily detected their import, and his
inferences received important confirmation from the subse-
quent discovery by Rigollot of similar flint-implements in
eravel sections near Amiens. But the observations of both
were either scorned or discredited. At the same time a
quiet observer, of matchless sagacity and indomitable per-
severance, Mr. Prestwich, was making the gravels in England
an object of special investigation. Engaged during a long
course of years upon the study of the European Tertiaries,
he gradually worked his way up to the superficial deposits.
Mr. Prestwich’s researches upon the Tertiaries, which have
only been partially published, have earned for him the repu-
tation of being one of the ablest geological observers of his
time. But in the Quaternary sands and gravels he was un-
rivalled. Men have been in the habit of saying, in mingled
earnest and raillery, that ‘point out a broken pebble amongst
a thousand others in a gravel-pit, and there is one who will
tell you the point of the compass from which it came, the
stratum which yielded it, the distance it had travelled, the
amount of rolling it had undergone, and the time it had
occupied in the journey.’ The power thus acquired was
soon to be applied with clenching authority to the proofs of
the antiquity of man yielded by those deposits. But to
be productive, complex researches of this nature must be
subjected to the economic law of division of labour, which
controls alike the skilled results of artisanship in the manu-
facture of pins and buttons, and the intellectual efforts
which contribute to the advancement of knowledge; special
application in either case determines the excellence of the
products. The mollusca occurring in the clays and gravels
had undergone the severest scrutiny; but although com-
monly of such importance in geological chronology, their
value as indicative agents in this case was but trivial, all
the forms being of existing species. The fossil bones had
also been collected and named. The authority of Cuvier in
this latter walk of research was still paramount, and in most
cases governed the specific determinations of the remains of
the large extinct Mammalia which abound in the brick-
earths and gravels. In 1844 I made excursions to various
localities in the Valley of the Thames, in company with my
friend the late Edward Forbes, and arrived at the conclusion
that they contained the remains of extinct species that were
then unrecognized. These results applied more especially
to the large fossil Ungulata, and to the Elephants in par-
ticular. In Parts I., IL., and III. of the ‘Fauna Antiqua
Sivalensis,’ published in 1846 and 1847, figured evidence

AND HIS COTEMPORARIES. 585

was put forward that the Pliocene and Post-Pliocene forma-
tions in England yielded three distinct species of Elephant
besides the Mammoth, namely Hlephas meridionalis of Nesti,
H. antiquus, and H. priscus. It was inferred also that cor-
responding results would follow a close study of the other
Ungulate genera, such as Rhinoceros, but a compulsory de-
parture to India on service, in 1847, prevented me from pur-
suing the investigation at the time. On my return to
Europe, in 1855, fresh from the study of the fossil remains
of the ‘Ancient Alluvium’ of the Jumna, and of their possible
bearings on the antiquity of man (antea, p. 580), I deter-
mined to take up the examination of the extinct Mammalia
of the Pliocene and Quaternary periods in Europe as a
monographic labour; in short, to apply to the deposits,
palzontologically, the same kind of analysis which Mr.
Prestwich was applying to them stratigraphically, but with-
out concert with or even personal knowledge of him at the
time. The object held in view was, first to endeavour to deter-
mine with precision the specific characters of the Mammalian
forms which occur in the Pliocene strata of the Sub-A pennines,
and in equivalent formations elsewhere; next, to ascertain
their range of existence in time, and then to deal in the same
manner with the forms which make their appearance in the
Quaternary deposits. In this way, it was expected that it
would be seen when the older forms ceased, and when the
newer forms came in; or whether the latter passed gradually
into the former. The Mollusca appeared to fail here in an
important respect as guides, the percentage of extinct spe-
cies in the newer Quaternaries being either nil or excessively
small, while the Mammalia promised to furnish a test of
greater significance: Ist, because the genera and species
are everywhere shown to be of more limited duration in time
than the Mollusca; 2nd, because from the vastly greater
complexity of their relative functions, they are much more
susceptible of being affected by the altered climatic con-
ditions which are necessarily involved in every great phy-
sical change, and which conduce most to the extinction of
species.

Another object was held in view, upon which it was anti-
cipated that the investigation might throw important light—
namely, the permanence or mutability of species. Between
the Pliocenes of the Sub-Apennines and the newer Quaternary
formations an enormous lapse of time had intervened—
amounting to many hundreds of thousands, if not millions, of
years—during which a great portion of the Continents of
Europe, Asia, and America had been chilled down to an ex-
tent to which the past history of our planet furnished no

586 PRIMEVAL MAN,

Known parallel. How fared it with the large Mammalia
during this mighty change? Were the older forms killed
off, and newer species intercalated, or did any of the former,
and if so, what number, survive through it, and how were
they affected? In this case the argument of the imperfec-
tion of the geological record, which has been so powerfully
handled by Darwin, could not be urged; the materials were
abundant, and the deposits which marked the successive
changes of dry land, submergence, and re-emergence, were
amply represented. I was well acquainted, through the un--
reserved communication of intimate friendly imtercourse,
with the gradual development of his views on the origin of
species; and it seemed that here was a case where quad-
rupeds which were either cotemporaries of man or close
upon his period, could be traced back into remote time, and
thus furnish a test of the mutability or persistence of specific
characters, of much higher value than that yielded by ob-
servation upon living animals, necessarily limited to a brief
lapse of time—in fact, to the personal experience of the
observer.

The results of these inquiries were, in part, communicated
to the Geological Society, in 1857, in two memoirs, the first
of which appeared in extenso, the second in abstract, in its
Quarterly Journal.!. In the former it was attempted to be
shown that the principal species of the large fossil Ungulata,
such as Mastodon Arvernensis, Elephas meridionalis, Elephas
antiquus, H. priscus, Rhinoceros leptorhinus and Hippopo-
tamus, might occur alike in the Sub-Apennine Pliocenes of
Ttaly and in the Norwich Crag, indicating the formations to
be of the same age. In the second, it was stated that the
same extinct species, with the exception of Mastodon Arver-
nensis, are found in the ‘ Hlephant-bed,’ and Lacustrine
blue clays of the Norfolk coast, which underlie the ‘ Boulder-
clay,’ proving them also to belong to the Pliocene period.
Finding that the same Hlephas antiquus and the variety called
E. (Loxodon) priscus, along with Rhinoceros leptorhinus and
Hippopotamus major, although unaccompanied by Elephas
meridionalis, were abundant at Grays Thurrock and Brent-
ford, I was led by the Mammalian evidence to consider that
the fluviatile deposits of the Valley of the Thames were
also of an earlier age than any part of the Boulder-clay. This
inference my subsequent researches in the caves proved to
be erroneous; and in a recent paper I have shown that the
true Mammoth, EF. primigenius, had its range of existence
stretching as far back in time as the preglacial forest-bed of
Norfolk, and the deposits of the ancient lake-formation of

1 See antea, pp. 1 and 76.—[Eb.]

AND HIS COTEMPORARIES. 587

the Seven Hills upon which Rome is built, ‘bearing his
organs of locomotion and digestion all but unchanged,’
through an enormous lapse of ages.!

V.

Another important branch of the same investigation re-
mained to be followed up—namely, an analysis of the fossil
fauna of the ossiferous caverns of Great Britain, conducted
upon the same plan; and as the reagitation of the question
of the remote antiquity of the human race in Hurope, and
the establishment of the evidence, to the conviction of
geologists, arose out of it as immediate results, I must be
pardoned for entering in some detail upon the history of the
question.

Bone caves occur generally over Europe, but they are
especially abundant in England, Germany, France, ltaly, and
Sicily. The fossil remains found in them had been carefully
studied by Cuvier and by various other paleontologists, in
different countries, since his time. The theory of the process
of their fillmg up, with their organic contents, had been
keenly discussed by Dr. Buckland and by Constant Prévost,
who entertained diametrically opposed views upon the ques-
tion. A great number of other cryptological observers, after
them, took a share in the discussion; and an excellent
epitome of the whole case was brought out by Desnoyers, in
a monograph which appeared in 1849. But at the period
to which I refer, 1856, the views of geologists, respecting
the epoch when the fossil quadrupeds whose remains occur
in them lived and were there introduced, were in a very
unsettled state. Desnoyers left the question as he found it,
undetermined. For the prevalent opinions held in Eng-
land, I shall refer to standard systematic works, by two
eminent living authorities, Professor John Phillips and Sir
Charles Lyell. The former, to whom we are indebted for
the systematic use of the terms PrEGLAcIAL and Post-
GLACIAL, which define with such precision the Quaternary
deposits before and after the submergence of the Glacial period,
in his ‘ Manual of Geology,’ published in 1855, expresses his
opinion, in the clearest language: ‘To the Preglacial era
belong, we think, the greater number of ossiferous caves
and fissures, containing Elephant, Hippopotamus, Hyzena,
and other extinct species of animals.’ (Op. cit. p. 411.) Sir
Charles Lyell, in the fifth edition of his ‘ Manual,’ published
in 1855, takes a similar view, and classes the caves in
Chap. xiii. among the newer Pliocene formations, below or

1 See antea, p. 239.—[Ep.]

588 PRIMEVAL MAN,

anterior to the Boulder-clay of the glacial submergence. In
the supplement to the fifth edition, published in 1859, the
following is the only passage having reference to the age of
the caverns: ‘To what part of the Pliocene period the cave
animals of Great Britain should be referred is still a vexed
question. There seems, however, no reason at present to
suppose any of them more ancient than the Norwich Crag ;
and many caves may have remained open during the Glacial
and Post-Glacial eras, while the fauna was gradually chang-
ing, so that the remains found in them may not always belong
to strictly contemporary quadrupeds.’ (Op. cit. p. 8.) No
language could have been used better calculated to express
the conjectural and hesitating views then entertained re-
specting the caves : they might be of any age before the Glacial
period or after it. The late Joshua Trimmer, an experienced
observer of weight on all that concerns Quaternary deposits,
expressed the following opinion in his ‘ Generalizations re-
specting the Norfolk Erratics’: ‘With regard to Mam-
malian remains, I believe that we have two Elephantine
groups, one preceding the submergence of the erratic period,
and the other inhabiting the country at the close of the period
of elevation. To the former are to be referred the Mam-
malian Crag, and the remains of the bone-caverns in
general,’ !

The determination of the Cave fauna was, in some respects,
in a similarly backward state. Not a suspicion even was
entertained, so far as published evidence shows, that any
other Elephant than the Mammoth, or Rhinoceros other than
the tichorhine species, had left their remains in the caverns,
although Hlephas antiquus, Fale., and Rhinoceros hemitechus,
Falc., were subsequently discovered in them in abundance,
these two species, along with Hippopotamus major, consti-
tuting at the present time the criteria by which British
geologists test the more or less remote antiquity of the
works of man occurring in Drift deposits.?

In May 1860 I communicated to the Geological Society a
memoir ‘On the Gower Caves,’ which was read at the meet-
ings held on the 31st May and 13th June. The caverns of
the Gower peninsula, Paviland being one of them, are very
numerous, and correspondingly rich in fossil remains; while
in other respects they present favourable conditions for de-
termining the question, then unsettled, such as are nowhere
else to be found in Great Britain. Their floor in many
cases is covered with marine deposits, containing shells,

* Quarterly Journal Geol. Society, ® Vide Sir C. Lyell’s ‘ Antiquity of

1850, vol. vil. p. 25. Man,’ passim.
3 See antea, p. 498.—[Eb.]

AND HIS COTEMPORARIES. 589

above which the ossiferous strata are piled in beds alter-
nating with numerous thick layers of stalagmite ; they are
in close juxtaposition with a boldly-developed raised sea-
beach, first observed and investigated by Mr. Prestwich;
and on the high land of Gower, which backs the caves, the
same able observer detected remains of Drift deposit, indica-
tive of the submergence of the Glacial period. Availing
myself of the rich and unrivalled collections formed by my
friend Lieut.-Colonel Wood, of Stout Hall, during many years
of patient inquiry, and following up their indications, in
further researches conjointly with him, I endeavoured to
trace the successive appearance of the newer and newer
forms, through the magnificent sections furnished by the
Gower caves, regarded as a series, until we were conducted
to the period when man presented himself as a tenant of the
caves. Some of the Gower caverns, such as ‘ Minchin Hole,’
were of an amplitude comparable to that of a cathedral, with
an open-vaulted mouth. Below they were strewed with sea-
sand and shells, above which ochreous cave-earth, crammed
with bones, and layers of stalagmite were disposed in nume-
rous successive layers. These alternations implied repeated
changes of the physical conditions which held in their
interiors, dependent probably on the shocks involved in
movements of upheaval or depression, by which long-esta-
blished lines of subterraneous drainage were for a time inter-
cepted and then reproduced ; the periods of action and repose
being respectively indicated by the layers of ochreous loam
and stalagmite. Had the coast line in which the caves are
chiefly situated ever been submerged, after the osseous
remains were introduced, under the sea of the Glacial period,
it was impossible that marine deposits should not have been
introduced somewhere between these alternations, significant
of the event. The cave called ‘Bosco’s Den,’ placed at a
higher level, presented an enormous depth of ochreous cave-
loam, impacted with horns and bones of the Reindeer, and
overlaid at the top by a thick cake of stalaginite, upon the
surface of which a bed of peat was accumulated. In no one
of these instances was there the slightest indication of Drift
or Boulder-clay above the ossiferous strata, ancient or
modern. After balancing the various classes of evidence,
and instituting a comparison of the Gower caves with those
of the other cave districts of England in particular, and of
Europe in general, the following conclusions were announced
in the memoir above referred to :—

‘1. That the Gower caves have probably been filled up with their
mammalian remains since the deposition of the ‘“ Boulder-clay.”

‘2. That there are no mammalian remains found elsewhere in the

590 PRIMEVAL MAN,

ossiferous caves in England and Wales referable to a fauna of a more
ancient geological date.

‘3. That Hlephas (Loxodon) meridionalis and Rhinoceros Etruscus,
which occur in, and are characteristic of, the “ Submarine Forest-hed”
that immediately underlies the Boulder-clay on the Norfolk coast, have
nowhere been met with in the British caverns.

‘4. That Elephas antiquus with Rhinoceros hemitechus and FE.
primigenius with Rh. tichorhinus, though respectively characterizing the
earlier and later portions of one period, were probably contemporary
animals; and that they certainly were companions of the Cave-Bears,
Gave-Lions, Cave-Hyznas, &c., and of some at least of the existing
Mammalia.’

The inferences arrived at were at least definite, in indi-
cating the fauna of the ossiferous caves of England to be
Post-Glacial. They were at the time unexpected, and, when
communicated, they encountered lively opposition in dis-
cussion as being opposed to the views then commonly enter-
tained regarding the age of the fauna of the caves, and of
the Quaternary deposits of the river valleys which yield
remains of the same Mammalia in corresponding association.
But their soundness has not been impugned since, and the
subsequent discovery by Mr. Prestwich, at Sudbury and Bed-
ford, of an analogous deposit containing bones of Hippopo-
tamus and other mammalian species characteristic of the
Thames Valley beds, superimposed upon the glacial drift,
proved that the brick-earths and gravels of the Thames
Valley were Post-Glacial ; while my own researches upon their
mammalian remains showed that their fauna was identical
with that of the caves, which I had inferred from another
class of evidence to be Post-Glaciai. Mr. Prestwich had long
and consistently held the opinion that the Grays Thurrock
and Ilford beds were later than the Boulder-clay;! but it was
not until after many years of ineffectual search that the
Sudbury and Bedford sections enabled him to prove the case
stratigraphically. The conclusions above cited, regarding
the Gower caves—exclusive of the second, which is irre-
levant—are at the present time applicable categorically to
the Post-Glacial deposits of the Valley of the Thames, in so
far as their age and mammalian fauna are concerned.

VI.

These cave researches extended over several years, and
involved an examination of every cave district in England,
besides numerous caves on the Continent. They demanded
also the careful study of every accessible collection of

1 Mr. Joshua Trimmer held the same | opinion that the same fauna in the caves
view, but coupled with it the erroneous | was preglacial. (Supra, p. 583.)

AND HIS COTEMPORARIES. 591

fossil bones in England, public or private, bearing on the
question. In the spring of 1858, while thus occupied, I
made an excursion to the cave districts of the Mendip hills,
Devonshire, and South Wales, in company with my friend
the Rev. Robert Everest, with whom I have been associated
in similar objects, off and on, during thirty years. At Tor-
quay, when examining ‘ Kent’s Hole,’ we heard of the recent
discovery of a new cave near Brixham, where we proceeded
on the 17th April, after receiving from Mr. Pengelly every
information that he could give us for our guidance. The
cave was then intact. We examined it, and made overtures
to the owner about an arrangement to explore it. On my
return to London, being then on its Council, I addressed a
letter regarding the Brixham Cave to the Secretary of the
Geological Society, dated the 10th May, 1858.! The following
is a condensed abstract of its contents :—

The decline of interest among geologists in the cave ques-
tion since the time of the ‘ Reliquize Diluviane’ of Buckland
—the backward and unsettled state of opinion respecting the
age of the ossiferous caverns—the erroneous views commonly
entertained on the subject in the cave districts, and the evil
effects resulting therefrom—the imperfect manner in which
cave explorations had hitherto been conducted in most in-
stances in Hngland—the huddling together in collections of
the fossil specimens without reference to the order in which
they occurred—the dispersion of some of the most important
and classical of English collections—the necessity of retriev-
ing past mischief and past errors, by the carefully organized
exploration of a virgin cave—a description of the newly-dis-
covered cave at Brixham as it then stood, and the favour-
able opportunity which it offered. The necessity was urged
‘of a combined effort among geologists to organize operations
for having it satisfactorily explored, before mischief is done
by untutored zeal and desultory work.’

In order to give weight to the appeal, I communicated
some of the new and unpublished results, to which my in-
quiry so far had led :—

1. The detection in certain of the caves of the remains of
a species of Rhinoceros, equally distinct from the tichorhine
species of Siberia, and of the Glacial period generally ; and
distinct also from the leptorhine Rhinoceros of Cuvier, of
the Sub-Apennine ‘ Hlephant-bed’ and lacustrine deposits of
the Norfolk coast. This is the species now known as Rhi-
noceros hemitechus, Fale.

2. Abundant evidence in all the cave districts of two
extinct species of Elephant ; viz. H. primigenius of the Glacial

1 See antea, p. 487.— [Ep.]

592 PRIMEVAL MAN,

period, and LH. antiquus of the Sub-Apennine period (the
Astesan and the ‘Norwich Crag’), the former commonly
associated in the English caves with Rhinoceros tichorhinus,
the latter with Rhinoceros hemitechus.

3. The absence of Hlephas meridionalis and the form called
A. priscus (since abandoned!) in all the cave collections;
and the association in one of them of Hlephas antiquus,
Rhinoceros hemitechus, and Hippopotamus major, without the
admixture of other species of the same genera.

4, The absence from all the caves of the Rhinoceros lepto-
vhinus of Cuvier, as I regarded the species to be limited.

Other results were communicated which need not be cited
here. The letter concluded by calling attention to an im-
portant speculation recently advanced by M. Lartet, that the
Mammalian fauna of the Glacial period in Europe was made
up of two distinct geographical elements: the one a northern
division, pushed southwards from Siberia; the other a sou-
thern division, projected northwards from Mauritania. The
proposed exploration might throw light on this question also.

The letter dated the 10th was on May 12th submitted
to the Council of the Geological Society, by whom the case
was warmly entertained; a recommendatory resolution was
passed, and entrusted to the able advocacy of the veteran
Mr. Horner, who brought it before the Council of the Royal
Society on May 13; and that learned body the same day
gave a grant of 1001. towards the object. Miss Burdett
Coutts, whose expansive munificence is alive to every object
affecting the improvement, material, moral, or intellectual, of
her country, contributed a large donation. So earnest was
the co-operation of men of science on this occasion, that in
little more than twenty-four hours after the case was
brought forward the means were provided for carrying the
design into execution. A committee at my suggestion was
appointed in London; Mr. Prestwich took charge of the
financial and business details; I was intrusted with laying
down the plan and giving the instructions upon which the
exploration was to be conducted; a local committee was
formed at Torquay, but for unity of action the responsible
direction of the explorations speedily centred in Mr. Pen-
gelly, of whom I am bound to say that never, I believe, was
an inquiry of the kind carried out with more conscientious
care or greater ability than by him. The deposits were
taken up horizontally, like sheets of paper; every object met
with had its position fixed in reference to definite points,
vertically and horizontally ; the collections of each day were

' See antea, p. 251, note 1.—[Ep.]

AND HIS COTEMPORARIES. 593

labelled and set apart, and a diary of the work kept. Ex-
cellent data of the most trustworthy character were thus
obtained. When sufficient progress had been made, Pro-
fessor Ramsay, now President of the Geological Society, and
myself proceeded early in September 1858, to examine and
report on the cave. The fossil remains were identified by
me, and the flint flake-implements determined on the spot
by reference to the outline figures of Abbeville specimens in
M. Boucher de Perthes’ ‘ Antiquités Celtiques;’ Professor
Ramsay made a rough sketch plan; Mr. Pengelly supplied
all the data of the workings; and a joint report was drawn
wp on September 9, in which it was stated that. human in-
dustrial remains occurred in the Brixham Cave, idiscrimi-
nately mixed with remains of Rhinoceros, Hyena, and other
extinct forms, in the undisturbed ochreous Cave-earth; and
that we failed to discover that they had been introduced
into the cavern by different agencies, or that they were of a
different age. That report was submitted to the London
Committee, and adopted on September 9,1858. It was then
forwarded to the Royal Society, which upon the strength of
it gave another grant of 100/. to prosecute the exploration.
From that period dates the wane of scepticism among
scientific men in England, respecting the geological evidence
of the antiquity of man as a cotemporary of the external
fauna of the Post-Glacial period. The humble share which I
claim to have had in the case is, that the exploration of the
Brixham Cave was led up to by the general cave inquiry
above referred to; that the existence of the cave was brought
to the knowledge of men of science by me; that its explo-
ration for a specific object was taken up at my suggestion,
and carried out on the plan laid down by me, with my
constant co-operation or advice throughout as the final re-
feree; that the fossil remains and other objects discovered in
it were identified or determined by me; and that the report
in which the important conclusions were announced was
drawn up on those determinations.’ The only other point
requiring notice is, that while the excavations were in pro-
gress, Mr. Bristow, F.R.S., was deputed from the staff of the
Geological Survey, on the requisition of the London Com-
mittee, to make a careful plan of the cavern, which he ac-
companied with a descriptive report, and Mr. Prestwich was
requested to look into some of the general physical phe-
nomena.

1 Professor Phillips, then President | all members of the committee, can con-
of the Geologcal Society, Professor | firm the statement; as do also the
Ramsay, now President, Mr. Prestwich, | documents in the records of the Geo-
Mr. Pengelly, and Dr.John Percy, F.R.S., | logical Society.

VOL. II. QQ

594 ; PRIMEVAL MAN,

VII.

To revert to the general cave investigation. It has been
already mentioned that the Gower caverns presented con-
ditions favourable for determining some of the doubtful ques-
tions, such as are nowhere else to be found in England, from
the fact of their floors being covered by marine deposits con-
taining shells. The only corresponding case in Europe then
generally known was the celebrated ossiferous cavern of San
Ciro, near Palermo, described by the Abbate Scina, Hoffmann,
and Turnbull-Christie, which I determined to examine before
making public the results arrived at regarding the geological
age of the English caves. The Gower caves had also yielded
in considerable variety and abundance relics of man, both
osseous and industrial, the latter consisting of wrought flints,
bone-weapons or implements, and ivory ornaments. ‘These
had been discovered chiefly in the Paviland Cave, above or
below the skeleton of the ‘Red Lady of Gower,’ celebrated
through the researches of Dr. Buckland. I became early
familiar with every form in which they occur, in the original
collections preserved in Gower, and formed at the time of
the exploration. My fellow-labourer and friend, Colonel
Wood, had amassed a considerable collection of wrought
flints and bone-weapons, a part of which were subsequently
the product of our repeated joint visits to the cavern. He
had also discovered human bones in ‘Spritsail-Tor’ Cave,
and in the ossiferous fissure of ‘ Mewslade.’ I was acquainted,
from observations made on the spot, with the extinct mam-
malia with which they were associated, and with the cireum-
stances under which they occurred. After the arrangements
for carrying on the exploration of the Brixham Cave were
completed I left England for Sicily, at the end of October
1858, taking Montpellier in my route, in order to study the
cave collections of the south of France, and the human re-
mains investigated by Christol, Tournal, Marcel de Serres,
&e., which had been the subject of such contested discussion
thirty years before. I then proceeded to Nice, to examine the
brecciated mass of human bones discovered near St. Hospice,
and thence to the ‘ Baussi-Raussi’ or Rocco Rosso caverns,
near Mentone, which have yielded such abundant relics of
long-continued human occupation, upon the exploration of
M. Francois Forel. On reaching Palermo I found that the
contents of the San Ciro cavern had been so disturbed
during the operation of the Government Commission, under
the direction of Abbate Scina in 1830, that not a trace of the
marine deposit, originally found upon the floor, could be de-
tected, in consequence of the materials having been thrown
up in a confused mass of talus, extending backwards to near

AND HIS COTEMPORARIES. 595
the roof of the grotto cavern. But the protracted submer-
gence of the wide fissure under the sea was distinctly proved
by the polished eastern wall, and by a band thickly drilled
by Pholad-borings. It thus became necessary to search
elsewhere along the coast, and after examining the other
ossiferous caverns in the immediate vicinity of Palermo, I
discovered the ‘ Grotta di Maccagnone,’ until then unknown
to the Sicilians.'. An account of this remarkable ossiferous
cave, dated Palermo, March 21, 1859, and extracted from a
letter addressed to Sir Charles Lyell, was communicated to
the Geological Society on the 4th of May; and on the 22nd
June a more detailed description was delivered to the same
body, when the collection of specimens illustrative of the
history of the cavern was exhibited. They excited lively
interest, from the striking manner in which they confirmed
the results arrived at by the exploration of the Brixham
Cave. The ‘Grotta di Maccagnone’ is an open rock-chamber
in Hippurite limestone, which had formerly been filled up to
the roof with alluvial matters and fossil bones, probably intro-
duced from above by rills through refts and crevices in the
fissured rock-strata. That the action had been of a tranquil
nature was clearly indicated by the presence of large fragile
Helicine shells, mingled in perfect integrity with the other
materials in the bone-breccia. The long-continued operation
of stalagmitic drip had by infiltration solidified the layer
immediately in contact with the roof into a thick mass of
hard breccia, in which large quantities were discovered of
broken siliceous stone-knives, composed of chalcedony, and
of the ordinary flake-pattern, presented by the modern obsi-
dian razor-knives of Mexico. These were intermixed with
entire and comminuted shells, bone splinters, teeth of horses
and ruminants, pieces of burnt clay, bits of charcoal,
&c., clearly indicative of the presence of man. In another
part of the same roof-breccia abundant remains were dis-
covered of the coprolite of a Hyzna, which has long been
extinct in Europe; and on the alluvial floor, below the flint-

knife breccia, the molar tooth

1 In my memoir on the Grotta di Mac-
cagnone, I premised the account of it
with a description of the physical and
geological characters of that portion of
the north coast of Sicily between ‘ Ter-
mini on the east and Trapani on the
west,’ in which ‘the ossiferous caves
abound.’ The author of the ‘ Antiquity
of Man’ (op. cit. p. 174) asserts that
‘geologists have long been familiar with
the fact, that on thg northern coast of
Sicily, between Termini on the east and

QQ

of an Hlephant was found, of

Trapani on the west, there are many
caves containing the bones of extinct
animals. Sir Charles Lyell has not
shown in what geological works the
asserted familiar knowledge is to be
found before the appearance of the
memoir aboye referred to. The exist-
ence close to Palermo of the ossiferous
caves of San Ciro, Belliemi, and Ben
Fratelli was of course well known.
2 See antea, p. 543.—[Ep.]

9

“4

596 PRIMEVAL MAN,

a species which, also, has been long extinct in Europe. The
inferences drawn from the whole case were, that the cavern,
originally full of alluvial matter, had been cleared out by
physical changes involved in upheavement, or other altera-
tion of level, with the exception of the solidified portion im-
mediately in contact with the ceiling. Following up the
investigation afterwards, strong presumptive proof came out
that the date of man’s occupation, in the savage state, of
Sicily went back to a period extremely remote, as compared
with accepted chronology, Biblical or profane, when the
Mediterranean was bridged over by land connecting Sicily
with Africa as a promontory of that continent. The evidence
upon this head will be considered in a subsequent chapter.’
Soon after the date of this communication another branch of
the evidence, proving the remote antiquity of the human race,
the agitation of which arose out of these cave investigations,
was launched into discussion among geologists. To this part
of the subject I shall now briefly refer.

WALT.

In September of 1856 I made the acquaintance of my
distinguished friend M. Boucher de Perthes, on the intro-
duction of M. Desnoyers at Paris, when he presented to me
the earlier volume of his ‘ Antiquités Celtiques,’ &c., with
which I thus became acquainted for the first time. I was
then fresh from my examination of the Indian fossil remains
of the Valley of the Jumna; and the antiquity of the hu-
man race being a subject of interest to both, we conversed
freely about it, each from a different point of view. M. de
Perthes invited me to visit Abbeville, in order to examine
his antediluvian collection, fossil and archeological, gleaned
from the Valley of the Somme. This I was unable to accom-
plish then, but I reserved it for a future occasion.

In October 1858, having determined to proceed to Sicily,
IT arranged by correspondence with M. Boucher de Perthes,
to visit Abbeville on my journey through France. I was at
the time in constant communication with Mr. Prestwich
about the proofs of the antiquity of the human race yielded
by the Brixham Cave, in which he took a lively interest ;
and I engaged to communicate to him the opinions at which
I should arrive, after my examination of the Abbeville col-
lection. M. de Perthes gave me the freest access to his
materials, with unreserved explanation of all the facts con-
nected with the case that had come under his observation ;
and having considered his Menchecourt Section, taken with
such scrupulous care,’ and identified the molars of Hlephas

1 See antea, pp. 552, 559.—[ Ep. ] Celtiques,’ tome i. p. 234, was made by
2 The section ijn question, ‘ Antiquités | M. Ravin.

AND HIS COTEMPORARIES. 597

primigenius, which he had exhumed with his own hands
deep in that section, along with flint-weapons, presenting the
same character as some of those found in the Brixham Cave,
J arrived at the conviction that they were of contempora-
neous age, although I was not prepared to go along with
M. de Perthes in all his inferences regarding the symbolical
hieroglyphics, and an industrial interpretation of the various
other objects which he had met with. The results of my
impressions were communicated to Mr. Prestwich, in a letter
dated Abbeville, November 1858, of which the following is a
transcript :—
[Copy.] et
Abbeville, Noy. 1, 1858.

My pear Prestwicu,—As the weather continued fine, I came on
here to see Boucher de Perthes’ collection. I advised him of my
intention from London, and my note luckily found him in the neigh-
bourhood. He good-naturedly came in to receive me, and I have
been richly rewarded. His collection of wrought flint-implements, and
of the objects of every description associated with them, far exceeds
anything I expected to have seen, especially from a single locality. He
has made great additions, since the publication of his first volume, in
the second, which I have now by me. He showed me ‘flint’ hatchets
which he had dug up with his own hands mixed indiscriminately with
the molars of Elephas primigenius. I examined and identified plates of
the molars and the flint-objects which were got along with them.
Abbeville is an out-of-the-way place, very little visited, and the French
savants who meet him in Paris laugh at Monsieur de Perthes and his
researches. But after devoting the greater part of a day to his vast
collection, I am perfectly satisfied that there is a great deal of fair pre-
sumptive evidence in favour of many of his speculations regarding the
remote antiquity of these industrial objects, and their association with
animals now extinct. M. Boucher’s hotel is, from ground-floor to
garret, a continued museum filled with pictures, medieval art, and
Gaulish antiquities, including antediluvian flint-knives, fossil-bones, &c.
If, during next summer, you should happen to be paying a visit to
France, let me strongly recommend you to come to Abbeville. Iam
sure you would be richly rewarded. You are the only English geolo-
gist I know who would go into the subject con amore. I am satisfied
that English geologists are much behind the indications of the materials
now in existence relative to this walk of Post-Glacial geology, and you
are the man to bring up the lee-way.

