DENTAL MORPHOLOGICAL STUDIES OF TRUE

UNGULATES FROM THE TYPE LOCALITY OF

NAGRI FORMATION, CHAKWAL DISTRICT,

PUNJAB, PAKISTAN

TASNEEM IKRAM

DEPARTMENT OF ZOOLOGYUNIVERSITY OF THE PUNJAB

LAHORE, PAKISTAN

2010

DENTAL MORPHOLOGICAL STUDIES OF TRUE

UNGULATES FROM THE TYPE LOCALITY OF NAGRI

FORMATION, CHAKWAL DISTRICT, PUNJAB,

PAKISTANBy

TASNEEM IKRAM

A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE

REQUIREMENTS FOR THE DEGREE OF

Doctor of PhilosophyIn the Faculty of Life Sciences, University of the Punjab, Lahore,

Pakistan

Supervisor

Prof. Dr. Muhammad Akhtar

Co-Supervisor

Dr. Muhammad Akbar Khan

DEPARTMENT OF ZOOLGY

UNIVERSITY OF THE PUNJAB

LAHORE, PAKISTAN

2010

“The more a man has the wealth of

knowledge and faith the higher will be his

status near God”.

(Al-Quran)

“Keep your thought well composed, and search

the fact of wisdom for it, otherwise mind gets

weary and the people get weary, so ponder into

the knowledge and science thoughtfully and

search for new facts and ideas.”

(Holy Prophet)

(Peace be upon him)

CERTIFICATE

It is hereby certified that the thesis entitled, “Dental Morphological Studies of True

Ungulates from the Type Locality of Nagri Formation, Chakwal District, Punjab,

Pakistan” is based on the original research work carried out by Mrs. Tasneem Ikram that

has not previously presented for the higher degree. Mrs. Tasneem Ikram has done her

research work under our supervision. He has fulfilled all the requirements and is qualified

to submit the accompanying thesis according to prescribed format for the degree of

Doctor of Philosophy in Zoology.

Supervisor Co-Supervisor

Professor Dr. Muhammad Akhtar Dr. Muhammad Akbar Khan

Department of Zoology, Department of Zoology,

Punjab University QA Campus, GC University,

Lahore, Punjab, Pakistan. Faisalabad, Punjab, Pakistan.

Dedicated to

The Love and Affection of

My Mentor

Prof. Abdul Hameed Ch.

My Husband

Ahmad Suhail Farooq

&

My beloved Son

Essa Salah-Ud-Din

ABSTRACT

The early Late Miocene type locality of the Nagri Formation from the Indo-Siwaliks has

yielded remains of the true ungulates that are today extinct to the south Asian

biogeographic realm. In this thesis, thirteen species recognize including

Brachypotherium, Hipparion, Listriodon and the bovids, of the true ungulates from the

village Sethi Nagri, district Chakwal, Punjab, Pakistan. The thirteen taxa of the true

ungulates are described and discussed in details. Quantitatively, the taxa of the bovids are

the most predominant. But Brachypotherium, Hipparion, Listriodon, tragulid and giraffid

fossils are approximately as common as each other at the type locality. Pachyportax,

Dorcabune, Miotragocerus and Gazella seem to be uniformly rare at the Sethi Nagri. The

tooth positions of all thirteen species are documented. The new findings from the type

locality are the Giraffokeryx’s hemimendible and the deciduous premolar of

Dorcatherium minus. The newly recovered hemimandible and deciduous premolar

enlarge our knowledge on the anatomic features of the Nagri true ungulates.

The Nagri type locality mammalian local fauna has similarities to late Miocene Eurasian

faunas. The investigation comprises extensive taxonomic descriptions of all species

represented and an interpretation of the palaecology based on an analysis of the

community structure. It seems that the abundance of Hipparion, giraffids, rhinocerotids

and bovids suggests a woodland to savannah environment at or near the type locality

during the early Late Miocene. There is little evidence to suggest that there was a humid

closed canopy forest interspersed with temporary and perennial waters and accompanying

open areas forest in the vicinity at the time of deposition.

i

ACKNOWLEDGEMENTS

Countless praise and thanks to Almighty Allah (The most beneficent, most merciful),

who enabled me and bestowed on me his countless blessings to complete this thesis

work, took troubles away from me and blessed me with courage and strength. Each and

every moment of my life is devoted to praise Hazrat Muhammad (Peace Be Upon Him)

who enlightened my conscience with essence of faith in Allah.

I feel great pleasure in expressing my deep regards and gratitude to my respected teachers

and research supervisor, Dr. Muhammad Akthar and co-supervisor Dr. Muhammad

Akbar Khan for their precious guidance, unprecedented help, vital instructions, continued

interest and encouragement throughout the progress of this work. Inspite of their diverse

engagements in life they spared time for having read the manuscript with special

attention and endowed me constructive criticism and suggestions. In particular Dr.

Akhtar is thanked for assignment of the topic, guidance, advice, support and continuous

encouragement during the progress of the work, and Dr. Akbar Khan is thanked for his

highly valuable suggestions to solve my significant problems, critical reading of

manuscript, useful comments for its improvement and giving the thesis its final shape.

Lots of thanks and gratitude to my father, my mother (late), my brothers and my sisters,

without their prayers I could not have done it. Special thanks to my father in law (late)

Professor Abdul Hameed Chaudhry and my husband for urging me to go for PhD. My

acknowledgments seem incomplete without paying homage to my mother in law; sisters

in law and my brothers in law whom moral support was always with me.

Dr. Abdul Majid Khan, Dr. Abdul Ghaffar and Dr. Umar Farooq are thanked for their

encouragement and regular visits as well as for their invaluable advice. I would like to

thank Adeeb Babar for his support and help in photography and preparation of maps.

Thanks to my good class fellows Abdul Majid, Khizar, Samiullah and my seniors Hafiz

Muhammad Nazir and Mehboob Ahmad that have supported me all the years.

My special acknowledgments to my very nice friends Humaira Muqaddas, Sadia Imtiaz,

Huma Javed, Faiza Shahana Abbas, Samreen Akhtar and Sadaf Rehman for their constant

support and for their caring and encouraging attitude.

ii

I sincerely thank to Maskeen Ali, Maqsood Ahmad, Sajid Shah, Abdul Razaq and Imran

for joining me in my every field trip for the fossil collection and for serving me in the

laboratory. I would like to thank all the people from the village Sethi Nagri for their

hospitality and their help.

Tasneem can’t refrain from citing the faculty members of Al-Suffah Girls College and

Govt. College for Women Farooq Colony, Sargodha who continued to support me

throughout. If I did forget to mention somebody that I ought to thank, it was not

intentional and I do apologize. So lots of thanks to those I have missed and if someone is

angry with me but always keeps praying for my good future.

May Allah bless everyone with a nice circle of people as I have (Ameen).

Tasneem Ikram

December, 2010

iii

LIST OF ABBREVIATIONSCa Circa

Myr Million years

Ma Million years ago

MN European Mammal Neogene zone scale

N Numbers

GPTS Geomagnetic Polarity Time Scale

GRTS Geomagnetic Reversal Time Scale

mm/yr Millimeter per year

ky Thousand of year

AMNH American Museum of Natural History

BMNH British Museum of Natural History

PMNH Pakistan Museum of Natural History

PUPC Punjab University Palaeontological Collection, housed in

the department of Zoology, Punjab University, Lahore,

Pakistan

PC-GCUF Palaeontological Collection of Government College

University, Faisalabad, Punjab, Pakistan

I Incisor

C Canine

P premolar

M11 First upper or lower molar

M22 Second upper or lower molar

M33 Third upper or lower molar

GSI Geological survey of India

GSP Geological Survey of Pakistan

DP Deciduous Premolar

W/L Width/Length ratio

r Right

l Left

iv

CONTENTSAbstract i

Acknowledgements ii

List of Abbreviations iv

List of Tables vii

List of Figures viii

Chapter 1

INTRODUCTION 1

Potwar Plateau of Northern Pakistan 5

Studied Section 7

Methodology 10

Tooth Morphology 11

Objectives of Present Study 14

Thesis Layout 14

Chapter 2

Literature Review

Siwalik Geology 16

Siwalik Biostatigraphy 17

Siwalik Statigraphy 19

Chronostratigraphy 26

Siwalik Sedimentology 29

Siwalik Lithology 31

v

Taphonomy 35

Palaeoenvironment of the Siwaliks 36

Siwalik Faunas 38

Chapter 3

Systematic Palaeontology

Genus Listriodon 41

Genus Selenoprotax 53

Genus Pachyportax 62

Genus Tragoportax 69

Genus Miotragocerus 77

Genus Gazella 82

Genus Giraffokeryx 88

Genus Giraffa 97

Genus Dorcatherium 101

Genus Dorcabune 110

Genus Hipparion 116

Genus Brachypotherium 129

Chapter 4

Discussion 138

Conclusions 147

References 148

Appendices 173

vi

List of Tables

Chapter 1

INTRODUCTION

Table 1: Mammalian faunas of the Nagri.

Chapter 2

Literature Review

Table 1: Stratigraphic section of the Siwalik Group illustrating formations and zones.

Chapter 3

Systematic Palaeontology

Table 1: Comparative measurements (mm) of the cheek teeth of the Listriodon

pentapotamiae.

Table 2: Comparative measurements (mm) of the cheek teeth of Selenoportax cf.

vexillarius.

Table 3: Comparative measurements (mm) of the cheek teeth of Pachyportax cf.

latidens.

Table 4: Comparative measurements (mm) of the cheek teeth of Tragoportax

punjabicus.

Table 5: Comparative measurements (mm) of the cheek teeth of Miotragocerus cf.

gluten.

Table 6: Comparative measurements (mm) of the cheek teeth of Gazella cf. lydekkeri.

Table 7: Comparative measurements (mm) of the cheek teeth of the Siwalik

Girrafokeryx and Giraffa.

Table 8: Comparative measurements (mm) of the cheek teeth of Dorcatherium cf.

minus and Dorcatherium cf. majus.

Table 9: Comparative measurements (mm) of the cheek teeth of Dorcabune cf.

anthracotherioides.

Table 10: Comparative measurements (mm) of the cheek teeth of Hipparion theobaldi.

Table 11: Comparative measurements (mm) of the cheek teeth of Brachypotherium

perimense.

vii

List of FiguresChapter 1

IntroductionFigure 1: Distribution of Siwalik sediments along the foot hills of Himalayas.

Figure 2: Map of the Potwar plateau showing main fossil localities in the Punjab,

northern Pakistan.

Figure 3: A.Location of the Potwar Pleatu in northern Pakistan. B. Simplified geology

map of the Nagri area.

Figure 4: Hipparion upper molar.

Figure 5: Rhino lower molar.

Figure 6: Suid lower third molar.

Figure 7: Ruminant lower third molar.

Chapter 3

Systematic Palaeontology Figure 1: Listriodon pentapotamiae’s studied sample.

Figure 2: Scatter diagram showing dental proportions of Listriodon pentapotamiae’s

studied sample.

Figure 3: Selenoportax cf. vexillarius’s studied sample.

Figure 4: Selenoportax cf. vexillarius’s studied sample.

Figure 5: Scatter diagram showing dental proportions of Selenoportax cf. vexillarius’s

studied sample.

Figure 6: Pachyportax cf. latidens’s studied sample.

Figure 7: Scatter diagram showing dental proportions of Pachyportax latidens’s

studied sample.

Figure 8: Tragoportax punjabicus’s studied sample.

Figure 9: Tragoportax punjabicus’s studied sample.

Figure 10: Scatter diagram showing dental proportions of Tragoportax punjabicus’s

studied sample.

Figure 11: Miotragocerus cf. gluten’s studied sample.

viii

Figure 12: Scatter diagram showing dental proportions of Miotragocerus cf. gluten’s

studied sample.

Figure 13: Gazella cf. lydekkeri’s studied sample.

Figure 14: Scatter diagram showing dental proportions of Gazella cf. lydekkeri’s studied

sample.

Figure 15: Giraffokeryx punjabiensis’s studied sample.

Figure 16: Giraffokeryx punjabiensis’s studied sample.

Figure 17: Giraffokeryx punjabiensis’s studied sample.

Figure 18: Scatter diagram showing dental proportions of Gazella punjabiensis’s

studied sample.

Figure 19: Giraffa cf. priscilla’s studied sample.

Figure 20: Scatter diagram showing dental proportions of Giraffa cf. priscilla’s studied

sample.

Figure 21: Dorcatherium cf. minus’s studied sample.

Figure 22: Dorcatherium cf. majus’s studied sample.

Figure 23: Scatter diagrams showing dental proportions of Dorcatherium’s studied

sample.

Figure 24: Dorcabune cf. anthracotherioides’s studied sample.

Figure 25: Scatter diagram showing dental proportion of Dorcabune

anthracotherioides’s studied sample.

Figure 26: Hipparion theobaldi’s studied sample.

Figure 27: Hipparion theobaldi’s studied sample.

Figure 28: Scatter diagrams showing dental proportions of Hipparion theobaldi’s

studied sample.

Figure 29: Brachypotherium perimense’s studied sample.

Figure 30: Brachypotherium perimense’s studied sample.

Figure 31: Scatter diagram showing dental proportions of Brachypotherium perimense’s

studied sample.

ix

INTRODUCTION

Ungulate refers to any animal with hooves however, the “True Ungulates” are considered

the members of Artiodactyla and Perissodactyla. Subungulates (Paenungulates) comprise

Sirenia, Proboscidea and Hyracoidea. In addition to hooves, most Ungulates have

developed reduced canine teeth, bunodont molars due to herbivorous condition.

Ungulates diversified rapidly in the Eocene, but are thought to date back as far as the late

Cretaceous (Gentry and Hooker, 1988). Most Ungulates are herbivores and some

commonly known examples of Ungulates living today are the goat, sheep, giraffe, deer,

antelope, gazelle, camel, hippopotamus, horse, zebra, donkey, cow and rhinoceros. The

Nagri type area of the Nagri Formation, Middle Siwaliks has yielded very rich

assemblage of the true ungulates mainly recorded by Pilgrim (1913, 1926, 1937, 1939),

Anderson (1927), Colbert (1935), Lewis (1937), Pascoe (1964), Thomas (1977, 1984),

Akhtar (1992), Barry et al. (2002), Farooq (2006) and Khan A. M. (2010). The fauna

mainly consists of crocodiles, chelonins, proboscidians, rhinocerotides, artiodactyls,

carnivores and primates (Table 1).

The term Siwaliks denotes the Neogene terrestrial sediments which are found in widely

separated areas all along the foot hills of Himalayas. The Himalaya rose, and the

sedimentary rocks of the Siwaliks were deposited, because of the collision between the

Indian and Asian plates 40 to 50 Ma (Kumaravel et al., 2009). The Siwalik hills are

located in the political boundaries of Pakistan, India, Nepal, and Bhutan, and range

between 6 to 90 km in width (Acharyya, 1994). They gradually become steeper and

narrower in relief and width respectively, from northern Pakistan to Bhutan (over 2000

1

km in length). The fluvial sequence of the Siwaliks is situated along the Himalayan

foothills from Pakistan in the west of Myammar in the east for about 1689 km (Fig. 1).

These sedimentary deposits are over 6000 meters in thickness and provides an amazing

opportunity to palaeontologists, geologists and natural history researchers to study fluvial

dynamics, palaeomagnetic dating, palaeoclimatology, stratigraphic correlation, isotope

geochemistry, and vertebrate biochronology across the last 20 Ma (Andrews and Cronin,

1982; Pilbeam, 1982).

On the basis of lithology, Medlicott (1864) divided the Siwaliks into Lower, Middle and

Upper subgroups and used the term “Siwalik Series” for the first time. Oldham (1893)

and Holland (1926) also used the term Siwalik Series. Pilgrim (1910) showed that such a

division was also possible on the basis of fauna. On palaeontological basis Pilgrim (1913)

further differentiated the six zones, with lithological characteristics in the three divisions.

He (Pilgrim, 1913) described these rock units as Pinjor, Tatrot, Dhok Pathan, Nagri,

Chinji and Kamlial faunal zones. He also applied the lithostratigraphic classification of

Upper, Middle and Lower Siwaliks. Later, Anderson (1927) and Cotter (1933) applied

the names in the sense of lithostratigraphic units but referred them as stages. Lewis

(1937) modified this term as Chinji Formation, Nagri Formation and Dhok Pathan

Formation, while Kravtchenko (1964) used Soan Formation for Pinjor and Tatrot zones.

The Siwalik Group is a thick sequence of fluvial clastic rocks shed southward as the

Himalaya was uplifted, beginning in the late Oligocene. The Neogene sedimentary

deposits extend from western Pakistan to eastern India. At both extremities the mountains

turn southward around the edges of the Indian plate and form the prominent Himalayan

syntaxes.

2

Table 1: Mammalian faunas of the Nagri. Many species are under taxaonomic revision. Artiodactyla

Bovidae

Tragoportax browni T. salmontanusT. perimense T. punjabicusMiotragocerus gluten Elaschistoceros khauristanensisSelenoportax vexillarius S. lydekkeriPachyportax latidens P. nagriiGazella lydekkeri

Giraffidae

Giraffokeryx punjabiensis Giraffa priscilla Giraffa punjabiensis

Anthracotheriidae

Merrycopolumus nanus M. dissimilis

Tragulidae

Dorcabune anthracotherioides D. nagrii

Dorcatherium majus D. minusAnthracotheriidae

Chocromeryx silistrense Merycopotamus dissimilis

Suidae

Propotamochoerus uliginosus P. hysudricus

Perissodactyla Equidae

Hipparion theobaldi H. nagriensisH. perimense

RhinocerotidaeChilotherium intermedium C. blanfordi

Subchilotherium intermedium Alicornops sp.

Brachypotherium perimense

Chalicotheriidae

Chalicotherium salinum

ProboscideaTriophodon angustidens var. palaeoindicusPentalophodon falconeri Dinotherium indicum

CercopithecidaeSivapithecus sivalensis S. indicus

Ramapithecus punjabicus

RodentiaRhizomys sivalensis Rhizomys sp.

CarnivoraProgenella sp. Pathyaena sivalensePercurocuta carnitex Sivaelurus chinjiensis

3

The Siwalik deposits are one of the most comprehensively studied fluvial sequences in

the world (Lydekker, 1876, 1878; Matthew, 1929; Colbert, 1935; Pilgrim, 1937, 1939;

Hooijer, 1958; Pilbeam et al., 1977, 1979; Shah, 1980; Thomas, 1984; Hussain et al.,

1992; Flynn et al., 1995; Barry et al., 2002, 2005; Dennell et al., 2006, 2008; Nanda,

2002, 2008; Khan, 2008; Sheikh et al., 2008; Khan et al., 2009a, b, 2010a, b). The

deposits are composed of mudstones, sandstones and coarsely bedded conglomerates

deposited at times when the region was a colossal basin during Middle Miocene to Upper

Pleistocene times. Rivers flowing southwards from the Greater Himalayas, resulting in

extensive multi-ordered drainage systems, deposited the sediments. After this deposition,

the sediments were uplifted through intense tectonic regimes commencing in Upper

Miocene times, subsequently resulting in a unique topographical entity – the Siwaliks

(Chauhan, 2003). The Siwalik Group rocks extend along the base of the broadest outcrop,

occurring in the Potwar Plateau of Pakistan (West et al., 2010).

Figure 1: Distribution of the Siwalik sediments along the foot hills of Himalayas.

4

Potwar Plateau of Northern Pakistan

The Potwar Plateau (Lat. 33° 00´ N; Long. 72° 30´ E) is situated in the northern Pakistan

(Fig. 2). It is an elevated area comprising some 20,000 km2 bounded in the north by the

Kala Chita and the Margala hills, in the south by the Salt Range, in the east by the Jhelum

River and in the west by the Indus River (Badgley et al., 2008) (Fig. 2). The Neogene’s

strata of the northern Pakistan have been divided into the Kamlial, Chinji, Nagri, Dhok

Pathan and Soan formations. All these formations typically consist of gently tilted strata

that form shallow strike-valleys and laterally extensive channel sandstones form higher

ridges as the surface expression of the large structural synclinorium underlying the

Potwar Plateau. Fossils come out of these strata due to erosion and accumulate on the

outcrop surfaces between the ridges, providing best conditions for sampling within well-

defined stratigraphic intervals (Pilbeam et al., 1977, 1997).

The Neogene Siwalik sequence from the Potwar Plateau, northern Pakistan, is a

particularly good example of a long record of land mammals. This long faunal sequence

records numerous vertebrate taxa and biotic events in the South Asian biogeographic

realm (Pilbeam et al., 1997). The most extensive of the Neogene sediments, the Siwalik

formations are widely distributed through Pakistan (Figs. 1-2) (Keller et al., 1977;

Opdyke et al., 1979; Johnson N. et al., 1985; Barry et al., 2002). Within Pakistan they are

best exposed in the Potwar Plateau (Fig. 2). The Potwar Plateau biostratigraphic and

paleomagnetic framework continues to build on work published since the late 1970’s.

Many key stratigraphic sections measured and sampled (Opdyke et al., 1979; Pilbeam et

al., 1979; Tauxe, 1979; Barry et al., 1980; Johnson N. et al., 1982, 1985; Tauxe and

Opdyke, 1982) have been supplemented by radiometric dates and microstratigraphic

5

studies in the Potwar Plateau (Johnson G. et al., 1982; Badgley, 1986; Behrensmeyer,

1987; Tauxe and Badgley, 1988; Badgley and Tauxe, 1990; Flynn et al., 1995).

Consequently, the Potwar Plateau biostratigraphy is refined which represents almost the

entire Neogene from about Middle Miocene to Pleistocene (Barry et al., 1982, 1990,

1991, 2002; Jacobs et al., 1989, 1990; Flynn et al., 1990; Barry et al., 2002).

Figure 2: Map of the Potwar Plateau showing main fossil localities in the Punjab,

northern Pakistan.

6

Studied Section

The described specimens in this thesis are recovered from the outcrops nearby the Sethi

Nagri village (Lat. 32° 25' N: Long. 72° 14' E), a type locality of the Nagri Formation of

the Middle Siwaliks. The type locality is designated nearby the Sethi Nagri village of the

Chakwal district, Punjab, Pakistan (Fig. 3). The deposits consist mainly of thick, massive

sandstone with occasional shale beds. At few places fine and coarse-grained beds may be

encountered. In general the sandstone is immature and poor to moderately sort. The

sandstone bodies are mainly composed of different storeys stacked both vertically and

laterally (Shah, 1980). The cross-bed thickness varies from a few centimeters to one

meter in the lower part of the Formation. The basal surface of these cross-beds is usually

erosional. The colour of sandstone varies from greenish gray to light gray and dark gray

very rarely off white or gleaming white colours may be seen. Occasional interclast

pebbles are also present within sandstone bodies. The conglomerates with varying

thickness are present along different horizons (Pilbeam et al., 1997). Some limonitic

staining is also present. It mainly shows a salt and pepper texture. The shales are reddish,

brown, pale orange and sometimes chocolate coloured. The palaeochannels are very

common within the outcrops (Barry et al., 2002).

The fossiliferous area is situated in the south of the Sethi Nagri village (Fig. 3). The

average thickness of the deposits is about 650 m (Barry et al., 2002). Regionally the area

is situated in the north of the Ghabir River (Fig. 3). The section from which the remains

were excavated represents a typical sequence of fluvial sedimentation and consists of

bluish grey, massive and coarse sandstone with purple and orange clay and thick brown

sandstone. Sites surrounding the Ghabir River present an abundance of vertebrate fossils

7

that represent almost large size mammals. The areas are thoroughly excavated and the

discovered sites are indicated by ‘SN’ (abbreviation for the Sethi Nagri outcroping).

During excavations fifteen sites (SN1-15) are found that are mostly situated the north of

the Ghabir River (Fig. 3B). The three sites only are excavated from the south of the

Ghabir River. The sites towards the east are more fertile than those of towards the west.

The recovered specimens from these sites are characterized by large size mammals and a

few sites represent small size mammals. The assemblage displays the regional

characteristics of the Nagri Formation of the Middle Siwaliks. The fossils are mostly

fragmentary in nature and the postcranial fossils are more abundant than the cranial ones.

The weathering cracks, abrasion marks and byte marks are noted frequently while

observing the specimens. Some sites (SN5-6) are highly fossiliferous and seems to

expose for the long time. The fauna mainly consists of artiodactyls and perissodactyls.

Lithofacies suggest a fluvial depositional environment of the type locality.

Barry et al. (1982) indicated an age for the Sethi Nagri type locality between 7.4 to 9.5

Ma. Johnson N et al. (1985) date 10.8 Ma for the type locality, based on the fission-track

dating of the volcanic ash near the type locality. Pilbeam et al. (1997) calculated the age

10.7 Ma for the ‘Hipparion’ Datum, which is the oldest occurrence of Hipparion in the

Siwaliks. This date is estimated for the localities on the top of the Ghabir kas long normal

interval.

8

Figure 3: A. Location of the Potwar Plateau in northern Pakistan; the studied areas are encircled (map is modified from Behrensmeyer and Barry, 2005 and the boundary dates are from Barry et al., 2002; Dennell et al., 2008 and Nanda, 2008). B. Simplified geology map of the Nagri area indicating the fossiliferous sites along the Ghabir River (SN – abbreviation for Sethi Nagri) from where the studied material is recovered (the map is modified from Colbert, 1935).

Methodology9

The fossil remains of the true Ungulates include isolated dentition, mandible and maxilla

fragments. The specimens are recovered from the Sethi Nagri type area of the Nagri

Formation (Fig. 3). Surface collection was the primary method to recover the fossils from

the type locality. Some fossils were exposed and easily available for the collection.

Piercing instruments like chisels and geological hammers were employed for the

excavation of partially embedded fossils. In due course numbers of field trips are carried

out to the various fossilized sites of the Sethi Nagri village and the buried specimens

were dug out with the help of the light hammers, chisels and fine needles. Careful

measures were taken so as to prevent the fossils from disintegrating during excavation.

Each specimen was wrapped with a cotton piece to avoid the shocks of transportation.

Eventually the collected specimens were brought in the laboratory for taxonomic and

morphological analysis.

In order to remove dust particles and prepare the specimens for clear observation, the

specimens were carefully washed and cleaned in the Palaeontology laboratory of the

Zoology Department of the Punjab University, Lahore, Punjab, Pakistan (institutional

abbreviation – PUPC). Some specimens present in the Palaeontology laboratory of

Zoology Department, GC University, Faisalabad (institutional abbreviation – PC-GCUF)

are included in this study. Clay and other hardly adjoined sedimentary particles were

removed with the help of fine needles and brushes. Accidentally broken fragments of the

specimens were rejoined by using gums and resins such as Magic Stone, Elfy, and

Fixings etc. A hand lens was used for keen observation of very small and ambiguous

morphological characters.

10

The measurements of the specimens were taken in millimeters (mm) with the help of

metric Vernier Caliper. The morphological and metrical characters of the specimens are

described and their systematic determination is discussed. The catalogue number of the

specimens consists of series i.e., yearly catalogued number and serial catalogue number,

so figures of the specimen represent the collection year (numerator) and serial number

(denominator) of that year (e.g. 09/12). Uppercase letters with superscript stand for upper

dentition (e.g. M1) and with subscript number stand for lower dentition (e.g. M1). In the

discussion comparisons are made with fossils from the Natural History Museum, London

(BMNH), the American Museum of Natural History (AMNH), the Geological Survey of

Pakistan (GSP), the Geological Survey of India (GSI) and the specimens of Palaeontolgy

laboratory of the Zoology department of the Punjab University (PUPC). The studied

material is the property of the Palaeontology laboratory of the Zoology Department of the

Punjab University, Lahore, Pakistan.

Tooth Morphology

Tooth cusp nomenclature in this thesis follows that of Heissig (1972), Janis and Scott

(1987a, b), Pickford (1988), Akhtar (1992), Gentry (1994), and Cerdeño (1995) as shown

in the figures 4-7. An entostyle can be found on the center of the lingual side of the upper

molar and ectostylid is found on the buccal side of the lower molar, completely or partly

separate from the rest of the occlusal surface. Tooth length and breadth were measured at

occlusal level. Heights were measured on the mesostyle of the upper molar, the

metastylid of the lower molar and the protoconid of the lower premolar.

11

Mesostyle

EctolophMetacone

Metastyle

Postfossette

Hypostyle

HypoglyphHypocone

Metaloph ProtoconePli-caballin

Protoconule

Prefossette

Protoloph

ParastyleParacone

Ectoloph

Figure 4: Hipparion upper molar.

Figure 5: Rhinoceros lower molar.

12

Figure 6: Suid third lower molar.

Figure 7: Ruminant third lower molar.

13

Objectives of Present Study

My PhD project is multidisciplinary in nature. The main aim of this study has been to

provide the first complete documentation of true ungulates found in the vicinity of the

type locality of the Nagri Formation by tackling tooth morphology, taxonomy, and

palaeontology of the Siwaliks of Pakistan. An ecologically important group, the

ungulates, was selected for the study as the collected ungulate material presented notable

diversity and thus could provide significant taxonomic and palaeoenvironmental and

information.

Thesis Layout

This thesis consists of four distinct, separate and autonomous thematic units structured in

a format as directed by Doctoral Programme Coordination Committee of the University

of the Punjab, Lahore, Pakistan. Consequently, repetition of description, discussion and

systematics about the genera does occur. The first chapter entitled, “Introduction”

includes mainly about Siwaliks and the type locality Sethi Nagri as well as tooth

morphology of the ungulates by which identification of the studied specimen has been

made. This chapter elaborates a map of the Sethi Nagri, showing the excavated sites of

the type locality. The second chapter named, “Literature Review” includes geology,

biostratigraphy, stratigraphy, chronostratigrpahy, lithology, sedimentology,

palaeoenvironment and fauna of the Siwaliks. The third chapter entitled, “Systematic

Palaeontology” elaborates the taxonomical and morphological features of the recovered

material. The fourth chapter “Discussion” in which the Sethi Nagri ungulates’s

14

correlation, biostratigraphy and palaeoenvironmental requirements are interpreted.

Finally, conclusions are given at the end of the chapter.

The references and appendices are given at the end of the thesis. The references are

compiled to follow the pattern of the Pakistan Journal of Zoology, published by the

Zoological Society of Pakistan. The studied material is provided in appendix 1. One

additional paper which comprised description of Dorcabune nagrii has been already

published (Appendix 2). My contribution to this paper has been the description and the

discussion of Dorcabune nagrii. The published research work has been included in

appendix 2 (reprint), as directed by the Doctoral Programme Coordination Committee of

the Punjab University, Lahore, Pakistan.

15

LITERATURE REVIEWSiwalik Geology

Geologically the Siwaliks is a foreland basin of the Himalayas filled with molasses-type

sediments of the Neogene and early Quaternary age, developed at the foot hills of the

Himalayan mountain belt. The stratigraphic sequence preserves a continuous record of

the continental sedimentation (sediment thickness > 6km at places) as well as an equally

comparable continuous record of vertebrates, especially of the mammals (Sarage and

Russell, 1983; Nanda and Shani, 1990; Flynn et al., 1995; Scott et al., 1999; Metais et

al., 2000, 2001, 2004; Geraads et al., 2002; Flynn, 2003; Barry et al., 2002, 2005; Bernor

et al., 2003; Franz et al., 2003; Kaiser, 2003; Kaiser et al., 2003, Kaiser and Fortelius,

2003; Kappelman et al., 2003; Raymond et al., 2004; Basu, 2004). The Tertiary

continental formations of the northern Pakistan have yielded some of the richest fossil

mammalian faunas from South Asia. Fossil remains from this area have been known

since the nineteenth century and large mammals have been the object of several

monographies (Lydekker, 1883a, b, 1884; Pilgrim, 1910, 1912; Colbert, 1935).

The Siwalik Group in Pakistan can be clearly divided, according to the lithological

characters, into the usual three subgroups- Lower, Middle and Upper, and further into

their formation scale lithostratigraphic units. The Lower Siwaliks (Kamlial, Chinji

formations) consists of a sequence of sandstone-mudstone couplets with a marked

dominance of the mudstones over the sandstones. The development of paleosol horizons

is also fairly frequent. The Middle Siwaliks (Nagri, Dhok Pathan formations) are

dominantly arenaceous, consisting of multistoried coarse to medium-grained, blue-gray,

massive sandstones (30 to > 60m) with subordinate representation of clays, mudstones

16

and siltstones. The Upper Siwalik (Tatrot, Pinjor, Boulder Conglomerate formations)

subgroup is classified into three lithostratigraphic formations. The three units comprise

the sequences of the sandstone-mudstone couplets, the Parmandal Sandstone and the

Boulder Conglomerate Formation, the upper most lithostratigraphic unit the Boulder

Conglomerate Formation, the upper most lithostratigraphic unit (Quade and Cerling,

1995).

Siwalik Biostratigrapy

The Miocene sediments are entirely fluvial in origin, having been deposited by large river

systems. Some of the sections exceed 3000 m of such accumulated sediments and which

are now exposed on the surface. These sedimentary rocks are usually divided into time

successive formations, with the classic sequence of the Potwar comprising the Murree,

Kamlial, Chinji, Nagri, and Dhok Pathan formations of Pilgrim (1910, 1913) and what

the Geological Survey of Pakistan refers to as the Soan Formation (Cheema et al., 1977).

It is often difficult to delineate the boundaries between the formations, however, from the

geological or the sedimentological perspective it is best to view the Siwalik sequence as a

single genetic unit. Neverthless, the Siwalik formations have always been cryptic

chronostratigraphic units and from the paleontological point of view recognition of the

formations and their boundaries has been a crucial step in dating the fossils (e.g. Colbert,

1935). This practice has in the past produced much confusion and sterile debate, but it is

now possible to assume that with the contribution of magnetostratigraphy as a means of

dating the rocks, this era of confusion belongs to the past.

17

Pilgrim (1910, 1913) first recognized a series of successive “faunal zones,” initially using

the term in a manner comparable to modern usage of the “Stage” concept. Pilgrim’s unit

(Kamlial, Chinji, Nagri and Dhok Pathan formations) were based on a mixture of

contained fauna and lithological criteria. In most instances their superpositional

relationships could be demonstrated, but the boundaries of the faunal zones were not

delineated and, because of mistakes in correlation, the faunal content of some zones

could never be adequately differentiated. Subsequently, as stratigraphic concepts and

nomenclature became more précise, Pilgrim’s faunal zones came to be used primarily

either as lithostratigraphic formations, or as chronostratigraphic “zones” or even as some

confusing combination of the two (Pilbeam et al., 1979; Barry et al., 1980, 1985; Flynn,

1986). Occasionally, they were recognized as being essential biostratigaphic units.

Because of this change in the usage and the resulting confusion, Barry et al. (1980, 1982)

first advocated in restricting Pilgrim’s terms to the lithostratigraphic formations and later

proposed a new series of biostratigraphic zones in order to replace the Middle and Upper

Siwaliks “faunal zones.” So, each zone now had a well-defined base, and included the

entire stratigraphic interval below the base of the succeeding zone. Murphy (1977) has

noted the distinctions between the operations of definition, characterization, and

identification in the practice of stratigraphy. The biostratigraphic interval-zones of Barry

et al. (1982) were defined and characterized in reference sections and criteria were stated

for identifying or recognizing them in other sections. Because they were related directly

to stratigraphic sections, the interval-zones and their boundaries can, like stages, be

correlated to other geological phenomena, such as sedimentological or geochemical

events, magnetopolarity zones, or to geologic time.

18

However, the Siwalik interval-zones should not be confused with stages, which are

chronostratigraphic unts. Each interval-zone’s lower boundary is defined by a biological

event, not a stratigraphic level with a specific age. At the time the defining taxa were

selected, the stratigraphic levels and the ages of their first appearances were thought to be

accurately known. Consequently the Startigraphy Committee of Pakistan formalized the

Siwalik group to include “Soan, Dhok Pathan, Nagri, and Chinji formations.

Siwalik Stratigraphy

Danilchik and Shah (1967) applied the lithostratigraphic rank “Group” for the Siwaliks.

The Stratigraphic Committee of Pakistan formalized the Siwalik Group to include “Soan,

Dhok Pathan, Nagri and Chinji formations”. The type localities of the component

formations have been recommended to represent the type locality of the Group. The

Siwalik Group is divided into Lower, Middle and Upper Siwalik subgroups (Table 1)

based on faunal assemblages (Pilgrim, 1910) and lithostratigraphic variations (Pilgrim,

1913). These subgroups are further divided into individual “Formations” (when the

division relies upon lithostratigraphy) or “zones” (when the division is on the basis of

faunal assemblage).

Lower Siwalik Subgroup

The Lower Siwalik Subgroup (18.3-11.2 Ma) comprising Kamlial and Chinji formations

is approximately 4000 feet in thickness and is characterized by depositional environments

of high sinuosity meandering streams with broad flood plains. At outcrop level, Lower

Siwaliks comprise highly indurate, fine to medium grained, grey to greenish blue and

purple sandstones inter bedded with reddish brown to grey and hard concretionary

19

mudstones and paleosoles (Kumaravel et al., 2005). Sandstones of the Kamlial and the

Chinji formations contain abundant quartz with subordinate feldspar, variable proportions

of lithic grains, accessory amounts of micas and traces of a number of heavy minerals.

Feldspar contents mostly ranges from 18 to 30% and 24 to 28% in the Kamlial and the

Chinji sandstones, respectively (Kaleem Ullah et al., 2006).

Table 1: Stratigraphic section of the Siwalik Group illustrating formations and zones

(boundary dates are from Barry et al., 2002).

Siwalik Formations Siwalik Subgroups Siwalik Zones Age (Ma)

Soan Upper Siwalik Subgroup

Tatrot, Pinjor andBoulder Conglomerates

3.5-0.6

Dhok Pathan Middle Siwalik Subgroup

Dhok Pathan 10.1-3.5

Nagri Nagri 11.2-10.1

Chinji Lower Siwalik Subgroup

Chinji 14.2-11.2

Kamlial Kamlial 18.3-14.2

Kamlial Formation: The name “Kamlial Formation” was used for the rocks exposed

near the Kamlial village in the Attock district. These rocks were previously named

“Kamlial Stage” by Pinfold (1918). The name “Kamlial Formation” was proposed by

Lewis (1937) and later accepted by Stratigraphic Committee of Pakistan in 1964. Type

locality (Lat. 33° 15' N: Long. 72° 30' E) is located near the Kamlial village, Potwar

Plateau, Pakistan. The Kamlial Formation (18.3-14.2 Ma) consists of sandstones, usually

more than 5 m thick, of brick red to purple grey color inter bedded with red shale and

20

pseudo-conglomerates of light red color. Mudstones are common in middle and upper

parts and are often maroon and purple colored. The thickness of the Kamlial strata is

approximately 1700 feet. This Formation is transitional with overlying Chinji Formation

and underlying Murree Formation. The basal 10-30 m of the Kamlial Formation contains

abundant broken foraminifera and marine bivalve shells.

Chinji Formation: Pilgrim (1913) suggested the term “Chinji Zone” for this rock unit

that was later modified by Lewis (1937) as “Chinji Formation”. This name was also

reformed as “Chinji Stage” by Pascoe (1964). The Stratigraphic Committee of Pakistan

recommended the name “Chinji Formation” with type locality (Lat. 32° 4' N: Long. 72°

22' E) near Chinji village located at Rawalpindi-Sargodha road, approximately 25 km

from Talagang in the Chakwal district, Punjab, Pakistan. The Chinji Formation (14.2-

11.2 Ma), the strata approximately 2300 feet thick, is dominantly composed of bright red

and brown orange siltstones inter bedded with soft, ash grey sandstones. The proportion

of sandstones and siltstones varies at different places throughout the Formation (siltstone

and sandstone ratio = 4:1 in the type section). Sandstone is medium to coarse grained and

thick bedded to massive. Scattered pebbles and conglomerates along some

palaeochannels are present along different horizons of the formation. Facies of the

formation are predominantly composed of Argillites. Chinji Formation is widely spread

in Kohat and Potwar Plateau. At the type locality, the lower contact with Kamlial

Formation is gradational and upper contact is conformable with Nagri Formation (Basu,

2004). Badgley et al. (1995) studied the taphonomy of small-mammalian remains from

Chinji Formation of stratotype area and found that the Formation contains the fossils of a

mixture of arboreal, burrowing, and terrestrial small mammals with the dominance of

21

terrestrial herbivores. There is a little variation in the taxonomic composition of these

mammals among different sites.

Middle Siwalik Subgroup

Middle Siwalik subgroup (11.2-3.5 Ma) is approximately 6000 feet in thickness and

consists of two distinct formations, lower Nagri Formation and upper Dhok Pathan

Formation (Table 1). Unconsolidated sandstones called “salt and pepper” dominate the

subgroup. These formations in the eastern Potwar Plateau, northern Pakistan, comprise

relatively thick (tens of meters) sandstone bodies and mudstones that contain thinner

sandstone bodies (meters thick) and paleosols (Khan et al., 2006). Near the Indus River

the uppermost 3000 feet of the Middle Siwalik sequence contain thick beds of

conglomerates that die out eastwards and southeastwards into sandstones and clays (Gill,

1951). Thick sandstones are composed of channel bar and fill deposits of low-sinuosity

(1.08–1.19), single-channel meandering and braided rivers that formed large, low-

gradient sediment fans. Trace fossils and body fossils within all facies indicate the former

existence of terrestrial vertebrates, molluscs (bivalves and gastropods), arthropods

(including insects), worms, aquatic fauna (e.g. fish, turtles, crocodiles etc), trees, bushes,

grasses, and aquatic flora (Zaleha, 2006).