I saw no flint specimens in his collection so completely whitened
through and through, as our flint-knives (7.e. from the Brixham Cave),
and nothing resembling the mysterious hatchet which I made up of the
two pieces. What I have seen gives me still greater impulse to per-
severe in our Brixham explorations.—Yours very truly,

Joseph Prestwich, Esq. H. Fatconer.

The interest excited by this note was further stimulated
by the accounts which reached England, in April 1859, of
the results of the ‘Maccagnone’ exploration; and Mr.

598 PRIMEVAL MAN,

Prestwich, joined by Mr. John Evans, proceeded to Abbeville
during the Easter recess, at the end of the same month.
The evidence yielded by the Valley of the Somme was gone
into with the scrupulous care and severe and exhaustive
analysis which are characteristic of Mr. Prestwich’s re-
searches. The conclusions to which he was conducted were
communicated to the Royal Society on the 12th May, 1859,
in his celebrated memoir, read on the 26th May, and pub-
lished in the ‘ Philosophical Transactions’ of 1860, which,
in addition to researches made in the Valley of the Somme,
contained an account of similar phenomena presented by the
Valley of the Waveney, near Hoxne, in Suffolk. Mr. Evans
communicated to the Society of Antiquaries a memoir on
the character and geological position of the ‘ Flint Imple-
ments in the Drift,’ which appeared in the ‘ Archeologia’ for
1860. The results arrived at by Mr. Prestwich were ex-
pressed as follows :—

‘1st. That the flint-implements are the result of desion
and the work of man.

‘2ndly. That they are found in beds of gravel, sand, and
clay, which have never been artificially disturbed.

‘3rdly. That they occur associated with the remains of
land, freshwater, and marine Testacea, of species now living,
and most of them still common in the same neighbourhood,
and also with the remains of various Mammalia—a few of
species now living, but more of extinct forms.

‘4thly. That the period at which their entombment took
place was subsequent to the Boulder-clay period, and to that
extent Post-Glacial; and also that it was among the latest
in geological time—one apparently immediately anterior to
the surface assuming its present form, so far as it regards
some of the minor features.’

The European reputation of Mr. Prestwich stamped these
conclusions as well-weighed and established facts in science.
What has since followed has been the confirmation of the
results successively arrived at by Boucher de Perthes,
Rigollot, and Prestwich, the last having relieved them from
the trammels of Diluvian agency. Geologists, French, Eng-
lish, and Continental generally flocked to Amiens and Abbe-
ville, some for a time maintaining cautious doubts, but the
ereat majority convinced by the evidence. The case was
fully proved, when some eminent British geologists tendered
their acceptance of the geological evidence of the antiquity
of man, derived from his embedded works: Sir Charles Lyell,
at the Aberdeen meeting of the British Association in Sept.
1859 ; followed a year later by Sir Roderick Murchison at the
meeting of the same body held at Oxford in Sept. 1860. The
puble mind, led in its convictions by the favoured few whom

AND HIS COTEMPORARIES. 599

it elected to follow, was craving for accessible information
on the subject; and a well-considered digest of the evidence
of the case, in all its bearings, was the only thing wanting
to secure for it the accepted belief of mankind.

To recapitulate, I have endeavoured to show in the pre-
ceding remarks, that as far back as 1836 I was engaged on
questions concerning the remote antiquity of man in that
part of the globe which has been generally believed to have
been the cradle of the human race; that, in 1844, I had put
forward speculations on the subject, founded upon paleonto-
logical grounds ; that in 1854, by published evidence, I was
occupied with the same subject; that since then I have been
concerned with researches which have borne directly or in-
directly upon the same point; and that the cave inquiry which
I undertook led immediately to the re-investigation of the
whole evidence respecting the antiquity of man by geologists.

My aim in the above narrative has been to satisfy my
readers that, as a humble labourer in the field, I have not
been unfamiliar nor unconnected with the subject of the
present work. But in reading it over, I experience the
“surgit aliquid amari,’ in the feeling that the claims of the
individual float too prominently on the surface throughout.
This, doubtless, is an unseemly trait, which bears with it its
own corrective in the impression which it is calculated to
leave behind. There are occasions, however, in which the
retired and unobtrusive student is justified in speaking out ;
and this has appeared to me to be one. In support of all
that I have stated I have cited either published or authentic
documents. The important record of history sooner or later
apportions to each his due; while the true reward that the
inquirer into the truths of nature has lies in the rational
gratification which the pursuit carries inseparably with it,
and in the satisfaction that in his day and generation
he has contributed his mite to the advancement of that
natural knowledge with which the destiny of the human
race is so intimately bound up. After a few years the little
he may have done may be absorbed in a single line or sen-
tence of the great code, or it may be fused with it, ‘7m
succum et sanguinem, without a trace even of the source
from which it emanated. But while the observer ‘frets his
brief existence on the stage,’ he should take care, in the ab-
stract interest of truth, apart from personal considerations,
that what he may have done shall not be passed over by de-
fault of his own appearance to record it. This is the only
apology which I have to offer for the unseemly trait alluded
to. Much of what has been stated may appear to be beside
the question ; but there is no royal road to the extension of

€00 PRIMEVAL MAN.

natural knowledge. It is only to be accomplished by sus-
tained and patient labour. The generalized results to which
the tedious details lead, and which alone go to increase the
common stock, once arrived at are readily dressed up in a fit-
ting garb, and when skilfully combined along with cognate
materials, they produce the same pleasurable effect which is
yielded by the contemplation of exquisite forms in fragile
porcelain, artistically disposed in an elegant cabinet. But
the intelligent observer will not rest satisfied with the mere
esthetic effect of the ensemble; he will be curious to know
something of the nature of the crude material, of the pro-
cesses by which they have been elaborated, of the hands
which have moulded them into shape or laid on the har-
monious colours, and of the ateliers which have sent them
forth as finished works. And so the reflecting reader, whose
interest has been enlisted in some newly-explored region,
will not be indifferent to a narrative of the devious routes
and toilsome journeys, by which the travellers themselves
have acquired the knowledge which now engages his atten-
tion. In this spirit, I trust they will regard the preceding
remarks as an excusable vindication of the share that Mr.
Prestwich and myself have had, first, in the investigation of
the Quaternary deposits and ossiferous caves, which have
led up to that of the localities in which human relics of great
antiquity occur; second, in the investigation of the relics
themselves, and the conditions under which they are found.
Fic. 9.

HUMAN JAW, ETC., OF MOULIN-QUIGNON.

601

XXV. ON THE EVIDENCE IN THE CASE OF THE
CONTROVERTED HUMAN JAW AND FLINT-
IMPLEMENTS OF MOULIN-QUIGNON:!

Tue published procés-verbaux embody the leading points of
the evidence brought out during the investigation by the
late Conference,” held in Paris and at Abbeville, on the cir-

1The MS. of this essay was found |
among Dr. Falconer’s papers.—[ Ep. ]

2 It may be useful to give here a brief
resumé of the circumstances which led to
the meeting of the Conference. Fash-
ioned flint-weapons, unquestionably of
very remote antiquity and as certain
proofs of human agency as the watch in
the illustration of Paley, had turned
up in surprising abundance in the old
grayel-beds of Amiens and Abbeville,
but not a vestige of the bones of the men
who shaped them into form. Why it
should be so was a mystery, for human
bones are as enduring as those of deer,
horse, sheep, or oxen, and fossil bones of
extinct animals were not unfrequent in
the Somme Valley deposits. At last it
was thought that the objects so long
sought for in vain had been discovered.
To pass over minor incidents, on
March 28, 1863, a ¢errassier brought to
M. de Perthes, from the bottom of the
gravel-pit of Moulin-Quignon, two flint
haches and a fragment of bone, which,
on detaching the dark matrix enveloping
it, he found to be a human tooth. On
March 28, M.de Perthes was summoned
to the same gravel-pit (described by Mr.
Prestwich in his memoir in the ‘ Philo-
sophical Transactions’) to examine im
situ what appeared to be a portion of
bone projecting from the section, close
to its base. The specimen was carefully
detached with his own hands by M. de
Perthes, and proved to be the entire half
of an adult human lower jaw, quite per-
fect, and containing one back tooth—
namely, the penultimate, or last but one.
The sockets of the other teeth were all |

present, and filled with matrix, with the |
exception of the antepenultimate, the |
socket of which was effaced, the tooth |
having been lost during life. The soli- |

tary molar present was hollow from
caries, and this hollow was also filled
with the matrix.

The deposit from which the jaw was
extracted was the ‘black seam flinty
gravel,’ so called from its intensely dark
(blueish-black) colour, arising from ox-
ides of iron and manganese. It rested
immediately upon the chalk, and _be-
longed to what Mr. Prestwich calls the
‘high level’ series, being the oldest of
the Somme Valley beds. A thin layer
of black mangano-ferruginous clayey
matter was interposed between the chalk
and the gravel. If the jaw had proved
to be an authentic fossil, and had come
out of the alleged position, it indicated
the existence of man, by an actual bone,
at a period of extremely remote anti-
quity. A single detached human molar
was found at the same time, correspond-
ing exactly in appearance and in the
matrix with which it was covered ; and,
to complete the case, a flint hatchet, co-
vered with black matrix, was extracted
from the same spot by M. Oswald Dim-
pre, who accompanied M. de Perthes.

|The details were all given by M. de

Perthes in the ‘ Abbeyillois’ of April 9,
1863, and ina note communicated to
the Academy of Sciences on April 20.
Mr. Prestwich, Mr. Evans, and Dr.
Falconer were in France at the time, and
hearing of the asserted discovery, they
determined to visit Abbeville. The two
former proceeded there on April 13,
when their suspicions were instantly
aroused. They pronounced the haches
said to have been yielded by the ‘ couche
noire’ to be modern fabrications. Dr.
Falconer followed a day later, when they
had left, and also got several haches from
the ‘black-seam gravel,’ which, upon
closely examining them upon his return

602

HUMAN JAW AND FLINT-IMPLEMENTS

cumstances attending the alleged discovery of a human jaw
in the gravel deposits of Moulin-Quignon. The fact was first
announced by M. Boucher de Perthes in the ‘ Abbevillois’ of

to London, he considered to be spurious.
Having been obligingly permitted by M.
de Perthes to examine the jaw, he was
struck with the unusual combination of
peculiar anatomical characters which it
presented, and was thus led to the im-
pression that it was of fossil antiquity.
That impression he communicated on the
14th to.Dr. Carpenter, and on the 15th
to M. de Quatrefages, at Abbeville, but
subject to the reserve of a more detailed
study of the materials, and on the 15th
he wrote to the same effect to his friend
M. Lartet, to whom the jaw was con-
signed in Paris.

On April 16 Dr. Carpenter communi-
cated a short paper to the Royal Society,
supporting the authenticity of the disco-
very; and during the debate, Dr. Fal-
coner, in the absence of Dr. Carpenter
and himself, was unauthorizedly cited
as entertaining the same opinion. On
April 20, M. de Quatrefages communi-
cated to the ‘Academy of Sciences’ a
note by M. Boucher de Perthes, followed
by descriptive remarks by himself, con-
veying the high authority of his opinion
in fayour of the jaw being a true fossil
of geological antiquity. On April 18
Dr. Falconer, immediately after his re-
turn to London, commenced the delibe-
rate scrutiny of the materials which he
had brought with him from Abbeville,
and on the 21st, in conjunction with, or
aided by, Mr. John Evans, Mr. Prest-
wich, Mr. Busk, and Mr. Tomes, he ar-
rived at results opposed to the authen-
ticity alike of the ‘detached molar’ of
the jaw, and of the flint haches. That
day, without the delay of a post, he com-
municated his suspicions to M. Lartet,
requesting him to make them, and the
grounds upon which they were founded,
known to M. de Quatrefages. But the
latter had already given in his affirma-
tive memoir to the ‘Institut’ on the
previous day (20th), followed on April
27 and May 4 by successive notes in the
same sense. On April 25 a letter by
Dr. Falconer, written before he was
aware of M. de Quatrefages’ first com-
munication, appeared in the ‘Times,’
questioning the authenticity of the ‘jaw’
and of the haches. Men of science in
France and England were thus suddenly
placed at direct issue on a graye and
important point of great general interest.

But, happily, from the frankness and
rapidity of the communications inter-
changed, there existed the most cordial
relations, and the conviction of loyalty
and good faith on both sides. The
French savants, the more they went into
the case, were the more convinced of the
soundness of their conclusions; while
their English opponents, the more they
weighed the evidence before them, were
the more strengthened in their doubts.
As a wordy discussion would but have
wasted time and must have been pro-
tracted, and as a personal conference
held out the best prospect of a speedy
settlement of the questio , a ‘7éunion’
of men of science, to be h:ld at Paris,
was proposed by the French savants.

The English aeputation, consisting of
Dr. Falconer. Dr. Carpenter, and Prof.
Busk, reached Paris on May 9, and im-
mediately proceeded to business, being
joimed on the following day by Mr.
Prestwich. The French members con-
sisted of M. de Quatrefages, Member of
the Institute ; M. Lartet, Member of the
Geological Society of France and Foreign
Member of the Geological Society of
London; M. Delesse, Professor of Geo-
logy to the Ecole Normale, Paris; and
M. Desnoyers, Member of the Institute.
The following savants also took a share
in the proceedings throughout, and af-
forded the utmost aid in the investiga-
tion, viz.:—M. LiAbbé Bourgeois, M.
A. Gaudry, and M. Alphonse Milne-
Edwards. At the request of the Eng-
lish members, M. Milne-Edwards, Mem-
ber of the Institute, and the eminent zoo-
logist, courteously agreed to preside over
the Conference.

The procés-verbauc of this Conference,
which extended over five days, were
printed in detail in the ‘ Natural His-
tory Review,’ for July, 1863, with an
introduction and appended notes, by
Dr. Falconer, Mr. Busk, and Dr. Car-
penter. The conclusions arrived at by
the Conference are embodied in Dr. Fal-
coner’s essay, which is now for the first
time published.

The above particulars are derived from
the introduction to the procés-verbaux
above referred to, and from two letters
by Dr. Falconer which appeared in the
‘Times’ newspaper for April 25 and May
21, 1863.—[Eb. ]

OF MOULIN-QUIGNON. 603

April 9th. On April 16th it was communicated to the Royal
Society, and on the 19th to the Academy of Sciences. On
April 25th the authenticity of the jaw and haches was contro-
verted in the ‘ Times.’ On May 9th the Conference opened
at Paris, and it closed at Abbeville on the 13th. On May 18th
M. Milne-Edwards and M. de Quatrefages severally commu-
nicated notes to the Academy of Sciences, on the results at
which the Conference had arrived. The ink with which the
convention was signed was hardly dry when M. de Beaumont
struck at the whole fabric a blow, ‘qui semble enlever toute
valeur scientifique a la machoire dont on s’est tant occupé.’
At that meeting of the Academy, this eminent geological
authority refused to admit that the Moulin-Quignon gravels
were older than the modern period. He referred them to the
age of the Peat-beds of the Valley of the Somme, in which
human skeletons with implements of stone and iron abound ;
and at the next following meeting of the Academy, when he
resumed the subject, he stated that a corresponding peat-
bed covered the remains of an ancient Roman way in the
Département du Nord.

That ‘le cercle de la discussion relative au gisement de
Moulin-Quignon est peut-étre bien loin d’étre épuisé,’ is
clear from the next incident in the order of events. At the
meeting of the Academy held on May 25, M. Hébert, in a
review of the opinions advanced by M. de Beaumont, as-
signed a different geological position to these beds, which he
removes from the ancient Quaternary deposits of the Valley
of the Somme, such as Menchecourt and St. Acheul, placing
them two stages even above the Loess, but below the Peat-
bed alluvia, while Mr. Prestwich, in a recent communication
to the Geological Society, after a fresh survey of the field,
maintains his original opinion that the Moulin-Quignon beds
belong to the ‘ high level’ gravels, or oldest Quaternary de-
posits of the Somme, formed immediately after the exca-
vation of the valley began. We have thus three different
phases of opiniou by competent authorities on the geological
age of the beds, and it is difficult to conceive the divergency
being carried further, or of more being entertained, respect-
ing the case.

The contested haches of the couche noire have not fared
better. Accepted at last by the great majority of the Con-
ference as authentic, they have been severely assailed in a
communication addressed to the ‘ Atheneum’ on June 2, by
Mr. John Evans, who, unable to be present at the Conference,
to which he was invited, paid a fresh visit to Abbeville at
the end of last month, in order to reinvestigate the circum-
stances of that part of the case. He contends that the sus-

604 HUMAN JAW AND FLINT-IMPLEMENTS

pected haches are modern fabrications, and thinks that he has
detected the proof of their having been introduced into the
gravel section. After detailing the results of his observation
he adds: ‘There remains therefore not the slightest doubt
on my mind that a fraud, and a most ingenious and successful
one, has been practised by some of the Abbeville terrassiers.’

Nor has the celebrated jaw itself had immunity from dis-
cussion conferred upon it by the verdict of the Conference.
An eminent French Anthropologist, M. Pruner-Bey, well
known for his familiarity with the characters which distin-
guish the races of mankind, has submitted the jaw to exa-
mination in the aspect of its natural history peculiarities,
and arrived at the opinion that, ‘La mdchoire de Moulin-
Quignon appartenait a un individu brachycéphale, de petite
taille, de Page de pierre.’ He found that it corresponded
with a Swiss lower jaw of the brachycephalous type, of the
age of iron in every respect, except that the coronoid process
in the latter was more elongated. M. Pruner-Bey does not
state to what division of the ‘Stone-age’ the jaw is, in his
opinion, referable ; and his remarks do not necessarily remove
it from the position assigned to it by the verdict of the Con-
ference ; but they are suggestive of further inquiry as to the
direction of its affinities.

It is evident, therefore, that to the conviction of men of
science, the labours of the Conference have not definitively
settled any one of the moot points which came before it.
But the mass of evidence which it collected and the conflict
of opinion which it elicited will doubtless assist materially
in conducting to a final judgment hereafter. The case
throughout maintained a perplexed and contradictory cha-
racter, not to be surpassed probably by any cause célébre on
record.

The evidence was of two classes: 1st. The imtrinsic, yielded
by the Flint haches and by the jaw, regarded per se through
their physical characters. 2nd. The eatrinsic, or that con-
nected with the beds in which the objects were asserted to
have been found.

In many important respects the first was directly at vari-
ance with the second, which appears to have weighed most
with the Conference in determining the conclusions at which
it arrived. On the present occasion it is proposed to give a
summary and analysis of the evidence as set forth in the
proces-verbauz, together with the results of some later inqui-
ries, where it was incomplete.

The Flint haches and the human jaw, although in their
nature so widely different, were intimately connected in this —
case, from the fact that they were both said to have been

OF MOULIN-QUIGNON. 605

yielded by the same deposit; and the most important pieces
among the former were covered with the same matrix as the
latter. If either proved to be open to suspicion upon the
internal evidence, doubt was pro tanto reflected upon the
other. I shall therefore consider the evidence of the haches
as well as of the jaw.

1. Intrinsic Evidence of the Haches.—And first as regards
the haches. Suppose a mass of a compact hard, homogeneous,
non-crystalline, mineral substance like Obsidian or Flint,
presenting a flat surface. Let a blow be struck perpendicu-
larly to the latter, with moderate force, by a light convex
iron hammer, anywhere near the centre, and an invariable
result follows. Although not visible externally, a partial
solution of continuity takes place in the interior of the
mass, in the form of a cone of which the point of impact
constitutes the apex. Suppose the mass to have the shape
of a polygonal block, and a similar blow to be struck
near the periphery, where there is a line of least resist-
ance, a flake will be detached; and as from the given
conditions there is not sufficient thickness of substance for
the vibrations being propagated uniformly all round, instead
of a cone, the flake presents a conoid surface, the immediate
apex of which above, corresponding with the point of impact,
is a true cone. The general convex surface which results
may be compared to that of one of the valves of a convex
bi-valve shell; and the hollow which it leaves on the block
is commonly known under the name of ‘ conchoidal frac-
ture.’ The phenomena have long been observed, but although
manifestly dependent on a definite physical cause, they have
not yet, so far as I have been able to ascertain, been made
the subject of mathematical analysis.

Suppose the same conditions in another block, but let the
blows be struck with a piece of flint or other hard stone; the
same effects are produced, but modified. In the one case the
resultant cone is more depressed, and in the other the
conoid convexity, ‘or bulb of impact’ on the flake, is less
prominent and extended over a broader surface. A necessary
consequence is that the conchoidal facets produced by the
repeated detachment of flakes, or splinters, when stone-struck,
are broader and shallower, and the dividing ridges less ele-
vated ; while iron-struck facets are deeper, narrower, and
more pronounced, with more elevated and more angular
dividing ridges. I put the case in the most general terms,
as founded on observation. The principal cause of the
difference has still to be made out; but one thing is certain,
that commonly the facets, upon the most ancient haches, are
shallower and less pronounced than in modern imitations.

606 HUMAN JAW AND FLINT-IMPLEMENTS

The Flint-objects, which are accepted as proofs of human
industry of very remote antiquity, are therefore of two kinds.
Ist. The more simple, or flakes. 2nd. The haches, which
have been shaped out of blocks into form by the repeated
detachment of small fragments or éclats.

Where haches occur, flakes designed for use commonly
occur also.

Flint haches, of undoubted antiquity, bear certain marks,
the presence of one or more of which is considered necessary
to guarantee their genuineness. Ist. They are weather-
stained or discoloured, according to the nature of the embed-
ding deposit. 2nd. They present a general polish or vitreous
glimmer, which is not seen on the dull fresh fracture of the
same flint. 3rd. They are patched over with a patina, con-
sisting of a crust of carbonate of lime, or of other mineral
matter, or they bear dendrites.

Besides these, there are other indications which, regarded
per se, are less certain and constant, but which assist mate-
rially in judging on the genuineness of haches. 1st. Com-
monly they are more or less used and bhunted or indented at
their sharp edges, and they are frequently abraded by roll-
ing. 2nd. The conchoidal facets, caused by the detachment
of flakes, are broad and shallow, and the dividing ridges but
little conspicuous. 3rd. Residuary films are either absent
from the facets, or, if present, their margins are broken,
and colouring matter in matrix is generally interposed
between them and the body of the flint, through capillary .
action.

The broad and shallow facet and the low dividing ridges
have been closely imitated by using stone hammers at the
present time; but in certain cases haches, considered upon
other grounds to be of genuine antiquity, present conchoidal
facets and raised dividing ridges. In others, the points or
edges are uninjured. In these instances the vitreous glim-
mer, dendrite, or other kind of patina, are relied upon as
tests of their genuine character. Occasionally the discrimi-
nation, except by practised experts, is difficult and per-
plexing.

Counterfeit haches of modern fabrication are commonly
distinguishable with readiness, by the absence of all the
characters above indicated, and by the presence of certain
peculiar marks. Ist. They present no vitreous glimmer
different from a fresh fracture; they are unstained, un-
weathered, and free from dendrites or other patina; and they
bear no marks of rolling. 2nd. The facets are more con-
choidal, narrower, deeper, and commonly more numerous;
and the dividing ridges well pronounced, angular, and

OF MOULIN-QUIGNON. 607

unworn. 3rd. The edges all round are sharp, fresh, and
unworn; or, if indented, the breaks are unequal, with crushed
fractures. 4th. The facets present more or less numerous,
broad and entire, residuary films, free from underlying ma-
trix. 5th. They are frequently finished by vertical blows on
the broad surface, to reduce some inconvenient salient in-
equality, leaving rude crushed fractures, which occasionally
present streaks of metallic iron. 6th. They are commonly
ruder in form, and an experienced eye will detect the differ-
ence; but this indication is not to be relied on.

The Flint-implements from Moulin-Quignon were of two
kinds. Ist. A few haches, and a corresponding number of
flakes, which were universally accepted as being unquestion-
ably genuine. The haches were rolled, highly weathered, and
uniformly deeply stained with iron; polish vitreous; facets
broad and shallow, without films; ridges low and incon-
spicuous ; edges much worn and rounded. In short, they
presented more decided marks of age and rolling than most
of those found in the other localities of the Valley of the
Somme. The flakes-were also tinged with iron, and equally
genuine and ancient in appearance. The majority of both
appear to have been got before the present year, and to have
been yielded by the ferruginous gravel. Not a single speci-
men of either, derived from the ‘ black seam,’ was laid before
the Conference. 2nd. A large number of haches, without
the admixture, so far as I remember, of a single flake, and
all, when denuded of their matrix and well washed, present-
ing the most marked contrast to the genuine specimens from
the same locality. The English members of the Conference
produced about twenty, most of them procured within the
last three months, from the ‘black seam’ or overlying
ferruginous gravel ; and a considerable or nearly equal number
was presented by the French members. When washed they
yielded the following characters :—Outline generally of the
spear-headed pattern; a few oval or almond-shaped, and the
majority of them coarsely wrought, with a great sameness of
pattern, as if the production of one or two hands; surface
with the dull appearance of a recent fracture, and without
glimmer, dendrites, incrustation, or patina of any sort;
colour, even where greyish-white and porous, unaltered, and
substance perfectly free from ferruginous infiltration ; facets
conchoidal, with angular dividing ridges; films numerous,
broad, and entire, without underlying matrix; edges sharp,
continuous, not rolled, and either unindented, or where inter-
rupted showing recent angular fractures. Where a new frac-
ture was made before the Conference, the fresh surface differed
in no respect from that of the rest of the flint. Such at

608 HUMAN JAW AND FLINT-IMPLEMENTS

least was the view held by the English members, as set forth
in the procés-verbaua.

Further, two of these haches, procured by me at Moulin-
Quignon, on the 14th and 15th April, presented on the
middle of the thickest part an appearance which Mr. Evans
and myself interpreted as being probably the result of a
vertical blow applied with an iron hammer, to reduce an in-
convenient inequality. One got by Mr. Nicholas Brady, on
the 17th April, upon being carefully washed by me yielded
immediately afterwards a distinct streak of metallic iron. It
was observed at the time by Dr. Charles Murchison, Messrs.
Antonio and Nicholas Brady, and myself. It was examined
on the following day, under the microscope by Dr. Carpenter,
who confirmed the observation. It was suggested at the
Conference that the iron marks in this instance may have
been caused by the pick-axe of the pitman; but the explana-
tion was insufficient, inasmuch as the covering of matrix
was carefully examined before washing, and it presented no
external marks corresponding with the streak. These haches
were compared with modern counterfeit haches of known
English origin, and so far as surface, facets, sharp edges,
and fresh look are concerned, they exhibited no appreciable
difference of material value.

All these flint-implements, of the suspected character,
whether covered with ferruginous gritty sand or with ‘ black-
seam’ matrix, were coated superficially by a subjacent thin
dark film, which by slight washing was reduced to a bronze-
tint, and entirely removed by scrubbing with a brush and hot
water. The scrutiny of the Conference was mainly directed
to those which came out of the ‘ black seam’ that yielded
the jaw. The coating of matrix upon them was regarded by
M. de Quatrefages, resting upon the opinion of M. Delesse,
subsequently confirmed by Professor Delafosse, as being
natural, from the fact that there was an appearance of
minutely mammilliform aggregations of ‘ Limonite of iron’
upon the surface of the haches. But they had no firm adhe-
sion to the flints, and although presenting occasional points
of metallic brilliancy, they were removed with the utmost
facility by washing. The English members, throughout the
sittings devoted to this part of the case, saw no reason to
regard the matrix as. other than a coating artificially laid on.
Mr. Evans found by experiment, that the fine dry powder of |
the ‘ black seam’ gravel, consisting of earthy matter charged
with oxides of iron and manganese, when moistened and
rubbed on with the finger, exactly reproduced the same
appearance, and also the bronze tint remaining upon the
suspected haches when half-washed. Where the black matter

OF MOULIN-QUIGNON. 609

was insinuated into the pores of the whitish-grey incompact
parts of the flints, there was some difficulty in removing the
specks with the brush.

The characters of the flint-gravel of the ‘black seam,’ an
important part of the evidence, were submitted to examina-
tion. Specimens were handed in on both sides. The great
majority of the pebbles, and more especially the larger ones,
were deeply stained with iron or covered with dendrites. The
French members produced some fragments, mostly small
and angular, which when washed they considered to be
nearly free from tinting. The precise numerical ratio of the
latter to the former was not determined at the Conference.
The English members held, that the exceptions were so few
as not materially to alter the complexion of the evidence.
Dr. Carpenter, by a written question which he handed in to
be put to Professor Delafosse, implied that while the sus-
pected haches were ‘free from coloration or dendrites, the
ordinary flints of the same bed universally presented such
indications.’ Mr. Evans states that ‘the pebbles in the
“black band,” in which the haches are said to have been
principally found, are without exception more or less stained
by the ferruginous matrix, a stain which cannot be removed
by washing.’ Mr. Prestwich, during his subsequent visit to
Abbeville, took the opportunity of making the experiment
omitted by the Conference. He washed a portion of the
gravel containing 135 flint-fragments, and found that 108 of
them were completely stained and coloured, 22 partially so,
and only 5 (all small) not at all altered : the last thus forming
less than 4 per cent. of the whole. He admits ‘that the rarity,
therefore, of unaltered flints in this bed is in contradiction to
the unaltered condition to the totality of the flint implements
of the new type,’ i.e. the suspected ones.

The thin layer of argillaceo-metalliferous matter, constitut-
ing the ‘ couche noire’ or ‘black seam’ proper, was also ex-
amined. It lay immediately upon the chalk, portions of which
adhered to the under surface of ten specimens exhibited. No
detailed chemical analysis of this substance was submitted to
the Conference ; but a qualitative analysis was communicated,
showing that it consisted of fine earthy matter highly charged
with oxides of iron and manganese, and that it contained no
organic materials. Professor Williamson, of University
College, has since caused a careful analysis of the substance
to be made, in the ‘Birkbeck Laboratory,’ by his able
assistant Mr. Haughton Gill. The results, published in one
of the explanatory notes appended to the procés-verbauz,
establish that more than half of the substance of the ‘ couche
noire’ consists of matter insoluble in hydrochloric acid, and

VOL I. RR

610 HUMAN JAW AND FLINT-IMPLEMENTS

the residue, chiefly of alumina and oxide of iron, and of oxide
of manganese, in variable quantities in different portions of
the matrix ; and that it contained no humus or other organic
constituent. The matters analyzed consisted of the ‘ black
seam’ proper; of the same derived from the black flinty
gravel; and of portions of the black covering of the jaw and
of matter extracted from the hollow tooth. They were all
shown to be made up of the same constituents, although in
different proportions, and they are regarded as being of the
same nature.

M. de Quatrefages exhibited to the Conference a flake-lke
fragment, of a white, hard, and solid substance, extracted by
him from the matrix of the ‘couche noire,’ and remaining
uncoloured. He contended that as this body was unstained,
the absence of staining in the suspected haches might in a
similar manner be accounted for. The precise nature of the
body in question was not ascertained; it was first regarded
as being probably a portion of a tooth, and then, when exa-
mined by the Conference, as possibly portion of a shell (?).
Hairs, coloured and uncoloured, were also disengaged from
specimens of the ‘couche noire.’ They were regarded as
being of Rodents or of Bats, but there is good reason to
suspect, according to Mr. Busk, that their presence was
purely accidental.

The general results yielded by the examination of the
haches and accompanying circumstances, apart from the
gisement, were differently regarded by the different sides.
M. Milne-Edwards, President of the Conference, in his com-
munication made to the Academy of Sciences on May 18th,
puts the case thus :—

‘ Tous les membres de la réunion ont été d’accord pour admettre que
dans beaucoup de cas, 4 raison de l’existence de certains caracteres qui
semblent ne pouvoir étre imprimés que par le temps, on peut, par la
seule inspection d’une hache en silex, constater son authenticité, c’est-
a-dire son origine ancienne. Mais les avis ont été partagés au sujet
des bases d’un jugement légitime en sens contraire.

‘MM. Falconer, Prestwich, Carpenter et Busk pensaient que l'absence
de tout signe évident de vétusté et l’existence de certaines particularités
dans la forme ou dans les fractures de ces haches étaient.des preuves
irrécusables de leur fabrication récente. Ces savants se considéraient,
par conséquent, comme fondés a nier l’authenticité des haches dont la
surface ne présentait ni patine ni incrustations, dont les arétes étaient
trés-vives et dont la forme s’éloignait plus ou moins de celle des haches —
reconnues vraies. Puis, faisant l’application de ces principes aux
haches tirées des diverses couches du terrain de transport de Moulin-
Quignon ou d’autres lieux, ils admettaient l’authenticité des unes,
tandis quwils déclaraient fausses beaucoup d’autres, notamment toutes
2elles provenant de la couche noire ot M. de Perthes avait trouvé la
miachoire humaine.