Nagri Formation: Lewis (1937) introduced the name “Nagri Formation” which was later

accepted by Stratigraphic Committee of Pakistan. Nagri village is designated as the type

locality (Lat. 32° 25' N: Long. 72° 14' E) which is situated at about 20 km south of

Talagang, Chakwal district, Punjab, Pakistan. Johnson N. et al. (1982) established six

magnetic polarity sections over the Potwar Plateau region of Pakistan. In all the six

sections the dominant feature of the magnetic polarity stratigraphy was a long normal

22

polarity zone, which was contained within the Nagri Formation. This conspicuous normal

polarity zone had been radiometrically dated as 9.5 ± 0.6 million years.

The Nagri Formation (11.2-10.1 Ma) consists mainly of massive sandstones; usually 15

m in thickness, with mudstone inter beds and occasional shale beds. Sandstone is of

greenish grey, grey, or brownish grey in color, medium to coarse grained in size, highly

thick and cross-bedded. It has a salt and pepper pattern produced by magnetite and

ilmonite. Claystone is brown, reddish grey and orange and is sandy or silty. The thickness

is about 500-900 m (Sheikh et al., 2008). The shales are reddish, brown, pale orange and

sometimes chocolate colored. Conglomerates of varying thicknesses are present along

different horizons. These are represented by some rounded pebbles of igneous and

metamorphic rocks, in the upper part of the formation. Paleosols vary from place to place

in central salt range and are red in color containing calcium carbonate. The lower contact

of the Nagri Formation with underlying the Chinji Formation is gradational and with that

of the overlying Dhok Pathan Formation is conformable.

Dhok Pathan Formation: Pilgrim (1913) proposed the name “Dhok Pathan” which was

later reformed by Cotter (1933) as “Dhok Pathan Formation”. The Stratigraphic

Committee of Pakistan accepted this name for application in the Kohat-Potwar Province.

The type locality (Lat. 33° 07' N: Long. 72° 14' E) is situated at Soan River about 75 km

from Rawalpindi on Rawalpindi-Sargodha road. The Dhok Pathan Formation (10.1-3.5

Ma) consists mainly of sandstones and claystones while conglomerates and shales can

also be found. Fine to medium grained sandstone is light grey, gleaming white and

reddish brown in color. The beds are moderately thick and cross-bedded. Conglomerates

are represented by pebbles present along the palaeochannels. Shales are variously colored

23

ranging from bright orange, brown, greenish and chocolate colored. Claystone is orange

red to chocolate brown, often hard and compact. The measured thickness is about 500-

825 m. The unit is overlain conformably by the Soan Formation (Sheikh et al., 2008).

Upper Siwalik Subgroup

Upper Siwalik subgroup (3.5-1.0 Ma) is approximately 6000 feet in thickness and is

divisible, biostratigraphically, into three zones namely Tatrot, Pinjor and Boulder

Conglomerates. On account of lithostratigraphy these “zones” are collectively known as

Soan Formation (Lat. 32° 22' N: Long. 72° 47' E). All the zones are named after nearby

geographical localities except Boulder Conglomerates that is a descriptive geological

term. The subgroup comprises of conglomerates with subordinate sandstones, siltstones

and claystones. Conglomerate clasts range in size from pebbles to boulders. Sandstone is

greenish grey to light grey, coarse grained, soft and cross-bedded. Claystone is orange,

brown, pale and pink in color. The exposed thickness is 200-300 m. The color of shale

varies from brown, orange, pinkish and reddish brown. Appearance of the fast-invading

or migratory forms of Pinjor fauna (e.g. Equus, Camelus) is the characteristic faunal

change in the Upper Siwalik subgroup.

Tatrot Zone: The Tatrot type locality is situated in district Jhelum about 60 km south of

Jhelum city. The formation is characterized by fine, medium and coarse-grained grey

sandstones, variegated mudstones and siltstones deposited by low sinuosity streams,

mainly of trunk river system. Upper contact with Pinjor Formation is transitional. Grey

beds present in the transitional zone gradually disappear while brown sandstones and

mudstones become prominent. Lower contact is not well exposed (Dennell et al., 2008).

24

Pinjor Zone: Dr. G. E. Pilgrim of the Geological Survey of India first recognized Pinjor

zone of the Upper Siwaliks in 1913. The type rocks of this zone were first of all

differentiated near the village Pinjor in Indian Punjab. The zone consists of brown to

greyish brown, fine, medium and coarse-grained sandstones, multistory sandstones,

pebbly sandstones, pedogenic and non-pedogenic over bank facies probably deposited by

high gradient low sinuosity streams. Upper contact with lower Boulder Conglomerates is

transitional. Pebbly beds gradually increase while sandstone and mudstone beds

gradually decrease (Nanda, 2002, 2008). Dennell et al. (2008) studied the taphonomic

record from Pabbi hills (Pinjor Stage) and found that the fauna was dominated by

herbivores (particularly bovids) > 100 kg adult body size. The largest fossil

concentrations were found in silts and fine sands in abandoned and shallow stream

channels. The main agents of fossil accumulation were large predators and streams. The

Pinjor mammalian fauna marks the end of the record of the Siwalik vertebrate faunal

succession since the overlying Boulder Conglomerates Formation, the youngest

formation of the Siwalik Group, is devoid of fossils. About 49 mammalian taxa are

restricted to the Pinjor Formation (Nanda, 2002). Pinjor mammalian fauna, ranging in age

from 2.58 to 0.6 Ma, is the youngest fauna of the Siwalik Group and is very rich,

comprising of 98 species. The process of faunal extinction and migration started at

1.79 Ma and after 0.6 Ma there is no record of this fauna from the foothills of Himalaya

(Nanda, 2008).

Boulder Conglomerates: It is distributed throughout the Siwaliks and does not have any

type locality. It comprises quartzite pebbles, cobbles, and boulders that are of varying

size, type, density, and orientation. The Boulder Conglomerates is essentially divided into

25

Upper and Lower Conglomerates, both being noticeably distinct units. Lower Boulder

Conglomerates contains brown to greyish-brown, fine, medium and coarse-grained

sandstones. Brown mudstones and pedogenic horizons can also be found (Shah, 1980).

Conglomerates are stratified with large scale cross stratification. The sediments were

probably deposited by braided river channel and proximal alluvial fan. Upper contact

with Upper Boulder Conglomerates is transitional. Alterations of mudstones, sandstones

and conglomerates transitionally pass into thick and massive Boulder Conglomerates.

Upper Boulder Conglomerates is thickly bedded and has massive conglomerates with

pebbles, cobbles and boulders embedded in sandy to silty matrix. Interstratified

sandstones and mudstones can also be located (Cotter, 1933).

Soan Formation: The three lithostratigraphic units (Tatrot, Pinjor, and Boulder

Conglomerate formations) of the Upper Siwaliks are collectively formalized by the

Stratigraphic Committee of Pakistan as Soan Formation (Shah, 1980). The Soan

Formation is composed of pale pinkish orange brown, clay stones, brown grey siltstones

and shales, greenish grey, fine to medium grained sandstones and interbedded dark grey

conglomerates (Shah, 1980; Badgley and Behrensmeyer, 1980; Badgley, 1986).

Chronostratigraphy

During the last two decades, it has been demonstrated that the chronology of the Mio-

Pliocene fluvial sediments of the Siwalik Group from Pakistan, India and Nepal can be

deciphered through correlation of magnetic polarity reversal patterns to the GPTS

(Opdyke et al., 1979; Tauxe and Opdyke, 1982; Johnson G. et al., 1982). These data are

invaluable for regional chronological correlation throughout the Sub-Himalayan belt. The

26

Siwalik localities can usually be tied to the Geomagnetic Reversal Time Scale (GRTS)

with a precision of about 100 ky (Flynn et al., 1990). Of course, locality datings are

dependent on the accuracy of the particular time scale. Permanent forests and woodlands

with some interspersed grasses (mostly C3) were present about 9 Ma, after which wooded

grasslands became widespread on floodplains (Quade et al., 1989; Morgan et al., 1994).

The environmental transition from 9 to 6 Ma in the Potwar, as documented by isotopes,

may have been a response to both regional and climatic shifts. There is no evidence for

any significant Miocene environmental patchiness within the Potwar Plateau; indeed it is

likely that the Plateau is representative of a significantly larger South Asian faunal

province. An extensive set of paleomagnetic determinations has provided good

correlation within and between the various Siwalik sequences, and between the Siwaliks

and, through the GRTS, global record.

The Siwalik fluvial systems winnowed and scattered the faunal remains, and as a result

the sediments contain mostly unassociated and incomplete fossils. Thus, it is difficult to

know how closely a known stratigraphic range approximates the true stratigraphic ranges

for all but the most common taxa. In this regard, Pilbeam et al., (1997) have conducted

intensive biostratigraphic surveys in order to carefully document the first or last

occurrences of a few common taxa (for example, hipparionines and hippopotamids). On

the other hand, there are correlation problems created by several factors. Given the ages

for the boundaries in the time scale used (Barndt, 1977; Berggren et al., 1985; Cande and

Kent, 1995) and the accumulation rates the calculated date for the “Hipparion” Datum in

the Siwaliks is 10.7 Ma. This new date replaces previous estimates of approximately 9.5

Ma (Barry et al., 1982; Barry and Flynn, 1989), and like those dates, is based on

27

stratigraphically older occurrences of this taxon than those used by Badgley (1986), who

interpolated an age of 9.2 Ma. It is of more than historical interest that this “date” has

varied with changes in the GRTS. For example, using the same stratigraphic occurrences

under the Mankinen and Dalrymple (1979) time scale the date for the “Hipparion”

Datum in the Siwaliks was 9.5 Ma, while with Berggren et al., (1985) it was 10.1 Ma and

later with Cande and Kent (1995) it became 10.5 Ma.

Barry et al. (2002) have published a comprehensive paper on Late Miocene of northern

Pakistan in which they developed chronostratigraphic framework of the Siwaliks. Their

study in based on 20 measured sections that range between 250 and 3200 m thick, as well

as 27 shorter sections, 19 of the 20 long sections have determined magnetic-polarity

stratigraphics and 16 of the shorter have at least some magnetic determinations. Apart

from the two sections at Rohtas and Jalalpur (Opdyke et al., 1979; Johnson G. et al.,

1982), the included sections form three regional networks: one on the northern limb of

the Soan Synclinorium at Khaur, another on the southern limb near Chinji, and a third at

the eastern end of the Potwar Plateau near Hasnot. However, because of the distance and

absence of continuous linkage exposures between regions, chronostratigraphic

correlations between the three areas depend on the magnetic-polarity stratigraphy. No

sections were correlated on the basis of biostratigraphy. In the Siwaliks individual

stratigraphic sections of sufficient thickness that contain six or seven magnetic transitions

(on the Potwar typically about 250 m) can usually be correlated to the Geomagnetic

Polarity Time Scale (GPTS) (Johnson and McGee, 1983; Tauxe & Badgley, 1988).

Although sections recording fewer transitions normally cannot be independently

correlated to the GPTS, they can still be confidently placed in the chronostratigraphic

28

framework by determining their stratrigraphic relationships to the longer and better-

constrained sections. This has been done by tracing isochronous lithological horizons

laterally, such as sandstone bodies or paleosol horizons (Behrensmeyer and Tauxe, 1982;

Badgley and Tauxe, 1990; Kappelman et al., 1991). Johnson N et al. (1988) established

six magnetic polarity sections over the Potwar Plateau region of Pakistan, including the

major stratotypes of the Siwalik Group. In all six sections the dominant feature of the

magnetic polarity stratigraphy is the long normal polarity zone, which is contained within

the Nagri Formation. This conspicuous normal polarity zone has been radiometrically

dated at 9.5 ± 0.6 Myr, which identifies it as magnetic Chron 9. They indicate that the

Nagri Formation have nominal age ranges of 7.9-10.1 Ma.

Siwalik Sedimentology

The Siwalik sediments are extensive in Pakistan especially on the Potwar Plateau, where

they are exposed in a folded belt. Clay-mineral suites incorporated in fluvial deposits are

mostly detrital in nature and are a useful tool to understand provenance of the fine-

grained sediments, and composition and climate of the source terrains (Chamely, 1989).

River channels often constitute only transit environments, since the high transport energy

of running water hardly permits the abundant deposition of small, light clay particles,

except in specific environments like downstream alluvial plains and floodplains. Many

workers (Bhattacharya and Misra, 1963; Bhattacharya, 1970; Chaudhri and Gill, 1983;

Bagati and Kumar, 1994; Biswas, 1994; Raiverman and Suresh, 1997; Raiverman, 2002)

have discussed the clay mineralogy of the Neogene sediments of the Middle Siwaliks.

The clay minerals of the Middle Siwaliks have abundant illite and smectite (Bagati and

Kumar, 1994; Raiverman and Suresh, 1997; Raiverman, 2002). The first two drainages

29

were major rivers draining through Higher and Lesser Himalaya, whereas the piedmont

drainage, tributaries of the major river, drained through Sub-Himalayan region (Kumar et

al., 1999; Ghosh et al., 2003). Interfingering of channel deposits was recognized on the

basis of sand body geometry, colour, framework composition and plaeoflow pattern.

However, it is difficult to segregate the floodplain deposits of these different drainages

due to no apparent colour difference among the mudstone.

The section has thick pile of fluvial deposits, comprising sandstones, mudstones and

conglomerates of the Middle (Dhok Pathan Formation) and the Upper Siwalik (Tatrot,

Pinjor and Boulder Conglomerates) sub-groups. This succession is characterized by

stratigraphic coarsening up, stratal thickening and distinct changes in the channel body

geometry and variation in the percentage of overbank to channel deposits. The basal

400m stratigraphic succession (prior to 5.5 myr), dominated by multistory gray sheet

sandstones has abundant erosional surfaces, minor overbank mudstone, lack of lateral

accretion surfaces and low paleocurrent variability. The sizes of cross-bedded set (up to 2

m) and bank-derived intraclast lags indicate braided river system. Southeast paleoflow

suggests the palaeodrainage was parallel to present Himalayan trend and hence represents

the axial trunk drainage system (Kumar et al., 1999). Between 400 and 600 m (5.5 and 5

Ma), the abundance of mudstone increased while the size of the sandstone bodies

decreased. Presence of lateral accretion surfaces and channel plug deposits in these

bodies as well as the abundance of overbank mudstones suggest a meandering river setup.

It is inferred that the sediments of Himalayan foreland basin and contemporary Bengal

Fan have high proportion of smectite during late Neogene (7.4 – 0.5 Ma). Despite

intensification of monsoon in the Himalayan foreland basin at about 8 Ma, Quade et al.

30

(1989) and Burbank et al. (1993) inferred a decrease in the sediment accumulation rate in

the Himalayan foreland basin and the Bengal fan. Einsele et al. (1996) suggested that the

decreasing accumulation rate was probably controlled by exogenic factors rather than by

slowing tectonic activity. Mechanical denudations in the mountain ranges may have

declined as a result of increasing slope stabilization from dense plant cover (Burbank et

al., 1993) under intensified monsoonal climate (Kroon et al., 1991; Quade et al., 1989).

The variation in climatic conditions in the source area may produce different suites of

clay minerals. Illite and chlorite formation predominate during less hydrolyzing, cold-dry

glacial periods, whereas smectite and kaolinite predominate during more hydrolyzing

warm-humid conditions during interglacial stages. The increase in warm humid climate

can trigger chemical weathering. This suggests that not only the narrow belt of basic

rocks exposed in the Lesser Himalayan hinterland but also a higher rainfall may have

played an important role in the distribution of smectite in the Middle and the Upper

Siwalik succession. However, the rarity of smectite in the piedmont river sediments,

which also experienced higher rainfall, is probably due to pre-dominance of sedimentary

rocks in the Sub-Himalayas.

Siwalik Lithology

The sediments of the Middle Siwaliks were deposited by coexisting fluvial systems, with

the larger emergent Nagri system followed by an inter-fan Dhok Pathan system. In

comparison to Nagri floodplains, Dhok Pathan floodplains were not well drained, with

smaller rivers having more seasonally variable flow and more frequent avulsions.

Paleosol sequences indicate reorganization of topography and drainages accompanying a

31

transition to a more seasonal climate (Fatmi, 1973). A few paleosols may have been

formed under water logged, grassy woodlands, but most were formed under drier

conditions and more closed vegetation. The Miocene Siwaliks were deposited in a very

large-scale fluvial system, one comparable in size to the modern Indus or Ganges

systems. These modern rivers occupy divergent basins paralleling the northern and

western mountains and bounded by the Indian craton to the south and east. This

arrangement of mountains, basins and major drainages is a result of the tectonic

movement of the subcontinent against Asia and existed throughout the Miocene.

However, paleochannel directions and paleocurrents indicate that the paleo-Ganges

system may have extended farther west draining the Potwar region (Beck and Burbank,

1990), which now is connected to the Indus drainage basin. The reconstructed Miocene

Indo-Gangetic system extended over 2000 km to the east and 1000 km to the south, with

floodplain widths on the order of 100 to 500 km. Thus, the Potwar Plateau encompassed

only a small part of this ancient foreland basin, providing only limited information on the

whole system (Willis and Behrensmeyer, 1995).

The preserved features of the fourth-order floodplain streams indicate that they were

typically cut into preexisting floodplain deposits, had single rather than braided channels,

and were much smaller (10 to less than 100 m wide) than the individual channels of the

third-order streams. The preserved deposits are mostly ribbon shaped, indicating that the

channels did not migrate laterally, perhaps because they were stabilized by vegetation

(Willis, 1993a). Flow in these smaller channels was apparently slower and more episodic

throughout the year than in the major channels and may have ceased at times, as the

smaller channels became sites of pounded water (Willis, 1993a).

32

In all three formations the smaller floodplain channels are exceptionally important as

sites of vertebrate bone accumulation. Most of the larger fossil localities are from such

depositional environments or from fine-grained fills associated with the tops of second or

third-order channels (Barry et al., 2002). Floodplain paleosols are well developed,

common, and varied throughout the whole of the Siwalik sequence, although they have

been studied in detail only locally (Retallack, 1991). Both the Chinji and the Dhok Pathan

formations have more well differentiated paleosols than the Nagri Formation, having

overall differences in proportions of channel and overbank deposits. None of the

formations have paleosols that are readily assigned to modern soil types, although they

are broadly similar to the modern soils of the Indo-Gangetic plain (Retallack, 1991;

Behrensmeyer et al., 1995; Quade and Cerling, 1995), especially those forming under

25oC mean annual temperature and 1400 mm/yr precipitation (Behrensmeyer et al.,

1995). The nine paleosol series described by Retallack in 1991 differed primarily in

topographic setting, frequency of disturbances, degree of maturity, drainage, and parent

material as well as inferred vegetation and precipitation. At all stratigraphic levels,

however, the paleosols give evidence of seasonal differences in the water table that

produced waterlogging with formation of Fe-Mn nodules, followed by desiccation with

leaching and precipitation of calcium carbonates. Leaching of matrix carbonate was

essentially complete on coarse deposits and well-drained sites. There is also evidence for

intense oxidation at the time of soil formation, which depleted organics and presumably

helped destroy bone in mature paleosols as well (Willis, 1993a; Behrensmeyer et al.,

1995; Zaleha, 1997).

33

Temporal changes in the Siwalik fluvial system have been documented, but the degree to

which they are related to climate or subsidence, or to which they are simply due to the

autocylic dynamics of the fluvial system, is not clear. The transition between the Chinji

and the Nagri formations has been interpreted as due to the progradation of a second-

order emergent system over a smaller, third-order interfan system (Willis, 1993a; Zaleha,

1997). The Nagri and the Dhok Pathan transition, on the other hand, seems a well-

documented case of the coexistence of two contemporaneous systems (Behrensmeyer and

Tauxe, 1982). The Chinji floodplains were apparently more poorly drained than those of

the Nagri Formation, with the abundance of lacustrine deposits and iron-concretion-rich

paleosols indicating greater seasonal waterlogging. Floodplain deposition in the Chinji,

and perhaps the Dhok Pathan Formation was also more episodic than in the Nagri, with

longer-lived small floodplain channels. Initially, flow in these small channels was

continuous year-round, but in later stages it became more episodic with periods of

subaerial exposure as the channels were abandoned and later infilled. By contrast, in the

Nagri Formation even the initial stages of channel flow may have been more seasonal,

with deposition of sediments occurring mostly during floods. Specific and important

differences between the rivers of the Nagri and Dhok Pathan formations include

decreased channel size and discharge in the latter, as well as decreased avulsion period

(Zaleha, 1997).