OF MOULIN-QUIGNON. 611

‘MM. de Quatrefages, Desnoyers et Lartet, ainsi que les autres
naturalistes francais qui prirent part a cette partie de l’enquéte, soutin-
rent qu'il fallait étre plus réservé; que tres-rarement, peut-étre méme
jamais, des particularités de forme, une apparence de fraicheur ou
d’autres caractéres intrinséques du méme ordre, ne pouvaient suffire
pour bien établir la fausseté d’une de ces haches en silex; que des
caracteres de ce genre pouvaient inspirer des doutes, et qua défaut
d'autres données ces doutes devaient peser beaucoup dans nos juge-
ments; mais que les considérations tirées du mode de gisement de ces
instruments et des circonstances dans lesquelles leur découverte a eu
lieu devaient avoir 4 nos yeux une valeur bien plus grande; enfin, que
des preuves d’authenticité obtenues de la sorte doivent toujours l’em-
porter sur les soupcons que pourraient faire naitre les particularités dont
je viens de parler. Ainsi ces naturalistes furent unanimes dans le
jugement qwils portérent sur l’une des haches trouvées dans la couche
noire de Moulim-Quignon par M. de Quatrefages: maleré la facilité
avec laquelle la surface lisse de ce silex se laissait dépouiller de sa gan-
gue, malgré sa forme, la vivacité de ses arétes, et malgré son aspect de
fraicheur, ils n’hésiterent pas & en admettre l’authenticité, par cela seul
que les circonstances dans lesquelles ce savant l’avait découvert dans le
sein de la terre leur paraissaient exclure toute idée de supercherie. Par
conséquent, MM. Desnoyers, Lartet et Delesse, aussi bien que tous les
autres naturalistes francais qui assistaient & cette discussion, ont déclaré
que dans leur opinion le jugement porté sur les haches de la couche
noire de Moulin-Quignon, par M. Falconer, ne pouvait légitimer aucune
conclusion touchant l’introduction frauduleuse de la michoire humaine
dans le dépot de gravier ou M. Boucher de Perthes avait trouvé cet os.’

The grounds of the doubts maintained by the English
members, during the three sittings of the Conference de-
voted to the subject of the haches may be enumerated thus:

The haches formed two sets, exhibiting the most contrasted
differences, the one consisting of haches and flakes highly
weathered, discoloured or iron-tinted, and occasionally much
rolled, both bearing every mark of antiquity, and unanimously
accepted as genuine. The other set absolutely fresh-looking,
having sharp edges, unweathered, unstained, free from marks
of rolling, and unaccompanied by a single flake ; there was a
wide chasm between the two sorts, and no intermediate links
to connect them. The recent-looking series were yielded
by all parts of the section below the Loess, alike by the
ochreous gravel and by the ‘black seam.’ The flint frag-
ments in both deposits were, as a general rule, more or less
stained with iron, while not a single specimen among the
suspected haches was appreciably tinted. The gravel and
sands of the section are freely permeable to water down to
the chalk, and the mangano-ferruginous layer of the ‘ black
seam’ also readily imbibed moisture. According to Mr.
Prestwich, the most certain of all the tests of the authen-
ticity of a hache is its identity in mineral character with that

RR2

612 HUMAN JAW AND FLINT-IMPLEMENTS

of the component flints of the beds in which it occurs. How
then, it may be asked, could these haches, jointly exposing
several square feet of surface, have remained for a length of
time under such conditions and escaped staining or dendritic
incrustations? If it were assumed from their sharp edges
and freedom from rolling that they had been manufactured
in the neighbourhood, how was the entire absence of corres-
ponding flakes to be accounted for? Genuine haches, accord-
ing to Mr. Evans, have been comparatively rare at Moulin-
Quignon, where they have been found for upwards of twenty
years. Not one of this description, universally so admitted,
from the ‘black seam,’ was laid before the Conference, while
haches of the suspected character have turned up in extra-
ordinary abundance, all of a sudden, within a late period. It
was stated at the Conference that more had been yielded at
Moulin-Quignon within the last few months than of the
genuine kind during several previous years. Nor were they
confined to Moulin-Quignon. Specimens were submitted,
professing to have been yielded by the gravel-pits of Mautort
and Saint Gilles, and exhibiting the rudest attempt at fabri-
cation, but disguised by a coating of black matrix smeared
over them. One of these, from Mautort, according to a
manuscript note of Dr. Carpenter, was ‘ given up as facti-
tious ;’ and on another from St. Gilles, Mr. Busk pointed
out that the matrix contained ‘unquestionable fragments of
recent vegetable structure.’ One ingredient of fraud in the
ease of the haches, clearly made out, was sufficient to cast
doubts on the rest of the series, which of themselves were
intrinsically suspected upon other grounds. The French
members, according to M. Milne-Edwards, held by one of
the haches, found by M. de Quatrefages in the ‘ couche noire.’
But its genuine character was not accepted on this intrinsic
evidence by the English members, during the three sittings
of the Conference at Paris. And, even if admitted to be
genuine, it would have gone for little as evidence in the main
case; for, on the hypothesis that the great majority of the
questionable specimens had been artificially placed before-
hand by the pitmen in the situations where they were found,
it might have been expected that some genuine specimens
would have been introduced to give countenance to the others.
Further, it is well known that in all countries where coins,
antiquities, or archeological objects are in demand, counter-
feits are offered abundantly in the market. That Abbeville
should have formed an exception to the general law was, in
the abstract, more than could have been expected. But it
was notorious that an active fabrication of counterfeit flint-
implements was carried on both at Amiens and Abbeville,

OF MOULIN-QUIGNON. 613

to meet the lively demand caused by the authentic discoveries
made by M. Boucher de Perthes and others in the Valley of
the Somme. Mr. Evans, whose verdict on this point is of
the highest authority, states that a year ago he purchased
“an indisputably forged drift-implement from one of the
terrassiers of Abbeville.’ Numerous other instances of the
same kind might be cited if necessary. Some of the imple-
ments from Moulin-Quignon and Mautort, professing to be
ancient, were upon the internal evidence as certainly modern
counterfeits as an inference of this nature could be esta-
blished, in default of the testimony of eye- witnesses to the
act of manufacturing them. That the traffic in counterfeit
haches is Pelmmerative to some of those concerned in it is
sufficiently proved by the fact, that five francs are commonly
demanded of strangers at Amiens and Abbeyille for the
coveted privilege of detaching from its bed a hache professing
to be found im situ. It is within my knowledge, from the
direct statements of the parties, that this has occurred to
two Englishmen during the month of last April: to the one
at Amiens, to the other at Abbeville. Hach was asked, and
each paid, five francs. In the latter case, the specimen was
submitted to the Conference as one of the series of counter-
feits from Moulin-Quignon ; it having been regarded in that
light when brought to England and examined by Mr. Evans,
Mr. Prestwich, and myself. The great demand for flint-
implements arises from the number of strangers who now
visit Amiens and Abbeville, attracted by the general interest
which the subject has of late years excited. The supply of
genuine implements proved insufficient, and the natural result
followed. Considering the facility with which counterfeits
can be made, half a france per hache would upon a consider-
able sale be amply remunerative, apart from the larger sum
derived from specimens professing to occur in situ.

2. Intrinsic Evidence of the Jaw.—So much time was spent
on the question of the flints, which occupied the greater part
of the three meetings of the Conference at Paris, that little
space remained for the examination of the principal object,
the jaw itself; and this part of the case was gone through in
a much more summary manner. The ‘detached molar,’
covered with matrix of the ‘black seam,’ which had been
sawn up in London, and upon which so much weight was at
the time vested, was, by the consent of both sides, withdrawn
from the case, as being cpen to question on the score of
identification. But it gave rise to a discussion as to the
value of the quantitative presence of gelatine as a test of fossil
teeth. Two specimens, from the Diluviuwm of Anvers, were
produced by M. Delesse, the one the canine and the other

614 HUMAN JAW AND FLINT-IMPLEMENTS

the lower fang of a molar of Hyena spelea, as proofs of the
retention of gelatine in undoubted Quaternary fossils. But
the former adhered very strongly to the tongue, having lost
a very considerable portion of its gelatine, and the latter,
although non-adherent, had altered much in colour and
become yellow. Further, they hardly bore on the particular
case, inasmuch as the conditions of the beds may have been
very different, preservative in their nature in the one, and
not so in the other. For it was not shown that the Anvers
specimens had been embedded in a deposit charged with
mangano-ferruginous oxides.

The matrix covering the jaw was first examined. It was
regarded on the one side as being a natural deposit; on the
other as a coating which had been laid on, or otherwise pro-
duced artificially. Besides MM. Delesse and Hébert, M.
Delafosse, Professor of Mineralogy, who examined it in the
presence of the Conference, expressed his opinion that the
coating was natural, indicating that the bone had long been
embedded in the matrix. He expressed the same opinion
regarding the matrix of some of the flint haches. He ‘was
shown a hache from Mautort, which had been previously given
up as factitious, and pronounced the gangue to be a natural
and ancient deposit.? The doubts of the English members,
that the coating was artificial, were not shaken by this ver-
dict upon the Mautort and other haches detailed in the
ee -verbaux, one of them having exhibited in its matrix

‘ unquestionable fragments of recent vegetable structure.’
Chemical analysis, undertaken since the close of the Con-
ference, has shown that the matrix of the ‘ black seam’ and
the coating upon the suspected haches and upon the jaw are
alike in their composition, although differing quantitatively in
the constituents. The English members rested their doubts :

Ist. Upon the absence of staining by the matrix both upon ~

the haches and upon the jaw. 2nd. On the facility with
which the coating upon both was removed by washing. The
French members appear to have relied on the presence of
occasional brilliant points, upon the granular clusters of
‘limonite de fer,’ as indicating a natural origin.

The jaw was carefully sawn across by Mr. Busk, imme-
diately in front of the solitary penultimate molar, and the
section was so conducted as to include vertically a portion of
one of the fangs. The specimen consisted of the left ramus,
perfectly entire, of a human subject, estimated to have been
between 60 and 70 years of age. Although yielded by a bed
of flint gravel there was not the slightest appearance of
crushing « or mark of rolled action, the condyle and thin apex

of the coronoid being as perfect as in a recent bone. The

Pile,

OF MOULIN-QUIGNON. 615

black coating of the supposed natural matrix was washed off
the surface of one of the segments with great facility. But
small specks of the same fine black matter remained in some
of the minute hollows and miliary pores. These were re-
moved without difficulty by the application of a tooth-brush.
The general colour of the washed surface was a light buff,
mottled with faint brownish stains, which were persistent,
resembling those frequently observable in bones that have
lain long buried in the earth. But the outer surface was
tolerably smooth and perfect, presenting little indication of
the superficial erosion commonly seen on old bones derived
from cemeteries. There was no appearance either on the
exterior or within of dendritic deposits. This is a point in
which the jaw differed remarkably from other bones of un-
doubted fossil antiquity, found in the Somme gravel deposits,
and which contrasted also strongly with the jaw in the
ereater amount of alteration by loss of gelatine which they
had undergone. The substance of the bone was dry and
friable towards the alveolar border, but on the whole it was
tolerably firm under the sand and but little altered, and the
fresh section afforded a distinct odour of sawn bone. Inter-
nally the structure was absolutely free from any sign of mine-
ral impregnation. The cortical layer, which was remarkably
thin and condensed, was nearly white, and the walls of the
empty cancelli of a faint brownish tint. The upper aperture
of the dental canal was covered over with the black coating,
which had penetrated well into the mental foramen. The
most remarkable appearance in the section was the lining,
as it were, of the middle portion of the same canal, with a
thin layer of fine grey sand, easily removed by the point of
a needle, and which, when examined under the microscope,
was seen to consist of minute grains of white quartz or silex,
intermixed with a few particles of oxide of iron, but with-
-out a speck of the black matrix. The section of the fang
showed that the dentine, so far as exposed, was perfectly
white, full of gelatine, and in no respect different in appear-
ance from that of a recent tooth. In short, it reproduced
exactly the characters which I had previously described as
yielded by the section of the fang of the ‘ detached molar,’
and which had impressed me and others so strongly with the
conviction of its being a recent tooth. The body of the
molar present in the jaw was hollow from caries, but the
enamel was white and brilliant, and without the slightest
appearance of alteration. The crusta petrosa, very little of
which was present, was very faintly coloured. The socket
towards the upper part was not completely filled by the fang,
- and the interval was partially occupied by the black matrix

616 HUMAN JAW AND FLINT-IMPLEMENTS

of the coating, and by sandy particles. One important part
of the intrinsic evidence still remained to be determined,
namely, the chemical analysis of the bone; but the Conference
was too pressed for time to wait for it, and the recent charac-
ters were in other respects so pronounced, that no one thought
of urging the point. The opinion expressed by Professor Busk,
after his examination of the bone was completed, was ‘ that
the bone is of considerable but not of very high antiquity,
and that it presents no character which may not be found in
cemetery bones.’! Professor Busk’s long study of, and well
known familiarity with, the subject invest his opinion on this
point with the stamp of authority. The jaw was compared
on the one hand with a specimen in the possession of Professor
Busk, derived from a Gallo-Roman cemetery, near Amiens,
from which it differed chiefly in being a little more altered,
and in the latter showing no layer of sand lining the dental
canal, when sawn across. On the other hand, it was com- ~
pared with a very remarkable lower jaw, found in a coprolite
pit near Ipswich, and now belonging to Dr. Robert Collyer,
which, although retaining a portion of its gelatine, is infil-
trated through and through with iron. The section of the
cortical layer is dark; oxide of iron is seen filling the Ha-
versian canals; a dark crust of the same metal cover the
walls of the cancelli; coarse grains of sand, with red oxide
of iron, line the walls of the dental canal; and a vertical sec-
tion of one of the fangs of a molar shows that the dentine is
partly infiltrated with iron. The precise age of the Ipswich
jaw is not known, but it is conjectured not to exceed that of
the Roman occupation of England. Mr. Busk has ascertained,
since the Conference, that part of a human pelvis and other
bones, belonging to two bodies, in the collection of M. Boucher
de Perthes, and inferred by Mr. Prestwich and Mr. Evans to
have been interred in an open trench in the gravel below
the Loess, near Mesniéres, probably during the early part of
the Celtic period, were all more or less marked with dendrites.
The Moulin-Quignon jaw, although reported to have been
yielded by a deposit of mangano-ferruginous earth, presented
the remarkable and exceptional phenomenon of being abso-
lutely free from metallic infiltration or dendritic patches.
Those who are conversant with fossil bones will not readily
believe that the age of the individual, probably above sixty
years, and the density of the bone this implied, constituted
an adequate cause of resistance to the process. That bones
are very susceptible of staining and infiltration by metallic
matter, even during the act of maceration, is well known. Nor

1 Copied verbatim from Dr. Carpenter’s ‘notes of the proceedings, taken at the
time.

OF MOULIN-QUIGNON. 617

is the action limited to iron and manganese. Professor Huxley
has described remains of Macrauchenia Boliviensis, from the
copper mine of Santa Rosa, in Bolivia, ‘in which the bones
are allin the same, and that a very peculiar, mineral con-
dition, the Haversian canals being for the most part filled
up with threads of native copper, so that the fossils are not
only exceedingly dense, but in consequence of their internal
flexible metallic support, their thinner and more delicate
parts bend, rather than break, when force is applied to them.’ !
Mr. David Forbes, in his account of the geological structure
of the region, states ‘that the bones themselves are in some
instances almost converted into copper, or at least the pores
are filled with that metal—a circumstance easily accounted
for in strata so highly impregnated with it.’ ?

The labours of the Conference in Paris terminated at this
point, and the published procés-verbaux show that the three
English members who took a share in the discussion main-
tained, upon the intrinsic evidence throughout, that the black
coating on the human jaw was not a natural deposit, and
that the recent condition of the bone was not consistent with
its having been yielded by the ‘black seam’ as an original
occupant of the bed. The layer of gray sand in the dental
canal, and the absence of penetration by the black matrix,
suggested the inference that the former was the result of a
previous sepulture, and that the latter had been laid on.

M. Milne-Edwards, in his communication to the Academy
of Sciences, already cited, puts the divergence of opinion
between the English and French members thus:

‘De l’ensemble de ces faits, MM. Falconer, Prestwich, Carpenter et
Busk conclurent qu'il y avait eu fraude au sujet de cet os aussi bien
que pour les haches de la couche inférieure du terrain de Moulin-
Quignon; que tous ces objets devaient étre considérés comme trés-
récents et que, suivant toute probabilité, les ouvriers de la carriere,
apres les avoir enduits artificiellement avec de la matiére terreuse
provenant de cette couche noire, les avaient enfouis dans une excavation
de la carriére, ot leur présence aurait été ensuite signalée a M. Boucher
de Perthes comme une découverte inattendue.

‘M. de Quatrefages et les autres membres frangais de la réunion ne
crurent pas devoir tirer les mémes conclusions des faits observés. Ils
constaterent que des caillonx ordinaires tirés de la couche noire de
Moulin-Quignon, pour servir 4 l’entretien des routes, se laissaient quel-
quefois nettoyer par le lavage non moins facilement que la machoire, et
que tous les arguments déja présentés au sujet de l'influence des
différentes conditions de gisement sur le degré d’altération des fossiles
étaient applicables a cet os aussi bien qu’a la molaire isolée.’

? Quarterly Journ, Geolog. Society, 1861, vol. xvii. p. 74.
2 Idem, p. 47.

618 HUMAN JAW AND FLINT-IMPLEMENTS

To summarize the result of the intrinsic evidence in the
view here taken.

1. The suspected haches showed the characters which are
found in modern fabrications, and they were wanting in all
those which stamp haches of unquestionable antiquity.

2. The human jaw exhibited the physical characters
which belong to ancient bones of the modern period, and it was
wanting in those commonly seen in Quaternary fossil bones
of undoubted antiquity occurring under analogous conditions.

One part of the case which was entered for inquiry upon
the programme forwarded from London before the meeting of
the Conference, was not gone into at all, viz. the anatomical
character of the jaw, as significant of race-distinction. The
issue was narrowed to fossil, or non-fossil; and on the present
occasion I shall not cumber this communication by any
remarks on the subject, although fully impressed with its
importance.

[The precise measurements of the jaw were as follows: ?

Dimensions of Fossil Lower Jaw (Moulin-Quignon) Inches
1. Extreme length of ramus from chin to pai edge of ee Vk
eallipers. 5:0
2. Length of horizontal ramus from anterior marg. coronoid to incisive
border . : . : 5 Fa
3. Height of ascending ramus to apex of coronoid a 2:2
4, Height of sigmoid notch measured parallel to posterior border ‘along
middle of ascending ramus . - : - - : Re kt)
5. Width of ascending ramus. at constriction - 1-4
6. Antero-post. extent of gpd notch ps ee eee of ceondyle to ditto
coronoid) . : 1°3
7. Height of sigmoid notch : : : ; : : : » 0:53
8. Height of jaw at incisive border . : motes
9. Ditto at middle exact where antepenult. alveol, filled up 1:2
10. Ditto behind last molar alveolus : : : : : Bie ISH
11. Transyerse diameter of condyle. 5 é : ; : ome OCS
12. Antero-posterior of condyle . . 0-4 |

B. Extrinsic EHvidence.—The canton characters of the
Moulin-Quignon haches of the ‘black seam,’ insisted upon
by the English members of the Conference, were contested
seriatim, and more especially the films, by the French mem-
bers. M. Desnoyers, at the second sitting, in view of this
difference of opinion, and of the admitted difficulty in certain
cases of distinguishing the genuine from the counterfeit, urged
that the material proof was, above all, a question of gisement ;
and that if the position of a hache in the deposit were proved
to be unquestionable, the hache ought to be accepted as
genuine, whatever might be the characters which it presented.
Others of the French savants advocated the same view. One
of the English members replied that he fully admitted the
importance of the evidence yielded by the gisement ; but that

1 Extracted from Author’s Note-hbook—[Ep.]

OF MOULIN-QUIGNON. 619

it was liable to so many sources of fallacy and errors of
judgment on the observation, that it must be received with
great caution; and that if the French members of the Con-
ference refused to entertain as evidence the itrinmsic cha-
racters on which he relied so much, and rested entirely upon
the circumstantial, he should not feel warranted in advancing
further into the inquiry, and must withdraw from the Con-
ference. In the end, however, the gisement determined the
conclusions arrived at.

The Conference, augmented by a considerable number of
French savants, proceeded to Abbeville on May 12, and
commenced operations on the gravel-pit of Moulin-Quignon.
The face of the section had been worked back about 10 or 12
feet from the spot where the jaw was asserted to have been
found, and the debris was piled up in mass upon the spot.
The ‘black seam’ proper, consisting of a thin layer of man-
gano-ferruginous oxides, proved to have been a very local
deposit or pocket, which had been worked out. The face of
the gravel section showed abundant signs, either of disturb-
ance or of derangement, natural or artificial, and if the latter,
not of a very recent date. Three chasms, interpreted to be
‘pusards’ or ‘sand-pipes,’ separated by no great interval, were
seen descending from the surface to the chalk; near the
middle of the section masses of sand presented the appearance
of having been let down into the gravel, and the ferruginous
gravel at the south end of the section exhibited no distin-
guishable marks of stratification. The section of one of the
so-called ‘puisards’ pointed in a direction which, if con-
tinued outwards, would not have been more distant than a
few feet from the spot where the jaw was asserted to have
been found. M. Hébert, the eminent professor of geology to
the Sorbonne, after examining the gravel-pit, insisted ver
strongly to the French geologists around him that the
Moulin-Quignon deposit, if not remanié, did not belong at all
to the series of Diluviwm, or ‘ Ancient Quaternary’ sands and
eravels of the Valley of the Somme, such as those of Menche-
court and St. Acheul; but that it was a formation of a
much later date. Thin seams of ‘ gray sand’ were observed
a few feet above the chalk, in that part of the section which
corresponded with the direction of the spot where the jaw
was found. It was held by some of the members of the Con-
ference that the sand might account for the ‘lining’ of the
dental canal; that the jaw might have been first in this’
alluvial sand before it got into the ‘black-seam.’? There was
no means at the time of comparing the two sands under the
microscope, and taking all the circumstances of the case into
account, the conjecture did not appear to me to furnish a

620 HUMAN JAW AND FLINT-IMPLEMENTS
satisfactory explanation. A large party of workmen were
employed, under the inspection of the Conference and of
other French savants, in search of flint haches, five of which
were yielded during the operations of the day; and in no
case was any circumstance observed or noted which would
justify the opinion that they had been fraudulently introduced.
I was a witness to the appearance of two. One of them was
disclosed by a fall of a part of the section undermined by the
pick-axe, and on being summoned to the spot, I saw it, not
engaged in the gravel, but supported on the hand of M. Al-
phonse-Edwards, before it fell, at the spot where it was
observed. On moving his hand it dropped without leaving
any impression. The other was brought up to M. Bert from
near the edge of one of the so-called ‘ puisards,’ when I was
standing above, close by him; but no mark of its impression
could be seen in the incoherent gravel. Four of these haches
were afterwards handed over to Mr. Prestwich for detailed
examination, one of them having been retained in the con-
dition in which it was found, by the president, M. Milne-
Edwards. The four which were brought to England have
been washed and carefully examined by Mr. Prestwich
and myself. They maintain the same character of contra-
dictory and incompatible evidence, which has been evinced
throughout in this extraordinary case. One of them bears
the most pronounced marks of antiquity, in form, rolling,
facets, weathering, glimmer, and uniform staining; while
the other three exhibit every positive and negative character,
which we have held to throughout, as stamping haches of
modern fabrication! There was no intermediate condition.
I append a descriptive note of each.!

These haches were examined, in a cursory manner, by the

1 Notes on the specimens of flint-
implements found at Moulin-Quignon,
in presence of the Commission, May 12,
1863, dictated by H. F. to Mr. Prest-
wich :—

No. 1. Labelled ‘ Found in situ’ (A.
Edwards).—Form and contour narrow,
ovato-lanceolate, sharp all round the
edge and at the base ; thick ; facets very
irregular and deep, like those regarded by
me and Mr. Evans as unauthentic; ridges
separating the facets high and angular ;
surface where washed, like that of arecent
tiint ; nodendrites, no patina, no incrus-
tation, and no glimmer of rolling; thin
films, of considerable extent, without any
fracture of margin—of the unauthentic
character.

No. 2. Washed on one side ; covered
with ochreous sandy matrix on the

other; in form ovate; point broken off,
but fracture rounded ; body thin; facets
shallow, like the true St. Acheul type;
discoloured surface with a bright glim-
mer; margin all round more or less ob-
tuse—of the authentic type.

No. 3. Ovate in outline like the last,
but point entire. In all other respects,
as regards facets, fracture, &c., like No.
1.—of the unauthentic type.

No. 4. Pointed ovate, but in surface,
fracture, &c., fresh-looking, like No. 3;
tacets excessively rude ; ridges high, and
upon the thick part, on one side, a rude
irregular pit, as if caused by a vertical
blow. Some of the films very large, ex-
cessively thin, quite entire, and not the
least mark of matrix below—of the sus-
pected or unauthentie type.

OF MOULIN-QUIGNON. 621

Conference, late at night ; yet it will be seen, that the fact of
finding them to all appearance i sitw in an undisturbed
position determined the verdict given in conclusion No. 3,
that ‘the greater part, if not the whole, of the haches from
the lower part of the gravel-pit of Moulin-Quignon are
authentic.’

The evidence of the eye-witnesses to the finding of the jaw
was then taken, and it was so consistent and direct that no
flaw was detected, and it satisfied the Conference.

The operations on the gravel-pit of Moulin-Quignon were
hurried through in a single day, and although the results
were relied on, it is but right to indicate that with such a
number of workmen, dispersed over different parts of the
section, and with so many observers, the conditions were not
the most favourable for the close and severe scrutiny required
in a case of critical observation.

The Conference had now reached the fifth day of its
labours, and it was necessary to come to some conclusion.
The English members were compelled, by business engage-
ments, to return to London. Assuming the Moulin-Quignon
deposit to be of the age assigned to it by Mr. Prestwich, 7.e.
a ‘high-level’ gravel of the oldest Quaternary period in the
Valley of the Somme, the intrinsic evidence yielded both by
the flint haches and by the jaw was wholly irreconcileable
with the direct results arrived at under the eyes of the Con-
ference, and with the testimony of witnesses as to the find-
ing of the jaw; but if it was a disturbed or comparatively .
late deposit, although still embarrassed by formidable difii-
culties, there might be a way to a possible explanation. To
have persisted in attributing fraud in the case, without direct
evidence, whatever suspicions might be entertained, would
have been unwarrantable. The following conclusions were
arrived at: the French members and Mr. Prestwich being
unanimous in their assent to all :—

‘1. La machoire en question n’a pas été introduite frauduleusement
dans la carriere du Moulin-Quignon ; elle existait préalablement dans
Yendroit ou M. Boucher de Perthes l’a trouvée le 28iéme mars dernier.
Cette conclusion a été adoptée a lunanimité.

‘2. Tout tend a faire penser que le dépdt de cette michoire a été
contemporain de celui des cailloux et autres matériaux qui constituent
lamas argilo-graveleux, désigné sous le nom de “ Couche Noire,” laquelle
repose immédiatement sur la craie. Cette conclusion a été adoptée par
tous les membres présents, 4 l'exception de MM. Falconer et Busk, qui
réservent leur opinion jusqu’a plus ample informé.

‘3. Les silex taillés, en forme de haches, qui ont été présentés a la
réunion comme ayant été trouvés vers la méme époque dans les parties
inférieures de la carriére du Moulin-Quignon, sont pour la plupart, si
non tous, bien authentiques.

622 HUMAN JAW AND FLINT-IMPLEMENTS

‘ Cette 3ieme conclusion a été adoptée par toutes les personnes pré-
sentes sauf par M. Falconer, qui réserve son opinion jusqu’a plus ample
informé.

‘4, Il n’y a aucune raison suffisante pour révoquer en doute la con-
temporanéité du dépét des silex taillés avec celui de la machoire
trouvée dans la “ Couche Noire.”

‘Cette proposition est adoptée par tous les membres de la réunion
sauf par MM. Falconer et Busk, qui désirent réserver leur opinion.’

My verdict on the whole case, which is embodied in the
proces-verbaux, was as follows :-—

‘I am of opinion that the finding of the human jaw at Moulin-
Quignon is authentic; but that the characters which it presents, taken
in connection with the conditions under which it lay, are not consistent
with the said jaw being of any very great antiquity.’

Professor Busk expressed his in the following terms :—

‘Mr. Busk desires to add, that although he is of opinion, judging
from the external condition of the jaw, and from other considerations of
a more circumstantial nature, that there is no longer reason to doubt
that the jaw was found in the situation and under the conditions re-
ported by M. Boucher de Perthes, nevertheless it appears to him that
the internal condition of the bone is wholly irreconcileable with an
antiquity equal to that assigned to the deposits in which it was found.’

Throughout he questioned the antiquity of the jaw.

The first clause of my verdict, ‘that the finding of the
human jaw at Moulin-Quignon is authentic,’ was intended to
absolve the workmen from the imputation of having fraudu-
lently introduced the bone into the bed, when no direct proof
could be adduced to support it; while by the second clause
it was meant to express my opinion that the bone was not of
fossil antiquity, z.e. not reaching further back than some
date in the modern period. I think it necessary to indicate
this clearly, as a wider meaning appears to have been
attached to my use of the term ‘authenticity of finding,’
in the communications which were made to the ‘ Academy
of Sciences’ by M. Milne-Edwards and by M. Quatrefages,
on the 18th May, than I intended the words to convey. They
did not include an admission of the ‘ authenticity of the jaw’
as a true fossil bone. The ‘ procés-verbaux’ show that from
first to last I entertained ans adverse opinion on this head.

The necessarily hurried proceedings and unexpected re-
sults at the close of the Conference were not favourable to a
well considered and deliberate verdict, under the perplexing
contradictions of the case. I now believe that I committed
an error of judgment in not reserving my opinion on all the
moot points, instead of reserving it upon three only. I must
bear the blame which this admission carries with it, consider-
ing how strong my negative convictions, founded upon the

OF MOULIN-QUIGNON. 623

intrinsic evidence, were. I have since carefully reviewed the
facts and opinions set forth in the ‘ procés-verbaua,’ and sub-
mitted to the closest examination the numerous suspected
flint haches, yielded by the ‘black seam’ and ferruginous
gravel of Moulin-Quignon immediately before the Confer-
ence and while it sat there; and the result is an irresistible
_impression, that there is some mysterious complication in
the case which remains to be solved. M. Elie de Beaumont,
on the occasion when the notes on the conclusions arrived at
by the Conference were communicated, on the 18th May, to
the ‘Academy of Sciences,’ by M. Milne-Edwards and M.
de Quatrefages, insisted that there was a fundamental error
in the view taken of the age of the Moulin-Quignon beds.
He denied that they belonged to the Terrain Erratique, or
Diluwiwm properly so called, but regarded them as ‘dépéts
meubles sur des pentes,’ or superficial modern deposits, of the
age of the peat-beds, in which he would not be surprised if
human bones and industrial objects were found (*‘ Comptes
Rendus,’ 18 Mai, p. 956). At the next following meeting of
the Academy, held on the 25th ultimo, the same distin-
guished geologist is stated to have maintained that these
beds were not of an age exceeding that of the Lacustrine
habitations of the Swiss Lakes, that they might even corres-
pond with certain peat-beds, below which the vestiges of old
Roman roads had been met with; and that he would not be
astonished if the Moulin-Quignon deposits were found to
contain ‘ Gallo-Roman coins.’

This opinion, if it could be sustained, would appear at
first sight to furnish a ready explanation of the contradictory
complications of the case. But in reality it does not, for the
question is reduced by the evidence to a much narrower
issue. I have already dwelt upon the absolutely modern
appearance of the suspected flint haches, so numerous and so
uniform in their character, their rude form, iron-struck
facets, high ridges, sharp unworn edges, and fresh unstained
surfaces. Had they been yielded by a bed of peat it is con-
ceivable that these signs of freshness might have been re-
tained for an indefinite time; but, with our present lights,
is it conceivable that they should have remained for fifteen
or eighteen centuries in a highly-ferruginous bed of gravel,
freely permeable to water, where every large flint is stained,
as a natural result, with iron, and that they should have
invariably escaped, even where their substance is whitish and
highly porous? Or could they have remained in the ‘ black
seam’ uniformly without staining or incrustation, while the
flints around them are covered with dendrites? Again, as
regards the human jaw. I have stated in the preceding

624 HUMAN JAW AND FLINT-IMPLEMENTS

pages that it presented every physical character of preserva-
tion of bones buried within the historical period, and that it
wanted all those which are commonly seen in fossil bones.
Is it conceivable that this bone could have lain embedded in
a mineral layer of manganese and iron, permeable to water,
for fifteen or eighteen centuries, and that it should have con-
tinued but little altered and so absolutely free from impreg-
nation with these metals that the covering layer of black
matrix could be washed off with as much facility as if it had
been smeared on during the present year, and that not a
single dendrite should have been visible on the washed sur-
face? I have studied fossil bones during upwards of thirty
years under various conditions, within the tropics, in the ~
Pliocene and Quaternary beds of Europe, and in the caves;
such a case has never come within my knowledge or under
my observation. The mute and emphatic eloquence of the
persistent intrinsic evidence of the Moulin-Quignon jaw and
flints impresses me with a force that far countervails the
results of the fallible observation made upon the beds. In
this view it seems to me that the real issue, as regards the
jaw and the haches, is not the geological age of the beds, but
when and by what means, within a comparatively recent
period, these objects got introduced where they were found ?
That many of the haches are modern fabrications is as cer-
tain as a question of the kind can be, without the testimony
of eye-witnesses to the act of making them. Several which
I now possess from the ‘black seam’ of Moulin-Quignon
and Mautort are of this character. That the suspicions
which were expressed in England about the bone were war-
rantable is sufficiently proved by the fact, that M. Boucher
de Perthes, with characteristic candour, has put on record
the fact that some workmen of Abbeville, before 1857, had
attempted, and for some short time with success, to pass off
the skull of a recent horse as a fossil specimen from Menche-
court. They had first embedded it in a paste of chalk and
water, and then covered it with the sand of Menchecourt.!
Although the workmen were absolved, by the decision of the
Conference in the present instance, the fact ought to be borne
in mind, in weighing the different aspects of this inscrutable
case. The trick which was committed once might be repeated
again, and those who pass off fictitious haches as genuine,
would not be scrupulous about a bone.?

‘<«Antiquités Celtiques, &c., tom. ii. / Quignon and Abbeville, and on the pe-
p. 486. culiar character of some of the Flint-

? The ‘ Quarterly Journal of the Geo- | implements recently discovered there,’
logical Society, for November, 1863, | which concludes with the following para-

contained an essay by Mr. Prestwich, | graph :—
entitled: ‘On the Section at Moulin- Note.—Further and more deliberate

OF MOULIN-QUIGNON.

inquiry, on the part of myself and others, |

than was possible on the occasion of the
Conference at Abbeville, leads me now to
revert to my original opinion, and to be-
lieve that we were mistaken in conclud-
ing on that occasion that no fraud had
been practised. In addition to the ob-
jections originally urged, I found, on
washing a portion of the gravel contain-
ing the flint-implements (an experi-
ment contemplated, but unaccountably
omitted in May last), the discordance
between the mineral condition of the
flint fragments and the flint imple-
ments to be so great, as to render it evi-
dent that the two could not possibly
have been subjected during the same
time to the same influences. Further,
instead of being confined to a special
bed and a special level, we found, on a
subsequent visit, that specimens had
been brought by the men from Eparg-
nette (a bed on a yet higher level
and hitherto unproductive); and again,
we were given, at Mautort, at a low-
level valley-gravel-pit, three flint-imple-
ments of precisely the same type and in

VOL. II,

625

the same condition as those of Moulin-
Quignon; whilst, from the indications
given by the men, the specimens would
have been taken from a bed of gravel
subordinate to the Loess, and not even
part of the mass of fluviatile low-level
gravel. Our verdict in this case re-
specting the flint-implements (leaving
apart the question of the jaw) will,
therefore, I fear, have to be reconsidered.
The precautions we took seemed to ren-
der imposition on the part of the work-
men impossible; still, although it re-
mains undetected, I cannot, with the
strong and increased doubts (not one of
them since removed) attached to the
point, continue to accept the authenticity
of these specimens. The essential fact,
however, of the occurrence of genuine
flint-implements at Moulin-Quignon,
the Champ de Mars, and Menchecourt,
receives additional confirmation from
every fresh investigation, and places M.
Boucher de Perthes’s important origi-
nal discovery beyond all doubt.—J. P.,
Oct. 1863.—[ Ep. ]

SS

626 WORKS OF ART BY

XXVI. WORKS OF ART BY PRIMEVAL MAN IN
EUROPEHE.!

Srycez the exploration of the Brixham Cave in 1858, an im-
mense impulse has been given all over Europe to the search
for, and study of, the material proofs of the antiquity of the
human race. The public mind is now craving for informa-
tion on a subject which a few years back was condemned by
the general verdict of men of science, and hardly mentioned
except in a whisper. Fresh evidence is being brought to
light, day after day, of the most interesting and important
character, although not tending to carry man back, in every
particular instance, to a period of very high geological anti-
quity. The south of Europe is the quarter whence the current
is now flowing, and the ossiferous caves the springs whence
itissues. Professor Busk,in a recent communication (‘ Reader,’
January 30), has given a very clear and excellent account
of discoveries made within the last year in a bone-cave in
Gibraltar. The materials, not all yet arrived in England,
are now under investigation, and give promise of results of
high import. But the most interesting additions have been
yielded, very lately, by caves in central France, where what
may be called works of art, of primitive execution, have
turned up in considerable abundance, which prove that
savage man, of the unground and unpolished Stone period,
was able, in advance of the use of metals, to sculpture on
deer’s horns, and to grave on stone, figures of quadrupeds his
contemporaries that are now extinct in that region. My
friend M. Lartet, on behalf of himself and Mr. Henry Christy,
his collaborateur in the work, communicated to the Academy -
of Sciences on the 29th ultimo an account of these relics,
which, when exhibited, produced an unusual sensation among
the learned Academicians. I purpose now giving a brief
sketch of this new and certainly very ancient walk of art,
drawn mainly from M. Lartet’s paper, which will speedily
appear in the ‘Comptes Rendus’and from figures of the objects.
The proofs of the remote antiquity of man are derived
from two sources: —1, the ancient, or ‘quaternary,’ river

1 This communication appeared as a letter in the ‘Times,’ on March 25, 1864.
—T[Eb.]

PRIMEVAL MAN. 627

eravel-deposits ; 2, the ossiferous caves. The former, handled
with the severe caution of Mr. Prestwich, carries man furthest
back in time, and with the greatest certainty; but it is of
the most meagre and restricted character, consisting merely
of flint-weapons or implements, hardly ranging beyond a few
patterns. Not a single instance has yet occurred of a frag-
ment even of an unquestionably authentic human bone
having turned up in these deposits. On the other hand, the
evidence yielded by the caves, although less certain as an
index of remote time, is infinitely more varied and instruc-
tive. It tells us, in certain cases, the division of the human
race to which man, the early tenant of the caves, probably
belonged; what was his stature and what his physical
powers ; what the animals which were his contemporaries ;
what the molluscs, fish, flesh, and fowl upon which he fed ;
that he cooked his meat by fire ; that he extracted the marrow
from the bones, and how he did it; how and with what
weapons he killed his game; how he flayed and dressed the
hides; that he scraped the meat off the bones; that he
carefully cut the sinews of his slaughtered deer for harpoon-
lines, or for the fibre of sewing thread for his fine-pointed
pierced needles; where and in what direction he cut the
sinews; what the implements and weapous—in stone, bone,
and deer’s horn—which he used; what his ornaments; and
how he disposed of his dead. It is now beginning to en-
lighten us on what he was capable of achieving in the way of
art, and that in music he had got the initial length of a bone
whistle limited to a single note. The cave evidence has been
disparaged by cursory observers and light reasoners, upon
the grounds that the caves have been occupied at different
times, and their contents often disturbed by the latest tenants,
thus forming what are called remamé deposits. But the short-
comings lay with the objectors themselves. When the
profound paleontological knowledge, rare sagacity, and phi-
losophic caution of M. Lartet are applied to what were
sources of doubt and embarrassment to them, the supposed
difficulties are converted into aids in unravelling the tangled
clue, and into indices of ulterior truths. In short, beside the
bare fact that primeval man existed during the early ‘ Fluvi-
atile Drift period in Europe,’ all that we know of him—
exclusive of the later ‘kitchen-middens,’ and ‘ pile-habita-
tions °—is derived solely and entirely from the ossiferous
caves.

The caverns which, on this occasion, were the objects of
exploration by M. Lartet and Mr. Henry Christy, occur in
the department of the Dordogne (the ancient province of

Ss 2

628 WORKS OF ART BY

Périgord), and in the arrondissement of Sarlat, in the south-
western part of Central France. The most productive locali-
ties were the cave of ‘ Les Eyzies,’ in the commune of Tayac,
the cave of ‘ Le Moustier,’ and the shelter-recesses under the
projecting cliffs of ‘Laugerie-Haute,’ ‘ Laugerie-Basse,’ and
‘La Madeleine,’ in the Valley of the Vezére ; the rock-forma-
tion consisting of indurated chalk. The floor of ‘ Les Hyzies ’
cavern is overlaid by a continuous sheet of breccia, composed
of a base of cinders and ashes, mingled with charcoal; frag-
ments of bones either in the natural state, or split, scorched,
or burnt; outside pebbles; flint-cores with numerous frag-
ments of flint-flakes or knives, invariably of wrought forms,
and associated with other implements or weapons fabricated
out of reindeer’s horns ; the whole consolidated in a confused
mass, which had never been disturbed since the period of
deposition. This was established by the state of the materials
and by the fact that in several cases long bones were found
with their heads in articular continuity, and vertebre of
reindeer in sequence.

The principal objects of art were as follows:—In ‘ Les
Hyzies,’ among numerous fragments of a hard slate, foreign
to the district, two plates were found, each bearing an en-
graved representation of a quadruped. One of them, mutilated
by an ancient fracture, presents the fore-quarter of an herbi-
vorous (?) animal, the head of which was apparently invested
with horns, so far as the faint lines of the engraving at this
part admit of judging. The other bears the figure of a head,
with the nostrils sharply defined, and the mouth half opened ;
but the profile lines of the frontal region are interrupted in
consequence of erasure by subsequent friction. On one side
and a little in front is engraved the figure of the palm of a
large horn, inferred by MM. Lartet and Milne-Edwards, with
reserve, to be that of a Moose Deer. These specimens are
regarded by M. Lartet as being the earliest known examples
of engraving on stone, by primeval man, of the Reindeer
period in Europe.

The most striking part of the collection, consisting of
sculptured objects, was discovered in the shelter recesses,
under the cliffs of La Madeleine, Laugerie-Haute, and Lau-
gerie-Basse, amidst accumulations of bone-refuse and other
rejectamenta, mingled with an immense quantity of flint-flakes
and the cores from which they had been struck off. These
spots were evidently the kitchens and manufactories of the
ancient savage. The bones indicated the animals on which
he fed; being the horse, ox, ibex, chamois, reindeer, birds,
fish, &c, The common stag was rare, as were also the boar
and the hare. Some detached molar teeth were discovered

PRIMEVAL MAN. 629

of the extinct Irish elk, and also detached plates of the
molar teeth of the Mammoth.

Laugerie-Haute would seem to have been especially the
locality where fiint-implements were made, and Laugerie-
Basse that where reindeer horns were converted into spear-
heads, harpoons, daggers, arrow-heads, needles, and other
implements. Here an enormous accumulation of reindeer
horns were discovered, nearly the whole of which bore the
marks of a stone-saw, by which pieces had been detached
suitable for conversion. Here also were found the principal
sculptured objects, some of which, considering the period and
the nature of the tools, are marvels both of artistic design
and of execution.

The mest remarkable is a long dagger or short thrust-
sword, formed out of a single horn. The handle represents
the body of a reindeer, the parts in fair proportion, and
treated with singular skill and art-feeling, in subservience to
the use for which it was intended. The forelegs are folded
easily under the body; the hindlegs drawn out insensibly
into the blade; the salient horns and ears are cleverly applied
to the chest by giving an upward bend to the head; and a
convenient hollow for the grip of the hand is produced by a
continuous curve extending from the rump to the muzzle.
M. Lartet remarks that the hand for which it was designed
must have been much smaller than that of the existing
European races. The weapon was evidently left by the
artist-savage unfinished ; but, as a design imbued with taste,
it will bear a very favourable comparison with Oriental
dagger-handles cut in ivory.

Another specimen is described as a handle terminating at
one end in a spear-point, and bearing in partial relief the
heads of a horse and of a deer, probably reindeer. Others
are ornamented with longitudinal and parallel wavy lines,
&c. A distinct class consists of palmated portions of reindeer
horns, bearing representations of animal forms—some exe-
cuted in graved lines, others in bas-relief or in high relief.
One of these palmations exhibits a figure of a large herbi-
yvorous animal which has been conjecturally referred to the
Aurochs. Another is supposed to represent an ox, probably
Bos primigenius (2). The collection, judging by the drawings
which I have seen, is very rich in spear-heads, barbed
harpoons, arrow-heads, and finely pomted slender needles,
drilled with an eye-hole. The harpoons bear a close resem-
blance to the Esquimaux patterns. On one object the figure
of a scaly fish is distinctly represented. The ornaments
consist of canines of wolf, incisors of ox and other animals,
with ear-bones of horse or ox, all drilled for suspension. One

630 WORKS OF ART BY

curious object is the first digital phalanx of a ruminant,
drilled to a certain depth by a smooth cylindrical bore on its
lower surface near the expanded upper articulation. This is
supposed to have been a whistle or call, and a shrill sound is
yielded on applying it to the lower lip and blowing into it.
Three of these whistle phalanges are of reindeer, one of
chamois. One relic of surpassing interest consists of the
lumbar vertebra of a reindeer, pierced through and through ~
by a flint weapon, which still remains embedded in the bone,
fixed by calcareous incrustation. This is an object of great
significance and extreme rarity. Human bones, although
found, were very scarce; but M. Lartet has refrained from
alluding to them, with a reserve the reason of which is in-
dicated by M. Milne-Edwards. In forming an estimate of the
value of the relics of art, the reader will bear in mind that
they are the productions of the unpolished and unground
‘Stone-period,’ the tools employed having been thin chips
aud delicate flakes of flint. Such at least is the fair inference
drawn with our present lights from the negative evidence,
not a trace of metal in any shape having been met with in
the Dordogne caves. But if primeval man really had made
such progress in the conceptions of art, without having yet
attained the knowledge of metals, it will be as curious an
anthropological phenomenon as are the art objects them-
selves, which express that degree of luxury which ease,
leisure, and comfort beget. Reindeer’s horn is notoriously
the most worthless and incompact of cervine antlers; it is
readily whittled by a knife, which is not the case with stag’s
horns.

The labours of M. Lartet and Mr. Henry Christy on the
Dordogne caves commenced in August 1863. They have
been continued ever since, and are still in progress. Valuable
and instructive as is the Dordogne collection, it is surpassed
in certain respects by another, from the ‘ Bruniquel Cave,’ in
the south of France, more recently formed by other observers.
The Bruniquel series, it would appear, does not embrace the
same range of art, but it is richer in the department of
weapons and implements, such as harpoons, spear-heads, &c.,
which are larger, more numerous, better finished, and in
better preservation. These precious materials were offered
in succession to the French Government and to the British
Museum. ‘ Perfide Albion’ has got them: they are now in ~
the national collection. The result does infinite credit to the
zeal, enterprise, and activity of the administration of the
British Museum. But the satisfaction which so valuable an
acquisition necessarily excites is not wholly unmixed. The
investigation of truth is above and beside national predilec-

PRIMEVAL MAN. 631

tion. The ‘Bruniquel Cave’ series is now divorced from
the collections in France, of which it forms a complement,
and upon which M. Lartet has been engaged since 1861,
when he published his important researches on the Sepulture
Cave of Aurignac. Those who take an interest in the ad-
vancement of our knowledge of the subject would have con-
eratulated themselves if the Bruniquel materials had been
placed in his practised hands, to be included in the work
which he and Mr. Henry Christy are about to publish on the
remains of man of the ‘ Cave’ period in France. |

One circumstance in the case deserves to be generally
known. The instinct of a collector is to amass, hoard, and
retain. Mr. Henry Christy is the possessor of one of the
choicest private archeological collections in Hurope. M.
Lartet and he explored the Dordogne caverns on a large
scale, with the object—first of exhausting the ground, and
next of distributing duplicates. They have presented huge
slabs of the floor-matrix, containing embedded every variety
of object, to all the principal museums in Hurope, and
selected sets to persons of all countries having a recognized
position as labourers in the same field; and this, too, before
their own researches were published. In their case a higher
impulse extinguished the mere collector’s mstinct. No com-
ment is required.

P.S. At the meeting of the Academy of Sciences (Feb. 29)
a note on the same and cognate subjects was communicated
by the Marquis de Vibraye, who has laboured so merito-
riously on the ossiferous caves of France.

632 ANCIENT FLUVIATILE DEPOSITS

XXVII. ON THE ASSERTED OCCURRENCE OF HU-
MAN BONES IN THE ANCIENT FLU-
VIATILE DEPOSITS OF THE NILE
AND GANGES; WITH COMPARATIVE
REMARKS ON THE ALLUVIAL FORMA-
TION OF THE TWO VALLEYS!

I. FLUVIATILE DEPOSITS OF THE NILE :——1. GENERAL REMARKS—2. FOSSIL
HIPPOPOTAMUS—=3. ASSERTED DISCOVERY OF HUMAN BONES—4. ANALOGY
OF THE FLUVIATILE DEPOSITS OF THE NILE WITH THOSE OF THE GANGES
—II. FLUVIATILE DEPOSITS OF THE GANGES :—1. PHYSICAL FEATURES
OF THE VALLEY OF THE GANGES—2. MAMMALIAN FOSSILS—3. FOSSIL
MOLLUSCA—4. GENERAL INFERENCES—III. ANTIQUITY OF MAN IN INDIA:
CONCLUSION.

§ I. Frouviatiie Deposits or THE NILE.

1. General Remarks.—The object of this communication is
to bring together the few instances on record of the occur-
rence of Mammalian fossil remains in the Valley of the Nile,
and to institute a comparison between the Nilotic alluvial
deposits and those of the upper part of the Valley of the
Ganges which have come under my own observation. Fossil
human bones have, according to certain statements, been met
with in both of these sub-tropical valleys; and it may be
useful, at the present time, to consider to what general infe-
rences the cases lead, as a guide to future observation.

The explorations conducted by the French authorities in
Algeria have brought to light numerous remains of Hippo-
potamus and of other Mammalia, extinct or living, from the
later deposits of that part of Africa; but it is not a little
singular that the Valley of the Nile has heretofore been so
unproductive, considering the stream of intelligent travellers
that flows up the river every season from Alexandria to the
Cataracts, and the not insignificant number of accomplished
explorers, German, French, and English, who have traversed

1 The author was engaged in writing ; Journal of the Geological Society’ for
this paper immediately before his fatal | November 1865, from which it is now
iliness. What was completed was pub- | reprinted.—[Ep.]
lished after his death in the ‘ Quarterly

OF THE NILE AND GANGES. 633

the country, as high as the confluence of the ‘Blue’ and
‘White’ rivers, and latterly above it. The alluvial deposits
along the banks appear in many places to be developed in
great force; and the lowermost present characters which
would refer their origin back to a high antiquity. But
although fossil wood and shells of land- and fresh-water Mol-
lusca have been very generally met with in these deposits,
Mammalian remains have been but very rarely observed, and
the instances on record only cursorily described. These will
be referred to in the sequel. Any case which is calculated
to direct attention to this neglected walk of observation
deserves to be noticed.

2. Fossil Hippopotamus.—The specimen sent to the Geolo-
gical Society by Dr. Leith Adams consists of a fragment of
the left maxiHary, containing in situ the two last upper true
molars of a very large Hippopotamus. Of these the penulti-
mate is far advanced in wear, the crown-divisions having
been ground down to the common nucleus of ivory, leaving
only two small islets of enamel upon the depressed dise. The
last molar is but partially worn, the two pairs of trefoil com-
prising the crown-surface being distinct both in the longi-
tudinal and transverse directions.

The molar teeth present the ordinary characters of the
existing Hippopotamus of the upper part of the Valley of the
Nile and Senegal, but in size they equal those of the great
extinct form of Europe, Hippopotamus major of Cuvier. TI
have compared them with the corresponding molars of the
largest specimen of the living species that has come under
my observation, being the huge male skull No. 3,405 of the
Hunterian collection, in the Museum of the Royal College of
Surgeons, and with a set of specimens of H. major from
the Val d’Arno, presented by Mr. Pentland to the British
Museum.

The following are the comparative dimensions :—

wm
Bos HS . os . (>) >
224 | 228 |884|/ see) £8 | 88a
0 eer ev lan rq) -
2am [rigd [nse [ri<d | G6 | nies
Joint length of the two last true | in. in. in in. in in.
molars. : ox) |) a 4°3 4:3 43 4°65
Length of penultimate : .| 195 | 2-0 2°3 2:2 2°25 | 2:5
Width of penultimate . ¢ 5 20) 2-2 21 2°3 2°2 2-2
Length of last molar . |) 2PLL 2-1 21 2°2 2:2 2-2
Width of last molar . ¢ a || ZPIl PASS) || 2°3 215 | 2:25

From these figures it will be seen that the molars in the
Kalabshee specimen are as large as the majority of those of

634 ANCIENT FLUVIATILE DEPOSITS —

the extinct form with which it was compared, and that of
the latter there is only one, No. 28,785, in which the dimen-
sions are greater, while the Kaldbshee specimen slightly
exceeds the proportions yielded by the existing Hippopo-
tamus. It is at the same time to be remarked that the latter
is as large as No. 28,790 of H. major; and it has still to be
shown that the bones of the skeleton of the latter form sur-
pass those of the living species more than do the fossil bones
of Bison priscus those of the existing Aurochs, which is gene-
rally regarded as being of the same species.

In H. major, the basal cingulum of the molar is commonly
more salient and crenately lobed than in those of the living
species. The Kalabshee specimen in this respect agrees with
the latter form; but it is at the same time to be observed that
the cingulum is not well preserved in the Nile fragment.

The evidence yielded in the present case is too limited to
warrant any well-founded opinion regarding the species;
but, notwithstanding the large dimensions of the molars, I
have failed to detect any diagnostic characters which would
justify the separation of the Kalabshee specimen from the
Senegal variety of the hving H. amphibius. It is inferred to
have been yielded by a large and old male. In mineral con-
dition it appears to be as well fossilized as specimens of H.
major from the Val d’Arno and Auvergne. The cancelli of
the bone are filled throughout with matrix resembling Nile
mud; the ivory of the molars has lost a large portion of its
gelatine. Professor Busk, who has analyzed the specimen,
found that the earthy salts of the bone yielded a very large
proportion of carbonates ; but he failed to detect more than
a very faint trace of fluorine, so commonly met with in bones
of great antiquity.’ A calcareous crust covers the enamel of
the teeth. Dr. Leith Adams does not indicate, in his paper,
the precise stratum near Kalabshee out of which the frag-
ment was exhumed, this being a point of much importance
in the case.

Geologists may be reminded that, although rarely observed,
this is not the first instance in which fossil or sub-fossil re-
mains of Hippopotamus have been procured from the Valley
of the Nile. Dr. Rippell brought to Europe, in 1827, the
remains of a species of Hippopotamus from above the Cata-
racts. They were deposited in the Senckenberg Museum at
Frankfort, and indicated a species in size between H. am-
phibius and the existing small H. Liberiensis of St. Paul’s

1 The following gives Professor Busk’s Phosphates, &e. . : . 85:0
analysis :— Iron . ; : . a trace.
Organic matter. c neice Fluorine . scarcely an indication,

Earthy carbonates . . o7d

OF THE NILE AND GANGES. €35

River, in Western Africa. I had an opportunity of examining
them in 1849; and in the synopsis which I contributed to
Dr. Morton’s account of the Liberian Hippopotamus, the
species was referred to, under the provisional name of H.
annectens, as intermediate between the two living species.!
It may prove to be of H. Pentlandi, the extinct species pre-
vailing in Sicily, Malta, and Candia, which was then un-
known to me.

Of H. major, the huge fossil form of the Val d’Arno and
Auvergne, abundant remains have been found in deposits,
either Pliocene or Quaternary, in Algeria. A fine series of
specimens derived from that region is exhibited in the
Museum of the Ecole des Mines in Paris.

3. Asserted Discovery of Human Bones.—The next case of
Nilotic fossil remains is of still higher interest, being the
asserted discovery of human bones in one of the conglomerate
or older beds of the Nile-valley alluvia, at a time when the
antiquity of the human race did not engage the attention of
men of science as it does at the present day. In Leonhard
and Bronn’s ‘ Jahrbuch ’ for 1838 a series of letters appeared,
in which Russegger gave some account of the results of his
explorations then in progress in Nubia and Sudan. In one
of these letters, dated Sennaar, March 23, 1838, he de-
scribes the structure of the alluvial banks of the Blue Nile
from Khartoom up to Sennaar, and thence to Roserres, and
adds that, ‘In the alluvia of the Blue Nile at Duntai we
found human bones. The structure of these bones was
perfectly preserved, but the animal matter had disappeared.
Their surface was polished and of a blackish-brown colour ;
the substance very hard, but not yet petrified’? (Jahrbuch,
1838, p. 403). In the second volume of his travels, published
in 1843, Russegger enters at greater length into the details of
the case, and states that the alluvial formation of the Blue
Nile, from Khartoom to Sennaar, consists of freshwater beds
thrown down by the river itself, and that, regarded as a whole,
they are divisible as follows, from above downwards :—

1. Ordinary fluviatile mud, the result of modern periodical
inundation, analogous in its external characters to the Nile-
mud of Egypt, and containing embedded nodules of calcareo-
argillaceous concretions (nodular kankar ?).

2. Friable, fine and coarse conglomerate, composed of

1 «Observations on a new living Spe- | schenknochen. Das Gefiige der Knochen
cies of Hippopotamus, H. Liberiensis, of | war vollstandig erhalten, der Thier-
Western Africa.’ By Dr. Morton, Phil- | materie aber zerstért. Die Aussenfliche
adelphia, 1849, p. 8. [See antea, p. 404, | war glinzend und schwarzbraun ge-
—Ep.] firbt, die Masse sehr hart, aber’ noch

2 *In den Alluvionen des blauen | nicht versteinert.’
Flusses bei Dundai fanden wir Men-

636 ANCIENT FLUVIATILE DEPOSITS

quartz-grains and pebbles, cemented by ancient mud, form-
ing a kind of sandstone-grit, and yielding caleareous and
marly concretions.

3. Ancient Nile-mud, indurated, and containing embedded
iron shot clay, siliceous limestone, &c., full of calcareous
and marly concretions in the ferruginous portions.

4, Fine and coarse quartzose conglomerate, with the
materials united by ancient Nile-mud and calcareo-argilla-
ceous cement, very hard, used as a building stone, and con-
taining embedded masses of saline clay and of ordinary clay
and marl, full of clay ironstone, ferruginous sandstone, and
of calcareous and marly concretions,

5. Freshwater limestone (travertine, or slab kankar ?) of a
dark-grey colour, hard and sonorous, occasionally having a
marly appearance, with here and there a tendency to a con-
centric and generally crystalline structure.

The beds are described as horizontal, of very variable
thickness, attaining sometimes, in Nos. 1, 2, and 3, as much
as five or six fathoms (Jahrbuch, 1838, p. 408). According
to Russegger, with the exception of the uppermost deposit,
they contain very generally fossil vegetable remains, chiefly
the wood of Mimosas (Mimosa Nilotica) and stems of Asclepia
(Calotropis) procera; the former are either converted into
lignite or have their core exhibiting a concentrically disposed
and radiating crystalline structure, derived from the embed-
ding matrix; the latter have the bark preserved, but the
spongy core occupied either by calcareous matter or con-
glomerate. These alluvia presented very commonly shells of
the Mollusca now living in the waters of the Nile, both bi-
valves and univalves, together with some land species.
Among the most common was Aitheria Caillaudi, occurring
frequently in heaps or oyster-banks, together with species of
Unio, Iridina, and Anodonta. In the alluvium of Sennaar he
found Ampullaria ovata and a species of Helix. He adds,
that Mtheria Caillaudi was also abundant in the deposits of
the White Nile.

M. Lefévre, writing at the same time, confirms the obser-
vation of Russegger about the occurrence of the conglome-
rate. ‘Near Khartoom, on the Libyan side, you meet on
the redans of the White Nile with a modern conglomerate
composed of fragments of sandstone united by a calcareous
cement, either deposited by the water of the river or filtered
through the alluvial soil. This concretionary deposit is ex-
hibited equally on the banks of the Blue Nile, and it is well
seen on either side where the banks are perpendicular.’ He
adds that the alluvial escarpment is nowhere higher than
from 50 to 56 feet.

OF THE NILE AND GANGES. 637

In his ‘ Reisen,’ Russegger describes the occurrence of the
human bones at Duntai in terms somewhat different from
those employed in his ‘ Briefe’ in the ‘Jahrbuch.’ ‘The
freshwater alluvia occurring at Karkodije and Seru extend,
with the slight modifications already indicated, up the Blue
River as far as Roserres, forming a hillocky alluvial track ;
and in this instance I must observe that in the ancient mud-
conglomerate of Geivan we found portions of Mimosa-wood
completely converted into lignite, and, at the village of
Duntai, near Seru, calcined (verkalkt) human bones in the
incipient stage of bitumenization.! A kind of bitumen, or
rather highly bituminous lignite, also occurs, although spa-
ringly, at Geivan, and, according to my observation, only in
small elongated nodular pieces, which in their transverse
fractures display a concentrically laminated structure, resem-
bling the annual growth of wood, and burning with a brief
flame only, but emitting a very smoky and bituminous odour.’
On neither occasion does Russegger specify what these fossil
human bones were, nor am I aware that any detailed identifi-
cation of them has been published. But the case is of sufti-
cient importance to demand the attention of future explorers
of the Nile valley to a walk of observation which may yield
results of high importance. Captain Grant informs me that
neither Captain Speke nor he ever met with fossil bones along
their route. Besides the asserted human bones at Duntai
(about halfway between Sennaar and Roserres), Russegger
mentions, cursorily, that in the conglomerate of Woadd
Medineh, also on the Blue Nile, he encountered a kind of
sandstone mass containing bones, which he took to belong to
the foot of a young camel. Dr. Murie, who accompanied
Mr. Petherick on his return to Soudan, has shown me speci-
mens of an alluvial fine grained siliceous grit or conglomerate
from the White Nile, above Khartoom, which is full of shells
of Oyrena fluminalis. The observations of Russegger, Mr.
Leith Adams, and Dr. Murie agree as to the abundance of
the shells of Mollusca in the Nilotic alluvia, mud, or con-
glomerate, from the neighbourhood of the First Cataract
upwards, and along the course both of the Blue and White
Rivers—oyster banks of Aitheria Caillaudi occurring through-
out.

4. Analogy of the Fluwviatile Deposits of the Nile with those
of the Ganges.—In some of the phenomena observed by all the

1 The original passage stands thus:— | Braunkohle umgewandelt war, und am
‘Und ich glaube diessfalls nur bemer- | Dorfe Duntai bei Seru Menschenknochen
ken zu miissen, dass wir in dem Altern | fanden, verkalkt und im Zustande einer
Flussschlamm-Konglomerate bei Geivan | beginnenden Verkohlung.’ (Russegger,
Stiicke von Mimosenholz, das ganz in ! ‘ Reisen,’ Band ii. pt. 2, p. 717.)