34

Taphonomy

The Siwalik Group of the Potwar Plateau, in the northern Pakistan, contains a rich

vertebrate fossil record in predominantly fluvial deposits that spans for most of the

Neogene. The sequences contain a broad range of fluvial deposits and the frequency of

facies changes markedly over time and space (Badgley et al., 1995). Taphonomic

research to date has focused on the distribution of fossil localities among depositional

environments, inferring to condition of mortality and accumulation, and reconstructing

the profusion of taxa in the original community (Badgely and Behrensmeyer, 1980; Raza,

1983; Badgley, 1986; Behrensmeyer, 1988). At the level of depositional system, the

particular distribution of local environments determines the habitats available to

organisms. The rates and processes of preservation may vary greatly among these

environments in relation to biotic and abiotic components. Recognition of changes in

taphonomic selectivity facilitates the distinction between apparent and real changes in

original biotas (Koch, 1987; Badgley and Gingerich, 1988). Three vital aspects of fossil

assemblages verify the reliability of inference regarding the original faunal composition,

the associations amongst taxa, the rates of morphological evolution within lineages and

the patterns of immigration and extinction. Changes in the preservational bias may

expose significant environmental changes that can be correlated with changes in biotic

composition or fossil productivity (Behrensmeyer, 1988; Kidwell, 1988; Badgley et al.,

1995). Fossil assemblages from the Siwalik deposits reveal features indicative of fluvial

transportation and deposition of abraded bones, bones dispersed through the sediment

matrix, absence of skeletal association of the fossilized animals and lack of the more

35

transportable elements such as vertebrae and ribs. Teeth and jaws bones are the major

constituents of the assemblages.

The taphonomic study of the fossil material collected from the Siwaliks reveal a variety

of pre-burial and post-burial processes that affected the bones and teeth deposited in the

Siwaliks. Significant modifications were observed in the vast majority of the examined

specimens. Extensive weathering cracks are indicative of the long-term exposure of the

collected specimens on ground. Partly articulated, partly associated and mostly dispersed

skeletal parts point out the long transportation and the significant dispersal of the

occurred skeletal elements. Seismo-turbation and faulting caused the post burial

fracturing of various skeletal elements.

Palaeoenvironment of the Siwaliks

There are evidences that the palaeoclimate during deposition of the Siwalik Group was

warm, humid, sub-tropical to tropical, and monsoonal. These evidence comes from the

nature of the palaeosols (Cerling et al., 1993; Quade et al., 1989; Willis, 1993b; Zaleha,

1994), isotopic studies of marine microfossils (Wright and Miller, 1993), plant material

(Sahni and Mitra, 1980), and climate modeling (Iacobellis and Somerville, 1991a, b;

Kutzbach, et al., 1989; Prell and Kutzbach, 1992; Ruddiman et al., 1989; Raymo and

Ruddiman, 1992). The relatively constant thickness of the horizons of mature paleosols in

different formations (age 15-8 Ma) in the Chinji area was taken to imply constant mean

annual rainfall by Willis (1993b). However, studies of the isotopic compositions of

palaeosol carbonate nodules and fossil teeth (Quade et al., 1989; Cerling, et al., 1993)

suggest a major change in vegetation from dominantly trees and shrubs to dominantly

36

grasslands at 7-4 Ma. Morgan et al. (1994) has proposed the beginning of the change at

9.4 Ma. This change in vegetation was also associated with major changes in the fauna,

with less woodland-dependent fauna and more grazing fauna (Barry et al., 1985; Morgan

et al., 1994), and a cooler and drier climate. The proposed climate change around 7.4 Ma

is not reflected in changes in alluvial architecture, and it seems that climatic changes

were not important enough during the deposition of the Nagri and Dhok Pathan

formations to have a marked effect on deposition (Willis, 1993b; Zaleha, 1994). There is

evidence for accelerated formation of the Antarctic ice cap since c. 15 Ma, associated

with episodically falling sea level, decrease in atmospheric CO2, and general global

cooling (Klootwijk et al., 1992; Zaleha, 1994). There is a particularly major eustatic sea

level fall at c.10.8 Ma, near the base of the Nagri Formation and a vivid decrease in

atmospheric CO2, from 11 to 8Ma (Freeman and Hays, 1992). It is therefore possible that

there was at least a glacial period during deposition of the Nagri Formation, and it is

possible that the higher Himalayas were glaciated. Such global climatic change would not

essentially have a major effect on the climate of the Indo-Gangetic foreland due to its low

latitude and elevation. Zachos et al. (2001) have suggested similar transient climates in

the Oligocene. Evidence from the modern Indus valley near the Himalayas indicates that

aggradation rates increased by an order of magnitude during the last (Pleistocene) glacial

advance, and have progressively decreased up to now (Jorgensen et al., 1993). Hence,

increasing deposition rates in the Nagri Formation may be associated to increased erosion

rates and sediment supply from a partly glaciated locality (Khan et al., 1997).

37

Siwalik Faunas

The mammalian fauna of the Siwaliks was reported by various workers as Lydekker,

1876, 1878; Colbert, 1935; Matthew, 1929; Pilgrim, 1937, 1939; Pascoe, 1964; Sarwar,

1977; Shah, 1980; Akhtar, 1992, 1995, 1996, Akhtar et al., 1997; Barry et al., 1982,

1985, 1991, 1995, 2002, 2005; Farooq et al., 2007a, b; Khan, 2007, Khan et al., 2005a, b,

2006, 2009a, b; Iqbal et al., 2009; Khan A. M., 2010. The Siwalik faunas include the

youngest known creodonts and adapted primates, endemic radiations of murid and

rhizomyid rodents, viverrid, carnivores, tragulid and bovid artiodactyls and hipparionine

(equid) perissodactyls as well as cosmopolitan taxa (Pilbeam et al., 1979; Lipson and

Pilbeam, 1982; Barry et al., 1982). The following mammalian fauna is reported from the

Siwaliks:

Kamlial Formation: Primates – Anthropoids; Suids – Conohyus, Listriodon;

Proboscidea – Dinotherium, Trilophodon, Tetrabelodon, Telemastodon; Perissodactyla –

Brachypotherium; Hyanidae – Hyaenelurus; Anthracotheriidae – Merycopotamus;

Artiodactyla – Dorcatherium and Eotragus.

Chinji Formation: Primates – Sivapithecus sivalensis, S. indicus, Ramapithecus

punjabicus; Rodentia – Rhizomyoides punjabiensis, Copemys sp., Megacricetodon sp.,

Antemus chinjiensis; Carnivora – Hyanailouros bugtiensis, Dissopsalis carnifex;

Tubulidentata – Orycteropus pilgrimi; Chalicotheriidae – Chalicotherium salinum;

Suidae – Listriodon pentapotamiae, Conohyus chinjiensis, Lophochoerus sp.;

Anthracotheriidae – Merycopotamus pusillus; Tragulidae – Dorcatherium majus,

Dorcatherium minus, Dorcabune anthracotherioides; Bovidae – Miotragocerus gluten,

Kubanotragus sakolovi, Sivoreas eremite, Gazella sp.; Giraffidae – Giraffokeryx

38

punjabiensis, Giraffa priscilla. Rhinocerotidae – Brachypotherium perimense,

Gaindatherium browni, Aprotodon fatehjangense, Chilotherium intermedium.

Nagri Formation: Primates – Lorisidae indet, Sivapithecus sivalensis, S. indicus,

Ramapithecus punjabicus, Gigamopithecus sp., Sivapithecus parvada; Rodentia –

Rhizomyoides sp., Kanisumys sivalensis, Progonomys, Parapodemus sp.; Carnivora –

Viverridae, Viverra chinjiensis; Hyaenidae, Progenetta sp., Pathyaena sivalense,

Miohyuena sp., Percurocuta carnitex, Percurocuta grandis; Felidae – Sivaelurus

chinjiensis; Mustelidae – Martes lydekkeri, Mustlinae sp., Eomellivora sp., Sivaonyx

bathygnathus; Proboscidae – Deinotherium sp., Gomphotheriidae indet.; Equidae –

Hipparion small and large species; Chalicotheriidae – Chalicotherium cf. salinum;

Suidae and Tayassidae – Propotamochoerus hysudricus, Propotamochoerus sp.,

Conohyus sindiensis, Tetraconodon magnus; Anthracotheriidae – Merycopotumus nanus,

M. dissimilis; Tragulidae – Dorcabune nagrii, D. anthracotherioides, Dorcatherium

majus, D. minus; Bovidae – Gazella lydekkeri, Tragoportax punjabicus, Selenoportax

vexillarius, Pachyportax latidens, Elachistoceras sp., Miotragocerus gluten, Sivaceras

gradiens. Equidae – Hipparion small and large species; Rhinocerotidae – Aprotodon

fatehjangense, Chilotherium intermedium, Brachypotherium perimense, Caementodon

oettingenae, Gaindatherium vidali, Alicornops complanatum.

Dhok Pathan Formation: Primates – Cercopithecidae, Cercopithecus hasnoti, Macacus

sivalensis; Pongidae, Dryopithecus frickae, Palaeopithecus sivalensis, Palaeopithecus

sp.; Rodentia – Spalacidae, Rhizomys sivalensis, Rhizomys sp., Muridae; Hystricidae,

Hystrix sivalensis; Carnivora (Fissipedia, Canidae) – Arctamphicyon lydekkeri; Indarctos

punjabiensis, Promellivora punjabiensis, Vishnuictis salmontanus, lctitherium sivalense,

39

I. indicum, Lycyaena macrostoma, L. macrostoma-cinayaki, Crocuta carnifex, C.

gigantean, C. gigantean-latro, C. mordax, Mellivorodon palaeindicus, Acluropsis

anneclens, Paramachacrodus pilgrimi, P. indicus, Propontosmilus sivalensis, Felis sp.;

Proboscidea – Dinotherium indicum, D. angustidens (?), Trilophodon hasnotensis,

Tetralophodon falconeri, T. punjabiensis, Rhynchotherium chinjiensis, Synconolophus

dhokpathanensis, S. propathanensis, S. corrugatus, S. ptychodus, S. hasnoti, Anancus

perimensis, Stegolophodon latedens, S. cautleyi, Stegodon bombifrons, S. cliftii, S.

elephantoides; Perissodactyla – Hipparion antelopinum, H. theobaldi, H. perimense,

Aceratherium perimense, Aceratherium lydekkeri, A. blanfordi, Chilotherium

intermedium; Artiodactyla – Tetraconodon magnus, T. mirabilis, Listriodon

pentapotamiae, Propotamochoerus uliginosus, P. hysudricus, Dicoryphochoerus titan,

D. titanoides, D. vagus, D. vinayaki, Hyosus punjabiensis, H. tenuis, Sivahyus hollandi,

Hippohyus lydekkeri, H. grandis, Sus comes, S. adolescens, S. praecox, Chocromeryx

silistrense, Merycopotamus dissimilis, Dorcabune latidens, Dorcatherium majus, D.

minus, Cervus simplicidens, C. triplidens, Vishnutherium iraratieum, Bramatherium

perimense, Hydaspitherium megacephalum, H. grandi, H. magnum, H. birmanicum,

Giraffa punjabiensis, T. perimensis, Proleptobos birmanicus.

Soan Formation: Proboscidea – Mastodon sivalensis, Stegodon clifti, S. bombifrons,

Elephas (Archidiskodon) cf. planifrons, Anancus falconeri; Equidae – Hipparion sp.,

Equus sp.; Suidae – Potamocheorus palacindicus, Hippohyus grandis, H. lydekkeri; Sus

peregrinus; Hippopotamidae – Hippopotamus sp.; Giraffidae – Sivatherium giganteum;

Bovidae – Sivaonyx bathygnathus, Antilope cf. planicrnis, A. cervicapra, Hydaspicobus

auritus, Proamphibos lachrymans, Hemibos sp., Leptobos sp., Bos sp., Bubalis sp.

40

SYSTEMATIC PALAEONTOLOGY

Order Artiodactyla Owen, 1848

Family Suidae Gray, 1821

Subfamily Listriodontinae Gervais, 1859

Genus Listriodon Von Meyer, 1846

Type Species: : Listriodon splendens Von Meyer, 1846

Generic Diagnosis: Molars are lophodont. Tooth crests are perfect and have sharp

cutting edges. Teeth are smaller in size than other genera of the family Suidae. Talon in

M3 present and varies in size in different species of the genus and symphysis also present

(Colbert, 1935). The listriodonts are middle Miocene suids possessing features such as

primitive basicranium, unflared zygoma, parietal lines not widely separated, no canine

flanges, rounded snout and low glenoids. They possess a very elongated mandible,

achieved both by elongation of the symphysis as well as by retiring the ascending ramus.

In side view, the whole of M3 is visible as well as gap behind M3. The symphysis is

splayed outwards, so that the lower canines emerge almost horizontally. The incisive

margin is evenly curved and projects substantially in front of the canines. There is long

diastema between the canines and anterior premolar (P2). The borders of diastema lie well

below the occlusal surface of the cheek teeth. P1 is reduced or lost in most species. In

Listriodontinae the I1 is spatulate and occludes with I1-2. In Listriodon females, upper

canines are usually two rooted if they are not hypsodont although the lower canines seem

to be more nearly single rooted. I2 is a robust triangular tooth set vertically in the 41

premaxillae. The tip is triangular in lingual view, with a lingual cingulum and a central

rib. The crown is slightly offset from the root. There are two wear facets along the

occlusal edge of the tooth; there are two wear facets along the occlusal edge of the tooth,

the mesial facet corresponds to the outer portion of the scoop-shaped distal edge of I2,

while the distal one is caused by wear with the root ward half of the scoop in I2. In I2 the

facet caused by I2 is very prominent along the distal edge and in the body of the scoop,

while I1 occludes only at the tip. Unworn I2 has bifurcate tip. M1 is a square tooth with

four main cusps disposed in two lophs, with anterior and posterior cingula. The anterior,

median and posterior accessory cusps although present in all suids are very small in

Listriodon, and soon disappear with wear (Pickford, 1988; Van der Made, 1996).

Known Distribution: The genus Listriodon is known from Europe, Africa as well as

from the Lower Siwaliks and lower portion of Middle Siwaliks (Pickford, 1988). In

Europe (MN4 – MN7) it is known in the basal Middle Miocene deposits, in Africa it is

known from Ngorora Formation and from the Siwaliks known from the Chinji Formation

and the lower part of the Middle Siwaliks (Pilgrim, 1926; Pickford, 1988, 2001; Pickford

and Morales, 2003).

Listriodon pentapotamiae Falconer, 1868

Type Specimen: GSI B107, a complete right M2 and fragment of right M3; also right and

left P4.

Type Locality: Khushalghar, Pakistan (Pickford, 1988).

Stratigraphic Range: Lower Siwaliks and lower portion of the Middle Siwaliks

(Colbert, 1935; Pickford, 1988, Pickford and Morales, 2003; Khan et al., 2005b).

42

Diagnosis: A species of Listriodon similar in size to L. splendens of Europe, but in which

the upper central incisors are shorter mesiodistally and smaller; the upper canine shorter

and narrower; P1 usually present, but rudimentary; a large talon on the third molar, a

strong cingulum in the fourth premolar, the shortness and more slenderness of symphysis

(Pickford, 1988; Van der Made, 1996).

Studied Specimens: Upper dentition: PC-GCUF 10/04, left first upper incisor (I1); PUPC

07/73, a maxillary ramus with M1-2. Lower dentition: PC-GCUF 10/05, isolated left P4;

PUPC 07/72, almost complete mandible with the partial canines, the right hemimendible

with M1-3 and the left hemimendible with M2-3.

Description

Upper Dentition

The upper incisor, PC-GCUF 10/04 (Fig. 1(1)) is in early wear on its lingual aspect, but

is otherwise well preserved. A cutting edge is present mesially which forms due to

occlude with I1-2. The tooth is wide mesiodistally and its apical edge is divided into three

lobes, the central one being the narrowest. The first incisor is a spatulate tooth with

complete lingual cingulum. The occlusal tip has a deep sulcus near its mesial edge. The

root is narrower than the crown.

PUPC 07/73 have lophodont upper first and second molars in late wear (Fig. 1(2)). A

small part of the palatine is associated with the maxillary ramus. Cingulum is strong

anteriorly and somewhat weak posteriorly. An evidence of cingulum is also present on

the buccal as well as on the lingual sides. The buccal margin of the molars possesses

relatively a well developed cingulum. The large dentinal islets indicate the late age of the

animal. The posterior dentinal islet is more prominent than the anterior one. The buccal

43

cones are vertically higher than the lingual ones. The enamel is thick. The molars are

square shaped with four main cusps disposed in two lophs. The anterior, median and

posterior accessory cusps are disappeared due to the late wear. These are very small in

Listrodon and soon disappear with wear (Pickford, 1988). The upper first and second

molars are so worn that little occlusal morphology is preserved. The M1 and M2 have

almost same size and appearance.

Mandible

PUPC 07/72 is a complete mandible bearing partial canines with the M1-3 in the right

hemimendible and the M2-3 in the left hemimendible (Fig. 1(4)). The mandible has long

diastema and flat symphysis. The ascent begins well behind M3, so there is a gap between

the M3 and the ascending ramus. The mandible is deep and broken fromwhere the

ascending ramii retiring upwards. The symphysis is splayed outwards, so that the lower

canines emerge horizontally. The lower border of the jaw below the third molar

terminates in a prominent flange and lingual tubercle which is separated from the slightly

descending angle by a low crest of bone. The internal and external surfaces of the jaw

distal to the third molar are marked by well developed rugosities representing muscle

attachments. The length of the molar series is 64 mm. The length of the mandible from

anterior to posterior (PUPC 07/72) is 128 mm and the depth of the mandible at m3 is 44

mm.

Lower Dentition

The canines are broken at the apex in PUPC 07/72 (Fig. 1(4)). Both canines have

triangular cross section. They emerge almost horizontally and sweep outwards. The tooth

appears to have grown during all individual’s life span. The male lower canines are

44

permanently growing teeth of triangular section. In females the tooth is oval in section,

and has closed roots (Pickford, 1988). The incisors and premolars are missing in the

recovered mandible (PUPC 07/72). Nevertheless, the alveoli of the premolars are

preserved ((Fig. 1(4)).

The lower fouth premolar, PC-GCUF 10/05 is an isolated excellently preserved molar

(Fig. 1(3)). The P4 is rectangular in occlusal outline with a very prominent innenhugel

which is large and far offset from the main cusp. The cingula are large and the posterior

accessory cusp is very prominent, placed closer to the buccal side of the tooth. The

prominent talonid cusp joined lingually and buccally by a swollen cingulum.

The lower studied molars are early in wear ((Fig. 1(4)). The molars reflect lophodonty.

The transverse valleys are wide. The fact that the lophids appear higher is possibly

caused by a decrease of the antero-posterior diameter at the base of the lophid, resulting

in wider transverse valleys with steeper slopes and by an increase of the transverse

diameter at the top of the lophid. The M1 is a four conids tooth but fragile. The conids are

being disposed in two pairs forming lophs as in the upper molars. However, the lower

molar is narrower than the upper and has less lingual and buccal flare. The posterior

accessory cusplet is prominent and centrally placed ((Fig. 1(4)). The M2 is a larger

version of M1. The second molar is a four conid tooth with anterior, median, and

posterior accessory cusplets in the midline of the crown ((Fig. 1(4)). The posterior

accessory cusplet is prominent and centrally placed. The molar is bunodont with the usual

suid layout of four main cusps arranged in two lophs. The ectostylid is absent in the

transverse valley. The M3 differs from the M2 by the presence of talonid and wide

anterior lophid. The talonid of the M3 is simple. It is really an enlarged cingulum,

45

surrounding the posterior accessory cusplet. The comparative dental measurements are

provided in table 1.

Table 1: Comparative measurements of the cheek teeth of the L. pentapotamiae in mm

(millimeters). * The studied specimens. Referred data are taken from Colbert (1935),

Pickford (1988), Van der Made (1996) and Khan et al. (2005b).

Taxon Number Nature/Position Length Width L. pentapotamiae PC-GCUF 10/04* I1 17.0 9.70

PUPC 07/73* M1 19.5 20.0 M2 20.0 20.0

PC-GCUF 10/05* P4 13.0 11.5 PUPC 07/72* M1 17.0 12.0

M2 19.0 14.0 M3 28.0 17.0

PC-GCUF 08/22 I1 21.7 9.90GSP 1424 I1 22.7 11.3GSP 1378 I1 21.7 10.5K 15/777 I1 21.7 9.80K 13/772 I1 19.2 10.5K 13/770 I1 19.6 11.0K 13/767 I1

21.5 10.6K 15/535 I1 20.0 10.3K 13/774 I1 20.0 10.7K 15/813 M1 15.5 17.0M 13586 M1 17.3 16.3M 13590 M1 15.7 13.8AMNH 19644 M2 18.0 18.0 M 13257 M2 18.3 18.3M 13586 M2 19.7 19.5M 13590 M2 17.9 16.0M 31869 M2 20.1 19.7K 15/813 M2 19.6 20.0K 15/813 M3 23.0 21.0K 22/435 M3 26.7 24.0K 13/808 M3 23.0 19.0K 13/803 M3 22.9 20.0M 13257 M3 21.0 20.4M 31869 M3 23.5 20.3GSP 1606 M3 21.7 20.5 AMNH 29836 M3 23.0 20.0

46

Table1 (continued) K 13/808 P4 15.3 12.3

K 13/436 P4 17.4 11.3K 23/721 P4 16.1 12.5K 14/492 P4 16.5 11.8K 15/520 M1 19.0 14.2GSP 4412 M1 16.8 13.0GSP 4527 M1 15.7 14.0GSP 4413 M1 17.3 14.0GSP 949 M1 15.3 10.8M 31867 M1 16.5 13.2M 13587 M1 17.8 13.7K 15/520 M2 22.0 17.4GSP 4527 M2 22.0 17.0GSP 4413 M2 22.0 16.9GSP 4412 M2 21.0 16.6GSP 4423 M2 23.0 17.0GSP 4478 M2 23.4 18.3M 31873 M2 20.5 14.7M 13592 M2 21.5 16.7AMNH 19519 M2 19.0 16.0PUPC 99/18 M2 13.0 13.5AMNH 19432 M2 19.0 16.0K 41/858 M3 29.5 16.0K 41/862 M3 29.5 18.0K 41/870 M3 30.7 26.7K 41/841 M3 25.0 16.4K 19/138 M3 33.0 19.0K 13/206 M3 29.4 17.7K 13/806 M3 30.7 17.2K 23/512 M3 33.7 19.7GSP 4527 M3 36.5 20.0GSP 4413 M3 35.3 18.7GSP 4412 M3 32.5 18.8GSP 1360 M3 31.7 18.2M 31873 M3 29.5 16.8M 13592 M3 30.6 19.0AMNH 19424 M3 31.0 19.0

AMNH 19519 M3 28.0 16.0

47

Figure 1: Listriodon pentapotamiae. 1. PC-GCUF 10/04, lI1. 2. PUPC 07/73, right

maxillary ramus with M1-2. 3. PC-GCUF 10/05, lP4. a = occlusal view, b = lingual view, c

= buccal view. 4. PUPC 07/72, almost complete mandible with the partial canines, the

right hemimendible with M1-3 and the left hemimendible with M2-3: a = occlusal view, b =

buccal view. Scale bar equals 10 mm.

48

Comparison and Discussion

The sharp chisel shaped crest is a feature seen in the molar teeth of deinotheriid

proboscideans, lophodonts pigs and some metatheres. The tooth under discussion is too

small to be referred to any of the proboscideans. In lophodont metatheres the crest is

imperfect while in lophodont pigs i.e. listriodonts, the tooth crests are perfect with very

sharp cutting edges. All lophodont pigs are placed in a single genus, Listriodon that

consists of three species, of these, the species Listriodon pentapotamiae is the smallest

and is known from the middle Miocene of the Siwaliks (Pickford, 1988). Structurally, it

is the most primitive species.

The specimens examined here belong to species L. pentapotamiae and are comparable to

the specimens studied by Colbert (1935), Pickford (1988) and Van der Made (1996)

(Table 1; Fig. 2). The most striking feature of L. pentapotamiae’s mandible is its very

long diastema and flat symphysis which can be seen in the studied sample PUPC 07/72.

In Listriodon pentapotamiae the male lower canine has triangular cross section which is

observed in the sample. The lower molars are characterized by bilophids, possess

elongated crown, development of post talonid which is well raised and tuberculated, and

have a chisel shaped cutting edge. All the features correspond to species Listriodon

pentapotamiae and consequently the recovered sample is assigned to the Siwalik suid

Listriodon pentapotamiae which is very common in the Siwalik middle Miocene. But the

rare findings are found in the lower part of the middle Siwaliks (Pickford, 1988). The

new sample is also found from the lower part of the Middle Siwaliks, confirms its

stratigraphic range in the earliest late Miocene of the Siwaliks.

49

The genus Listriodon was founded by Von Meyer (1846) on dentition discovered from

molasses of Switzerland, which he described under the name L. splendens. Falconer

(1868) described the second molar of maxilla under the name Tapirus pentapotamiae. In

the year 1876, Lydekker studied two isolated molars from the Salt Range area of the

Punjab, and referred it to the genus Listriodon, one of which Falconer had named Tapirus

pentapotamiae. In 1884, he refigured and described these, together with certain additional

isolated molars, upper and lower, from the same area. He assigned his material to two

species, the original L. pentapotamiae and L. theobaldi Stehlin. Lydekker (1879) pointed

out that the two upper molars from the Laki Hills of Sind, figured by Lydekker under the

names (?) Hyotherium sp. and (?) Hyotherium sindiense, belong in reality to the genus

Listriodon. Colbert (1935) described and referred some maxillary and mandibular

fragments to the genus Listriodon. Lydekker (1876) distinguished Listriodon theobaldi

from Listriodon pentapotamiae on the basis of size. He documented that structurally no

constant distinction could be drawn between smaller teeth of L. theobaldi and the larger

teeth of L. pentapotamiae.

From the Siwaliks the genus Listriodon is known by three species L. pentapotamiae, L.

theobaldi and L. guptai (Pilgrim, 1926; Colbert, 1935). Listriodon theobaldi is much

smaller than the L. pentapotamiae. Pickford (1988) placed all the middle Miocene

Siwalik lophodont pigs in L. pentapotamiae which is considered the smallest and

primitive species of the genus Listriodon (Van der Made, 1996).

Pilgrim (1926) has referred to the existence of bunodont species of Listriodon in the

lower Siwalik horizon of Sind, the Kamlial zone; this may be compared with L. lockharti

and L. latidens of the Burdigalian and Vindobonian of Europe. Listriodon pentapotamiae

50

is stratigraphically fairly long ranging species, extending from the base of the Lower

Siwaliks to the lower portions of the Middle Siwalik beds (Pickford, 1988; Khan et al.,

2005b). Listriodon pentapotamiae is a fairly long ranging species, extending from the

base of the Lower Siwaliks well up into the Middle Siwalik beds. Several specimens in

the American Museum collection from the lower portion of the Middle Siwaliks should

definitely establish the persistence of this genus beyond its typical Chinji development, a

fact that was of some doubt to Matthew (1929).

Listriodon pentapotamiae is very close to L. splendens, from the Miocene of southeastern

Europe, a fact that was pointed out in detail by many earlier researchers (e.g. see Pilgrim

1926; Colbert, 1935; Pickford, 1988; Van der Made, 1996). It is evident that the species

L. pentapotamiae is allied to L. splendens and has reached the same stage of development

as regards the formation of the molar crests, but in those features in which it differs from

that species it seems to show a more primitive structure, which approximates to that of

the bunodont forms L. lockharti and L. latidens (Van der Made, 1996).

Listriodonts disappeared more or less simultaneously everywhere, around the arrival of

Hipparion. In Europe, the last L. splendens is known in MN 9/10 transition. In Pakistan,

L. pentapotamiae is known to co-occur with Hipparion (Hussain, 1971; in this thesis). In

Africa, Lopholistriodon kidogosana is found in Member D of the Ngorora Formation,

which is dated between 9.7 and 9.8 Ma and which has locally the first Hipparion. In

Europe, the density of data is greatest and indicate that L. splendens became extinct a

considerable period after the entry of Hipparion (the whole of MN 9). At about this time,

there was a marked drop in suoid diversity in Europe, but not in Pakistan (Van der Made,

1996).

51

The origin of the Listriodontinae is unknown. The oldest Suoidea known are the

Tayassuidae from the Oligocene of North America (Pearson, 1932), the Palaeochoeridae

from the Oligocene of Europe (Ginsburg, 1974; Van der Made, 1994) and the

palaeochoerid Odoichoerus from the Eocene (?) of China (Tong et al., 1986). The first

suids appear in Europe in MN 1 as immigrants. This indicates that Suidae probably

originated in Asia. The first record of Listriodontinae is from Africa in Set I (Faunal Sets:

Pickford, 1981) and in Bugti, Pakistan. They are absent in Meswa Bridge (Set 0) and in

Pakistan, there is no earliest Miocene or Oligocene record of mammals. Later members

of the subfamily are found in Europe and China, suggesting that the earliest listriodonts

evolved somewhere south of the Himalayas (Van der Made, 1996).

L. pentapotamiae

0

5

10

15

20

25

30

0 5 10 15 20 25 30 35 40

Length

Wid

th

Lower Fourth Premolar Lower First MolarLower Second Molar Lower Third Molar

Figure 2: Scatter diagram showing dental proportions of L. pentapotamiae’s studied

sample. Referred data are taken from Colbert (1935), Pickford (1988), Van der Made

(1996) and Khan et al. (2005b).

52

Family Bovidae Gray, 1821

Tribe Boselaphini Knottnerus-Meyer, 1907

Genus Selenoportax Pilgrim, 1937

Type Species: Selenoportax vexillarius Pilgrim, 1937.

Generic Diagnosis: Moderate to large sized Siwalik bovid; skull wide both at frontals

and occipital, face slightly bent down on the cranial axis; frontals moderately depressed

behind the horn-cores and form slightly elevated surface between the horn-cores;

hypsodont to extremely hypsodont teeth, upper molars quadrate with strong divergent

styles, median ribs well developed, entostyle strongly developed and ectostylid

moderately developed, enamel very rugose (Pilgrim, 1937). Crown is narrow at the base

and broad at the apex in Selenoportax whereas in Pachyportax the crown is not

constricted at the apex. Entostyle is strong and much extending transversely in

Pachyportax while in Selenoportax it is not much extending transversely. In Pachyportax

posterior median rib is flattened whereas in Selenoportax it is strong as anterior median

rib (Pilgrim, 1937, 1939).

Known Distribution: The genus Selenoportax is well known from the Nagri and the

Dhok Pathan formations of the Middle Siwaliks (Pilgrim, 1937; Akhtar, 1992; Khan et

al., 2009a).

Selenoportax cf. vexillarius Pilgrim, 1937

Type Specimen: AMNH 19748, a skull lacking maxilla and dentition and most of the

basicranium.

53

Type Locality: Hasnot, Jhelum, Punjab, Pakistan (Pilgrim, 1937).

Stratigraphic Range: Middle and Upper Siwaliks (Pilgrim, 1937, 1939; Akhtar, 1992;

Khan et al., 2009a).

Diagnosis: Cheek teeth large and strongly hypsodont, enemal very rugose. Upper molars

quadrate with strong and divergent styles near the neck of the crown, ribs quite large,

entostyle and ectostylid strongly developed. Central Cavities without indentations and

simple in outlines, transverse anterior goat folds developed at front of lower molars

(Pilgrim, 1937, 1939).

Studied Specimens: Upper dentition: PC-GCUF 10/07, isolated left M1. Lower

dentition: PC-GCUF 10/06, isolated left incisor (I1); PUPC 09/117, isolated right M1;

PUPC 07/135, a fragment of right mandible having M1-3.

Description

Upper Dentition

The first upper molar PC-GCUF 10/07 (Fig. 3(1)) is in an excellent state of preservation

and in middle wear. The enamel is finely rugose and the rugosity is more evident on the

lingual side than on the buccal side. The entostyle is strongly developed, exposing the

dentine at the apex. The principal cones are well developed and the buccal cusps are

higher than the lingual ones, which at this stage of wear are not attached to each other at

the transverse valley. The protocone is V-shaped. The styles and median ribs are well

developed. The central cavities are wide and no spur of enamel seems to project into

these central cavities.

54

Lower Dentition

The left lower incisor is in early wear (Fig. 3(2)). It has a simple outline. The incisor has

a wide cutting edge with the outer angle pulled outwards. In buccal view the crown is

slightly inclined upwards posteriorly.

PUPC 09/117 is a well preserved and in early wear (Fig. 3(3)). The appearance of the

molar indicates a high crowned and narrow tooth. The enamel is thick and shows fine

plications all over the crown. These plications are more prominent and distinct on the

buccal conids than on the lingual ones. The anterior transverse flange is developed on the

anterior side of the lower molar. The ectostylid is strongly developed and looks as an

isolated pillar in the transverse valley. As it is commonly observed the lingual conids are

higher than the buccal ones. The protoconid is crescentic in shape. The praeprotocristid is

larger than the postprotocristid. The metaconid is represented antero-lingually with two

slightly worn sloping cristids. The entoconid is slightly higher than the metaconid and

pointed in the middle. The wear is more distinct to the center of the entoconid than to the

sloping cristids. The hypoconid is more V-shaped than the protoconid. The metastylid

and the entostylid are strongly developed while the mesostylid is not distinct. The median

ribs are developed but these are distinct to the base of the crown. The central cavities are

moderately wide and deep, having no indentation (Fig. 3(3)).

The fragile mandible PUPC 07/135 has many vertical cracks and in a poor state of

preservation (Fig. 4(4)). It is broken anteriorly and posteriorly. A small part of ascending

ramus is present posteriorly behind the 3rd molar. The molars on the mandible are in an

excellent state of preservation but the premolars are missing. The roots of the P4 and the

P3 are preserved. The M1 has a long and wide transverse valley between the anterior and

55

posterior ribs. There is an ectostylid present in the transverse valley. The central cavities

are wide and deep. The anterior central cavity is compressed at its centre. The M2 is

comparatively larger than the M1. The ectostylid is not visible owing to the deposition of

sand stone. The anterior transverse flange is large enough to look as a goat fold. The

metastylid and the entostylid are strongly developed while the mesostylid is not distinct.

The 3rd molar crown is high crowned. The buccal side of the molar is covered with the

matrix and consequently, the occlusal and lingual views are available for the

morphological study. The major conids and hypoconulid are well developed. The

hypoconulid is attached to the ascending ramus posteriorly. The M3 also has strongly

developed metastylid and the entostylid as in the M1 and the M2. The mesostylid is not

prominent. The comparative dental measurements are provided in table 2.

56

Table 2: Comparative measurements of the cheek teeth of S. vexillarius in mm

(millimeters). * The studied specimens. Referred data are taken from Pilgrim (1937,

1939); Akhtar (1992), Khan (2008) and Khan et al. (2009a).

Number Nature/Position Length Width

PC-GCUF 10/06* I1 21.0 13

PC-GCUF 10/07* M1 20.0 20.5

PUPC 09/117* M1 20.4 12.0

PUPC 07/135* M1 21.0 14.0

M2 26.6 15.0

M3 32.0 15.0

M2 27.9 16.1

M3 31.4 16.0

PUPC 98/78 M2 25.0 16.0

M3 36.0 15.0

PUPC 85/40 M1 19.7 12.5

PUPC 04/12 M2 20.0 12.5

PUPC 87/90 M3 38.0 16.5

AMNH 10514 M3 33.0 15.0

AMNH 29917 M1 18.0 13.0

AMNH 19844 M2 25.7 24.0

AMNH 19844 M2 25.9 16.5

AMNH 19514 M2 22.0 15.5

AMNH 29917 M2 21.0 15.0

AMNH 19514 M3 33.0 21.5

PUPC 87/19 M1 24.2 21.5

57

Figure 3: Selenoportax cf. vexillarius. 1. PC-GCUF 10/07, lM1. 2. PC-GCUF 10/06, lI1.

3. PUPC 09/117, rM1. a = occlusal view, b = lingual view, c = buccal view. Scale bar

equals 10 mm.

58

Figure 4: 4. PUPC 07/135, a fragment of right mandible having M1-3. a = occlusal view, b

= lingual view, c = buccal view. Scale bar equals 50 mm.

59

Comparison and Discussion

The seleno-hypsodonty pattern of the studied material confirms its inclusion to

Ruminantia. The specimen has hypsodonty, greater strength of external lobes and ribs,

and fairly rapid increase in antero-posterior diameter from base to summit of crown. The

specimen morphology differs from tragulids, cervids and giraffids (Colbert, 1935;

Pilgrim, 1937; Bhatti, 2005; Farooq, 2006). The specimens, morphometrically clearly

indicate a large sized Miocene bovid. To this group belong Selenoportax and

Pachyportax of the Middle Siwaliks. Crown is narrow at the base and broad at the apex

in Selenoportax whereas in Pachyportax the crown is not constricted at the apex. The

general contour of the studied specimens, the rugosity of the enamel, the srong

entostyles/ectostylids, the prominent median ribs, the strong and divergent styles exclude

the studied specimen from the genus Pachyportax and favour its inclusion in the genus

Selenoportax. The recovered sample represents features of Selenoportax (Pilgrim, 1937,

1939; Akhtar, 1992; Khan et al., 2009a) and proves its inclusion the Siwalik genus

Selenoportax. The Siwalik Selenoportax is recorded by two species from the Siwaliks: a

small S. vexillarius and a large S. lydekkeri (Khan et al., 2009a). The studied specimens

correspond to species S. vexillarius morphometrically (Figs. 3-5; Table 2) and assign to

S. cf. vexillarius because of the insufficient material.

Pilgrim (1937) erected the genus Selenoportax, based on a collection from the various

Siwaliks localities of Pakistan and India. Akhtar (1992) added two species in it, one is S.

dhokpathanensis, based on a damaged cranium. It differs from S. vexillarius by its

gigantic size. The second is S. tatrotensis, based upon a maxillary ramus with right P3-M3

and left P4-M3. More recently, Khan et al. (2009a) reviewed the boselaphines from the

60

Middle Siwaliks of the Hasnot, Punjab, Pakistan, and they considered that S. vexillarious

and S. lydekkeri are valid species in the Middle Siwaliks of the subcontinent. Reviewing

the Siwaliks Selenoportax species, they (Khan et al., 2009a) synonymized S.

dhkopathanensis Akhtar, 1992 with S. lydekkeri and S. tatrotensis Akhtar, 1992 with S.

vexillarius.

S. vexillarius

0

5

10

15

20

25

0 5 10 15 20 25 30 35 40Length

Wid

th

Upper First Molar Lower First MolarLower Second Molar Lower Third Molar

Figure 5: Scatter diagram showing dental proportions of S. cf. vexillarius’s studied

sample. Referred data are taken from Pilgrim (1937, 1939); Akhtar (1992), Khan (2008)

and Khan et al. (2009a).

61

Genus Pachyportax Pilgrim, 1937

Type Species: Pachyportax latidens (Lydekker) Pilgrim, 1937.

Generic Diagnosis: Boselaphinae of small to large or very large size; closely allied to

Strepsiportax but differing from that genus by the much more massive skull, with horn-

cores longer, stouter, more twisted and less curved inwardly; occipital condyles and

foramen magnum larger; mastoid process and squamosal shelf more developed;

supraoccipital exposed on the upper surface of the occiput as a narrowly elliptical area

much extended transversely; basioccipital approaching a rectangular shape, with posterior

tuberosities not greatly expanded; upper molars strongly hypsodont but less so than in

Selenoportax, quaderate, with strong entostyle, external folds weaker and less divergent

than in Selenoportax, external ribs weaker than in Selenoportax, in particular the median

rib of the posterior lobe flattened, enamel rather thick, somewhat less rugose than in

Selenoportax, with traces of cement (Pilgrim, 1937).

Known Distribution: The genus Pachyportax is present in the Nagri and the Dhok

Pathan formations of the Middle Siwaliks (Lydekker, 1876; Pilgrim, 1937; Khan et al.,

2009a). It is also present in the Tatrot zone of the Upper Siwaliks (Akhtar, 1992). The

material under study comes from the type locality of the Nagri Formation of the Middle

Siwaliks, Pakistan. Gentry (1999) describes the species from Abu Dhabi.

Pachyportax cf. latidens Pilgrim, 1937

Type Specimen: GSI B560, a skull fragment (Pilgrim, 1939).

Type Locality: Nagri, Middle Siwaliks, Punjab, Pakistan (Pilgrim, 1939).

Stratigraphic Range: Middle Siwaliks (Pilgrim, 1939; Akhtar, 1992).

62

Diagnosis: A large Pachyportax, with quadrate upper molars and strong entostyle

extended transversely; the crown is not constricted at the apex, relatively strong styles

and ribs, enamel moderately thick and rugose with traces of cement. Crown is narrow at

the base and broad at the apex in Selenoportax whereas in Pachyportax the crown is not

constricted at the apex. Pachyportax has strong entostyle extending transversely while in

Selenoportax it is not much extended transversely. In Pachyportax posterior median rib is

flattened whereas in Selenoportax it is strong like anterior median rib (Pilgrim, 1937,

1939).

Studied Specimens: Upper dentition: PUPC 09/46, isolated right P3; PUPC 09/69,

isolated left M2.

Description

Upper Dentition

The recoverd material comprises only upper dentition. The P3, PUPC 09/46 is a

triangular tooth (Fig. 6(1)). The anterior median rib is closed to the parastyle forming a

narrow vertical groove on antero-buccal side of the premolar. The posterior groove is

wide and shallow. The premolar is in middle wear. The paracone is round. The metastyle

is prominent and narrow. The cingulum is absent on both the lingual as well as on the

buccal side and a slight indication is present buccally at the base of the metacone. A wide

cavity in the center of the tooth becomes extremely narrow anteriorly.

PUPC 09/69 is a well preserved second molar (Fig. 6(2)). It is in late early stage of wear.

The contact facets present antero-posterior sides of the molar. The enamel is rugose. The

molar has divergent styles. The entostyle is present in the transverse valley between the

63

protocone and hypocone of the molar. The protocone is relatively narrower transversely

than the hypocone with two running praeprotocrista and postprotocrista towards the

parastyle and mesostyle. The hypocone is slightly higher and more crescentic than the

protocone. The anterior and posterior central cavities are wide, isolated and deep. The

anterior rib is strong whereas the posterior one is flattened. The comparative

measurements are provided in table 3.