638 ANCIENT FLUVIATILE DEPOSITS

travellers above named, there is a striking analogy between
the alluvial deposits of the Valley of the Nile and those
occurring along the banks of the Ganges and Jumna rivers,
in the great alluvial valley of Hindostan. Of these the most
obvious is the great abundance of argillaceo-caleareous con-
cretions, forming an impure kind of travertine, and in the
lowermost beds horizontal deposits of more or less extent,
composed of the same kind of material. Russegger con-
stantly alludes to their frequent occurrence, both in the
conglomerates and in the indurated sand- or mud-deposits, in
the form of nodular concretions, varying in size from a pea
up to a quarter of a cubic foot, and having their centres
occasionally occupied by drusy cavities lined by crystals of
carbonate of lime. The lowermost bed, No. 5 of his section,
consisting of a hard dark-grey clinking limestone, appears to
be a modified kind of the same calcareous deposition. The
nodular form of these concretions is familiar to English ob-
servers in the ‘ Race,’ which so thickly studs the sections of
the brickearth-pits in many localities in the Valley of the
Thames.

§ II. Fuuviatire Deposits oF THE GANGES.

1. Physical Features of the Valley of the Ganges.—The vast
expanse of the plains of Hindostan consists of a fundamental
deposit of very ancient fluviatile sediment, which is developed
in great force, but varying in its detrital characters as we
follow the course of the rivers down to the sea. The valley
is longitudinally traversed, after their escape from the
Himalayah Mountains, by the Ganges and Jumna, which
unite at Allahabad. The segment, the Doab, constituting
the upper division of the plains of Hindostan, is that to which
the remarks which follow apply. It is comparable in some
respects to the tract through which the ‘ Blue’ and ‘ White’
Niles flow in the lower part of their course to their junction
at Khartoom. The Ganges at Hurdwar, where it debouches
from the Sewalik hills, is, according to the results given by
Sir Proby Cautley, 974 feet above the level of the sea, and at
Allahabad 269 feet, after running a course in a straight line
of 472 miles, giving an average fall of nearly 18 inches per
mile. From Allahabad to Rajmahal in Bengal, near the top
of the modern delta, the average fall, according to the in-
structive table given by Mr. Fergusson, amounts to about
6 inches along a stretch of 385 miles. The Jumna river,
where it escapes from the Sewalik hills at Rajghat, is a
little more elevated ; but it runs a nearly parallel course at
no great distance from the Ganges, and in the inclination of
its bed and other physical phenomena it resembles that river

OF THE NILE AND GANGES. 639

so closely that in the present sketch it is not necessary to
dwell on the points of difference.

Although the average inclination of the Ganges between
Hurdwar and Allahabad is about 18 inches per mile, it in-
creases considerably as we ascend the river. Thus the fall,
which in the distance of 122 miles between Cawnpore and
Allahabad diminishes to 13 inches, attains in the mean of
the 350 miles above it 18°3 inches, and so on upwards as we
ascend. ‘The sedimentary deposits and transported materials
vary, aS a general rule, in the same ratio. The northern
slope of the Sewalik hills is overlain with a thick mass of
boulder-gravel, inclined at a considerable angle, and con-
formable to the sandstone strata of which this Miocene range
is composed. The boulders vary from a few inches to upwards
of a foot in diameter ; they have undergone the utmost amount
of attrition, being constantly smooth and rounded into more
or less of a globular form. Their origin is distinctly shown,
as they are invariably composed of some of the rocks which
form the intramontane portion of the nearest river-channel,
transported by violent torrential action during the protracted
season of flooding. Modern boulder-dejections of precisely
the same character, and derived from the same rocks, are
seen in progress of formation where the rivers debouch into
the plains, constituting rude deltas, having a flattened conoid
surface, the base of which is ultimately confounded with the
plains. This gravel and boulder alluvium disappears from
the surface and along the beds of the rivers within a short
distance from the hills, and is replaced by a sand or clay
alluvium, which becomes the prevailing deposit down to the
confluence of the two rivers at Allahabad. It marks the
boundary of the habitat of some of the characteristic ver-
tebrate forms of the Ganges, such as the Gharial Crocodile
(Gavialis gangeticus) and the freshwater Porpoise (Platanista
Gangetica), which ascend to within thirty or forty miles of
Hurdwar, where the gravel-beds and rapids of the stream
terminate ; while the Crocodilus bombifrons is met with in the
dhoons or longitudinal valleys which lie between the Hima-
layahs and the Sewalik hills.

The rivers which traverse the alluvial plain of Hindostan
have produced the usual effects of powerful fluviatile action
operating during a long lapse of ages, aided by movements
of upheaval or depression, distinct evidence of which has been
brought to light by deep borings in the delta. The two
principal streams have gradually scoured their channels down
through the ancient alluvium to a depth of from 100 to 150
feet below the level of the adjacent plains, thus exposing a
very instructive section of great extent. At the lower part

640 ANCIENT FLUVIATILE DEPOSITS

of this section the rivers, as in the case of the ‘Blue’ and
‘White’ Niles, have intersected horizontal beds of argillaceous
or arenaceous travertine, or banks of aggregated nodular
kankar, which frequently form dangerous subaqueous reefs
or bars, obstructing boat-navigation. The Government of
India undertook in 1828 a series of operations, which ex-
tended over seven or eight years, for the removal of these
and other obstacles from the bed of the Jumna, in which
they are most prevalent. These were conducted by highly
instructed officers of the Bengal Engineers, one of whom,
Captain Edward Smith, published an account! of the most
striking facts which were observed on the occasion between
Agra and Allahabad. The upper half of the section, consist-
ing of beds of sand and clay, contained throughout, in more
or less abundance, the impure calcareous concretions called
nodular kankar. Near the base of the lower half these
calcareous deposits were developed in much greater force,
sometimes forming strata of rock kankar, from 1} foot to 2
feet thick, with a thinner bed of clay interposed. At one
point, Burldt, below the junction of the Chambal river, where
the bank is precipitous and 100 feet high, a stratum of rock
kankar, in the form of a granular concrete 2 feet thick, was
observed 60 feet above the lowest level of the stream. But
the most ordinary condition of the material is the concre-
tionary, in the form of nodular botryoidal stalactite or ramified
kankar. In some places the concretions, closely compacted
and connected by veins, are disposed in horizontal strata in
clay, at 10 or 12 feet above the level of the stream ; in others
the kankar presents itself in vertical seams in the scarped
front of the bank, or it ramifies in every direction through
the clay, literally lacing it together ; and occasionally ancient
surfaces of sun-cracked clay, where denuded, are seen with
the fissures filled with septarian plates of the same material.
At one point, Kareem-khan, slab-kankar, used for building-
purposes, and consisting of fine sand solidified by carbonate
of lime, is quarried at shallow depths from under the bed of
the river. Captain Smith (from whose memoir the above
particulars are for the most part drawn) has given an ex-
cellent series of highly instructive sketches, showing the
various modes in which the kankar occurs along the banks
of the Jumna.

2. Mammalian Fossils.—Fluviatile shells were either ex-
tremely rare, or they escaped the notice of individuals who
were not familiar with this walk of observation. Only two
instances are recorded—one an open Unio, embedded in a

1 Journ. Asiat. Soc. Bengal, vol. ii. p. 622 (1833),

OF THE NILE AND GANGES. 641.

perforated sandy clay near the level of the river; the other,
marks of shells in the granular concrete of rock kankar, found
at 60 feet above the stream, at Burlot. But fossil bones were
encountered in great abundance. In one case, unconnected
with the operations above referred to, the skeleton of a fossil
Hlephant was discovered in a bed of clay deposited on a
bottom of kankar, overflowed by the water of the river during
the floods, about three miles above Calpee. Some of the re-
mains were forwarded in 1828 to the Asiatic Society of Bengal.
In another case the skeleton of an Elephant, forming a great
mass, was observed by Mr. E. Dean lying amongst an im-
mense assemblage of kankar deposits, contained in the lowest
stratum of clay mtersected by the river, under the village of
Pauch-kowrie, near Korah Jehanabad. The stratum forms
a bank, there elevated 44 feet above the highest flood-mark,
and 80 feet below the summit of the cliff; and abreast of it
the Jumna has deepened its bed 25 feet. Numerous other
organic remains occurred in the masses of other deposits
surrounding the skeleton, but the precise kinds were not
ascertained. In a third case, a very large tusk ofan Elephant,
stated to have been 8 inches in diameter, was discovered
lying beneath a plate or slab of kankar in removing obstruc-
tions from the bed of the Jumna, near Adhae. The ivory
was fossilized, but not petrified; and the Sepoys engaged on
the work broke it up, and burnt it for pipeclay to whiten
their belts.

The great mass of the fossil bones which were discovered
during the first five years of the operations were unfortunately
lost, having been heedlessly thrown back into the deep
channel; and only those subsequently met with were pre-
served. They were found either embedded in the lowest
deposit of stiff clay, or in the shoals of kankar. Of the latter,
some were unquestionably of very modern origin, since they
yielded a sword and portions of a sunken boat. Their mode
of formation is obvious. Nodules of kankar and fossil bones,
detached from the alluvial cliffs by various denudating agen-
cies, are swept on by the floods until they meet with some
obstruction, where they collect and get commingled with
extraneous materials of modern origin, and the whole become
' solidified in a concrete, formed by the calcareous mud of the
kankar, aided by lime derived from the waters of the river.
They are therefore remamé deposits, wholly distinct from the
original kankar and fossiliferous clay-beds through which the
stream has cut its way down. The difference was clearly
made out by the engineer officers employed in the removal
of the shoals, who distinguished the two by the names of
natural and artificial kankar.

VOL. II. rp

642 ANCIENT FLUVIATILE DEPOSITS

The great majority of the bones were well fossilized, and
in most cases petrified.! Species of the following genera were
determined :— Elephas, Hippopotamus, Sus, Equus, Bos, Cervus,
Antilope, small Rodents, Gavialis Gangeticus, and freshwater
Chelonians. The specimens were commonly too mutilated,
and the materials then available for comparison too defective,
for certain specific determination in all cases: but among
them I identified molars of the extinct Hlephas Namadicus ;
a lower jaw with teeth, and a perfect astragalus, of the true
Indian Hippopotamus, H. (Tetraprotodon) Paleindicus; a
fragment of a jaw of the great fossil Buffalo of the Nerbudda,
Bos (Bubalus) Paleindicus; and jaws undistinguishable from
those of the living Gharial Crocodile. Both Captain Smith
and Mr. Dean, aided by medical officers more or less versed
in anatomy, thought that they had encountered human
bones among the Jumna fossils; and this opinion was pub-
lished at the time in an Indian scientific journal; but-the
identifications were negatived by Dr. Pearson and Dr. Evans,
the curators of the Museum of the Asiatic Society of Bengal ;
and on submitting the specimens to a close examination
several years afterwards, I could discover no determinable
human bones among them.

Another observation was made by Captain Smith, upon
which he was professionally competent to give an opinion
with authority—namely, that some of the fossil bones ‘ were
dug from depths of 6 to 18 inches in the firm shoal, which is
composed of substances (sic), kankar, bricks (vitrified clay ?),
more or less rolled and cemented by mud and clay. The cir-
cumstance is explicable on the modern accretion of some of
the kankar shoals above referred to, without involving a great
antiquity to the fragments of burnt clay.’

Of the fossil genera above named there are three well-
determined species, which are of much significance in the
history of the Doab alluvia. The first is Hlephas Namadicus,
an extinct form characteristic of the Pliocene fauna of the
Nerbudda. It belongs to the same group, Huelephas, as the
existing Indian Elephant; but it is broadly distinguished
from that species, and from all other known species, by a very
marked peculiarity in the form of the cranium, in addition to
dental and other characters. Among the vast quantity of
Miocene Proboscidean remains yielded by the Sewalik hills
not a trace of Hlephas Namadicus has ever come under my

1 Mr. James,Prinsep examined one of | Phosphate and carbonate of lime 17:5
the fossil bones of the Jumna, whichon | Water u é 5 ; E
a rough analysis yielded the following | Red oxide of iron with alumina 7675
results :— =

100-0

ee

ye ee ee ee ee ee

Marys

ie
jake * Cas

OF THE NILE AND GANGES. 643

observation. But I have seen perfect skulls of the species
from richly fossiliferous fluviatile deposits of Southern India.

The second important species of the Doab alluvium is the
huge ruminant, Bos (Bubalus) Paleindicus, also characteristic
of the Nerbudda fossil fauna. This form is closely allied to
the existing wild Buffalo, or Arnee of the Indian forests,
from which the domestic animal appears to have sprung.
Not a trace of it, or of any species of the same subgenus, has
yet been observed among the Sewalik fossil Mammalia ; nor
has its range in the fossiliferous beds of Southern India been
as yet accurately determined. But, in indicating these dis-
tinctions of the Miocene and Pliocene faunas, it is important
to remember that the rule is not absolute. I ascertained the
presence of the Miocene Proboscidean, H. (Stegodon) insignis,
of the Sewalik hills, among the Pliocene Mammalia of the
Nerbudda, where it was accompanied by a species of the
Miocene Hippopotamus, H. (Hexaprot.) Namadicus. The fossil
Buffalo here referred to existed in the same Nerbudda fauna
along with a huge Taurine species, Bos (Urus) Namadicus, of
which no close representative has been discovered among the
existing Indian Bovide. It differs alike from the Gour and
Gayal. Of it also no trace has been detected among the
Sewalik Mammalia.

The third fossil species, Hippopotamus’ Palwindicus, is,
perhaps, the most important in its indications. It belongs
to the subgenus Tetraprotodon, characterized by four incisors,
like the two African living species, and the European fossil
species, H. major and H. Pentlandi; but it is essentially
distinguished by constantly having the middle incisors
smaller than the outer pair, being the converse of what
occurs in the other. No well-authenticated case has as yet
been established of any fossil Tetraprotodon in Miocene strata.
A quadruped so remarkable for its size, form, and habits
must everywhere have forcibly impressed itself on the atten-
tion of mankind; and, struck with the close resemblance of
the Nerbudda fossil Buffalo to the existing species, the ques-
tion arose with me, May not this extinct Hippopotamus have
been a contemporary of man? and may not some reflection of
its former existence be detected in the extinct languages or
ancient traditions of India, as in the case of the gigantic
Tortoise? Following up the inquiry I ascertained from the
profound Sanscrit scholar, Rajah Radhakanta Deva, that
the Hippopotamus of India is referred to under different
Sanscrit names of great antiquity, significant of ‘Jala-Hasti,’
or ‘Water-Hlephant,’ in the ‘ Amaracosha’ and ‘Subda-rat-
nayali.? This view is confirmed by the opinion of two great
Sanscrit scholars, Henry Colebrooke and H. H. Wilson. The

Be

644 ANCIENT FLUVIATILE DEPOSITS

former, in his annotations on the ‘ Amaracosha,’ interprets
the words ‘Graha’ and ‘Avahara’ as meaning Hippopo-
tamus ; and the latter not only follows this version, but gives
two other words, ‘ Kariyadus’ and ‘ Vidoo,’ which he supposes
to signify the same animal. It is therefore in the highest
degree probable that the ancient inhabitants of India were
familiar with the Hippopotamus as a living animal; and it is
contrary to every probability that this knowledge of it was
drawn from the African species, imported from Egypt or
Abyssinia. Assuming that the quadruped was a contem-
porary of man in India, a very complex question is involved,
which is beyond the scope and limits of the present commu-
nication, namely, the ancient vocables above referred to as
the groundwork of the argument being of Aryan derivation.
Did the Aryan emigrants see the animal living on the northern
rivers of India, or was their knowledge of it derived from the
traditions of the more ancient indigenous races whom they
subjugated or displaced? After reflecting on the question
during many years, in its paleontological and ethnological
bearings, my leaning is to the view that Hippopotamus Na-
madicus was extinct in India long before the Aryan invasion,
but that it was familiar to the earlier indigenous races. I
may add that remains of the species have nowhere as yet
been observed in recent or comparatively modern deposits
in India; they have only been met with in a petrified con-
dition, deep in the alluvium of the Jumna, or in ancient
deposits in the Valley of the Nerbudda.

3. Fossil Mollusca.—I have already stated that our infor-
mation regarding the Mollusca occurring in the ancient allu-
vium of the Jumna is almost nil; until lately, this would have
applied to the fossil shells of what I have designated through-
out as the Pliocene deposits of the Valley of the Nerbudda.
But the operations of the Geological Survey of India have
already extended to that district; and I have been favoured
with a communication from Professor Oldham, dated Jan-
uary 8, 1858,in which he informs me that the evidence as to
the age of the formations is now becoming tolerably con-
clusive. A large collection of shells, comprising a consider-
able assemblage of species and a great quantity of individuals,
all proved to be of existing forms.

But there was this remarkable i in the group, that of many
of the commonest.living species some were exceedingly rare,
or even absent. Of Planorbis Coromandelianus, one of the
most prevalent Indian species and abundant in the Nerbudda
district, only two specimens were found in the fossil state;
while of Melanide, M. variabilis, and M. spinulosa, also com-
mon living forms, were not met with. The species, none

OF THE NILE AND GANGES. 645
being marine, in all amounted to twelve or thirteen. In

designating the formation as Pliocene, which I have done
during many years, I have been guided by the indications of
the Mammalian fauna, as intermediate between the Miocene
of the Ivrawaddi, Perim Island, and the Sewalik hills, and
that of the existing period.

4. General Inferences.—I shall now briefly indicate the
inferences to which the observations on the section of the
Jumna lead.

1. That the Doab alluvium, intersected by the Ganges and
Jumna, consists of fluviatile sedimentary deposits, the in-
ferior portion of high antiquity.

2. That there are no indications of its being anywhere
overlain by deposits resulting from marine submergence.

3. That during the progress of alluvial deposition the area
now constituting the plains of Hindostan was probably
subject to movements of upheaval and depression, analogous
to or corresponding with those which have been demon-

strated to have occurred in the delta of the Ganges.!

1 MS. Note written by Author in
1844. ‘The conditions under which
Oceanic deltas are formed, upon the
verge of the sea, preclude the occurrence
of sections, so as to show how the de-
tails of the daily progressing strata are
arranged. Our knowledge in such cases
is ordinarily limited to the superficial
banks of the rivers and creeks, through
which the tidal currents scour. Hap-
pily, the desired information in regard
to the delta of the Ganges has in some
measure been supplied by boring opera-
tions, which have of late years been car-
ried nearly 500 feet below the level of
Caleutta, the results having been care-
fully recorded in the Indian journals.
These borings were kegun as far back as
1804, and have been repeated at dif-
ferent periods, the last and deepest ex-
eavation haying been carried on between
the years 1836 and 1840, under the
superintendence of a committee of engi-
neers.

*Caleutta is situated in the midst of
the delta of the Ganges, upon the banks
of the Hooghly, within the range of the
tidal current. The surface soil consists
of ten feet of fine mud or inundation
sediment, which is succeeded in the de-
scending order by a bed of blue adhe-
sive clay, 40 feet deep. The lower por-
tion of this clay abounds in decayed
vegetable matter, part of which is in the
condition of peat mixed with fragments
of red-colourcd wood, hardly altered,

which Dr. Wallich was enabled to iden-
tify with the Soondree of the creek
forests. Next follows a bed of ten feet
thick of the eoncretionary argillaceous
limestone, which, under the name of
‘“‘kankar,” occurs abundantly in many
parts of India, in the form of irregular
nodular masses. Then succeeds a bed of
green siliceous clay, 15 feet thick, the
lower portion of which also furnishes
kankar-nodules. At 75 feet below the
surface, variegated sandy clay begins, and
continues to a depth of 45 feet. This is
succeeded by thin beds of clay marl,
loose friable sandstone, and a sandy clay
abounding in weathered fragments of
mica-slate, and waterworn nodules of
hydrated oxide of iron. At 175 feet a
soft conglomerate, 10 feet thick, is met
with, consisting of gravel cemented by
a clayey matrix, together with sharp an-
gular fragments of quartz and felspar,
larger than peas, and resembling in
every respect the debris of granitic rocks
which have suffered little attrition by
transport from a distance. The gravel
passes gradually into an indurated fer-
ruginous sandy clay abounding in scales
of mica, which continues with little va-
riation through a depth of 18 fect. A
thin stratum of soft sandstone occurs
next, succeeded by beds of sand and
sandy clay extending with great uni-
formity of character through a depth of
172 feet. The lower half of the hume-
rus of a ruminant, apparently a species

646 ANCIENT FLUVIATILE DEPOSITS

4. That the fossil remains occurring in the undisturbed
banks of clay and kankar, at the bottom of the section, are
of the same age as the deposits in which they occur.

5. That the ancient fossil Mammalia of the Gangetic
valley belong to the Pliocene fauna of the Nerbudda, as dis-
tinguished from the Miocene fauna of the Sewalik hills.

6. That of the Jumna fossil Crocodiles, some belong to
species which are now living; and that of the extinct Mam-
malia, some were probably contemporaries of man.

7. That no trustworthy cases of the occurrence of very
ancient human bones or industrial objects have yet been
established from the sections of the Jumna and Ganges, but
that they may be looked for on a more careful and extended
search.

8. That in the great abundance of calcareous concretiona:
deposits there is an analogy between the alluvial beds of the
valleys of the Ganges and Nile; but that in the property of
vertebrate remains, the latter, in so far as it has been ex-
plored, is a remarkable contrast to the former.

§ III. Anriguiry or Man in Inpra.

1. Introduction.—In discussing some of the speculative
points which have been raised in this paper, I have introduced
topics which are not usually brought before the Society.
But I make no apology. Geology has never disdained to

of Deer, and of the size of the common | fragment of vesicular basalt (?) is stated
Cervus porcinus of India, impregnated | to have been found. From the upper
with hydrate of iron and retaining but | portion of this gravel, near its junction
a slight trace of animal matter, together | with the superincumbent clay, frag-
with portions of the bony shell of a spe- | ments of excellent coal were brought up,
cies of Trionyx, resembling those now | a large field of this mineral occurring in
found existirg in India, were brought | the Burdwan district, 70 miles above
up from the lower part of this sand at | Calcutta. Fragments of lignite, re-
depths of 850 and 375 feet below the | sembling that found in the adjacent
surface. The sand is followed by a bed | hilly district of Cuttack, together with
of blue clay full of freshwater shells, | caudal vertebre, attributed to a Lacer-
which from their fragmentary and inco- | tine or small Crocodile form, and Chelo-
herent condition could not be specially | nian remains referable, like those from
identified. Under this lacustrine depo- | the superincumbent sand-beds, to species
sit, another “ dirt bed ” or dark-coloured | of the freshwater genus Trionyx, were
clay occurs, composed almost entirely of | yielded at different depths by the gravel.
decomposed or decayed woody matter, | Near the bottom of the bore, the auger
mixed up with argillaceous earth, ex- | passed through a log of waterworn de-
- tending through a depth of 10 feet. At | cayed wood, unearbonized, and so little
this point, about 400 feet below the sur- | altered as to resemble in every respect
face, a remarkable and abrupt change | “fragments occurring in the modern
takes place in the sedimentary character | alluvium of the Soonderbuns.” The
of the strata of the delta. Immediately | bore terminated at 481 feet beneath the
below the clay an enormous accumula- | surface by the auger becoming unfor-
tion of gravel is met with, consisting of tunately jammed, the gravel still con-
boulders and large rounded pebbles of | tinuing of the same character, although
quartz, felspar, limestone, and other | already penetrated to a depth of 85
primitive rocks, among which a rolled feet.’—[ Hp.]

OF THE NILE AND GANGES. 647

draw upon any department of human knowledge that could
throw light on the subjects which it investigates. Cuvier, in
the ‘Discours Préliminaires,’ exhausted the records and
traditions of every ancient people in search of arguments to
support the opinion that the advent of man upon the earth
dates from a comparatively late epoch. At the present time
the whole aspect of the subject is transformed. The science
is now intimately connected with archeological ethnology, in
searching for evidence of the hand of man in the oldest
Quaternary fluviatile gravels of Hurope. In other continents,
under different physical conditions, it may be possible to
interweave the indications of language and misty tradition
with the more certain results of paleontological research,
and thus to aid us in arriving at that ‘speck now barely
visible in the distance, which is our goal to-day, and may be
our starting-point to-morrow.’ I shall, therefore, not hesitate
to enter upon a complementary portion of the same walk of
investigation which is intimately connected with the thread
of the preceding speculations. .
[A portion of the paper is here omitted, as it corresponds
with a passage in the historical essay on ‘ Primeval Man,’
p- 573 to p. 579.—Ep. |
The large question which these reflections concern is at
the present time followed up with the keenest intelligence and
with the closest scrutiny over a large portion of Hurope.
But in the tropical regions, which promise to be the most
fertile of results, the ground has been barely broken. The
observations of Russegger in the Valley of the Nile would
seem to have fallen into the oblivion which shrouded the
shrewd observations of Frere on the Hoxne implements, until
they were brought to light by the researches of Mr. John
Evans. In India also the inquiry, begun so auspiciously
nearly thirty years ago, appears to have stagnated in later
days, and to require a fresh impulse. The important dis-
coveries of Captains Speke and Grant will assuredly attract
explorers, until the affiuents which feed the lake out of which
the White Nile flows are traced to their sources. It is
incredible that that great river should run for fifteen or
seventeen hundred miles, often through alluvial deposits,
ancient and modern, without yielding traces of its former
population. In the interest of the general investigation, I
have therefore thought it might be useful to bring together
the facts and speculations which are set forth in the preced-
ing observations, as a guide to future inquiry.

648 GLACIER-EROSION | THEORY

XXVIII. ON THE GLACIER-EROSION THEORY OF
LAKEH-BASINS.

J.—ABsTRACT OF AN EXTEMPORE SPEECH DELIVERED By Dr. FaLconer
AT THE Royat GroGgrapuicaL Society, oN January 27, 1864, AND
PUBLISHED BY THE SOCIETY.

Dr. Fatconer, after describing the progress of the Trigonometrical
Survey in India, next drew attention to the glacier system of the
Himalayahs. All the best observers—Dr. Thomson, Jacquemont, and
others—had been of opinion that there was but one great system of
mountains. There was no such thing as any break of mountain range,
or any distinct mountain chains. There were great rivers which cut
them across, rivers like the Indus, the Sutlej, and some feeders of the
Ganges ; but, regarded in one grand aspect, they constituted a series or
mass of mountains with ravines and valleys intervening. Viewed, then,
in this light, there were two great ranges which culminated to especially
great altitudes, and which bounded the Indus river to the south and
the north; and this being one of the points where the Himalayan chain
attained its greatest elevation, there the glacial phenomena were developed
in most grandeur and upon the loftiest scale. Captain Godwin-Austen
had referred to that part of the range which bounded the Valley of the
Indus upon the north, the Kara~-Korum or Mooztagh or the ‘ Icy range
of mountains,’ and the other great series of them were the mountains
which bounded the Indus upon the south. Although the glaciers upon
the Shiggur Valley and in the Valley of Braldoh, which he himself had
visited in 1838,! were of such surpassing grandeur and importance as
had been mentioned by Sir Roderick Murchison, it was but fair to
other observers to say that upon the northern side there were glaciers
which, so far as description went, were equally grand, if not grander.
Those to which he should especially refer were the glaciers at the head
of the Zanscar river, the sublime features of which had been so well
described by Dr. Thomson. Mr. J. Arrowsmith, from his labours on
the maps of Hugel, Thomson, and other explorers, was well acquainted
with the mountain-ridge to which he referred, and the glaciers which
arose from it. There was the river called the Chenab, and a mountain
range which stretched across between the Indus and the Chenab. ‘The
pass of the dividing ridge at this point was 18,000 feet above the level
of the sea; and upon either side, but more especially upon the north,
at the heads of the Zanscar river, were some of the grandest glacier
phenomena which were to be seen in any part of the world. There
were glaciers extending from a very great distance, which attained

1 See vol. i. p. 570.—[ Ep].

OF LAKE-BASINS. 649

enormous width—confluent into a sea of ice—and which, until the
description that had been given by Captain Godwin-Austen, had been
unrivalled by any glacial phenomena with which they were acquainted,
except the glacial formations in the Arctic regions, such as the Hum-
boldt glacier in Smith’s Sound, described by Dr. Kane.

With regard to the glaciers upon the north, the Indus ran through a
long depressed valley westward, receiving from the north three great
branches ; the first branch, called Shayook, from the Kara-Korum, next
the Nubra river, and also the Shiggur, which was the especial object of
Captain Godwin-Austen’s communication. Now, the Shiggur valley
was the third of importance of all the affluents of the Indus, and was
bounded by mountains of a great elevation. Some of them which had
been measured by Major Montgomery attained a very great elevation ;
one a height of 28,000 feet above the level of the sea. This naturally
entailed a prodigious amount of condensation of the moisture of the
atmosphere, and led to a very heavy fall of snow, the result of which
was seen in these glacial phenomena. Twenty-seven years ago he had
been up to Arindoh, the extreme termination of the western or Basha
branch, and from that point by a détour he got across upon the other
valley by the Scora-la Pass to the glacier of the Braldoh river, where
he saw all the phenomena which had been described by Captain
Godwin-Austen.

Having premised this much with regard to special details, there were
one or two points which he was desirous to bring before them. One
was, What were the peculiar characteristics of the Himalayahs, as well
as of all tropical mountains, as compared with our European mountain-
chains? There was one characteristic of the Himalayan chain so re-
markable that he should take the liberty of explaining it at some length.
He presumed that most of his audience had visited either the northern
or southern side of the Alps; and those who had been in the plains of
Italy, along the Valley of the Po, were well acquainted with the
numerous lakes which jutted out from the Alps into the plain of Italy.
Commencing on the west, they had got the Lago d’Orta, the Lago
Maggiore, the Lago Lugano, the Lago di Como, the Lago d’Iseo, and
the Lago di Garda; in fact, wherever a great valley projected itself
from the chain of the Alps at right angles to the strike of the chain,
there they had, with one or two exceptions, uniformly a great lake.
Regarding these lakes in a general way, without reference to detailed
phenomena, they found one thing which was constant about them—
‘they were invariably narrow, and some forty or fifty miles long, as
notably in the case of Maggiore, Como, and Garda.’ The next re-
’ markable thing about them was that they invariably radiated out at
right angles to the strike of the great chain of the Alps. The Alps
made a curve from the Pennine round to the Rheetian Alps. They
would also observe that these lakes were severally fed by a consider-
able river which proceeded from a high ridge of the chain, and which
was thrown forward into the plains of the Valley of the Po.

If they would consider the Himalayahs, or any tropical range of
mountains whatever, in a similar way, they would find that those lake
phenomena were invariably wanting. Great rivers like the Indus, the
Chenab, the Sutlej, and the Ganges, which passed through the Himalayan
Mountains and debouched into the plains of India, had got valleys of

€50 GLACIER-EROSION THEORY

infinitely greater importance than the valleys either to the north or
south of the Alps; but they were never connected with a lake.

The question then arose, What was the physical reason of this great
difference between the tropical mountains and those of temperate
Europe? Nearly thirty years ago, he was for ten or twelve years
rambling about the Himalayahs along a stretch of 800 miles, and he
used to open a map before him, and try to make out the comparative
features of European and Eastern mountains. He looked to the
numerous lakes to the north and south of the Alps; and he would put
the map of India alongside, where the same kind of rivers were de-
bouching into the plains, but where there was an utter absence of lakes
in connection with them; and he used to puzzle himself in trying to
discover a physical explanation of this difference. He was perfectly
satisfied there must be some secondary conditions which were not
common to the two, and he determined that, on his return to Europe,
he should make them the subject of special research; for at that time
the glacial investigations of Charpentier and Agassiz in the Alps were
unknown to a solitary wanderer in the Himalayahs. That intention he
had carried out, by repeated visits to the Italian valleys of the Alps.
There was the same kind ot elevation above the level of the sea, the
same kind of valleys, the same kind of fissures intersecting the great
ridges—What then was the explanation? This he would endeavour
to indicate. About two years ago, March 5, 1862, as his friend Sir
Roderick Murchison was aware, a paper was brought before the Geo-
logical Society of London by Professor Ramsay, which excited a great
deal of attention, and gave rise to a very animated discussion. The
theory of the paper was that, as a rule, lakes in all the temperate and
cold regions of the world were the product of glacial excavation ; that
is to say, that wherever a glacier descended from a high ridge of moun-
tains into a plain, it ploughed its way down into the solid rocks and
carved out a great lake. This was the theory, or rather hypothesis
which Professor Ramsay put forward, to explain the lakes which were
so abundant in the valleys of the Alps. A similar speculation, but
greatly more restricted, had been advanced by Mortillet a short time
before. He limited the action of the glacier to scouring out the silt of
the filled-up lake basins, the origin of which he attributed to antecedent
fissures the result of upheavement. An application of this theory was
made to the different physical phenomena which were connected with
the case; and it occurred to himself and many others (and he believed
Sir Roderick had an opinion in common with himself), that it was not
adequate to explain the phenomena; and on the occasion when it was
produced, he met it with the most lively opposition in connection with
his own experience in the Himalayan Mountains. The opposition
which he gave to it was upon these grounds. Many of them would
remember that the lakes Maggiore and Como were upon the edge of the
plains of Italy; that the glaciers, say that of the Ticino which came
down into the Lago Maggiore, descended along a steep incline, and
were at last delivered into that lake, which was about fifty miles long,
and only eight or nine miles wide at its widest point. Its prolongation
nearest to the Mediterranean attained a depth of about 2,600 feet below
the level of the sea. Where the river escaped out of the lake it was
not more than about 600 feet above the level of the sea. It was a

OF LAKE-BASINS. 651-

remarkable point in the case, that this glacier, by the hypothesis, should
have ploughed its way down, and actually dived into the bowels of the
earth 2,000 feet below the level of the Mediterranean, and then should
have again risen up along an incline at a rate of about 180 feet per
mile.

Without going into all the objections, he might state he believed the
principal one was, that the mechanical difficulties in the case were en-
tirely left out of sight by the supporters of that theory ; and on that oc-
casion, after very long study of the subject, he had endeavoured to bring
forward what occurred to him as the true explanation of the difference
between the Himalayan Mountains and the Alps. The difference he
believed to consist in this: that after the last upheaval of the Alps,
great fissures, or basins of lakes, were left there, with rivers running
into them, in the manner in which" the Rhone runs into the lake of
Geneva, bringing down vast quantities of silt, which, if you give a
sufficient number of ages, would have completely filled them up. But
before this was accomplished, what is called the Glacial period set in;
that is to say, there was an enormous projection of ice and snow, below
the limit that they now saw it in the Alps, out into the plains, both to
the north and south of that chain ; and, as the snow and ice came down,
they filled up those lakes, and formed a bridge, upon which the moraine
material was carried over, there being a certain measure of incline from
the summit of the Alps down to the plains of Italy | When once the
basins were filled with ice to the depth of 2,500 feet, they made, as it
were, a slide or incline, upon which all the solid material could be
transported ; and that being carried forward by the vis motrix of the
mass, formed the large moraine which we saw at Lake Maggiore, that
of the Brianza, and also the moraine which bounded Lake Garda, where
the battle of Solferino was fought. This was the secondary condition
that occurred in Europe. Precisely the same primary conditions oc-
curred in the great valleys of the Himalayahs, but without the same
glacial phenomena. These mountains were thrown up above the level
of the sea, and vast perpendicular fissures left, forming what constituted
at that time the basins of lakes. But in those tropical regions, the ice
never descended from the highest summits down into the plains of
India ; and instead of being filled up by snow, which afterwards melted
into water, these lake basins were gradually silted up by enormous
boulders and alluvium of every kind, which were transported down
from the Himalayan Mountains in prodigious quantities by the torrential
action of the periodical rains. The difference in the two cases was,
that whereas the ice filled up the lake basins in the Alps, constituting,
as it were, the conservative means by which those basins were saved
from being silted up by alluvial and other matters, in the Himalayan
Mountains this conservative action did not take place, and the lake
basins remaining open got filled up in the manner above described.
If they would look at the map of the Himalayahs, one of the most
remarkable things they would observe on the southern side of the
chain was, that there were no great lakes whatever—not one that
would compare with Lake Lugano, or with any of the second or third-
rate lakes in the Alps. But if they crossed to the northern side of the
chain where the temperature was much colder during the winter, there
they would find great lakes. The cold produced the same conservative

652 GLACIER-EROSION THEORY

action on the northern side of the Ilimalayahs, in preventing the lakes
being filled up, that it did in the Alps by restricting the silting action.

This was the main fact to call to the attention of the Society, with
reference to the great difference between the Himalayan and other
tropical ranges of mountains and those in Europe. The next point
was one of some interest and importance. There was a material well
known in commerce and arts, called borax, now largely employed in
ceramic products. It used only to be got from India as an export from
Tibet, and it was invariably found in connection with hot springs.
Within the last twenty years a remarkable change had taken place.
The late Count Lardarel, an original-minded and eminently philan-
thropic Frenchman, of Leghorn, aware of the presence of boracic acid
in the jets of steam which are emitted from the surface of the broken
soil in the ravines of Monte Cerboli, on the margin of the Maremma of
the Volterra in Tuscany, hit upon the happy idea of utilizing the
natural heat in lieu of fuel to effect the process of evaporation. Extem-
porized tanks fed by rills of cold water were employed to intercept the
jets of steam, until the fluid got charged with boracic acid; while other
jets of steam, tapped from the soil, were led off in pipes, and distributed
under the evaporating pans. An unbounded supply of boracie acid
was the result. As a consequence, the borax of Tibet fell in value
from 87/. or 40J. a ton to nearly half that price, until at length borax
was exported from England at the rate of 10/. per ton, to displace the
native article from the bazaars of India. In Tibet the mineral occurs
in the form of biborate of soda, that alkali in many places abounding in
the soil; while in Italy it is yielded in the form of boracic acid. In
both cases its appearance was coincident with a region of hot springs,
which occurred at great elevations in the Himalayahs; and for the best
account of their connection with borax he could refer to Dr. Thomson’s
‘Travels in Tibet.’

Connected with the Himalayahs, there was also a physical and vital
phenomenon of still greater importance. Ienry Colebrook, the first
who, along with Colonel Crawford, measured the heights of the Dwa-
lagiri, procured from the plateau of Chanthan in the Himalayahs, at a
height of 17,000 feet above the sea-level, fossil bones, which were
brought down and exported as charms into India, to which the natives
attributed a supernatural origin, and called them ‘ lightning or thunder
bones.’ At the present time, during eight months of the year, the
climate differed in no important respect from that of the Arctic circle,
and in the whole of the district there was not a single tree or shrub
that grew larger than a little willow about 9 inches high. The grasses
which grew there were limited in number, and the fodder, in the shape
of dicotyledonous plants, was equally scarce. Yet notwithstanding this
scantiness of vegetation, large fossils were found, of the rhinoceros, the
horse, the buffalo, the antelope, and of several carnivorous animals; the
group of fossil fauna as a whole involving the condition that, at no very
remote period of time, a plateau in the Himalayan Mountains, now at
an elevation exceeding three miles above the level of the sea, where we
get the climate of the Arctic regions, had then such a climate as
enabled the rhinoceros and several sub-tropical forms to exist. It
would occupy too much time to explain the details of this complex
phenomenon. He would briefly state that the only rational solution

OF LAKE-BASINS. 653

which science could suggest was that, within a comparatively modern
period, a period closely trenching upon the time when man made his
appearance upon the face of the earth, the Himalayahs had been thrown
up by an increment closely approaching 8,000 or 10,000 feet.!

IJ.—Extracts rrom Letters sy Dr. FALconer IN THE ‘ READER’ FOR
Fesruary AND Marca, 1864.

Terrestrial mechanical phenomena are susceptible of mathematical
analysis on mechanical principles. A very eminent mathematician,
and one of the most distinguished physical geologists of our time,
Mr. Hopkins, has undertaken the mathematical investigation of the
problem of mountain elevations, viz., ‘to determine the nature of the
effects produced by a general elevatory force acting at any assigned
depth on extended portions of the superficial crust of the earth, and
with sufficient intensity to produce in it dislocations and sensible
elevations.’ Suppose a given area, bounded by parallel lines, and in-
definitely long as compared with the width, to be subjected to an
expansive subterraneous force, acting simultaneously upon every point,
and sufficient to bear up the superincumbent- mass, the strata will at
first rise in the form of an arch, limited by their extensibility, and at
length give way to the strain. The resultant fractures will take place
simultaneously in two directions; those of the first class, parallel to the
line of greatest extension (the strike); the second, at right angles to the
first, 7.e. across the strike. Local and partial causes—such as ‘ irre-
gularity in the intensity of the elevatory forces, and in the constitution
of the masses on which they are supposed to act’—will cause the effects
to deviate from strict geometrical laws.

Such, broadly and very imperfectly put, is the mechanical theory of
Mr. Hopkins, which, so far as it has gone, has been regarded by mathe-
maticians as a great step in the dynamics of geology, resembling that
which affected astronomy when mathematical reasoning and calculation
were directed to it ‘at the turning point of its splendid career.’ (Dr.
Whewell, ‘ Anniversary Address,’ G.S., 1838, p. 27.) Practically ap-
plied, it has been accepted by geologists as remarkably in unison with
the phenomena of mineral veins in a definite known area; and the
special application of it to the phenomena of the Weald and the Bas
Boulonnais by Mr. Hopkins has been regarded as a very successful
illustration of the truth of the theory. ‘The remarkable breaks in the
bounding chalk-ranges which give passage to the rivers flowing from
the Wealden northward and southward are shown to correspond in
situation with cross fractures, indicated by the theory, and sometimes
rendered probable, and occasionally proved, by observation.’ (Professor
Phillips’s ‘ Manual,’ 1855, p. 598.) Mr. Hopkins has not yet given to
the world the extension of the mechanical theory to the phenomena of
great mountain-chains like the Himalayahs, where angular elevation, so
to speak, is what we now see; but he distinctly admits: 1. That
‘along the flanks of elevated ranges longitudinal valleys are not unfre-

1 See voi. i. p. 177 to 185. The views | which follow are taken from Dr. Fal-
above expressed led to an animated dis- coner’s letters which appeared on that
cussion in the ‘ Reader.” The extracts ' occasion.—[Ep. |],

654 GLACIER-EROSION THEORY

quently found running nearly parallel to the general axis of elevation.’
2. Transverse valleys.—‘ Deep valleys are sometimes found of which
the directions are nearly at right angles to that of the general eleva-
tion.’ He admits that these latter may frequently be due to the effects
of erosion, but states that in some instances they appear to have
been obviously formed by the elevation of the strata on either side of
them; and he cites a beautiful example of this kind of formation in a
river in Wales. So far as I am aware, no attempt has yet been made
to investigate how much of the expansion of a transverse valley near
the axis of a lofty chain like the Himalayah is due, theoretically, to a
fissure of upheavement, and how much to subsequent erosion.

Suppose a given area of the bed of the ocean to be thus upheaved, so
as to elevate part of it, in the shape of islands, above the sea-level, and
leave part of it submerged, but with the dislocations and fractures and
fissures of both orders within the scour of tides and currents: the
abrading and solvent action of water, with other atmospheric agents, in
the former case, and the movements of the ocean in the latter, ope-
rating during long periods, will produce enormous alterations of the
exposed surfaces, denude, widen, and deepen the rents, fissures, and
other inequalities. Let the action of ice be superinduced, as exem-
plified on the shores of Spitzbergen, Smith’s Sound, and the ancient
fiords of Norway, or upon mountain chains, the effects will then be
produced on a scale of very great magnitude. But the fractures,
fissures, foldings, and dislocations, and the uptilting of the beds caused
by the original movements of upheavement have determined the di-
rection in which the subsequent abrading and denuding causes have
operated. The same line of reasoning applies to the upheavement into
mountain chains of land already above the level of the sea. Some-
thing like the foregoing would embody the combined results of the
mechanical theory and empirical reasoning—The Reader, Feb. 27,
1864.

The Valley of the Jordan, regarded as a whole, is one of the most re-
markable narrow rectilinear indentations of which physical geography has
taken cognizance as occurring on the surface of the earth: a deep trench
extending from the mouth of the Orontes at Antioch to the Gulf of
Akabah, and bounded on either side by considerable elevations, but
interrupted at two points—first, by Lebanon and Hermon, dividing the
watershed of the Orontes from that of the Jordan second, by the high
ground bounding the basin of the Dead Sea to the south. The trough
is thus divided into three segments, from north to south: 1st, in the
Valley of the Orontes; 2nd, the Ghor, or valley proper of the Jordan ;
8rd, the Valley cf Arabah. The second of these segments is that to
which attention is now called. The Jordan arises out of springs near
Hushbeyah, on the western flank of Mount Hermon, at an elevation of
about 1,700 feet above the level of the Mediterranean. After receiving
some afiluents, it runs due south into the Lake of Merom (El Htileh), a
triangular basin about 44 miles long by 3} wide, and elevated about
50 feet above the level of the sea. ‘The lake occupies the southern end
of an intramontane plain 15 miles long by 5 wide, indicating the former
boundaries of the sheet of water. From Lake Merom the river descends
by a contracted channel to plunge into the Lake of Tiberias, with a fall
of about 600 feet in a distance of 9 miles, being 663 feet per mile.

2

OF LAKE-BASINS. 655

That lake is 13 geographical miles long by 6 in width; its surface is 653

feet below that of the Mediterranean, and its greatest ascertained depth
165 feet, according to the cursory soundings made by the American
expedition under Lieutenant Lynch. From the Lake of Tiberias the
Jordan runs due south in a trough, as strait as an arrow, and about 7
miles wide from wall to wall, to plunge into the chasm of the Dead Sea
—a distance of about 60 miles—during which it falls 983 feet, or at
the rate of about 16-4 feet per mile. The immediate bed of the river,’
eroded by fluviatile action, is so excessively tortuous as, by calculation
(Lynch), to triple the length of its course, which throughout is a nearly
continued cataract.

The Dead Sea is 46 English miles long by 104 wide, of a narrow
oblong form, and its area has been estimated to cover about 250 square
miles. The depression of its surface below the Mediterranean has been
variously estimated from 1,446 feet (barometrical, von Wildenbruch)
to 1,512 and 1,316 feet by trigonometrical and levelling operations, the
former by Captain Symonds, the latter by Lieutenant Lynch. The last
two estimates may practically be considered to be identical, as the
height of the lake varies at different seasons, and Lieutenant Lynch’s
does not appear to have started from any bench-mark or fixed datum.
Captain Symonds’s estimate of 1,312 feet is adopted here.! The depth
of the lake has been variously described. Messrs. Moore and Beeke
assign a maximum of 2,400 feet; while Lieutenant Lynch, with a party
of sailors accustomed to using the lead, only got a maximum, by sound-
ings, of 1,308 feet between Ain Terdbeh and Wady Zirka, about 9
miles from the mouth of the Jordan. This, added to Captain Symonds’s
result, would give a depression of 2,620 feet below the level of the
Mediterranean. That this depth was originally very considerably
greater can hardly be doubted, since the Dead Sea, with no outlet, has
had poured into it the silt of the Jordan and of the other streams
that flow into it, and have flown from all points of the compass
during a long lapse of ages, of which there are no data for arriving
at even an approximate estimate. The southern end of the lake has
been filled up by the deposition of inpoured silt. That the chasm is
precipitous is distinctly proved by Lynch’s sections, he having ascer-
tained a depth of 900 feet, close to the margin, on th2 eastern side.
The southern extremity is bounded by the Khashm Usdum, or salt-

1 Up to the time of his fatal illness,
Dr. Falconer was engaged in collecting
information on the physical geography
and geology of the ‘ Dead Sea.’ The last
occasion on which he ventured out was
to attend a meeting of the Council of
the Royal Society, for the purpose of
urging the propriety of voting a sum of
mouey, for finally determining the vexed
question of the relative levels of the
Dead Sea and of the Mediterranean.
This investigation was subsequently con-
ducted by Col. Sir Henry James, and
from a Report presented by him to the
Royal Society on May 3, 1866, it appears
that on March 12, 1866, the level of the
Dead Sea was 1,292 feet below that of

the Mediterranean, and it was ascer-
tained that during the early summer the
level of the sea falls at least six fect
below the level at which it stood on the
day the levelling was taken. The sound-
ings in the Dead Sea by Lieut. Vignes of
the French Navy gave a maximum depth
of 1,148 feet, making the depression of
the bottom of the Dead Sea 2,446 fect
below the level of the Mediterranean.
The soundings in the Mediterranean,
midway between Malta and Candia, by
Capt. Spratt, R.N., gave a depth of 13,020
feet, or a depression of the bottom five
times greater than that of the bottom of
the Dead Sea.—[ Ep. ]

656 GLACIER-EROSION THEORY

ridge, on the south-west, and due south by a confused mass of unim-
portant low hillocks, composed of chalky indurated marl. From this
point the trough rises along the Valley of Arabah to a point 35 miles
north of Akabah, where it attains a height, in round numbers, of about
500 feet above the level of the western arm of the Red Sea, or 1,812
feet above that of the Salt Lake. For an excellent summary of the
facts the reader may be referred to Mr. George Grove’s article on the
“*Salt Sea’ (‘Smith’s Dictionary of the Bible,’ vol. iii. p. 1173). The
geological formation of the tract is nummulitic and jurassic, with dykes
of basalt, which abound along the Lake of Tiberias.

The physical features of the tract have led almost all good autho-
rities to infer—1. That a long straight fissure or crevasse, caused by
mechanical disturbance, extended from Antioch to the Red Sea. 2.
That a strait connecting the Mediterranean with the Gulf of Akabah
was scoured out along this fissure forming the trough of the Jordan.
3. That the extremities of the trough were elevated, cutting them off
from the two seas. 4. That the Jordan probably flowed at one time
into the Gulf of Akabah. 5. That simultaneously with the upheaval
of the Akabah ridge, the valley of the Jordan was depressed through a
violent mechanical convulsion. 6. That, as a result of the last opera-
tion, the chasm constituting the precipitous deep basin of the Dead Sea
was left. How was this chasm formed? That it was a volcanic crater
is negatived by all the facts known on good authority; for details vide
Dr. Anderson’s Report and Russegger’s Sections. Let us apply to the
conditions the hypothesis so stoutly advocated, of glacier-erosion as the
formative agent of lake-basins. The excavation to such a depth of a
rocky area of 250 square miles, would involve an enormous terminal mo-
raine, without reference to transported blocks and detritus, derived from
the head and upper course of the glacier. Nothing corresponding has been
observed by any explorer at the southern end of the Dead Sea. The
salt-ridge of Khashm Usdum, the detached pillars of salt, and the unim-
portant eminences of chalky indurated marl, traversed by the Wady el
Jeib, are not in accordance with the usual physical character of a moraine.
To an argument so conclusive it may be needless to add others. But the
lowest point to which Dr, Hooker traced the moraines of ancient glaciers
on Lebanon was 6,172 feet above the sea. Could the ridges at the
head of the Jordan have yielded a stream of ice which must have
travelled 150 miles to reach the sultry chasm of the Dead Sea, and
then scoop it out? How explain the string of the three lakes, E] Hfleh,
Tiberias, and the chasm of the Ghor, with the sixty-mile trough
between the two last? Erosion by marine or fluviatile action might
account for the general excavation of the trough along the line of a
fissure ; but how can it be made to explain the sudden, irregular, pre-
cipitous chasm, so much below the general level, which is met with in
the basin of the Salt Lake? That chasm occurs, at the point of greatest
depression, where, to use a figurative expression, the back of the trough
appears to have been broken during the convulsion which caused the
subsidence. But, for the present argument, it is sufficient to indicate
that the Valley of the Jordan presents along its course three well-
marked lake-basins, the formation of which seems irreconcileable with
the hypothesis of glacier-erosion.

I shall now revert to the lakes of Lombardy, which I had in view

OF LAKE-BASINS. 657

while comparing the perpendicular or transverse valleys on the southern
sides of the Alps and Himalayahs.

Take the Lago Maggiore, which is about 52 English miles long, and
3} to 4 miles wide, at the broadest part of the central portion between
Luino and Laveno. It is fed by the Ticino, consisting of two main
branches given off across the strike of the chain. The basin forms a
long, narrow, flexuous hollow, bounded by low hills, but open to the
south. Its greatest depth, opposite Santa Caterina, at one of its con-
tractions, about 124 miles from its southern extremity, is 2,605 feet,
giving a general gradient of about 200 feet per mile for the rise of the
bottom to the surface between the two points. Taking Lynch’s esti-
mate for the Dead Sea (1,308 feet), the extreme depth of the chasm
occupied by water in the Lago Maggiore is double that of the former,
and the depression of its bottom below the surface of the Mediterranean
is 1,927 feet, being only about 678 feet less, or the height of the lake
above the sea-level. It is even deeper than any soundings yielded in
the Gulf of Genoa (vide Admiralty Chart). ;

The basin of the Lago di Luguno is so irregular in form, branched,
and ‘ polype-like’—to use the phrase of Desor—that its erosion by a
pro- ssive glacier is not easily conceivable, but is readily so on the sup-
position of an intersection of fissures combined with aqueous erosion.

The Lago di Como is about forty miles long, and its body or northern
half three miles wide; its surface above the sea-level is 695 feet; its
greatest depth is 1,926 feet. Near the middle it divides into two long
diverging arms (like the legs of a pithed frog), pointing southwards, and
separated by the promontory of Bellaggio. The distance across between
Lecco and Como, the extremities of the two branches, is no less than
fifteen or sixteen miles. Upon the glacier-erosion hypothesis how is
this great fork to be explained? If the glacier of the Adda Valley was
adequate to the erosion of the profound lake, how did it spare the com-
paratively insignificant barrier of the promontory and separate into two
diverging branches? If physical features are to be admitted as of any
weight in the argument, it would seem as clear that the fissures existed
before the descent of the glacier as if the eye of man had witnessed the
fact.

The Lago di Garda, the largest of the Italian lakes, is about fifty-five
miles long. Its form is irregularly gourd-like, the northern part or
neck being three or four miles wide, and the southern or expanded
portion fifteen miles, the whole covering an area of about 140 square
miles: surface 227 feet above the Adriatic. Its depth is variously
stated, as being 951 feet (H. de Beaumont), and 1,992 feet (Murray’s
Hand-Book). ‘The latter would be most in accordance with the depths
of the lakes Maggiore and Como, the Adriatic, as far south as Ancona,
nowhere exceeding fifty-nine fathoms. The basin is bounded on the
south by the enormous moraine stretching across between Lonato and
Sommacampagna. Although now fed by the Sarca, and disconnected
from the Adige, it was the great glacier of the latter valley which
formerly contributed most to fill the basin of the Garda. A remarkable
point in the physical geography of the case, and bearing on the argu-
ment, requires to be noticed. The Valley of the Adige, with the ex-
ception of the Pass of Calliano, is very open between Trent and
Roveredo, and more especially open on its west side towards the Valley

VOL. II. UU

658 - GLACIER-EROSION THEORY

of the Sarca. Below Mori it is crossed by a formidable barrier, com-
posed of a lofty ridge, thrown off from Monte Balbo. This ridge is
traversed by the narrow fissure and gorges of La Chiusa, bounded by
mural cliffs like those of the defile of the Avon at Clifton, but greatly
more contracted. The river Adige continuesits straight course through
the defile, to emerge above Volgarne and then pass on to Verona; but not
a trace of an erratic block derived from the head of the valley, or of
any other glacier indication, has been detected below the commence-
ment of the defile. The ancient glacier of the Adige appears to have
refused the gorge, and to have been deflected at right angles to its
previous course, to pass westwards into the Valley of the Sarca, and
thence into the chasm of the lake-basin of the Garda. The fact is
strongly dwelt upon by Mortillet (‘Ancient Italian Glaciers,’ p. 25), and
the rationale of the phenomenon is magnificently illustrated by the
chaotic accumulation of the blocks and debris of the secondary moraine,
called the Slovino di San Marco, a spot which struck Dante with awe,
but which does not yet appear to have attracted the notice of English
glacialists, who maintain the erosion-hypothesis of Alpine valleys and
lake-basins.

It has been shown above, that an hypothesis of the Dead Sea having
been eroded by glacier-action would be beset, to say the least, by very
formidable difficulties. Excluding the facts that it occurs in an area of
depression, and on a longitudinal instead of a transverse valley, there is
much in common between it and the basins of the Italian lakes, viz. :
extreme depth, great length compared with width, and chasm-shaped
transverse section. Those who have faith in the order of nature will
be strongly disposed to infer that the causes which operated on the pro-
duction of the one have had a share in that of the others.

Each of the Italian lakes, taken apart, offers difficulties to the glacier-
erosion hypothesis. The present Valley of the Adige, from its head
down to Verona, shows not a trace of a lake-basin. That an enormous
glacier formerly descended it, is beyond question; and equally so that
that glacier must have greatly widened and deepened the channel along
which it moved. But whatis the dynamic rationale of the new function
it acquired near the termination of its journey, of ploughing down into
the bowels of the earth to a depth of nearly 2,000 feet, and in the case
of Lago Maggiore 2,605 feet? Can such a result be mechanically
sustained by mathematical calculation applied to the conditions of the
problem? or does it accord with the observed effects of existing Alpine
glaciers near their terminations ?

I have already alluded to the difficulty in the case of the Lago di
Como. Without pre-existing fissures to guide its course, by what
means could a progressive glacier have split, so as to excavate, the
diverging branches of Como and Lecco? Why was the promontory of
Bellaggio left intact? In Lago Maggiore 12} miles intervene between
Santa Caterina, where the depth is 2,605 feet, and Santo Calande,
where the glacier, emerging from the lake, was delivered on the plains
of Lombardy, yielding, as already stated, a gradient of 200 fect per
mile. Without reference to the fact that the bottom of the glacier,
under the hypothesis in question, must have here performed the mecha-
nical duties of an auger or a plough-share, could the vis a tergo, under
the conditions of the problem, have propelled the bottom stratum of

OF LAKE-BASINS. 659

ice up such an incline? Before admitting this, it would be desirable,
in this case also, to see the practicability of the result established by
mathematical calculation on mechanical principles.

Further, one of the greatest of the ancient Piedmontese glaciers,
fed by the snows of Mont Blanc and Monte Rosa, descended the Val
d’ Aosta, to stretch far into the plain, and leave the enormous moraine of
Ivrea, close upon the Po. The course of this glacier is marked by no
great lake-basin, like those of lakes Maggiore and Como. If the latter
were eroded by glaciers, how is the absence of a lake-basin in the case
of the former to be explained? The same argument will apply to the
important moraine of Rivoli near Turin, and to the smaller moraines
between the Stura and the Dora Riparia, the whole of which are
unaccompanied by rock-basin lakes. Superficial moraine-tarns are
beside the question.

The expressions used by me did not differ materially from those
employed by that eminent philosopher Sir John Herschel :—‘ In the
upheaval of any extensive tract of land from the sea, hollows fitted for
lake-basins cannot fail to be left. If the upheaval be rude and pa-
roxysmal, resulting in the formation of mountain-chains, and accom-
panied with fracture and dislocation of the strata, such hollows will be
deep, precipitous, and narrow, in proportion to their length. Such is
the general character of lakes in mountainous regions—of the Swiss
lakes for instance, of those of North Italy, &c.’ (Physical Geography,
‘Encye. Brit.’ vol. xvii. p. §91). Mortillet also, who maintains the
questionable hypothesis that the lake-basins were first filled up by silt
and then scoured out by the descent of the glaciers, expresses himself
thus, in reference to the Italian lakes :—

‘Le dernier soulevement des Alpes, qui a eu une puissance énorme,
a dt produire dans les fonds des vallées de grandes inégalités de sol, de
nombreux bassins ; en effet nous voyons ces vallées étre une succession
plus ou moins fréquente d’étranglements et d’élargissements,’ &c.
(‘ Anciens Glaciers Ital.’ p. 17.)

My object on the occasion referred to was to endeavour to account
for the remarkable difference between the Himalayahs and the Alps in
the important physical feature of lake-basins, which are absent in the
transverse valleys on the southern sides of the former, while they abound
in the latter. With the exception of greater and lesser elevation, and
magnitude of development, the two chains repeat each other in all their
primary orographic conditions. It seemed clear that the difference
must depend upon some secondary attribute, which might or might not
be present in a mountain chain. The enormous frontal moraines which
bound all the Italian lakes to the south made it certain that their
basins had been in the track of glaciers of vast magnitude, and which
had existed during long protracted periods. Conversely, lake-basins
were wanting in corresponding situations in the Indian mountains, and
glaciers, according to every aspect of the evidence, had never descended
so low there. What then was the relation, in the positive and negative
sense, of glaciers to the two cases? The supporters of the erosion-
hypothesis maintain that the lake-basins were mechanically excavated
in the solid rock by the grinding action of the glaciers in motion,
working vertically.

The first objection to this view is that lake-basins are wanting imme-

uu2

660 GLACIER-EROSION THEORY OF LAKE-BASINS.

diately behind the enormous moraine of Ivrea and behind that of
Rivoli. If the cause was equal to produce the effect in the other
instances, why did it wholly fail in these? The onus probandi rests
with the advocates of erosion. The objections of the next class are
still more formidable. Glaciers move and erode under the law of
gravitation. Take the enormous Humboldt glacier of ‘Peabody Bay’
described by Dr. Kane. Between Capes Forbes and Agassiz it presents
a front of cliff 300 feet high along a stretch of sixty miles, forming a
stream of ice which is propelled uninterruptedly from the interior into
the sea. ‘The general configuration of its surface showed how it
adapted itself to the inequalities of the basin-country beneath. There
was every modification of hill and valley, just as upon land’ (Kane,
p: 358). The stream of ice in motion in this case exerts a planing action
upon the rocks below by grinding, smoothing, and polishing. But
there is no penetrating effect along contracted lines of excavation. The
actual results of past conditions of the same nature are disclosed to us
in Norway at the present day on surfaces which were in the track of
ancient glaciers moving seawards. The area is ground, scooped, and
polished, but the major inequalities remain.

The same conditions, under certain modifications, apply to glaciers
descending from mountain-chains. Instead of spreading out horizontally,
there they are pent up along contracted lines in Alpine valleys, accu-
mulated in masses several thousand feet in thickness, and therefore
exerting an enormous vertical pressure. The scouring, polishing, and
eroding effect is thus proportionately augmented, but the nature of the
action is not changed—it is still planing, not penetrating. Take the
chasm at the bottom of the incline, as in the case of the Lago Maggiore,
within a few feet of half a mile in depth. What new mechanical power
did the ice acquire there at the end of its journey of excavating so pro-
found an abyss? And, to repeat what I have already asked, could the
bottom stratum of ice, under the enormous vertical pressure to which
it was subjected, have moved up the incline in the given case? Ob-
servation on modern glaciers by the most experienced of all glacialists,
the Swiss philosopher, has shown that a glacier at its terminus pushes
the frontal moraine ahead of it, but that it exerts no peculiar excavating
power at that point. So far as I] am aware, not one of the Swiss geo-
logists has come forward as a convert to the erosion-hypothesis. Toa
man, when they have spoken, they have pronounced against it.

The view which I advocated was, that, both in the Alps and in the
Himalayahs, the lake-basins existed in the transverse valleys before the
descent of the glaciers; that, in the former case, they filled the lake-
basins, thus preventing them from being silted up, while in the latter,
glaciers not having descended, the basins remained open, and thus were
levelled by accumulated debris. Assuming that the basin-fissures
existed before the glaciers, the explanation would account for the con-
trary phenomena in the two mountain-chains.—The Reader, March
5, 1864.

INDEX.