Table 3: Comparative measurements of the cheek teeth of P. latidens in mm

(millimeters). * The studied specimens. Referred data are taken from Pilgrim (1937,

1939), Akhtar (1992), Khan et al. (2009a).

Number Nature/Position Length Width

PUPC 09/46* P3 16.0 15.0

GSI B218 P3 19.0 19.0

PUPC 09/69* M2 25.0 23.0

PUPC 98/59 M2 22.0 17.3

PUPC 96/40 M2 19.4 18.4

PUPC 96/3 M2 27.0 22.0

PUPC 86/37 M2 27.4 18.0

PUPC 86/36 M2 30.0 23.0

PUPC 83/718 M2 27.4 26.0

PUPC 83/646 M2 30.0 18.0

PUPC 83/744 M2 30.2 21.9

PUPC 86/210 M2 26 17.1

PUPC 00/100 M2 25.5 25.0

PUPC 04/14 M2 29.3 20.6

PUPC 98/60 M2 23.1 15.9

PUPC 97/103 M2 24.5 17.7

PUPC 86/203 M2 26.4 17.9

AMNH 29964 M2 28.0 25.0

AMNH 19730 M2 28.5 28.5

PUPC 96/42 M3 30.2 22.564

Table 3 (Continued).

PUPC 01/24 M3 28.4 25.0

PUPC 96/38 M3 34.4 29.0

GSI B219 M3 34.5 28.0

AMNH 29914 M3 36.0 34.0

AMNH 29913 M3 31.0 29.0

AMNH 19730 M3 29.5 27.0

PUPC 83/840 M3 31.9 23.0

PUPC 87/88 M3 27.2 16.6

PUPC 04/15 M3 28.0 21.2

PUPC 00/87 M3 25.9 17.6

AMNH 29913 M3 31.0 29.0

AMNH 19730 M3 29.5 27.0

Figure 6: Pachyportax cf. latidens.1. PUPC 09/46, rP3. 2. PUPC 09/69, lM2. a = occlusal

view, b = lingual view, c = buccal view. Scale bar equals 10 mm.

65

Comparison and Discussion

The Middle Siwaliks is represented by two species of Pachyportax, a large size P.

latidens and a small size P. nagrii (Pilgrim, 1937, 1939; Khan et al., 2009a). The

described fossil remains show all the basic features of the genus Pachyportax and

comparable to species Pachyportax latidens morphometrically (Figs. 6-7; Table 3).

Nevertheless the recovered material is scarce and the sample assigns to P. cf. latidens.

The described specimens give additional information about third premolar and second

molar of the species. The most prominent feature of the upper molars is the transverse

extension of the entostyle as mentioned by Pilgrim (1937). It varies to some extent in its

longitudinal dimensions towards the lingual side; in some it is slightly broader. The

second upper molar compares vary favorably to the referred specimens present in the

American Museum of Natural History (AMNH) and the previously described specimens

of Punjab University Palaeontological Collection (PUPC) present in the Zoology

Department of Punjab University, Lahore (Table 3). They have resemblance

morphologically and metrically in all the structural details like cusps, entostyles, styles,

median ribs and central cavities. The only slightly difference among the molars is found

in their dental measurements and this difference is too insignificant to be considered as of

taxonomic importance (Fig. 7). On the basis of these similarities the studied specimens

are being referred to P. cf. latidens.

Lydekker (1876) described a right M3 (GSI B219) under the name Cervus latidens. In the

same paper he described a lower molar (GSI 23) also and referred it to this species. Later

on, he (1884) realized that an upper and lower molar of a large ruminant from the

Siwaliks of the Punjab which were described and figured under the name Cervus latidens

66

do not belong to the family Cervidae. To the same species Lydekker referred a left

maxilla with P2 – M3 (GSI B218a) and provisionally assigned these three specimens to

the genus Oreas (?). Lydekker (1878) described and figured a horn-core under the name

Capra sp.

Pilgrim (1937) applied the generic term Pachyportax to all these specimens which were

described by Lydekker under the names Cervus latidens (1876), Capra sp. (1878) and

Oreas (?) latidens (1884). He referred these specimens to Pachyportax latidens

(Lydekker) Pilgrim, making the type specimen an isolated M3 (GSI B219) which was

described by Lydekker (1876) under the name Cervus latidens. Pilgrim (1939) reported

the occurrence of Pachyportax from Nagri by describing a new species Pachyportax

nagrii. The species is based upon a hornless female cranium. According to Gentry

(1974), Pachyportax nagrii is a probably invalid species. Akhtar et al. (1997) ascribed

Pachyportax nagrii from the Nagri Formation, based on the left maxilla PUPC 86/77.

Pachyportax nagrii is of smaller size than those of Pachyportax latidens (Akhtar et al.,

1997).

Pachyportax is a gigantic sized boselaphine (Lydekker, 1876, 1884; Gentry, 1999).

Pachyportax latidens although have been continuously present from the Middle Siwaliks

to Upper Siwaliks sequence but it is more abundant in the Hasnot succession (Khan et

al., 2009a). Bibi (2007) discussed the origin of the early bovines and grouped

Selenoportax and Pachyportax with them. Pachyportax have been recovered from the

late Miocene of the Middle Siwaliks (Lydekker 1876, 1884; Pilgrim, 1937, 1939; Akhtar,

1992, 1995, 1996; Khan, 2008; Khan et al., 2009a) and from the early Pliocene of the

Upper Siwaliks (Akhtar, 1992). The faunas of Negeringerawa, Namurungulea and Nakali

67

dated 10-8 Ma and the faunas from the Mpsida, dated 7-6 Ma in Africa do not have the

genus Pachyportax (Hill et al., 1985; Nakaya, 1994; Kingston et al., 2002). Pachyportax

is also lacking from localities of the same age such as the Afghani locality of Tagar dated

at 8.7-8 Ma (Sen et al., 1997) and Iranian locality Marageh dated 9.5-7 Ma (Bernor,

1986). Pachyportax is a typical Late Miocene taxon, occurring in the Nagri and the Dhok

Pathan formations of the Siwaliks (Akhtar et al. 1997). Recently, Khan et al. (2009a)

ascribed Pachyportax from the Middle Siwaliks of Hasnot, Pakistan. Pachyportax was

restricted to the Middle Siwaliks because Himalyan Mountains acted as a barrier in the

dispersal of Pachyportax out of southerns Asia prior to the late Miocene, isolating the

Siwalik faunas (Barry et al. 1982; Bernor, 1984).

P. cf. latidens

0

5

10

15

20

25

30

35

40

0 5 10 15 20 25 30 35 40

Length

Wid

th

Upper Third Premolar Upper Second Molar Upper Third Molar

Figure 7: Scatter diagram showing dental proportions of P. latidens’s studied sample.

Referred data are taken from Pilgrim (1937, 1939), Akhtar (1992), Khan et al. (2009a).

68

Genus Tragoportax Pilgrim, 1937

Type Species: Tragoportax salmontanus Pilgrim, 1937.

Generic Diagnosis: Moderate to large sized Eurasian bovid. Skull long and slender,

brain-case rather slender, temporal ridge very strong, occipital rather high and narrow,

lambdoidal crest prominent, occipital condyles large, basiocoipital short, subtriangular,

with a shallow median furrow, paraoccipital process elongate and narrow. Upper molars

hypsodont, quadrate, with small entostyles, enamel rugose, moderately strong styles and

ribs, central cavities connect at mid-wear, upper premolar series large and long, P2 long

with small parastyle; P3 with large hypocone in relation to protocone. Retention of

plesiomorphic dental features and is para-phyletic, large boselaphines from late Miocene

assemblages, greater size, slightly more reduced premolar rows, and more inflated p4

metaconids (Pilgrim, 1937; Spassov and Geraads, 2004.).

Known Distribution: Eurasia, Africa and the Indian subcontinent (Pilgrim, 1937, 1939;

Spassov and Geraads, 2004; Kostopoulos, 2009).

Tragoportax punjabicus (Pilgrim, 1910)

Type Specimen: GSI B486, skull.

Type Locality: Dhok Pathan, Middle Siwaliks, Punjab, Pakistan (Pilgrim, 1910, 1939).

Stratigraphic Range: Middle Siwaliks (Pilgrim, 1939; Akhtar, 1992).

Diagnosis: A species slightly smaller than Tragoportax browni, with relatively short

upper premolar series; P2 rather longer than P3; upper molars with small entostyle;

moderately developed styles and ribs; central cavities connect at mid wear and enamel

moderately rugose (Pilgrim, 1937).

69

Studied Specimens: Upper dentition: PC-GCUF 10/08, isolated left P3; PUPC 09/66,

isolated right M1; PC-GCUF 10/09, partial tooth probably M1. Lower dentition: PC-

GCUF 10/11, isolated right P3; PUPC 09/70, isolated left P4; PUPC 07/77, isolated left

M1; PUPC 07/86, isolated left M1; PUPC 07/138, a mandibular ramus with P4-M2.

Description

Upper Dentition

PC-GCUF 10/08 is in early wear and triangular tooth showing all the morphological

characteristics (Fig. 8(1)). The enamel is somewhat wrinkled and rugose. A prominent

central cavity is present. A small, very thin, transverse enamel layer connects the

posterior end of the protocone with the hypocone. The paracone is comparatively higher

than the protocone. The anterior median rib is very prominent and closer to the parastyle.

A furrow of moderate depth is present between the parastyle and the anterior median rib.

The posterior groove is wider than the anterior one. The hypocone is inflated lingually.

PUPC 09/66 is in an early wear and an excellent state of preservation (Fig. 8(2)). The

crown is quadrate. The crown height and width shows that it is a subhypsodont tooth. A

faint cingulum is present on the antero internal and postero-internal surface of the molar.

The entostyle near to hypocone is present. The cones are very well developed and broad.

The protocone and the hypocone are similar in their general appearance. They are

crescentric in shape. The metacone is higher than the paracone vertically. Both these

cones are spindle shaped, broad in the center and narrowing at the sides. The styles are

well developed and divergent. The metastyle is stronger than the parastyle. The mesostyle

is also well developed. The posterior rib of the molar is stronger than the anterior one.

70

The anterior cavity is deep and narrow while the posterior cavity is broader than the

anterior one. PC-GCUF 10/09 is a partial tooth and most of the crown portion is missing.

The occlusal view is somewhat available for the morphological study. The cavities and

the posterior median rib can be seen in the tooth. The cones are crescentic and the rib is

prominent in the molar. The styles, some parts of the protocone, the metacone and the

hypocone are missing.

Lower Dentition

PC-GCUF 10/11 is an isolated dainty premolar (Fig. 8(3)). The metaconid of the

premolar is backwardly directed. The entoconid of the premolar is stronger than the

metaconid. The paraconid of the tooth is stronger than parastylid and placed antero-

posterior axis of the premolar. The buccal surface appears flat and the lingul one presents

two vertical grooves. The anterior one is open.

The P4 PUPC 09/70 is in middle wear and has a postprotoconulidcristid, a metaconid, and

a postmetacristid (Fig. 8(4)). The p4 has a strong paraconid, metaconid and entoconid.

The entoconid is fused with the endostylid. The prominence of the hypoconid is

noteworthy and has a deep and narrow valley in front of it. The p4 is extended antero-

posteriorly. The metaconid of the premolar is larger than the P3. It is splayed lingually

forming T-shaped on the p4, with an open anterior valley (Fig. 8(4)).

PUPC 07/77 and PUPC 07/86 are the first molars of the left lower molar series (Figs.

8(5-6)). The protocone of PUPC 07/77 is missing and the other parts are available for the

crown study (Fig. 8(5)). PUPC 07/86 is broken anteriorly and posteriorly. The metastylid

and the parastylid are prominent whereas the mesostylid is absent. The ectostylid is

present but weak. The anterior transverse flange is present.

71

PUPC 07/133 with P4-M2 is well preserved and in an early wear (Fig. 9(7)). In the P4 the

enamel is finely wrinkled and it is thick on the lingual side. The p4 has a strong

paraconid, metaconid and entoconid and the metaconid of the P4 is splayed lingually

forming T-shaped with an open anterior valley (Fig. 9(7)). The M1 ectostylid is strong

and almost circular in cross section. The principlal conids are well developed and

crescentic. The metaconid and the entoconid are spindle shaped with narrowing borders.

The metastylid and the entostylid are prominent. The anterior and the posteriro median

ribs are present. The anterior and the posterior central cavities are narrow. The M2 is is

somewhat worn out on the lingual side. The overall contour indicates that it is

subhypsodont and narrow crowned tooth. A very small ectostylid is located in the

transverse valley between the protoconid and the hypoconid. The entoconid is highest

vertically than the other conids. The central cavities are narrow and with simple outlines

without any indentation. The metastylid and the mesostylid are more prominent than the

entostylid. The comparative measurements are provided in table 4.

72

Table 4: Comparative measurements of the cheek teeth of T. punjabicus in mm

(millimeters). * The studied specimens. Referred data are taken from Pilgrim (1939) and

Akhtar (1992).

Number Nature/Position Length Width

PC-GCUF 10/08* P3 14.0 11.5PUPC 09/66* M1 18.4 18.5PC-GCUF 10/11* P3 14.0 7.00PUPC 09/70* P4 15.5 8.20PUPC 07/77* M1 18.0 12.3PUPC 07/86* M1 17.4 11.5PUPC 07/138* P4 14.6 9.00

M1 17.3 12.0M2 19.7 12.0

GSI B486 P3 14.5 12.0P4 11.0 15.5M1 18.0 18.0

GSI B574 M1 18.0 19.0GSI B563 M1 19.0 12.5

M2 21.0 13.0GSI B564 P3 16.0 8.50

P4 17.5 10.0M1 17.5 12.0M2 20.5 14.0

73

Figure 8: Tragoportax punjabicus. 1. PC-GCUF 10/08, lP3. 2. PUPC 09/66, rM1. 3. PC-

GCUF 10/11, isolated right P3. 4. PUPC 09/70, lP4. 5. PUPC 07/77, lM1. 6. PUPC 07/86,

lM1. a = occlusal view, b = lingual view, c = buccal view. Scale bar equals 10 mm.

74

Figure 9: Tragoportax punjabicus. 7. PUPC 07/138, a mandibular ramus with P4-M2. a =

occlusal view, b = lingual view, c = buccal view. Scale bar equals 10 mm.

Comparison and Discussion

Being a squared and tetratuberculated sample it can be referred to some herbivorous

mammalian group. Crescentic cusps of selenodont nature represents that it can safely be

included in the order Artiodactyla (Zittel, 1925; Romer, 1974). The compressed outer

cusps favour its inclusion in the family Bovidae. The teeth are small size and selenodont.

The teeth may be distinguished at a glance from teeth of Pachyportax and Selenoportax

by their smaller size and the weaker basal pillar (Gaudry, 1865; Arambourg and Piveteau,

1929; Pilgrim, 1937). The studied P3 indicates inflated hypocone which is the feature of

the genus Tragoportax. The p4s display a T-shaped feature of Tragoportax (Spassov and

Geraads, 2004). The described character somewhat corresponds to numerous medium-

75

sized boselaphine Tragoportax from the Siwaliks to which this specimen could be

attributed. The teeth are same in size and general morphology to T. punjabicus (Fig. 10;

Table 4) and consequently, the sample can be assigned to T. punjabicus.

Tragoportax are known in the Siwaliks by five species namely T. perimensis, T. islami,

T. salmontanus, T. browni and T. punjabicus (Pilgrim, 1937, 1939). The species is

distinguished based on the horn-cores. Tragoportax perimensis and T. islami are

represented by the poor fossil record. Kostopoulos (2009) contributed in an extensive

systematic revision of the Samos bovids and synonymised T. curvicornis and T. browni

with T. punjubicus. Kostopoulos (2009) adapted Moya-Sola’s recommendations to

synonymies T. browni with T. punjabicus because both of them are indistinguishable and

have common stratigraphic origin from the Dhok Pathan Formation of the Middle

Siwaliks. However, a more findings are required from the Middle Siwaliks for the exact

specific determination of the Siwalik Tragoportax.

T. punjabicus

02468

101214161820

0 2 4 6 8 10 12 14 16 18 20 22 24Length

Wid

th

Upper Third Premolar Upper First Molar Lower Third PremolarLower Fourth Premolar Lower First Molar Lower Second Molar

Figure 10: Scatter diagram showing dental proportions of T. punjabicus’s studied

sample. Referred data are taken from Pilgrim (1939) and Akhtar (1992).

76

Genus Miotragocerus Stromer, 1928

Type Species: Miotragocerus monacensis Stromer, 1928.

Generic Diagnosis: Horn-cores triangular in cross section; above the orbits; not

particularly compressed; converging anteriorly and forming a pronounced frontal

buttress; not particularly twisted but with twist restricted to tips; gently diverging with

tips neither turned in nor out; posterior grooves; anterior keel blunt, stopping about two-

thirds from base forming at least one characteristic anterior demarcation (bump); horn-

core thinner and rounder in cross section from point at which keel stops to tip; horn-core

axis more vertical and bases broader anteroposteriorly, pedicals more poorly formed than

in Tragoportax. Anterior keel blunter, often with several distinct growth bumps in males;

less blunt, with a single bump in females. No postcornual pits; frontals strongly depressed

behind horns; basicranium not particularly angled in relation to palate; preorbital fossa

deep; supraorbital pits small, variable in number and position. Premolars longer in

relation to the molars than in Tragoportax; P2 long; P3 with small hypocone in relation to

protocone; upper molar central cavities connect at mid-wear; entostyle small on upper

molars. p4 cavity between paraconid and metaconid open; p4 paraconid tends to be larger

than parastylid. Differs from Mesembriportax in having less sinused frontals; differs from

Protragocerus in having longer, more compressed horn-cores with an anterior keel

(Solounias, 1981).

Known Distribution: Europe and south Asia (Pilgrim, 1937; Spassov and Geraads,

2004).

77

Miotragocerus cf. gluten (Pilgrim, 1937)

Type Specimen: AMNH 19746, Skull lacking face and most of the dentition.

Type Locality: West of Hasnot, upper boundary of the Chinji, Lower Siwaliks, Punjab,

Pakistan.

Stratigraphic Range: Lower and Middle Siwaliks (Pilgrim, 1937; Thomas, 1984).

Abbreviated Diagnosis: Low crowned teeth with strongly with strongly folded walls.

The upper molar central cavities connect at mid-wear and the entostyles are smaller than

in Tragoportax. The lower dentition is more primitive than Tragoportax. The p4 cavity

between the paraconid and the metaconid is open and therefore p4 is similar to p3. The

metaconids of P3-4 are weak. The lower molars have transversely situated protoconids and

hypoconids (Pilgrim, 1937, 1939; Solounias, 1981; Spassov and Geraads, 2004).

Studied Material: Upper dentition: PUPC 07/138, isolated left M1. Lower dentition: PC-

GCUF 10/12, isolated right P3.

Description

Upper Dentition

The upper dentition only includes one molar. The molar PUPC 07/138 is brachydont

(Fig. 11(1)). It is in early wear. The enamel is rugose. The molar has smart entostyle. The

buccal styles and ribs are well developed and strongly projected in the molar (Fig.

11(1c)). The anterior median rib is well projected than the posterior one. The mesostyle is

robust pillar like structure. The styles are divergent. The upper molar is with strong

folded walls.

78

Lower Dentition

PC-GCUF 10/12 has a weak groove between the paraconid and the parastylid. The P3 is

unworn and well preserved (Fig. 11(2)). The preprotoconulidcristid distinguishes from

the postprotoconulidcristid. The metaconid is located behind the protoconid. It has flat

lingual wall and enlarges from the tip to the base, tending to close the medial valley. The

anterior lingual valley is much wider than the posterior ones. The protoconid is the

highest among the conids. The postprotocristid leads to the entoconid through a thin

praemetacristid. The postentocristid is directed towards the hypoconid. The posterior

lingual valley is narrow relative to the anterior one. A short and moderately developed

entostylid is present adjacent to the entoconid. The anterior half of the tooth is elevated

relative to the posterior one. The cingulid is absent and enamel is rugose. The hypoconid

is separated from the protoconid through a buccal groove. The comparative

measurements are provided in table 5.

Table 5: Comparative measurements of the cheek teeth of Miotragocerus (Pilgrim, 1937)

in mm (millimeters). * The studied specimens. Referred data are taken from Pilgrim

(1937).

Taxa Number Nature/Position Length Width

M. cf. gluten PUPC 07/138 M1 17.0 17.0

PC-GCUF 10/12 P3 12.0 6.20

M. gluten P4 9.50 13.0

M1 13.0 15.0

M2 16.0 18.0

M3 16.0 16.0

AMNH 19993 M1 14.5 9.00

M2 16.5 11.0

79

Figure 11: Miotragocerus cf. gluten. 1. PUPC 07/138, lM1. 2. PC-GCUF 10/12, rP3. a =

occlusal view, b = lingual view, c = buccal view. Scale bar equals 10 mm.