Sane aint
ABB BOS
ee te, fossil human jaw, ii. | Attock, fossils from, i. 414
601, 613 Aucklandia Kostos, description of, i. lv

Aceratherium, characters of, ii. 361

African Elephant. See Hlephas

Agassiz, his glacial theory, 11. 581, 650

Aigoceros Faleoneri, i. 580

Algeria, fossils from, ii. 682

Alps, lakes of, ii. 648

American fossil Elephant, ii. 234

Amphibos acutiformis, i. 23, 280, 547,
dd4

— Antilopinus, i. 28, 280

— elatus, i. 28, 280

Amphicyon ? i. 416

Ampullaria glauca, i, 389

Amyxodon, i. 331

Anca, Baron, discovery of Sicilian caves,
ii. 551

Anisomerous, ridge-formula, ii. 10, 82,
144

Anoplotherium, remarks on genus, i. 196

— posterogenium, i. 191

— Sivalense, i. 190. See Chalicotherium

Anthracotherium Sewalik? i, 35, 149,
296

— Silistrense, i. 508

— Velaunun, i. 508

Antilope gyricornis, 1. 23, 281

— Paleindicus, i. 28, 281, 290, 555

— picta, i. 281

Antilopidee, Sewalik, i. 281, 555

Antiquity, remote of Man. See Pri-
meval Man

— — argued by Falconer in 1836, ii.
576

— — first accepted by leading geolo-
gists in 1859, 11. 598

— — in India, i. 1, 388; ii. 576, 646

Antoletherium, i. 416

Arctomys Tibetana, 1. 583

Aryicola, fossil from Brixham, ii. 497

— — Gower, ii. 525

Asiatic Soe. Bengal. See Museum

Assafeetida, description and discovery of,
1. ly, 573

Athalamia, a new genus of plants i. lvi

Atlas. See Vertebre

Aya. See Burmah

Aves, Sewalik, i. 23, 297, 554

Axis. See Vertebre

Aye-Aye, lower jaw, S&c., ii. 448, 446

ACON Hole. See Caves
Bacton Bear. See Ursus

Baker, discovery of Sewalik fossils, i, 5

— on Sewalik Antilopide, i. 290

— — — Antoletherium, i. 416

— — — Carnivora, i. 339

— — — Quadrumana, i. 298

— — — Rhinoceros, i. 158

Baker’s skull of Eleph. Ganesa, i, 453

— — Rhinoce. platyrhinus, i. 157

Balznidze of Red Crag, ii. 60

Balzenoptera Cortesii, ii. 61

— Cuvieri, ii. 61

Beaumont on gravels of Moulin-Quig-
non, ii. 603

Beckles, discovery of Purbeck fossils,
li. 408

Bijli ki har, i. 4, 174

Birds, Sewalik, i. 23, 554

Bison. See Bos

Bison latifrons, ii. 223

— priscus from Folkestone, ii. 567

— — — Gower, ii. 425

— — identical with existing Aurochs,
li. 634

— Sivalensis, i. 25, 280, 555

Blainville, dentition of Proboscidea, i.
49; 11.7

— on E, primigenius, ii. 79

— on fossil Rhinoceros, ii. 317, 320

— subdivision of Elephants, i. 65

Bleadon. See Caves of Mendip

Bologna. See Museum

Bos bombifrons, ii. 481

— Etruseus, ii. 481

— Namadicus, i. 28, 280, 286, 545; ii.
577, 643

— occipitalis, 1. 28, 280, 281

— Paleindicus, i. 23, 280, 284, 546; ii.
577, 642, 643

662

BOS

Bos priscus, ii. 481. See Bison

— fossil from Brixham, ii. 494

— — Tibet, i. 179

Bosco’s Den. See Caves

Botanical collections, i. xxxv, 575

— Memoirs, i. lv

Botany of Cashmeer, i. 576

Boucher de Perthes, flint-implements,
ii. 583, 597

— human jaw of Moulin-Quignon, ii. 602

— researches on Primeval Man, ii. 570

Bovey Tracey Coal formation, i. xliv

Bovidee fossil of Gibraltar, ii. 556

— Nerbudda, i. 21, 28, 280, 545; ii.
577, 643

— Sewalik, i. 28, 280, 546

Bowen’s Parlour. See Caves

Bramatherium Perimense, i. 399, 410,
045

— absence in Sewalik hills, i. 402

Brandt, on fossil Rhinoceros, ii. 319,
320

Bristol. See Musewm

British Museum. See Musewm

Brixham. See Cave

Bronn, distinction between Elephant and
Mastodon, i. 64; ii. 6

— opinion on fossil Elephants, il. 80

— — — Rhinoceros, ii. 320

Brougham, Lord, objection to fossil
Camel of Sewaliks answered, 1. 242

Bruniquel. See Cave

Buckland Dr., on Caye Fauna, 11. 587

Buist, Perim Island fossils, 1. 410

Burdwan Coal plants, i. lv

Burmah, discovery of fossils in, i. 5

Burmah fossils, Burney’s, i. 412, 461

— — Calder’s, i. 412

— — Crawfurd’s, ii. 5, 59

— fossil Crocodiles, i. 413

— — Elephant, i. 113

— — Equus, 1. 527

— — Hippopotamus, i. 142, 499, 529

— — Mastodon, i, 118, 413

— — Merycopotamus, i. 147

— — Rhinoceros, i. 172

— — Tortoises, i. 413

Burney, fossils from Burmah, i. 412,461

Busk, on analysis of fossil bones from
Nile, ii. 684

— — fossils from caves of Gibraltar, ii.
554

— — from caves of Malta, ii. 292

— marine incrustations of Lepralia, ii.
134

— Moulin-Quignon flints and jaw, ii.
610, 612, 614

ALDER’S Burmah fossils, i. 412
Cambridge. See Musewm
Camel, dentition of, i. 280

INDEX.

CAU

Camel, existing species of, i. 241

— Bactrian, 1. 241

— Boghdi, 1. 241

— dromedary, i. 241

— fossil of Sewaliks, 1. 22, 227, 532

Camelopardalis, existing species, 1. 207

— fossilin France, i. 205

— fossil in Perim Island, i. 207, 398,
544

— fossil in Sewalik hills, i. 22, 197, 543

Camelopardalis affinis, 1. 201

— Sivalensis, i. 197, 398, 543

— — bones of extremities, 1. 544

vertebree, i. 197, 398, 543

Camelus antiquus, i. 231

— Bactrianus, i. 241

— Dromedarius, i. 241

— Sivalensis, i. 227, 231, 245, 532

Canis fossil in Brixham Cave, i. 495

— — — Cefn Cave, 11. 542

— — — Gibraltar Caves, ii. 556

— — — Gower Caves, 11. 525

— — — Sewaliks, i. 339, 343, 553

— — — §pritsail-Tor, i. 462

— lupus, il. 525, 542

— occidentalis, ii, 462

— pictus, i. 462

— vulpes, i. 341; ii. 495, 525, 556

Canterbury. See Museum ;

Caoutchoue Tree of Assam, best mode
of tapping, 1. lvi

Capra, fossil in Gibraltar, 11. 556

— — — Tibet, i. 179

— Aigoceros, il. 556

— Falconeri, i. 580

— quadrimammis, i. 580

Carpal bones of Proboscidea, i. 484

Carnivora, fossil in Caves. See Cave,
passim

— fossil in Crag, ii. 58

— fossil in Sewaliks, i. 23, 339, 553

Cashmeer, description of, i. 566

— Deer, i. 576

— expedition to, i. 557

— fossils, i. 567

— Goats, i. 580

— Marten, i. 582 :

— voleanic tract, i. 567

Cautley, discovery of Sewalik fossils by,
i. xxvil, 5

— on Anoplotherium, 1. 191

— — Bear, i. 321

— — Camel, i. 227, 242

— — Crocodiles, i. 344

— — Giraffe, i. 197

— — Hippopotamus, i. 130

— — Mastodon, i. 126

— — Quadrumana, i. 292

— — Sivatherium, i. 247

— — Tiger, i. 315

— — Structure and Fauna of Sewalik
Hills, i. 30

INDEX.

CAV

Cave Bear. See Ursus

— Era, different opinions regarding,
ii. 587, 590

— Fauna, ii. 587

— Hyena. See Hyena

— Lion, ii. 455, 457

— of Bacon Hole, ii. 183, 325, 340, 349,

501

— — Bleadon. See Caves of Mendip

— — Bosco’s Den, ii. 510, 589

— — Bowen’s Parlour, ii. 517

— — Brixham, ii. 486, 591

— — Bruniguel, ii. 630

— — Cefn, ii. 210, 541

— —Crendi. See Cave of Krendi

— — Crow Hole, ii. 519 -

— — Deborah’s Den, ti. 467

— — Dordogne, ii. 627

— — Durdham Down, ii.178, 179, 323,

327, 349, 534

— — France, ii. 626

— — Genista, il. 554

— — Gibraltar, ii. 554

— — Gower, ii. 179, 181, 498, 589

—— Hutton. See Caves of Mendip

— — Kent's Hole, ii. 160, 177, 179,
210, 460, 489, 534

— — Kirkdale, ii. 176, 210, 399, 487

— — Krendi, ii. 300, 305

— — Laugerie-Basse, 11. 628

— — Laugerie-Haute, ii. 628

— — La Madeleine, ii. 628

— — Le Moustier, ii. 628

— — Les Hyzies, ii. 628

— — Long Hole, ii. 400, 525, 538

— — Lundwig, ii. 457

— — Maccagnone, ii. 546, 595

— — Malta, ii. 292

— — Mendip hills, ii. 159, 161, 167,
180, 452, 455

— — Minchin Hole, ii. 181, 184, 325,
340, 352, 507, 589

— — Nice, ii. 594

— — North Hill Tor, ii. 457

— — Oreston, il. 823, 353, 487

— — Paviland, ii. 477, 521

— — Pedrara, il. 465

— — Raven’s Cliff, ii. 519

— — Rocco Rosso, ii. 594

— — San Ciro, i. 544, 594

— — San Teodoro, ii. 465

— — Sicily, ii. 543

— — Spritsail-Tor, ii. 179, 462, 477,
522

— — Wookey Hole, ii. 172, 399

— — Zebbug, ii. 292, 299

Cayman, characters of, i. 347

Cefn. See Cave of Cefn

Ceryidee, existing of India, i. 587

— fossil of Bedford, ii. 487

— in Bosco’s Denin great number, ii. 515

— Clacton, ii. 478

663

COR

Cervidee of Crag, ii. 57, 479

— of Gibraltar, ii. 555

— — Gower Caves, il. 477

— — Mr. Gunn’s Museum, ii. 476

— of Italy, ii. 478

— — Le Puy, ii. 479

— — Norfolk Coast, ii. 479

— — Sewaliks, i. 23, 281, 555

— — Tibet, i. 179

Cervus Bucklandi, ii. 478, 525

— Capreolus, ii. 478, 525

— Cashmeerianus, i. 576

— Clactonianus, ii. 479

— Dama, ii. 480, 555

— Duvancellii, i. 587

— Elaphus, 11. 478, 480, 525, 555

— Eucladoceros, i. 587 ; ii. 472

— eurycerus, il. 477, 525, 542

— Guettardi, ii. 478, 525, 542

— intermedius, ii. 478

— Namadicus, i. 23, 281

— Paleeindicus, i. 28, 281

— Polignacus, ii. 479

— priscus, i. 478, 525

— Rusa, ii. 479

— Sedewickii, ii. 472

— Solilhacus, ii. 479

— Somonensis, ii, 479

— Strongyloceros, i1. 478, 479, 525, 542

— Tarandus, ii. 477, 478, 495, 525, 568

Ceselli See Musewms of Rome

Cetacea of Crag, 11. 60

Chalicotherium, remarks on genus, i.
195

— Goldfussi, i. 194, 221, 222, 223

— Sivalense, i. 22, 190, 208, 523

Cheiromys Madagascariensis, i1.443,446

Chichester. See Musewm of Chichester

Cheerotherium, i. 149

Christol on Rhinoceros megarhinus, il.
314, 319, 328

Christy, Archzeological collection, ii. 631

— Cave researches, ii. 626

Cinchona Plant, introduction into India,
1, XXxix

Clacton Deer, ii. 478

— Rhinoceros, ii. 310, 317, 351

Clay, Sewalik, i. 15, 37

Clift, fossil bones from Burmah, i. 5

Coal, Plants of Bovey Tracey, i. xliv

— — Burdwan, i. ly

Coal, Sewalik, i. 16, 560 s

Colchester. See Museum

College of Surgeons. See Musewm

Colliculi on teeth of Proboscidea, ii. 143

Colossochelys Atlas, i. 297, 359, 556;
li. 573

Conglomerate, Sewalik, i. 15

Cornalia on Cortesi’s cranium of Rhi-
noceros, ii. 814

Corse on dentition of Indian Elephant,
i. 47; ii. 150

664 INDEX.

COR

Cortesi, cranium of Rhin. leptorhinus,
ii. 113, 310, 313, 381

Costus, plant yielding, i. lv

Crag, Carnivora, ii. 58

— Cervide, ii. 57, 479

— Cetacea, ii. 60

— Elephant, i. 444; ii. 129

— Mastodon, i. 467; ii. 26

— Rhinoceros, ii. 56, 345, 354

— Tapir, ii. 56

Crane, fossil of Sewaliks, i. 28, 297, 554

Crawley Rocks Rhinoc., ii. 354

Crawfurd, discovery of fossils in Bur-
mah, i. 5, 59

Crimping in teeth of Elephant, ii. 147

Crivelli on Cortesi’s cranium of Rhi-
noceros, ii. 315, 320

Crocodiles existing of Ganges, i. 355;
li. 485, 639

— fossil of Burmah, i, 413

— — Perim Island, i. 409

— — Sewaliks, i. 344, 555

Crocodilus biporeatus, 1.345, 355 ; 11.485

— bombifrons, i. 355, 555; i. 485

— cataphractus, i. 356; ii. 483

— crassidens, i. 297, 355, 555

— Gangeticus. See Leptorhynchus

— Journei, i. 357; ii. 484

— leptodus, i. 355, 555

— marginatus, ii. 484

— palustris, i. 355; i. 485

— Schlegelii, i. 356; i. 484

— vulgaris, ii. 484

Croizet and Jobert on fossil Rhinoceros,
i. 815, 319

Crow Hole. See Cave

Crustacea, Sewalik, i. 23

Cryptolepis, reformed character of, i. lv

Cuvier on dentition of Proboscidea, i. 49

— distinction between Mastodon and
Elephant, i. 57

— opinion on fossil Proboscidea, i. 44;
li. 106

— — fossil Rhinoceros, ii, 318, 319

Cygnus Falconeri, ii. 300

— musicus, ii. 300

Cymatotherium antiquum, ii. 161

Cyrena, Sewalik, i. 389

ARMSTADT. See Musewm
Darwin, views as to origin of species,
li. 251, 253
Dasyurus, lower jaw, ti. 447
Dauntela Elephant, i. 477; ii. 258
Dead Sea, its level, depth, &c., ii. 655
Deborah’s Den. See Cave
Deccan, fossil Mastodon of, i. 124
Deer. See Cervus and Cervide
Desnoyers on Cave Fauna, ii. 587
Dichobunes, distinction from Anoplo-
therium, i. 225

ELE

Dinotherium, dentition of, i. 86, 406;
li. 4, 176 ;

— microscopic structure of bone, i. 410

— Bayaricum, i. 408

— Cuvieri, i. 408

— giganteum, characters, i. 406

— -— discovery, i. 61

— — section of teeth, i. 85, 425

— — varieties, i. 408

— Indicun, absence of in Sewalik hills,
i. 402

— — discovery, i. 396

— — lower jaw, i. 404, 466

— — teeth, i. 20, 396, 404, 409; ii. 5

— — tooth sections, i. 85, 425

— Keenigii, i. 408; ii. 5

— Pentapotamicum, i. 414; ii, 5

— proayum, i. 408

Dises in teeth of Proboscidea, ii. 146

Doreatherium moschinum, i. 23, 280

Dordogne. See Cave

Dormouse, fossil of Malta. See Myorus

Drepanodon, Blainyille’s specimen, ii,
457

— Bravard’s specimen, ii. 457

— cultridens, 1i. 459

— latidens, ii. 459

— specimens of British origin, ii. 459

— Sivalensis, i. 339, 550

Duckworth’s Perim Island fossils, 1. 409

Durand, discovery of Sewalik fossils, i. 5

— on Sewalik Carnivora, i. 339

— — Monkey, i. 298

— — Rhinoceros, i. 158

Durdham Down. See Cave

Duvernoy on fossil Rhinoceros, ii. 318,
320

CHINODON Becklesii, ii. 410
Edgeworthia, a new genus of plants,

i. lv

Elasmodon, i. 20, 476

Elastic Sandstone, i. 420

Elephants, classification, ii. 14

— dentition, i. 70; ii. 5

— distinction from Mastodon, i. 68, 70;
li. 5, 16

— fossil in America, ii. 234

— — England, ii. 204, 585

— — Sewaliks, i. 20, 43

— — geological age, ii, 188

— — persistence of specific characters,
ii. 251

— — plurality of species, i. 44

— Indian, i. 20, 428; ii. 14, 148

— structure of teeth, i. 70

— subdivision, i. 20; ii. 13

— transition into Mastodon, i. 26 ; ii. 18

Elephas Africanus, ii. 14, 89

— — cranium, ii. 123

— — food, ii. 282

ween ren

1

INDEX.

ELE

Elephas Africanus, fossil, ii. 94, 283, 552

— — lower jaw, i. 440, 441

— — plurality of species? 11. 266

— — ribs, ii. 257

— — ridge-formula, ii. 89

— — section of teeth, i. 75, 422 \

— — vertebrae, ii. 257

— Americanus, i. 56

— antiquus, 11. 14, 147

— — correction concerning, i. 443 ; ii
109

— — discovery of, ii. 81

— — femur, i. 490

— — localities, in Cefn, ii. 542

— — — England, ii. 204

— — — Gibraltar, ii. 557
— — — Gower, li. 525
— — — Maccagnone, ii. 545

— — — Rome, i. 443

— — — San Ciro, ii. 545

— — lower jaw, i. 488, 489, 440; ii.
185, 188

— — ridge-formula, ii. 176

— — skeleton, ii. 187

— — teeth, i. 440, 442, 447; i1.176

— — tusk, ii. 188

— Armeniacus, ii. 14, 247

— — in Italy, ii. 249

— — in Sicily, ii. 250, 552

— Asiaticus. See EF. Indicus

— bombifrons, i. 20, 110; ii. 14, 84

— — cranium, i. 458 ; ii. 123

— — lower jaw, i. 457, 459, 460

— — section of teeth, i. 81

— — teeth, i. 81, 110, 459
— campylotes, i. 55

— Cliftii, i. 20,113; i. 14, 84

—-— palate, i i. 461

— — ridge-formula, ii. 85

— — teeth, i. 81, 108, 461;

— Columbi, ii. 14, 211, 212

— — associated fauna, ii. 231

— — discovery of, 11. 213

— — localities and geolog. age, ii. 230

— — ridge-formula, ii. 227

— — section of teeth, 11. 222

— — skeleton, 11. 227

— — teeth, ii. 219

— Faleoneri, ii. 292

— Ganesa, i. 20; ii. 14, 88

— — cranium, 1. 453; i. 124

— — lower jaw, i. 452, 456, 457, 460

— — section of teeth, 1. 80, 423

— — teeth, i. 80, 423, 455

— Hysudricus, i. 20, 112, 425; 11. 14

— — cranium, i. 425, 428; ii. 123, 124,
126

— — lower jaw, i. 428, 429, 440

— — section of teeth, i. 77, 422

= —— teebhy 1. 44, 422) 426

— imperator, il. 229, 238

— Indicus, ii. 148

11. 84

665
ELE

Elephas Indicus, cranium, ii. 123, 149

— — food, ii. 277

— — lower jaw, i. 440

— — plurality of species? ii. 253

— — ribs, ii. 257

— — ridge-formula, ii, 10, 150, 157,
176, 260

— — sections of teeth, i. 78, 421

— — teeth, ii. 148

— — varieties of, i. 477;

— — vertebre, ii. 257

— insignis, i. 20, 109, 117; ii. 14, 85

— — cranium, i. 448, 451, ‘456, 457

— — from Nerbudda, i. 496; ii. 643

—— — lower jaw, i. 450, 452, 456, 460,
461

— — origin of species, i. 461 ;

— — ridge-formula, ii. 86, 176

— — section of teeth, i. 73, 423

— — teeth, i. 78, 428, 449; ii. 85

— Jacksoni, ii. 239

— Kamenskii, i. 55

— macrocephalus, i. 56

— Melitensis, ii. 14, 251, 284

— — discovery of, ii. 292, 300

— — femur, ii, 303

— — humerus, ii. 303

— — pelvis, ii. 303

— — ridge-formula, ii. 298

— — teeth, ii. 292

— — tusks, ii. 296

— — vertebra, ii. 251, 302

— meridionalis, ii. 14, 104

— — absence from Cave fauna, ii. 490,
535, 592

— — discovery of, ii. 80

— — cranium, ii. 121, 140

— — humerus, ii. 143

— — femur, li. 144

— — localities in England, i. 139, 204

— — lower j jaw, ii. 126, 140

— — pelvis, ii. 142, 144

—— ridge-formula, i. 118, 176

— — scapula, ii. 142

— — teeth.

A. British, i. 443 ; ii. 129
B. Tuscan, i. 447; ii. 109

— — tusks, ii. 119, 390

— minimus, i. 53

— minutus, ii. 80, 105

— Namadicus, i. 20, 115 ; 11.14, 108, 642

— — cranium, i. 485, 456; ii. 125

— — discovery of, i. 435

— — femur, i. 495

— — humerus, i. 480, 496

— — lower jaw, i. 437, 496

— — pelvis, i. 496

— — radius, i. 480, 496

— — resemblance to E.antiquus, ii. 108

— — resemblance to E. Indicus, i. 80;
li. 157

— — teeth, i. 437; ii. 577

li, 254

1. 85

666 INDEX.

ELE

Elephas Namadicus, tibia, 1. 496

— — ulna, i. 480

— odontotyranunus, i, 55

— Panicus, i. 55

— planifrons, i. 20, 111; i. 14, 91,

108

— — cranium, i. 427, 430, 437

— — lower jaw, i. 429, 430

— — premolars, i. 68, 433; ii. 6, 93

— — resemblance to E. meridionalis,
ii. 108

— — ridge-formula, ii. 91

— — sections of teeth, i. 75, 423

— — teeth, i. 431, 441

— primigenius, i. 79, 421; ii. 14, 158

— — cranium, ii. 123, 175, 188

— — earliest head-quarters, il. 245

— — femur, i. 490

— — food, ii. 284

— — in Gower Caves, li. 525

in Italy, ii, 170, 178)-176,, 241,
587

— — localities, ii. 204

— — lower jaw, i. 439; ii. 127

— — pelvis, ii. 175

— — range in time, ii. 239, 586

— — range of prevalence, il. 77

— — ridge-formula, ii. 119, 163, 175,
176, 236

— — section of teeth, i. 79, 421

— — teeth, i. 79, 421,441; i. 158

— — ulna, ii. 175

— priscus, il. 14, 94 ;

— — abandoned as a species, ii. 2651,
592

—— az variety of E. antiquus, 11. 15,
104, 241

— — discovery of by Goldfuss, i. 54;
li. 81, 94

— — localities, ii. 204

—. — teeth, i. 441; ii. 94

— proboletes, i. 55

— pygmeus, i. 55

— Rupertianus, 11. 238

— Sumatranus, ii. 255

— Texianus, li. 217

Emyde, fossil of Burmah, i. 413

— — Sewaliks, i. 23,35, 39, 882, 556

Emys tecta, fossil, i. 882, 556 ; 1. 574

Enamel plates of teeth of Proboscidea,

ii. 146

Enantia, a new genus of plants, 1. lv

Enhydra, dentition of, 1. 332

Enhydriodon ferox, 1. 331, 552

— Sivalensis, 1. 831, 552

Equide fossil of Brixham, ii. 495

— — Burmah, i, 527

— — Cefn, ii. 542

— — Crag, i. 57

— — Gibraltar, ii. 555

— — Gower, ii. 525

— — Nerbudda, i. 186, 525

FRO
Equidz fossil of Niti Pass, i. 580
— — Perim Island, i. 188, 402
— — Sewaliks, i. 22, 186, 524
Equus Asinus, ii. 525
— Caballus, ii. 525
— Namadicus, i. 186, 525
— palzeonus, i. 22, 186, 527
— Sivalensis, i. 186, 524
Esox Hindostanicus, 1. 589
Eucladoceros, i. 587 ; ii. 472
Euelephas, i. 477; ii. 10, 144
Euhyena. See Hyena
Eurycoronine molars, ii. 83, 146
Evans on flint-kniyes of Moulin-Quig-

non, li. 604

i pgemerrages a genus of plants, i.
XXXii

Fauna Antiqua Sivalensis, description of
published Plates, i. 421

— — — Letter-press, i. 43

Fauna fossil of Caves. See Cave Fauna

— — Nerbudda, i. 21; i. 577, 642, 646

— — Perim Island, i. 391

— — Sewaliks, i. 19, and vol. i. passim

Felide, Sewalik, i. 343

Felis catus, ii. 525

— cristata, i. 315, 548

— leopardus, i. 556

— paleotigris, i. xxi

— pardinensis, ii. 59, 556

— serval, ii. 556

— spelea, li. 455, 457, 525, 542

— from Kent's Hole, ii. 457

— from North Hill Tor, ii. 457

Femur of Proboscidea, i. 487. See also
Elephas and Mastodon

Ferishta, notice of fossil bones by, i. 4

Fish, fossil of Sewaliks, i. 23

Fitch. See Museum

Flint-knives, &c. distinction between
spurious and genuine, ii. 605, 611

— from Brixham Cave, ii. 495, 497

— — Grotta di Maccagnone, ii. 549, 595

— — Laugerie-Haute, ii. 629

— — Long Hole, ii. 538

— — Mautort, 11. 612

— — Moulin-Quignon, ii. 601

— — Olivella, ii. 552

— — &t. Gilles, ii. 612

— — Somme Valley, ii. 583, 597, 598

Florence. See Museum

Fluorine in fossil bones, i. 869, 371

Fluvio-marine Crag, its geological age,
ii. 64

Folkestone, fossils from, ti, 353, 564

Forbes, E., on Sewalik Mollusca, i. 26,
389

Fox fossil, Sewalik, i. 341

Freshwater shells of Sewaliks, i. 389

Frogs, Indian, predaceous habits of, i.
590

INDEX. 667

FRU

Fruit-trees, transportation of in ice, i.
lyi

Fulljames, fossils from Perim Island,
i. 5, 891

os a new genus of plants,
i. lv
Ganges Valley, alluvial formation of, i1.
632, 638
— — delta of, borings in, ii. 645
— — deposits, analogy to those of Nile,
il. 637
— — fossils of, ii. 640
— — physical characters of, i. 8; ii. 638
Gavial, fossil, i. 350, 555
Gavials, characters of, ii. 483
Garnett, discovery of fossils at Attock,
1.414
Gelatine, its presence a test of recent
teeth, ii. 613
Geneva. See Musewm
Genista Cave. See Cave
Gervais, opinions respecting fossil Ele-
phants, 11. 80
— — — Rhinoceros, ii. 318, 320
Gharial, i. 350. See Gavial
Gibraltar Caves. See Cave
Gibraltar, former connection with Africa,
li. 559
Giraffe, fossil. See Camelopardalis
Glacial theory of Agassiz, 11. 581
Glacier-erosion theory of Lake-basins,
i. 648
Goats of Cashmeer, 1. 580
Gold in Sewalik hills, i. 41
Gower Caves. See Cave
Gralle, Sewalik, i. 23
Granite in Himalayahs, i. 14
Grantham. See Musewm
Grass Cloth Plant, i. lvi
Grays Thurrock Elephants, i. 441; 11.96,
176, 177, 204
— — Fossils, ii. 199, 201
— — Rhinoceros, il. 310, 336
Greenstone Trap in Himalayahs, i. xxxi,
13
Gulo fossil of Sewaliks, i. 389
Gunn, Rey. J. See Musewm

ALITHERIUM of Malta, ii. 304
Hangool, i. 576
Hare of Tibet, i. 582
Hemibos triquitriceras, i. 23, 280, 546,
555

Hexaprotodon, i. 21, 140, 499; ii. 406
Himalayahs, absence of Lakes in, ii. 648
— aptitude for culture of tea,i. xxx, lv
— geological structure of, i. 11, 13
— physical characters of, i. 9
— upheavement of, i, 28,173, 182

HUM

Hipparion Horse of Sewaliks, ii. 577

Hippohyus Sivalensis, i. 22, 509

Hippopotami, classification, 11. 405

— fossilof Algeria, 11. 632, 635

— fossil of Burmah, i. 142, 147, 498,
529

— — Nerbudda, i. 21, 146, 498 ; 11. 406,
644

— — Nile, ii. 683

— — Sewaliks, i. 21

— — multitudes of, in caves of Sicily,
ii. 544, 548

— — traditions in India regarding, ii.
580, 643

Hippopotamus amphibius, i. 131, 140
i. 405

— annectens, ii. 406

— antiquus, i. 140; ii. 405

— Capensis, i. 406

— dissimilis. See Merycopotamus

— Iravaticus, i. 142, 498, 529; ii. 406

— Liberiensis, ii, 404, 635

— major, 1. 405°

— — canine, i. 502

— — from Algeria, li. 632, 635

— — — Auvergne, ii. 48, 635

— — — Cefn, ii. 542

— — — Folkestone, ii. 564

— — — Gower, il. 525

— — — Sicilian Caves, ii. 545

— — — Val d’Arno, ii. 47, 635

— — humerus, ii. 564

— — inferences from as to Pliocene
climate, ii. 207

— — radius, il. 564

— — scapula, il. 565

— — teeth, ii. 633

— — ulna, ii. 564

— — vertebree, ii. 565

— medius, i. 140; 11. 466

— minor, i. 140

— minimus, i. 140

— minutus from Krendi, ii. 307

— Namadicus, i. 21, 498; ii. 406, 644

— Paleindicus, i. 21, 146, 497, 502;
li. 405, 642, 643

— Pentlandi, ii. 405

— — from Krendi, ii. 301, 305

— — San Ciro, ii. 545

— Senegalensis, ii. 405

— Sivalensis, i. 21, 1380, 142, 499, 503

Horse. See Hgwus

Human bones fossil in Gibraltar Caves,
li. 559

— — in India? ii. 572, 642

— — — Nice Caves, ii. 594

— — — Nile Valley, ii. 635

— — — Payiland Cave, ii. 522

— — — Spritsail-Tor, ii. 524

— jaw of Moulin-Quignon, ii. 601, 613

— works of industry in Gibraltar Caves,
ii. 561, 626 1

668 INDEX.

HUM
Human works in caves of Rocco Rosso,
ii. 594. See Flint-Implements
Humerus of Proboscidea, i. 480. See
also Elephas and Mastodon
Flurdwar, discovery of fossils at, i. xxvill
Hyena Arvernensis? ii. 465
— brunnea, ii. 464
— — correction regarding, i1. 465, 556
— crocuta, il. 464, 552, 556
— fossil from Tibet, i. 179
— notes on living species, ii. 464
— Sewalik, two species, i. xxi
— Sivalensis, i. 343, 548
— spelwa, 1. 465, 495, 497, 525, 542
— striata, i. 464
— synopsis of living, ii. 464
Hyzenarctos Sivalensis, i. 821, 551
Hyenoid Wolf from Spritsail-Tor, ii. 462
Hypisomerous ridge-formula, ii. 9, 82,
144
Hyracotherium cuniculus, ii. 63
— leporinun, ii. 63
Hypsiprymuus, dentition of, ii. 420, 438
Hystrix, Sewalik, i. 23

BEX of Snowy range, i. 580
fossil of Gibraltar, 11. 566

Ilford, fossils from, ii, 160, 162, 165,
178, 370

Imola. See Musewm

India, animated creation beforeman, i. 3

— antiquity of human race in, 1. 1
388 ; li. 576, 646

— first notice of fossil bones in, i. 4.

— probable origin of human race in,
ii. 579

— supposed human fossils in, ii. 572,
580

— timber trees of, i. lvi

Indian Crocodiles existing, i, 355; ii.
485, 639

— — fossil. See Crocodiles

— Deer existing, i. 587.