Comparison and Discussion

The recovered sample indicates the medium size bovid. Morphometrically, these

specimens are typical of Miocene boselaphines in appearance; the divergent styles of the

teeth make their inclusion in boselaphines. There are many differences among the 80

boselaphines from the Dhok Pathan (Pilgrim, 1937, 1939; Thomas, 1984; Khan et al.,

2009a). Selenoportax and Pachyportax are large size boselaphines found in the Dhok

Pathan Formation (Khan et al. 2009a). The Helicoportax, Elachistocerus and Eotragus

are comparatively small size boselaphines (Pilgrim, 1937, 1939; Akhtar, 1992; Khan et

al., 2009a). The medium size boselaphines include Tragoportax and Miotragocerus. The

studied molar and premolar are well accentuated distinguished than those of Tragoportax

(Spassov and Geraads, 2004). The morphology of the teeth show that the samples reflects

the diagnostic features of Miotragocerus and differentiate them to Tragoportax, other

Siwalik medium size boselaphine of the common stratigraphic range, and should be

assigned to Miotragocerus (Figs. 11-12; Table 5). In Siwalik, Miotragocerus gluten

represents from the Chinji and the Nagri formations. The material resembles

Miotragocerus gluten and it can be assigned to M. gluten. However, the material is

insufficient for the specific determination and designates M. cf. gluten for the recovered

sample.

Upper First Molar

14.5

15

15.5

16

16.5

17

17.5

0 5 10 15 20Length

Wid

th

M. cf. gluten M.gluten

Figure 12: Scatter diagram showing dental proportions of M. cf. gluten’s studied sample.

Referred data are taken from Pilgrim (1937).

81

Tribe Antilopini

Genus Gazella Blainville, 1816

Type Species: Gazella dorcas Linneaus, 1758.

Generic Diagnosis: A Gazella of the size and type of the living G. bennetti but females

hornless; skull long and slender; face bent down on cranial axis at about 35°; occipital

rather high. Upper molars moderately hypsodont, styles narrow and strong with

entostyles very small or absent, enamel moderately thick and rugose, central cavities

narrow and deep, anterior median rib stronger than posterior one, premolar series rather

long. Lower molars extremely hypsodont approaching quadrate shape, with small

ectostylids, prominent goat folds, central cavities fairly simple in outline, stylids and ribs

moderately developed. Horn cores moderately long, spaced, slightly curved backward,

broadly elliptical in cross-section, fine ribs becoming rudimentary near the tips, one deep

furrow posteriorly (Pilgrim, 1937).

Known Distribution: The occurrence of Gazella is recorded from the Lower Pliocene of

Eurasia and several Pleistocene localities of Africa. It is abundantly found in the Lower

Pliocene fauna of Asia and the southern parts of Europe. It is also recorded from the

Siwaliks of the Subcontinent (Pilgrim, 1937, 1939; Thomas, 1984). Gazella is reported

from the late Miocene of Sivas, Turkey (Bibi and Gulec, 2008) and Greece (Kostopoulos,

2009).

Gazella cf. lydekkeri Pilgrim, 1937

Type specimen: AMNH 19663, a skull and conjoined mandible (Pilgrim, 1937).

Type Locality: Dhok Pathan (the Middle Siwaliks), Punjab, Pakistan (Pilgrim, 1937).

82

Stratigraphic Range: The Lower and the Middle Siwaliks (Pilgrim, 1937; Khan, 2008).

Diagnosis: Hypsodont upper molars, strong styles, entostyles very small or absent,

enamel moderately thick and rugose, central cavities narrow and deep, anterior median

rib stronger than posterior one, premolar series rather long. Lower molars extremely

hypsodont approaching quadrate shape, with small ectostylids, prominent goat folds,

central cavities fairly simple in outline, stylids and ribs moderately developed (Pilgrim,

1937).

Studied Specimens: Lower dentition: PC-GCUF 09/02, a right mandibular ramus with

M1-3; PUPC 07/71, isolated left M3.

Description

Lower dentition

The studied material comprises only lower dentition. A complete lower molar series is

preserved in PC-GCUF 09/02 but the hypoconulid in the M3 (Fig. 13(1)).

The ramus is poorly preserved and the molars are in early wear. The ectostylid size

gradually decreases towards posterior of the molar series and consequently, the large

ectostylid is in the M1 and the small one is in the M3 (Fig. 13(1c)). A goat fold is present

anteriorly in the molars. The stylids are moderately developed and the mesostylid is the

weakest. The anterior median rib is broad. The lingual conids are higher than the buccal

ones. The hypoconulid is broken distally in the M3.

PUPC 07/71 is in an early wear. The protoconid and the talonid are damaged slightly

(Fig. 13(2)). The enamel is moderately thin on the lingual side while it is thick and

shining on the buccal sides. The conids are well developed. The hypoconid appears to be

83

more cresentric in shape than the protoconid. The apex of the entoconid is damaged. The

metastylid is more developed than entostylid. The hypoconulid is well developed, high

with a wide and inflated area. It is some what damaged posteriorly. The central cavities

are narrow. The anterior central cavity is wider than the posterior one. The comparative

measurements are provided in table 6.

Table 6: Comparative measurements (mm) of the cheek teeth of G. lydekkeri. * The

studied specimens. Referred data are taken from Pilgrim (1937, 1939); Akhtar (1992) and

Khan (2008).

Number Nature/Position Length Width

PC-GCUF 09/02* M1 15.0 10.0M2 16.0 10.5

PUPC 07/71* M3 23.0 11.0PUPC 04/08 M3 20.0 9.00 AMNH 19663 M1 10.0 12.0

PUPC 84/133 M1 12.0 6.00

PUPC 84/67 M1 14.5 9.00

PUPC 86/04 M2 15.0 10.0

PUPC 87/ 162 M3 22.0 10.5

PUPC 67/42 M3 20.0 9.00PUPC 84/67 M1 15.0 9.00

M2 16.6 10.0

PUPC 87/160 M1 12.2 8.00

M2 14.0 9.00

PUPC 04/02 M1 11.0 8.20

M2 14.0 8.70

M3 19.0 8.60

84

Figure 13: Gazella cf. lydekkeri. 1. PC-GCUF 09/02, right mandibular ramus with M1-3.

2. PUPC 07/71, lM3. a = occlusal view, b = lingual view, c = buccal view. Scale bar

equals 10 mm.

Comparison and Discussion

The studied specimens present the typical structure of the molars for the genus Gazella

and these are the prominent median ribs, the narrow styles and the appearance of the

ectostylids in the lower molar. The presence of these characteristics in the studied

specimens clearly confirms their inclusion to Gazella. The morphometrical characters of

the described specimens resemble with the holotype of G. lydekkeri in the structure of the

conids, the development of the stylids and ribs, and the presence of the ectostylids (Figs.

13-14; Table 6). However, the material is not enough to the specific identification.

Therefore, G. cf. lydekkeri is assigned for the sample.

85

The genus Gazella was erected by Blainville (1816) and was recorded for the first time as

fossil horn-cores of G. stehlini in the European upper Vindobonian (Gentry, 1966). Most

species of Gazella from Europe are founded on fragmentary material such as isolated

teeth and horn-cores. All these species are small. Pilgrim and Hopwood (1928) simply

summarized all the nomenclature of fossil gazelles propagating the existence of a large

number of species. In their opinion the curvature, degree of divergence, and size of horn-

cores should be the main criteria for species identification. They did not consider age,

sexual variation and specimen deformation. Akhtar (1992) has noticed that these

characters change with the age of the individual of a species in the living forms.

Solounias (1981) has suggested that the type and degree of longitudinal grooving as well

as the shape of the cross-section might be better species features although their variability

is not known.

Pilgrim (1937) erected a new species G. lydekkeri from the Siwaliks of Pakistan. The

holotype is almost complete and comprising a skull and conjoined mandible (AMNH

19663). It is much more like the living forms of Gazella in having its longer and more

slender skull, the higher occipital and the shape and direction of its horn-cores. Gazella

capricornis, the best known of the European Pontian species is represented by more

material than G. deperdita. It differs from G. lydekkeri in having large size of the skull,

and less hypsodonty. In this respect, G. lydekkeri is more primitive than Gazella

capricornis, the most progressive feature of G. lydekkeri is its hypsodonty. The Chinese

Pontian gazelles are more progressive especially the species G. dorcadoides. In all the

Chinese species the skull seems to be less slender than in G. lydekkeri although the width

at the orbits is greater. The horn-cores seem to be identical in their morphology with G.

86

lydekkeri and the nasals are somewhat shorter. In G. dorcadoides and G. altidens the

teeth are still more hypsodont than in G. lydekkeri. Gentry (1970) and Solounias (1981)

considered G. lydekkeri as an invalid species. Recently, Khan and Farooq (2006)

described ruminant fauna from the Neogene of the Siwaliks of Pakistan. They discussed

the appearance of the ruminant species in the Siwalik hills of Pakistan. According to their

report, G. lydekkeri and G. padriensis Akhtar, 1992 appeared in the Middle Siwaliks of

the Siwalik hills of Pakistan.

G. cf. lydekkeri

0

2

4

6

8

10

12

14

0 2 4 6 8 10 12 14 16 18 20 22 24 26Length

Wid

th

Lower First Molar Lower Second Molar Lower Third Molar

Figure 14: Scatter diagram showing dental proportions of G. cf. lydekkeri’s studied

sample. Referred data are taken from Pilgrim (1937, 1939); Akhtar (1992) and Khan

(2008).

87

Family Giraffidae Gray, 1821

Subfamily Palaeotraginae Pilgrim, 1911

Genus Giraffokeryx Pilgrim, 1910

Type Species: Giraffokeryx punjabiensis Pilgrim, 1910.

Generic Diagnosis: Medium size giraffid with four horns, two at the anterior extremities

of the frontal and two on the fronto-parietal region. Posterior horn overhanging the

temporal fossa. Limbs and feet presumably of medium length. Teeth are brachydont with

rugose enamel as in the other genera of the Giraffidae. Giraffokeryx was a medium sized

member of the Giraffidae distinguished by two pairs of horn cores (ossicones) (Matthew,

1929; Colbert, 1935).

Known Distribution: Indian subcontinent and Eurasia (Pilgrim, 1910; Colbert, 1935;

Geraads, 1986; Janis and Scott, 1987a; Gentry and Hooker, 1988; Bhatti, 2005; Bhatti et

al., 2007).

Giraffokeryx punjabiensis Pilgrim, 1910

Lectotype: GSI B502, a third molar of the right maxilla.

Type Locality: Chinji, Lower Siwaliks, Punjab, Pakistan (Colbert, 1935).

Stratigraphic Range: Lower Siwaliks and the lower portion of the Middle Siwaliks

(Colbert, 1935; Bhatti, 2005).

Diagnosis: Larger than the other species of the genus. Upper molars are comparatively

large and subhypsodont. Parastyles and mesostyles are well pronounced. Accessory

column present blocking the transverse valley (Colbert, 1935).

88

Studied Material: Upper dentition: PUPC 07/88, isolated left P3; PUPC 09/67, partially

preserved isolated right P3; PUPC 07/133, isolated right M2. Lower dentition: PUPC

09/43, left hemimandible with M2-3, broken canine and alveoli of P3-M1; PUPC 07/90,

isolated right M3.

Description

Upper dentition

PUPC 07/88 and PUPC 09/67 are in middle wear. PUPC 07/88 is well preserved (Fig.

15(1)) whereas PUPC 09/67 is broken lingually and buccally (Fig. 15(2)). The premolars

are greater in length than in width, and each tooth is characterized by a strong parastyle,

and an internal posterior swelling (Fig. 15(1-2)). A well developed paracone rib is present

close to the parastyle forming two vertical grooves. The anterior groove is narrow and the

posterior groove is broad. The three rooted premolars have thick and rugose. The

cingulum is not developed. The central cavity is well developed. The internal side of the

buccal crescent is weakly divided into paracone and metacone. The premolars look

subquadrangular with an antero-lingual protuberance of the lingual wall (Fig. 15(1)).

The second molar is approximately quadrate, with the protocone and the metaconule of

about equal size and with strong parastyle and mesostyle (Fig. 15(3)). The molar is in

middle wear. The enamel is thick and rugose. All the major cusps are well developed and

prominent. The buccal cusps are slightly higher than the lingual ones. The entostyle is

absent. The cingulum is present anteriorly, posteriorly and lingually. The praeprotocrista

is narrow and touches to the parastyle. The general contour of the paracone is like spindle

with maximum width in the middle. The metacone resembles the paracone in general

89

shape and it is pyramidal with two sloping cristae. The hypocone is crecentic in shape

and it is connected with the metacone through a narrow ridge posteriorly. The anterior

central cavity is shallower than the posterior one. The styles are well developed. The

mesostyle is strongly developed and the parastyle is more prominent than the metastyle.

The anterior median rib is absent but the posterior one is very prominent and broad at the

tip of crown.

Mandible

PUPC 09/43 hemimandible is an incomplete specimen retaining only broken canine, two

last molars and the base of the ascending ramus (Fig. 16(4)). The anterior part of the

symphysis is broken. The symphysis is flat. The length of the hemimandible is 263 mm.

The body of the mandible is typical of giraffid species; it is bucco-lingually narrow. The

depth of the mandible below P2 is 47 mm and it is 61 mm below M3. The diastema

between the canine and the P2 is excellenty preserved having length of about 46 mm. It is

hitherto recovered for the first time for the species Giraffokeryx punjabiensis from the

Nagri type locality. Colbert (1933) restored diastema based on comparison with

Palaeotragus. The length of the lower molar series is 79 mm and the premolar series is

56 mm. The ventral edge of the horizontal ramus is thick. The posterior edge of the angle

of the mandible is thin and the masseteric fossa is deep (Fig. 16(4)). The lower molars in

the hemimandible are in the latest wear but they consist of the familiar ruminant

crescents, and in the third molar there is a talonid.

Lower dentition

PUPC 07/90 is a high crowned tooth with hypoconulid and rugose enamel (Fig. 17(5)).

Its length is much more than the transverse width. The cigulum is well developed

90

anteriorly. The protoconid is well developed and V-shaped. The metaconid is slightly

worn out and it is spindle shaped. It is slightly wider in the middle with the narrow

sloping cristids. The postmetacristid is overlaping with the praeentocristid. The

praehypocristid and posthypocristid are simple. A well developed ovate central cavity is

present in the hypoconulid. The mesostylid is well developed whereas the metastylid and

the entosylid are weakly developed. The ectostylid is present on the buccal side in front

of the hypoconid and a supplementary one is present between the hypoconid and the

talonid. The comparative measurements are provided in table 7.

Table 7: Comparative dental measurements of the cheek teeth of the Siwalik

Giraffokeryx and Giraffa in mm (millimeters). * The studied specimens. Referred data

are taken from Matthew (1929), Colbert (1935) and Bhatti (2005).

Taxa Number Nature/Position Length WidthGiraffokeryx PUPC 07/88* P3 24.0 21.3 punjabiensis PUPC 09/67* P3 25.0 ?22

PUPC 07/133* M2 25.6 27.0PUPC 09/43* M2 23.0 17.0

M3 38.0 20.0PUPC 07/90* M3 39.0 13.0AMNH 19475 P3 20.5 20.0

M1 22.0 24.0M2 25.0 27.0

AMNH 19334 M1 25.5 25.0AMNH 19311 M1 23.0 22.0AMNH 19930 P3 22.0 20.0AMNH 19472 M2 27.0 25.5AMNH 19587 M2 25.0 17.0

M3 37.0 17.0Giraffa priscilla PUPC 07/131* M1 25.0 25.0

PUPC 07/89* M1 27.0 27.0PUPC 02/99 M1 24.0 24.0

Giraffa punjabiensis GSI K 13/349 M1 30.0 24.0PUPC 86/84 M1 31.0 28.0PUPC 95/23 M1 31.0 27.0

91

Figure 15: Giraffokeryx punjabiensis. 1. PUPC 07/88, lP3. 2. PUPC 09/67, rP3: occlusal

view. 3. PUPC 07/133, rM2. a = occlusal view, b = lingual view, c = buccal view. Scale

bar equals 10 mm.

92

Figure 16: Giraffokeryx punjabiensis. 4. PUPC 09/43, left hemimandible with M2-3,

broken canine and alveoli of P3-M1. a = occlusal view, b = lingual view, c = buccal view.

Scale bar equals 50 mm.

93

Figure 17: Giraffokeryx punjabiensis. 5. PUPC 07/90, rM3. a = occlusal view, b = lingual

view, c = buccal view. Scale bar equals 10 mm.

Comparison and Discussion

The studied specimens include selenodont teeth and these may be referred to some

tylopods or ruminants. Since the specimens under study have very rugose enamel and this

fine rugosity is not seen in any tylopods so they can be referred to ruminants. In

ruminants, such a heavy rugosity is the characteristic of the giraffids (Pilgrim, 1911). The

Siwalik giraffids may be divided into two groups, one consisting of the large forms and

other small forms (Sarwar and Akhtar, 1987). The small forms include the genera

Giraffokeryx and Giraffa, while the large forms include the genera Bramatherium,

Hydaspitherium, Sivatherium, and Vishnutherium.

The teeth are small in size and can be included Giraffokeryx and Giraffa. Giraffokeryx

punjabiensis is very close to Giraffa priscilla in size (Colbert, 1935). The external folds

94

(parastyles) are comparatively more developed in the premolars of the Giraffa

punjabiensis which can be observed in the studied premolar. Moreover, the

anteroposterior length and transverse width of the premolars are same to the already

known material of the species Giraffokeryx punjabiensis (Table 7). The second molar

shows the typical morphology of the species Giraffokeryx punjabiensis. The stylids and

median ribs are less pronounced in the studied specimens, the feature of Giraffokeryx

punjabiensis. Morphometrically, the specimens resemble to the already described

samples of Giraffokeryx punjabiensis (Fig. 18) and should be assigned to Giraffokeryx

punjabiensis. The hemimandible with diastema is new to science from Nagri, Middle

Siwaliks of Pakistan and reports for the first time in this thesis.

Giraffokeryx was founded by Pilgrim (1910) on the genotype Giraffokeryx punjabiensis.

The genus and species has been found from the Siwaliks of Pakistan and India, and is

known from Turkey (Geraads et al., 1995). Pilgrim (1910) based Giraffokeryx upon a

collection from various Lower Siwalik localities of Pakistan and India. The collection

consisted of a skull, cranial fragments, mandibular fragments and many isolated teeth

from the Nagri Formation of the Middle Siwaliks and the Chinji Formation of the Lower

Siwaliks, which are described and figured by Pilgrim (1910, 1911) and Colbert (1935).

Giraffokeryx attributes of a giraffe ancestor and occupies the right evolutionary position.

Its features straddle its Palaeomerycine antecedents on the one hand and the

Palaeotraginae assemblage that seems to have arisen from them. Colbert (1935)

concluded from his analysis of its fossils that it had an elongated neck and drew it as a

small giraffe (e.g. see p. 331). The reconstruction by Savage and Long (1986) shows it

95

looking more like an okapi. According to Mitchell and Skinner (2003) is that

Giraffokeryx is a primitive palaeotragine and an ancestral species to Giraffa.

Giraffokeryx punjabiensis

0

5

10

15

20

25

30

0 5 10 15 20 25 30 35 40 45Length

Wid

th

Upper Third Premolar Upper First Molar Upper Second MolarLower Second Molar Lower Third Molar

Figure 18: Scatter diagram showing dental proportions of G. punjabiensis’s studied

sample. Referred data are taken from Matthew (1929), Colbert (1935) and Bhatti (2005).

96

Subfamily Giraffinae Zittel, 1893

Genus Giraffa Brunnich, 1771

Type Species: Giraffa giraffa Brunnich, 1771.

Generic Diagnosis: Medium sized giraffids with extremely elongated neck and limbs,

skull with a moderately large post-orbital development; basicranial and basifacial axes

inclined at a small angle. Paired parieto-frontal bony processes of small size and a

median naso-frontal protuberance in both sexes; in some species paired occipital

processes. A pre-lachrymal vacuity is present (Colbert, 1935). Dentition very brachydont,

enamel very rugose, enamel folds penetrating deeply into the crown and enamel islands

not formed until a late period of dentition, lobes very oblique to the axis. External ribs of

upper teeth very strongly marked, outgrowths of enamel from the crescents into the

central cavity. Length not in excess of breadth, tubercles variable, but generally

rudimentary, cingulum absent. Lower molars not elongated, tubercles in the external

valleys variables but a large one always present in M1 and generally in M3 (Matthew,

1929; Colbert, 1935).

Known Distribution: Indian subcontinent, Greeco-Iranian province and Africa (Pilgrim,

1910, 1911; Matthew, 1929; Colbert, 1935; Gentry, 1997).

Giraffa cf. priscilla Matthew, 1929

Type: GSI B511, a left M3.

Type Locality: Upper portion of Chinji, Lower Siwaliks, Punjab, Pakistan (Matthew,

1929).

97

Stratigraphic Range: Lower Siwaliks and lower portion of the Middle Siwaliks

(Matthew, 1929; Colbert, 1935; in this thesis).