— Elephant. See EHlephas Indicus

— Esox. See Hsor

— Fauna, analogy between existing and
extinet, 1. 24

— Frogs. See Frogs

— Mythology, i. 2,297, 367, 377; 11.574

— Otters, existing, i. 337

— — fossil, i. 331, 552

— Zoology, notes on, i. 587

Insectivora, Sewalik, i. 23

Ipswich. See Museum

Iron pyrites, Sewalik, i. 15

Trrawaddi, fossils. See Burmah

Iskardoh, visit to, i. 570

Isomerous ridge-formula, ii. 7, 82, 144

,

eee researches on geo-
logy of India, i. 13

LUS

Jager on fossil Rhinoceros, ii. 316, 320

Jordan, valley of, its physical characters,
il. 654

Jubbulpoor, remarkable fossil vertebra
from, i. 418

Jumua, fossils from, ii. 572, 580

Jumnootree, expedition to, i. xxxi

ALOWALA Pass, lignite in, i. 15,
33

Kankar of Nile, ii. 638
— India, ii. 572, 640
Kaup on Fossil Rhinoceros, ii. 316
Kent's Hole. See Cave
Kheeri Pass, i. 15
King, Rev. 8. W. See Musewm
Kirkdale Cavern. .See Cave
Koala. See Phascolarctus
Krendi. See Cave
Kustora, i. 579

AGO di Como, ii. 657
— — Garda, ii. 657

— — Ingano, ii. 657

— Maggiore, il. 657

Lagomys fossil in Brixham Cave, ii.
497, 534

Lakes, absence of in Himalayahs, i.
Xxxi; li. 649

Lake-basins, mode of formation, ii. 648,
660

Lakes of Lombardy, ii. 649, 656

La Madeleine. See Cave

Lamellibranchiata Sewalik, i. 389

Lartet, Caveresearches, ii.626; on Mam-
malia of Glacial period, ii. 490

Laugerie-Basse. See Cave

— Haute. See Cave

Laurillard on Fossil Rhinoceros, ii. 317,
320

Lawford Rhinoceros, ii. 348, 400

Layton, Rev. J. See Musewm

Leeds. See Musewm

Le Moustier. See Cave

Leptorhynchus crassidens, i. 297, 355,
555

— Gangeticus i. 344, 357, 555

— Leptodus, i. 355, 555

Lepus cuniculus, ii. 497, 525, 556

— timidus, ii. 525, 556

Le Puy. See Museum

Les Eyzies. See Cave

Lightning Bones of Tibet, i. 5, 174

Lignite in Sewaliks, i. 16, 17, 28, 560

Limestone in Himalayahs, i. 13

Litorina, species of in Gower Caves, ii.
504, 531

Long Hole. See Cave

Loxodon, i. 20, 477; ii. 10, 88

Lush, fossils from Perim Island, i, 5

—_——Tr er!

INDEX. 669

LUT

Lutra, dentition of, i. 332

— Indica, i. 337

— Paleindica, i. 331, 552

— vulgaris fossil in Gower, ii. 525

Lyell, Sir C., opinion on Cave Fauna,
ul. 587

Lyons. See Museum

Lyons Rhinoceros, 1. 369

ACCAGNONE. See Cave
Machairodus. See Drepanodon
Maidstone. See Musewm
Malaga Rhinoceros, ii. 310, 360

Malta Caves. See Cave
— Dormouse. See Myoxus
— Elephant. See Hlephas Melitensis

— former connection by land with Sicily
and Africa, 11. 553

Mammoth. See Hlephas primigenius

Marcel de Serres on Fossil Rhinoceros,
11.315, 319

Markhor Goat, i. 580

Marmot of Tibet, 1. 583

Marten of Cashmeer, i. 582

Mastodons, classification, i. 20, 87 ; ii.
11, 14

— dentition, i. 87; 11.5

— distinction from Elephant, i. 68, 70 ;
ii. 5, 16

— division into Trilophodon and Tetra-
lophodon, first proposed by Dr. Fal-
coner, li. 216

— fossil in America, ii. 74

— — Britain, i. 1

— — Europe, ii. 20

— — Burmah, i. 59; ii. 85

— — Perim Island, i. 402, 470

— — Sewaliks, i. 20, 43

— — Val d’Arno, ii. 47

— geological age, ii. 45

— plurality of fossil species, i. 45

— structure of teeth, 1. 70; ii. 5

— subdivisions, i. 20, 87; ii. 11

— transition into Elephant, i. 26; ii. 18

Mastodon Andium, i. 99; ii.14, 15

— — cement on teeth, i. 58, 84

— — cranium, i. 473

— — lower jaw, i. 99, 466, 473

— — teeth, 1.99, 466, 473; 11.8

— angustidens, i. 89; ii. 14, 21

— — different forms included under, ii.
26

— — geological age, ii. 45

— — in Winterthur Museum, ii. 73

— — in Zurich Museum, ii. 68

— — lower jaw, ii. 42

— — restriction to Simorre form, ii. 21,
26

— — ridge-formula, i. 98

— — skeleton, ii. 45

— — teeth, i. 89, 472; ii. 22

— Arvernensis, ii, 14, 26

MAS

Mastodon Arvernensis associated fauna
ii. 47, 50

— — discovery of, i. 59

— — geological age, ii. 45

— — localities in Britain, ii. 49

— — lower jaw, ii. 42, 126

— — ridge-formula, ii. 39

— — skeleton, ii. 45

— — teeth, 1. 467; 11. 28

— — upper Jaw, i. 467

— Australis, i. 64, 105; ii. 14, 271

— Borsoni, ii. 12, 14, 71

— breyirostre, ii. 15, 32

— Elephantoides, i. 40, 59, 74, 82, 129,
461; 11.9, 83, 85. See Hlephas Cliftii,
E. insignis, and Mast. latidens

— Gaujaci, ii. 24

— giganteus, i. 55, 407

— Humboldtii, ii. 14

— intermedius, i. 64

— latidens i. 40, 59, 402; ii. 14

— — lower jaw, i. 120, 463

— — palate, i. 118, 463

— — section of teeth, i. 83, 424

— — teeth, 1. 119, 128, 463, 471

— longirostris, i. 59, 107; ii. 14, 23

— — geological age, ii. 45

— — lower Jaw, i. 466, 472; ii. 41

— — skeleton, ii. 45

— — teeth, i. 107, 468, 472

— maximus, i. 55

— minutus, i. 57

— mirificus, ii. 229

— Ohioticus, i. 56, 84, 87, 424 3 11.14

— — calcaneum, i. 493

— — cranium, i. 476; ii. 6

— — discovery of, i. 55

— — lower jaw, i. 466 ; ii. 6

— — palate, i. 474 ; ii. 6

— — ridge-formula, ii. 176

— — section of teeth, i. 84

— — teeth, 1. 87, 466, 474; i1. 6

— — tusks, ii. 6

— Pandionis, i. 124; ii.
Plate xxxiv. of vol. i.

— Perimensis, i. 122, 440 ; ii, 12, 14

— — cranium, i. 470

— — lower jaw, i. 122, 464

— — palate, i. 464

— — teeth, i. 122, 472

— Podolicus, i. 64

— Pyrenaicus, ii. 14

— Simorrense, ii, 24

— Sivalensis, 1. 83, 117, 127; ii. 14,
25, 29, 44, 81, 87

— — cranium, i. 128, 464, 467, 471

— — lower jaw, i. 464, 467, 471

— — ridge-formula, ii. 88, 176

-— — section of teeth, i. 83, 425

— — teeth, i. 117, 465 ; ii. 44

— — upper jaw, i. 117, 465

— Tapiroides, i. 57, 103 ; ii. 14

12, 14, and

670 INDEX.

MAS

Mastodon Tapiroides of Nicolaieff, 11.
65

— Turicensis, i. 63

— Vellavus, ii. 20

— Vialettii, 11. 20

Mautort gravel-pits, li. 612

McEnery’s Plates of Kent's Hole Fos-
sils, 11. 177. 460, 489

Mecistops Bennettii, i. 356 ; ii. 483

Megalochelys Sivalensis, i. 359

Megantereon. See Drepanodon

Melania corrugata, i. 389

— thiarella, i. 389

Meles Taxus, fossil, li. 525

Mendip Caverns. See Cave

Meryecopotamus, characters of teeth, i.
138, 508

— derivation of term, i. 138

— from Attock, i. 416

— from Burmah, i. 147

— from Sewaliks, i. 21, 188, 146, 503

— relation to Anthracotherium, i. 508

— dissimilis, i. 21, 138, 146, 416, 503,
505 ; ii. 406

— — var. major, i. 505

— — — minor, i. 507

— nanus, i. 416 ; 11. 407

Metacarpal bones of Proboscidea, i. 487

Metatarsal bones of Proboscidea, i. 495

Mewslade Bay, raised beach of, 11. 501

Microlestes antiquus, 11. 414, 416

Milan. See Musewm

Minchin Hole. See Cave

Mollusca from Ganges deposits, ii. 640,
644

— — Gower Caves, ii. 504, 520, 525,
531

— — Italian fluvio-lacustrine deposits,
ii. 191

— — Maccagnone Cave, ii. 550

— — Nile deposits, ii. 636

— — Sewalik hills, i. 23, 26, 389

Monitor, Sewalik, i. Pl. xxxii.

Monkeys, fossil in Sewaliks, i. 292

— astralagus, i. 298 ; ii. 578

— canine, i. 304, 808 ; 11. 578

— lower jaw, 1. 300

— upper jaw, i. 298

Montpellier Rhinoceros, ii. 368

Mooztagh Mountains, visit to,i.570 ; ii.
648

Moschus aquaticus, separation of meta-
carpal bones in, 1. 196

Moulin-Quignon Deposits, opinions as to
age, 11. 623

— — flint-implements, ii. 605

— — human jaw, il. 613

— — — verdict on i. xlvii; ii. 621

Mukna Elephant, i. 477 ; 11. 258

Mule between Yak and Cow, i. 581

Mus, fossil of Sewaliks, i. 23

— rattus fossil in Gibraltar, il. 556

MUS

Museums and Collections, description of
Specimens.

Museum, Aberdeen, i. 213, 523

— Arezzo, iil. 1738, 379

— Asiat. Soc. of Bengal, i. 109, 117,
142, 169, 186, 206, 245, 273, 281, 357,
381, 512

— Belfast, ii. 482

— Bologna, ii. 363, 478

— Bristol, ii. 178, 179, 349

— British, almost all the specimens
described in vol. i., as figured in
the ‘Fauna Antiqua Sivalensis,’ i.
329 ; ii. 160, 161, 162, 177, 178, 222,
228, 247, 336, 370, 453, 456, 457,
461, 478, &e.

— Cambridge, ii. 162, 169, 170, 174,
182, 183

— Canterbury, i. 100 ; ii. 177

— Cautley. See vol. i. passim

— Chichester, ii. 181, 184, 187

— Christy, ii. 631

— Colchester, i. 62 ; ii. 178

— College of Surgeons, ii. 150, 153, 155,
160, 163, 162, 171, 220, 262, 302, 337,
353, 445, 461, &e.

— Darmstadt, i. 222; ii. 166,172,173,175

— Due de Luynes, ii. 195

— Duckworth, i. 409

— Ewer, i. 320, and Plates F. A. S.

— Fitch, ii. 33, 36, 847, 355

— Florence, ii. 109, 143, 144, 355, 379,
465, 481

— Frazer, Plates F. A. S.

— Fulljames, i. 391, and Plates F. A.S.

— Geneva, ii. 226

— Geological Society, i. 391; ii. 174,
178, 180, 183, 185, 459, 564

— Grantham, Mr., ii. 172, 401, 478

— Gunn, Rey. J., ii. 99, 128, 129, 140,
142, 143, 174, 175, 180, 182, 346,
849, 355, 398, 471, 476, 479, &e.

— Imola, ii. 394

— India House, i. F. A. S. ; ii. 149

— Ipswich, ii. 155, 183

— King, Rey. S. W., ii. 164, 467, 479

— King’s College, London, ii. 149,
152, 161

— Layton, Rey. J., ii. 345

— Leeds, ii. 403

— Le Puy, ii. 367, 399, 402, 479

— Lyons, ii. 369

— Maidstone, ii. 402

— Milan, ii. 101, 113, 198, 381

— Montpellier, ii. 368

— Munich, i. 223

— Naples, ii. 182

— Newsted, Mr., ii. 174, 457

— Nice, ii. 370, 466

— Norwich, ii. 134, 144, 188, 466, 467

— Oldham, i. 147, 414; ii. 5

— Oxford, ii. 179, 354, 460, 479

INDEX. 671

MUS

Museum, Paris, ii. 479

— Pisa, ii. 359, 394

— Practical Geology, ii. 169, 183

— Prestwich, ii. 165, 169, 478

— Rome, ii. 102, 170, 178, 175, 182,
183, 184, 185, 371, 376, 478

— Saffron Walden, ii. 178, 184, 265, 402

— Spurrell, Mr., ii. 398, 401, 457, 478

— Stuttgart, ii. 398, 401

— Swansea, ii. 325, 349

— Swayne, ii. 175

— Syracuse, ii. 182, 188

— Taunton, . 159, 161, 167, 172, 177,
179, 180, 452, 455

— Turin, ii. 162, 172, 1738, 187, 192, 380

— Verona, ii. 478

— Vicenza, ii. 397

— Winterthur, ii. 73

— Wyatt, ii. 478

— York, ii. 176, 457

— Zurich, ii. 68, 71, 173

Musk Deer, description of, i. 579

Mustela of Cashmeer, i. 582

Mustelide, Sewalik, i. 23, 331

Myoxus Melitensis, ii. 300, 306, 308

Mythology, Indian, i. 1, 297, 367, 377;

il. 574

AHUN, Fossils at, i. xxvii, 36
Namadicus, derivation, i. 485

Naples. See Museum

Nerbudda, discovery of Fossils, i. 5

— formation, Pliocene, i. 21; ii. 577,
642, 646

— fossil Bear, 1. 321

— — Bovide, i. 284, 545 ; ii. 642

— — Elephant, i. 115, 435, 496; ii. 14,
642

— — Elephas insignis, i. 496 ; ii. 643

— — Fauna, i. 21 ; ii. 577, 642

— — Hippopotamus, i. 498 ; 11. 642

— — Vertebra, i. 418

— Valley, physical characters of, i. 8

Nesti, discovery of Elephas meridionalis,
1.53; ii, 80, 104

— on fossils of Val d’Arno, i. 58; ii. 80

Newsted. See Musewm

Nice. See Cave and Museum

Nicolaieff Mastodon, 11. 65

Nile deposits, analogy to those of the
Ganges, i. 637

— — fossil mammalia in, ii. 633

— — human remains in? ii. 635

— — mollusea in, ii. 636

Niti Pass, geological formation, i. 175

— — Fossils, i. 174, 580

— — Fossil Rhinoceros, i. 177, 517

Norfolk Coast Fossils, ii. 199, 472, &e.

Norwich Crag, geological age, 11. 64

— — Elephant, i. 444; i. 129, 204

— — Mastodon, i. 467 ; ii. 26

— — Rhinoceros, ii. 310, 345, 847

PIG

Norwich Crag. See Museum
Northampton Rhinoceros, ii. 310, 351
North Hill Tor. Seo Cave

LDHAM, Indian Fossils.
sewm

Oreston. See Cave

Otters, Fossil, i. 331, 352

Owen on dentition of Proboscidea, i. 51;
li. 38

— — Distinction between Elephant and
Mastodon, i. 67

— — Elephas meridionalis, ii. 139

— — Fossil Elephants, i. 44, 52; ii.
3, 79

— — Mastodon of Crag, ii. 3

— — Plagiaulax, ii. 431

— — Rhinoceros leptorhinus, ii. 317,
320

Oxford. See Museum

See Mu-

ALAHOTHERIUM among Sewalik
Fossils? i. 25, 204

Paludina Bengalensis, i. 389

— unicolor, i. 389

Payiland. See Cave

Payiland ‘Red Lady,’ ii. 522

Pectenibranchiata, Sewalik, i. 389

Pedrara. See Cave

Pelvis of Eleph. meridionalis, ii. 142

— E. primigenius, ii. 175

Pelvis of Proboscidea, i. 490

Pentalophodon, i. 108 ; ii. 9, 86

Pepper Miss, Perim Island Fossils, i.
404, 464, 466

Perim Island, discovery of fossils in,
1. 5, 395

— — Fossil Antelope, i. 410

— — — Bramatherium,i. 399, 410,545

— — — Crocodiles, i. 409, 410

— — — Deer, i. 410

— — — Dinotherium, i. 396

— — — Giraffe, i. 207, 391, 398

— — — Hippopotamus, i. 391, 410

— — — Mastodon, i. 122, 391, 402,
410

— — — Rhinoceros, i. 121, 391, 410,
517

— — — Sus Hysndrieus, i. 402, 515

— — Fossils, i. 5, 391, 404, 409, 410,
464, 466

— — situation, i. 392

— — Strata, 1. 393

Phacocherus, dentition of, i, 86; ii. 38

Phalanges of Proboscidea, i. 487, 495

Phascolarctus cinereus, dentition of, ii.
436, 443

Phillips, opinion as to age of Caye Fauna,
ii. 587

Phoczena Cortesii, ii. 61

Pignano Skeleton, ii. 187

672 INDEX.

PIS

Pisa. See Museum

Plagiaulax Becklesii, ii. 410

— dentition of, 11. 411

— derivation of term, ii. 410

— disputed affinity, i. 480

— lower jaws, li. 416

— memoirs on, ii. 408, 430

— minor, ii. 410

Planorbis, Sewalik, i. 389

Plates, description of, in ‘ Fauna Antiqua
Sivalensis,’ i. 421

— in teeth of Proboscidea, ii. 143

Pliocene period, climate of, 11. 207

— — division into old and new unten-
able, li. 205

— species of Rhinoceros, ii. 309

Pomel on Fossil Rhinoceros, 11. 318, 320

Ponzi’s collection. See Musewm of Rome

Prangos Grass, description of, 1. 568

Pre- and Post-Glacial Eras of Phillips,
ii. 587 ,

Prestwich on Primeval Man, ii. 570

— Raised beach of Mewslade, 11. 536

— On Valley of Somme, ii. 598

— Researches on Quaternary sands and
grayels, ii. 584, 590. See Museum

Primeval Man and his Cotemporaries,
ii. 570

— — Works of Art by, ii. 626

Proboscidea, Classification of, i. 476; il.
11, 14

— bones of extremities, i. 486

— erania, 1. 475

— development of teeth in, i. 69

— diversity of opinion as to fossil forms,
i, 44

—generic distinctions and nomen-
clature, il. 4 -

— history of fossil forms, 1. 47

—number of fossil compared with
existing species, 1. 24

— Sewalik, i. 20, 43

— structure of teeth, i. 70 ; ii. 148

— Teeth. See Hlephas and Mastodon

— Tusks, i. 474

— Vertebre, i. 478

Pruner-Bey on human jaw of Moulin-
Quignon, ii. 604

Pulmonifera, Sewalik, i. 389

Purbeck fossils, ii. 408, 409

Putorius ermineus, li. 525

— yulgaris, 11. 497, 525

UADRUMANA, Fossil, correction of
published statements concerning,

i. 309

— Fossil of Sewaliks, i. 28, 2938; i.
578. See Monkey

Quaternary deposits, importance of as
regards Antiquity of Man, ii. 581

Quatrefages,on human jaw of Moulin-
Quignon, ii. 603

RHI

ADIUS of Proboscidea, i. 482, 483
Ratel, Sewalik, i. 339

Ravenscliff. See Cave

Red Crag, Elephant of, 11. 206

— — Fossils of, 11.54

— — Geological age of, 11. 64

— — Mastodon of, 11. 53

Red Lady of Payiland, 11. 522

Rein Deer. See Cervus Tarandus

— — Carvings on fossil horns of, 11. 629

Reptilia, Sewalik, 1. 23, 344, 359, 382,
500

Rhinoceros, classification of Pliocene and
Post-Pliocene fossil species, ii. 309

— antiquitatis, reasons for preferring
this name to R. tichorhinus, ii. 311

— — associated fauna, ii. 310, 525, 542

— — characters of, ii. 309, 399

— — dentition of, i. 399

— — localities of, i. 309, 399, 425,
494, 525, &e.

— — notes on, i. 309, 310, 399

— Aymardi, il. 49, 318, 320

— bicornis, ii. 328, 334, 335, 403

— brachypus, ii. 361 .

— cumus, ii. 403

— elatus, ii. 309, 315

— eriques, i. 413

— Etruscus, 11. 309, 354

— — absence of in Cave fauna, il. 535

— — associated fauna, ii. 288, 310

— — astragalus, 11. 367

— — calcaneum, ii. 367

— — characters of, ii. 309

— — cranium, ii. 355, 359, 363

— — dentition of, ii. 354, 364

— — discovery of, 1. 310, 345, 354

— — femur, ii. 367, 555

— — fibula, ii. 366, 367

'“— — humerus, il. 366, 390

— — localities, ii. 310, 348, 354, 555

— — lower jaw, il. 345, 359, 367

— — notes on, ii. 354

— — phalanges, ii. 367

— — tibia, ii. 345, 366, 367

— Goldfussi, ii. 361

— hemitcechus, ii. 309, 311

— — associated fauna, 11. 310

— — cranium, il. 323, 350, 357

— — dentition of, 11. 324, 349

— — discovery of, 11. 322

—- — derivation of specific name, ii.
312

— — femur, ii. 353

— — fibula, ii. 566

— — humerus, ii. 566

— — localities, ii. 310, 317, 325, 340,
349, 350, 351, 352, 497, 542, 566

— — lower jaw, il. 340, 352

— — memoir on, ii. 311

— — skeleton, ii. 353

— — tibia, ii. 353

“yi!

PP ee wre et

INDEX.

RII

Rhinoceros incisivus, ii. 315, 320, 361

— Keitloa, ii. 402

— Kirchbergense, ii. 316, 320, 398

— leptorhinus, 11. 309, 310, 368

— — associated fauna, ii. 113, 288

— — cranium, ii. 369, 381

— — Cuvier's, ii. 113, 310

— — dentition, ii. 370,.382, 395

— — humerus, ii. 390

— — localities, ii. 310, 368, 398, 554

— — lower jaw, ii. 368, 380, 392, 393,
397

— — metacarpus, il. 390

—- — notes on, il. 368

— — radius, ii. 397, 399

— — scapula, i. 390

— — ulna, il. 390

— — vertebra, ii. 390

— Lunellensis, 11. 309

— megarhinus, ii, 309, 310, 319, 828.
See L?. leptorhinus

— Merckii, ii. 309, 316, 320, 398, 399

— mesotropus, il. 309

— Monspessulanus, ii. 315, 319

— Pallassii, ii. 311

— Paleindicus, i. 157, 514, 515

— Perimensis, i. 157, 171, 391, 410,
517

— platyrhinus, i. 157, 613

— pleuroceros, ii. 361

— priscus, ii. 351, 390

— protichorhinus, ii. 319, 320

— Schleiermacheri, ii. 56, 820, 324

— Sivalensis, i. 157, 514

— Simorrensis, ii. 361

— Sondaicus, 1. 516

—tichorhinus. See 2. antiquitatis

— unicornis, ii. 334 a

— Fossil of Ava,i. 172, 413

— — British Bone Caves, ii. 810, 489,
536

— — Brixham, ii. 494, 497

— — Cefn, 11. 542

— — Clacton, ii. 310, 317, 351

-— — Cortesi, ii. 113, 310, 313, 381

— — Crag, ii. 56, 345

— — Durdham Down, ii. 349

— — Filippi, ii. 320

— — Folkestone, ii. 566

— — Forest-Bed, ii. 844, 398

— — Gibraltar, ii. 555

— — Gower, ii. 400, 425, 525

— — India, i. 21

— — Malaga, 11. 310, 360

— — Nerbudda, ii. 519

— — Niti Pass, i. 173, 177, 517

— — Norfolk Coast, ii. 310, 347, 355

— — Northampton, ii. 310, 351

— — Perim Island, i. 171, 517

— — Scinde, i. 172

— — Sewaliks, i. 21

— — Tibet, i. 173, 177, 517

VOL. II.

673

SEW

Rhinoceros Fossil of Val d’Arno, ii. 113
310, 347, 355

Rhynchotherium, ii. 75

Rice Paper Plant, description of, i. lvi

Ridge-formula of Dinotherium, 11. 176

— — Elephas Africanus, ii. 89

— —- — antiquus, 11. 176

— — — Cliftii, ii. 85

— — — Columbi, ii. 227

— — — Indicus, ii. 10, 150, 157, 176,
260

— — — insignis, 11. 86, 176

— — — Melitensis, ii. 298

— — — meridionalis, ii. 118, 176

— — — planifrons, 11. 91

— — — primigenius, ii. 119, 163, 175,
176, 236

— — Mastodon angustidens, i. 98

— — — Arvernensis, ii. 39, 176

— — — Ohioticus, 11. 176

— — — Sivalensis, ii. 88,176

Ridges in teeth of Proboscidea, 11. 143

Roceo Rosso, See Cave

Rodentia Fossil of Sewaliks, i. 23

Rome. See Museum

AFFRON WALDEN. See Museum
Saint Gilles Gravel Pits, 11. 612

Salt range at Kohat, i. 560

San Ciro. See Cave

San Teodoro. See Cave

Sandstone, elastic, i. 420

— of Sewaliks, i. 15, 32; 38

Saurzechmodon from Purbeck, ii. 410

Scapula of Proboscidea, 1. 482

Schists argillaceous in Himalayahs,i. 13

Schlegel, views as to plurality of Indian
Elephant, ii. 255

Scinde, fossil Mastodon, i. 124

— — Rhinoceros, i. 172

Sewalik, derivation of term, i. 6, 31

— Fauna climatal, inferences from, i. 28

— — extent and peculiarities of, 1.19,
21, 24, 27; i. 577

— — geological inferences from, 1. 28

— — mixing of extinct and existing
species in, 1. 25, 355, 388; 11. 574

— — of miocene age, i. 21; ii. 577,
639, 642

— formation between Punjab and Cash-
meer, i. 573 ;

— — extension to Rawul Pindee, i.
558

— fossil Antilopide, i. 281, 555

— — Aves, i. 23, 554

— — Bovide, i. 280, 545, 554

— — Camelide, i. 227, 530

— — Camelopardalis, i. 197, 543

— — Canis, 1. 339, 3438, 553

— — Carnivora, i. 339, 548, 553

— — Corvide, i. 281, 555

— — Crocodilia, i, 344, 555

xXx

674
SEW

Sewalik fossil Elephas, i. 48, 109, 425

— — Enhydriodon, i. 331, 552

— — Equide, i. 186, 524

— — Felis, i. 315, 339, 348, 548

— — Hippopotamus, i. 130, 142, 497

— — Hyena, i. 339, 348, 548

— — Luatra, i. 331, 552

— — Mastodon, i. 48, 117, 126, 464

— — Mollusca, i. 23, 26, 389

— — Monitor, i. Plate xxxii.

— — Mustelide, i. 331

— — Proboscidea, i. 48

— — Quadrumana, i. 292; 11. 578

— — Reptilia, i. 344, 359, 556

— — Rhinoceros, i. 157, 513

— — Sivatherium, i. 247, 537, 544

— — Suide, i. 22, 415, 508

— — Tortoises, i. 859, 382, 556

— — Ursus, i. 821, 551

— fossils, discovery of, i. xxvii, 5

— — in marl, i. 35

— — in sandstone, i. xli, 38

— — number and peculiarities, i. 19,
21, 24, 27; ii 577

— Hills, geological structure of, i. 12,
15, 19, 32

— — mode of formation, i. 8, 28

— — physical characters, i. 7, 14, 30

— — range and extent, i. 7, 30

— — section of, i. 19

— marl, i. 34

— sandstone, i. 15, 32

— strata of miocene age, i. 21; il. 577,
639, 642

— — no human remains in, i. 26

Sicily. See Cave

— former connection with Africa and
Malta, i. 552, 596

Sivatherium giganteum, i. 247, 537, 544

Smilodon, ii. 457

Somme, Valley of, ii. 570, 583, 597

Southwold, fossils of, ii. 181

Spermophilus concolor, ii. 453

— erythrogenoides, ii. 452

— erythrogenys, ll. 453

— Eversmanii, ii. 453

— fossil from Mendip hills, ii. 452

— — Salisbury, ii. 453

Spilsbury, discovery of Nerbudda fos-
sils, i. 5

Spritsail-Tor. See Cave

Spurrell. See Musewm

Stegodon, i. 20, 476 ; ii. 9, 82

Stenocoronine molars, ii. 83, 146

Stuttgart. See Museum

Suffolk Crag, its geological age, 11. 64

— — fossils of, il. 55

Suide: fossil of the Crag, ii. 57

— — Perim Island, i. 402, 513

— — Sewalik, i. 22, 415, 508

Sus, fossil from Folkestone, 11. 568

— — — Gibraltar, 11. 555

INDEX.

URS

Sus, giganteus, i. 508, 510

— Hysudricus, 1. 402, 510, 513
— pusillus, i. 415

— Serofa, i. 509; ii. 625

— Sivalensis, i. 509

Swansea. See Musewm
Swayne. See Museum
Syracuse. See Musewm

fig fossil of Crag, ii. 55, 56
Tarsal bones of Proboscidea, i. 492
Taunton. See Musewm
Tea, cultivation introduced into Hima-
layahs, i. xxx
Teak forests of Tenasserim, report on,
i. lvi
Teeth, constituent parts of in Probos-
cidea, i. 70
Tehr Goat, 7. 580
Testudo elephantopus, dimensions of, i.
363
— Sewalik, i. 381
Tetracaulodon, discussion on, i. 62
Tetraconodon magnum, i. 149
Tetralophodon, i. 20, 106, 476; ii. 7
Tetraprotodon, i. 21, 140, 497; ii. 405,
406
Thames Valley, deposits, ii, 582. See
Grays Thurrock
Thylacinus, lower jaw, ii. 447
Thylacoleo carnifex, dentition of, ii. 437
Tibet, expedition to, i. 557
— fossil bones from, i. 179
Tibetian Hare, i. 582
— Marmot, i. 583
Tibia of Proboscidea, i. 491
Tiger. See Felis
Timli Pass, fossils of, i. 16
Tortoise gigantic. See Colossochelys
Atlas
Tortoises, Sewalik, i. 359, 382
Transitional Mastodons, i. 68, 74
Triconodon from Purbeck, ii. 409
Trilophodon, i. 20, 87, 476; ii. 7
Trimmer, Jos., on Cave-era, ii. 588
Trionyx fossil from Burmah, i. 413
— — Sewalik, i. 23, 35, 39, 556
Trotter, fossils from Attock, i. 414
Tusks of Proboscidea, i. 474; ii. 188
Typhlodon,i.23 ~

LNA of Proboscidea, i. 482, 483
Unio favidens, i. 889
— marginalis, i. 389
Ursitaxus Sivalensis, 1. 339
Ursus Arctos, ii. 468, 525
— Arvernensis, ii. 58
— ferox, 11. 468
— Isabellinus, ii. 468
— maritimus, ii. 468

INDEX. 675

URS

Ursus Namadicus, i. 321, 552

— priseus, ii. 466 525

— Sivalensis, i. 331, 551

— speleeus, li. 466, 469, 495, 525, 542,
548

— notes on, ii. 466

— fossil from Bacton, ii. 469

— — Brixham, ii. 495

— — Cefn, ii. 542

— — Cromer, ii. 466

— — Deborah’s Den, ii. 467

— — Gibraltar, ii. 556

— — Gower, ii. 525

— — Kent’s Hole, ii. 469

— — Maccagnone, ii. 548

— — Nerbudda, i. 321, 552

— — Nice, ii. 466

— — Sewaliks, i. 321, 551

— — South America, i. 329

Urtica frutescens, fibre of, i. lyi

AL D’ARNO, fossil Elephant, i.
446; 11. 104
— — — Fauna, ii. 189
— — Hippopotamus, ii. 635
— — Mastodon, ii. 47
— — Rhinoceros, ii. 310, 345, 347
Valliculee in teeth of Proboscidea, ii.
143

THE END.

ZUR

Varanus Sivalensis, i. Plate xxxil.
Vertebre of Elephas Africanus, 1i. 257
— Elephas Indicus, ii. 257

— Elephas Melitensis, i. 251, 302

— Hippopotamus, i. 503

— Proboscidea, i. 478

Vicenza. See Musewm

Voleanic tract in Cashmeer, i. 567

ALKER’S, Dr. H., drawing of An-
toletherium, i. 416
Water-Elephant of India, ii. 580
Winterthur. See Museum
Wolf. See Canis
Wood’s, Col., researches on Gower Caves,
li. 498

Wookey Hole. See Cave

AK, Mule between it and Cow, i. 581
York. See Museum

EBBUG Fossils. See Cave
Zoology of Cashmeer, notes on, i.
576
— — India, i. 587
— — Tibet, 1. 581
Zurich. See Museum

LONDON

PRINTED BY SPOTTISWOODE AND CO.

NEW-STREET SQUARE

vs it
| % }) ts
i Vv
DAE:
he

tan ee Chee ee) ee ae ie
i
{
7 . “
. 3 x 1
4
.
\
|
j
~
.
n 4
:
e '
aod
¢
s :
-
* : ,