Diagnosis: The broad and more brachydont teeth than those of Giraffokeryx. The

metastyle and anterior rib are heavy; in M3 the more oblique-set inner crescents, broad

third lobe with strong accessory basal cusp in front of it, as well as shorter crown

(Matthew, 1929).

Studied Specimens: Upper dentition: PUPC 07/131, isolated left M1; PUPC 07/89,

isolated right M1.

Description

Upper dentition

The recovered material includes only upper dentition; one complete tooth and one partial

tooth (Fig. 19(1-2)). The molars are brachydont and square shaped. The enamel sculpture

is present. The buccal cones are higher than the lingual ones. The cusps are oblique to the

axis of the tooth. The praepotocrista and the praehypocrista are larger than the

postprotocrista and the posthypocrista Entotyle is present in the transverse valley. The

median ribs are prominent. The styles are heavy; para-, meso- and metastyles are

prominent in the buccal side. The anterior cingulum is present. The central cavity

between protocane and paracone is closed at its front, and a spur is present in it

posteriorly. The central cavity between hypocone and metacone is closed at its rear, and a

spur in the cavity is triangular shaped closing the posterior ridge of the cavity at its center

(Fig. 19(1-2)).

98

Figure 19: Giraffa cf. priscilla. 1. PUPC 07/131, lM1. 2. PUPC 07/89, rM1: occlusal

view. a = occlusal view, b = lingual view, c = buccal view. Scale bar equals 10 mm.

Comparison and Discussion

The molars are about the same size as in Giraffa priscilla (Table 7) and the size is in

practice the only distinguishing criterion in Giraffa (Gentry, 1997). The small size

brachydont Siwalik giraffids include Giraffokeryx and Giraffa (Colbert, 1935; Bhatti,

2005; Khan et al., 2010b). The styles are very weak in Giraffokeryx whereas these are

strong in Giraffa. Median ribs are absent or very weak in Giraffokeryx and these are well

99

pronounced in Giraffa (Matthew, 1929; Colbert, 1935). Furthermore, the crown is narrow

in Giraffokeryx and it is broad in Giraffa. The cusps are not straight line in the studied

molars. The recovered sample exhibits all the features present in Giraffa and the sample

can be assigned to Giraffa.

The Siwalik Giraffa is represented by three species G. sivalensis, G. punjabiensis and G.

priscilla (Colbert, 1935). The posterior half is reduced in G. sivalensis and G.

punjabiensis, however, it is much reduced in G. sivalensis. The metastyle is strong in G.

priscilla and weak in G. sivalensis, G. punjabiensis. Giraffa sivalensis and G.

punjabiensis are somewhat large species and G. priscilla is small species of the Siwalik

Giraffa. The recovered sample is pretty fit to G. priscilla in morphometrically (Fig. 20;

Table 7). Nevertheless the sample is insufficient for the exact specific determination and

assigns to G. cf. priscilla.

Upper First Molar

0

5

10

15

20

25

30

0 5 10 15 20 25 30 35Length

Wid

th

Giraffokeryx punjabiensis Giraffa priscilla Giraffa punjabiensis

Figure 20: Scatter diagram showing dental proportions of G. cf. priscilla’s studied

sample. Referred data are taken from Matthew (1929), Colbert (1935) and Bhatti (2005).

100

Family Tragulidae Milne-Edwards, 1864

Genus Dorcatherium Kaup and Scholl, 1834

Type Species: Dorcatherium naui Kaup and Scholl, 1834.

Generic Diagnosis: Bunoselenodont to selenodont teeth with more or less strong cingula

and cingulidae and mostly strong styli and stylidae at the molars. The upper molars

increase in size from M1 to M3. The lower molars show a special crest complex called

the ‘Dorcatherium-fold’. It is formed by the bifurcation of the posterior slopes of the

protoconid and the metaconid resulting in a ‘Σ’ shape. The premolars are comparatively

long and consist mainly of the buccal conids and crests. Lingual crown elements are

underrepresented. At the p4 the entoconid fuses with the postprotocristid. The p3 has

only a short lingual entocristid originating at the hypoconid. An exception is the P4,

which is shorter and does not have an anteroposterior longish shape. The cheek teeth are

high crowned. The upper molars bear strongly developed buccal styles. The lower molars

are characterized, either by well-developed ectostylid or by a vestigial ectostylid (Kaup,

1833; Rossner, 2010).

Known Distribution: Dorcatherium is known from the Lower Miocene of Europe by

Kaup (1833) and Arambourg & Piveteau (1929). It is also reported from the Miocene

deposits of East Africa by Lartet (1837), Arambourg (1933), Whitworth (1958) and

Hamilton (1973). It is distributed from middle Miocene to early Pliocene in Asia, and late

early Miocene to early Pliocene in Africa (Pickford et al., 2004). Dorcatherium is

recorded from the Siwaliks by Lydekker (1876), Colbert (1935), Prasad (1968), Sahni et

al. (1980), West (1980), Farooq (2006) and Farooq et al. (2008).

101

Dorcatherium cf. minus Lydekker 1876

Type Specimen: GSI B195, right M1-2.

Type Locality: Kushalgar near Attock and Hasnot, Punjab, Pakistan (Colbert, 1935).

Stratigraphic Range: Lower to Middle Siwaliks.

Diagnosis: A small species of the genus Dorcatherium with sub-hypsodont molar and

broad crowned molars having well developed cingulum, rugosity, styles, moderately

developed ribs and vestigial ectostylids (Colbert, 1935).

Studied Specimens: Upper dentition: PC-GCUF 10/10, isolated left dP4. Lower

dentition: PUPC 07/69, a right mandibular ramus with partial M1 and complete M2.

Description

Upper Dentition

The recovered upper dentition comprises only one deciduous premolar (Fig. 21(1)). The

tooth is in early wear and extremely brachydont. The central cavities are filled with

matrix. The front and back walls of the tooth are convergent. The anterior rib is strong

enough to form two vertical grooves: one between the parastyle and the anterior median

rib and the second is between the anterior median rib and the mesostyle. The occlusal

length of 15.5 mm buccally and it is 6.4 mm lingually. It has a notably large parastyle and

a cingulum on its lingual lobes (Fig. 21(1b)). The strong parastyle and the convergent

front and back walls of the tooth suggest it is a dP4 and not a molar.

Lower Dentition

PUPC 07/69 is a fragile mandibular fragment having a posterior half of the first molar

and the complete second molar (Fig. 21(2)). The second molar is hypsodont and narrow

102

crowned tooth. It is almost unworn tooth. The metaconid is pointed and higher than the

protoconid and the hypoconid. The entoconid is more pointed and higher than the

hypoconid. The rudimentary ectostylid is present. The posterior cavity is crescentic in

shape and opening out of the tooth at the lingual side. An anterior cingulid is present. The

anterior rib and stylid are well developed. A prominent and narrow posterior rib is

present but posterior stylid is weak. The Dorcatherium fold is present and directed

posteriorly. It is formed by the bifurcation of the posterior slopes of the protoconid

resulting formed a ‘Σ’ shape, a diagnostic feature of Dorcatherium. The comparative

measurements are provided in table 8.

Figure 21: Dorcatherium cf. minus. 1. PC-GCUF 10/10, ldP4. 2. PUPC 07/69, a right

mandibular ramus with partial M1 and complete M2. a = occlusal view, b = lingual view, c

= buccal view. Scale bar equals 10 mm. 103

Comparison and Discussion

The upper and lower molars show all the morphological features of the species

D. minus as described by Lydekker (1876) and Colbert (1935) e.g., the small sized upper

and lower molars. The upper molars are specifically characterized by their finely rugose

enamel, a comparatively weak mesostyle and well-developed lingual cingulum, whereas

the lower molars are characterized by the slight rugosity and the vestigial ectostylid

(Colbert, 1935; Farooq, 2006). The described specimens are found to closely resemble to

the type specimens regarding the measurements (Table 8). Consequently, the material

assigns to D. cf. minus, based on the morphometric features (Fig. 23). The above said

discussion confirms that the specimens belong to the species Dorcatherium cf. minus

undoubtedly.

Dorcatherium cf. majus Lydekker, 1876

Type Specimen: GSI B197, two upper molars (right M1-2).

Type Locality: Hasnot, Jhelum, Punjab, Pakistan.

Stratigraphic Range: Lower to Middle Siwaliks (Colbert, 1935; Farooq, 2006; Farooq

et al., 2008).

Diagnosis Dorcatherium majus is a species larger than Dorcatherium minus and is equal

in size to Dorcabune anthracotherioides. It is characterized by strong parastyle and

mesostyle, well-developed cingulum in upper molars and stoutly developed ectostylid

(Colbert, 1935).

Studied Specimens: Upper dentition: PC-GCUF 09/46, isolated right M2.

104

Description

Upper Dentition

PC-GCUF 09/46 (Fig. 22(1)) is a well preserved specimen, in early wear. The tooth is

quadrate in its general appearance. Transversely, it is more wide anteriorly than

posteriorly. The specimen under study is brachydont and broad crowned. The enamel is

uniformly thick and rugose. The cingulum is thick and well developed on the lingual side,

especially at the entrance of the transverse valley, whereas on the anterior and posterior

sides of the tooth the cingulum becomes thin and high. It is entirely absent around the

buccal cones. The anterior and posterior cavities are deep and wide. All four major cusps

are inclined towards the median longitudinal axis of the molar, although the degree of

inclination is greater in the lingual cusps than the buccal ones. The protocone is more

worn than the other cusps. It exhibits semi-crescentic shape, as its praeprotocrista is

longer than the postprotocrista. The praeprotocrista is linked with the parastyle through a

thin crista of the enamel and the postprotocrista on the other hand is free. The paracone is

higher than the protocone. The praeparacrista and the postparacrista seem to be equally

worn. The parastyle and the anterior median rib are well developed and linked together at

their base. The metacone is the highest cone among all cones. It is also equally worn

anteriorly and posteriorly. The posterior median rib is weaker than the anterior one. The

mesostyle is well developed, whereas the metastyle is comparatively weak. The

hypocone is more crescentic than the protocone, because the praehypocrista and the

posthypocrista are almost equal in length, exhibiting V-shaped structure. The

praehypocrista is not linked with the postprotocrista. The comparative measurements are

provided in table 8.

105

Figure 22: Dorcatherium cf. majus. 1. PC-GCUF 09/46, isolated right M2. a = occlusal

view, b = lingual view, c = buccal view. Scale bar equals 10 mm.

Comparison and Discussion

The studied specimens prove their inclusion in the family Tragulidae, based on the

selenobunodont to selenodont pattern with strong cingula (Rossner, 2010). There are two

genera Dorcatherium and Dorcabune of family Tragulidae present in the Siwaliks

(Colbert, 1935). The Dorcabune is a large extinct tragulid of the Siwaliks and close to the

anthracotherioides having the most bunodont molars. The anterior median rib is also

heavier in Dorcabune than Dorcatherium (Colbert, 1935; Farooq et al., 2007a, b). The

presence of the selenobunodonty cusp pattern, cingula and strong styles in the sample

proves that the material belongs to the genus Dorcatherium. The Siwalik Dorcatherium is

represented by three species D. majus, D. minus and D. minimus (Colbert, 1935; West,

1980) and show variation in size (Fig. 23). Dorcatherium minimus is erected by West

(1980), based on the single upper third molar and it is probably considered invalid

species (see Farooq, 2006). Dorcatherium minus is a small Siwalik tragulid and

Dorcatherium majus is the large Siwalik species of Dorcatherium (Farooq et al., 2008).

106

The studied specimen morphometrically pretty fit with D. majus (Table 8; Fig. 23) and

should assign to D. majus. The material is insufficient and assigns to D. cf. majus.

Table 8: Comparative measurements of the cheek teeth of D. majus and D. minus in mm

(millimeters). * The studied specimens. Referred data are taken from Colbert (1935) and

Farooq et al. (2007a, b, 2008).

Taxa Number Nature/Position Length Width

D. majus PUPC 09/46* M2 19.4 19.0

AMNH 19302 M2 18.5 21.5

GSI B198 M2 19.6 19.6

PUPC 85/15 M2 19.0 20.0

PUPC 85/21 M2 18.0 22.0

PUPC 87/328 M2 17.7 19.0

PUPC 67/191 M2 13.3 14.5

PUPC 68/33 M2 13.3 14.5

PUPC 68/250 M2 15.7 16.4

AMNH 19524 M2 16.0 11.0

GSI B593 M2 17.5 10.0

PUPC 63/243 M2 17.0 10.1

PUPC 84/115 M2 16.0 12.0

PUPC 86/02 M2 15.6 9.80

PUPC 86/05 M2 15.0 11.1

PUPC 86/152 M2 16.2 12.0

PUPC 98/61 M2 17.0 10.5

AMNH 19520 M2 17.0 10.5

D. minus PC-GCUF10/10* dP4 15.5 13.4

PUPC 07/69* M2 13.0 8.00

PUPC 68/41 M2 11.0 13.0

PUPC 68/355 M2 10.5 11.8

PUPC 86/81 M2 10.0 12.2

107

Table 8 (Continued).

PUPC 95/01 M2 10.0 11.0

PUPC 02/01 M2 10.5 11.6

AMNH 29856 M2 11.3 12.0

GSI B195 M2 11.0 12.0

PUPC 68/294 M2 11.0 6.40

PUPC 68/311 M2 10.0 6.60

PUPC 68/312 M2 10.0 6.20

PUPC 68/313 M2 10.2 6.70

PUPC 85/59 M2 9.50 7.00

PUPC 02/158 M2 12.7 8.20

AMNH 19365 M2 13.0 12.0

108

A.

Upper Second Molar

02468

1012141618202224

0 2 4 6 8 10 12 14 16 18 20 22Length

Wid

th

D. majus D. minus

B.

Lower Second Molar

02468

101214

0 2 4 6 8 10 12 14 16 18 20Length

Wid

th

D. majus D. minus

Figure 23: Scatter diagram showing dental proportions of Dorcatherium’s studied

sample. Referred data are taken from Colbert (1935) and Farooq et al. (2007a, b, 2008).

109

Genus Dorcabune Pilgrim, 1910

Type Species: Dorcabune anthracotherioides Pilgrim, 1910.

Generic Diagnosis: Primitive large tragulids having bunodont teeth. Isolated parastyle

and mesostyle, prominent cingulum and enamel rugosity are the diagnostic characteristics

of the upper molars, whereas lower molars are characterized by their broadness, a wide

talonid in the third molar, and a pyramidal protoconid with two posteriorly directed folds

(Pilgrim, 1910, 1915; Colbert, 1935). The upper molars of Dorcabune

anthracotherioides are characterized by their brachydonty and bunodonty. The inner

cusps of upper molars are truly selenodont, whereas the outer ones are quite bunodont

and absolutely conical in their general appearance. The median rib on the buccal face of

the paracone and metacone is so broad and prominent that it occupies almost all the space

between the styles. This feature is very much pronounced in the paracone, the buccal

surface of which is in fact entirely rib. The parastyle and mesostyle are strong, massive

and isolated, whereas the metastyle is very weakly developed. With wear, the mesostyle

clearly displays its closer association with the metacone instead of fusing equally to both

paracone and metacone. The protocone, instead of being a simple crescent, is more

pyramidal in shape and displays three equally strong folds, one proceeding forwards and

outwards, the second backwards and a third backwards with a tendency sometimes

inwards and sometimes outwards. A strong cingulum runs antero-posteriorly, but is very

much pronounced round the protocone. It often rises into a small tubercle at the entrance

of the transverse valley between the protocone and hypocone. The enamel is heavy and

has moderately fine rugosity (Pilgrim, 1915; Colbert, 1935). The lower molars are also

characterized by well pronounced brachydonty, bunodonty and presence of a typical

110

tragulid M structure at the rear of the tragonid. The anterior arm of protoconid terminates

on a broad shelf almost parallel to the anterior margin of the tooth. Entoconid is conical,

producing out anteriorly a short process in the direction of the mid line between the two

anterior cusps. The hypoconid is crescentic; its anterior arm touches to the external

process of the protoconid, while its posterior arm runs inward and completely encircles

the posterior base of the entoconid (Pilgrim, 1915; Colbert, 1935).

Known Distribution: The genus is found in the Lower Manchar of Bhagothoro,

Pakistan, and the Lower and the Middle Siwaliks, and China (Pilgrim, 1910, 1915;

Colbert, 1935; Han De-Fen, 1974; Farooq et al., 2007a).

Dorcabune cf. anthracotherioides Pilgrim, 1910

Type Specimen: GSI B580, a maxilla with molars.

Type Locality: Near Chinji, Chakwal, Punjab, Pakistan (Colbert, 1935).

Stratigraphic Range: Lower to Middle Siwaliks (Pilgrim, 1910, 1915; Colbert, 1935;

Farooq et al., 2007b).

Diagnosis: Dorcabune anthracotherioides is larger than Dorcabune hyaemoschoides and

almost equal to that of Dorcatherium crassum (Pilgrim, 1915; Colbert, 1935). The

mandible bears a fairly deep groove starting beneath the P4 and propagating towards the

posterior side behind the teeth. This groove exists also in Dorcatherium majus and minus

and in Dorcabune latidens but is absent in Dorcabune nagrii (Pilgrim, 1915). The upper

molars of Dorcabune anthracotherioides are very similar to that of Dorcabune

hyaemoschoides and differ only by the possession of prominent parastyle. The lower

111

fourth premolar (P4) is slightly shorter in length than the lower third premolar (P3). P4 is

broad and consisting of three lobes, of which middle one is the highest and longest,

whereas first and the last lobes are equal in length, though the third lobe is higher in

unworn condition. Third lobe is massive and crescent-shaped facing towards the inner

and the anterior sides. The posterior arm of the crescent is running out to a level with the

internal margin of the tooth. A small notch separates this arm from a long wing which

runs backward from the summit of the principal cusp and forms the inner wall of the

tooth. This wing is separated by a deep elongated cavity from the crest, which connects

the principal cusp to the hinder lobe (Pilgrim, 1915). Dorcabune anthracotherioides

differentiates Dorcabune latidens by characterizing a less deep mandible bearing

moderately broader molars and possessing much smaller size (Pilgrim, 1915).

Studied Specimens: Lower dentition: PUPC 07/87, isolated left M2.

Description

Lower Dentition

The bunodont second lower molar is in a good state of preservation (Fig. 24(1)). It is in

an early wear. It is hypsodont and narrow crowned. The enamel is heavy, thick and very

rugose. The cingulid is well developed on anterior and posterior sides but it is absent

buccally and lingually. A small singular tubercle is also present between the hypoconid

and the protoconid on the buccal side. The praeprotocristid terminates in a broad shelf,

almost parallel to the anterior margin of the tooth. The postprotocristid is bifurcated and

one limb of the bifurcation is attached to the postmetacristid while the other one is

attached to the praehypocristid producing M structure (Fig. 24(1a)). The hypoconid is

112

somewhat crescentic in shape; the praehypocristed touches the postprotocristid, whereas

the posthypocristid runs inwards and completely encircles the posterior base of the

entoconid. The metaconid is conical and bunodont. The entoconid is conical with a short

anterior process proceeding between the two anterior conids. There is a vertical groove

between the metaconid and the entoconid lingually. The comparative measurements are

provided in table 9.

Table 9: Comparative measurements of the cheek teeth of Dorcabune cf.

anthracotherioides in millimeters (mm). * The studied specimens. Comparative data are

taken from Colbert (1935) and Farooq et al. (2007a).

Number Nature/Position Length Width

PUPC 07/87* M2 22.4 16.0

PUPC 96/65 M2 20.30 13.30

PUPC 96/66 M2 19.00 12.00

PUPC 99/89 M2 19.60 11.55

AMNH 19355 M2 17.50 13.00

GSI B.682/683 M2 19.50 14.70

PUPC 85/28 M3 26.00 13.00

AMNH 19353 M3 28.00 14.00

GSI B682/683 M3 30.90 16.00

113

Figure 24: Dorcabune cf. anthracotherioides. 1. PUPC 07/87, lM2. a = occlusal view, b

= lingual view, c = buccal view. Scale bar equals 10 mm.

Comparison and Discussion

The recovered molar from the Nagri type area exhibits buno-selenodonty pattern. The

selenodonty are found in families of Bovidae, Cervidae, Giraffidae and Camelidae and

the semi-selenodonty with bunodont pattern is found in family Tragulidae (Colbert,

1935). The studied specimens reflect semi-selenodonty with bunodont pattern and belong

to family Tragulidae. The Siwaliks are represented by two tragulid genera Dorcatherium

and Dorcabune (Colbert, 1935; Farooq et al., 2007a, b). Dorcabune reflects bunodonty

pattern and Dorcatherium is somewhat selenodont (Fig. 24). The bunodont conical cusp

pattern of the studied samples confirms its inclusion to Dorcabune (Fig. 25). The molar

has the same size of the already recovered sample of D. anthracotherioides. The molar is

114

comparable with the holotype and the earlier described specimens (Table 9). Therefore,

the molar assigns to D. cf. anthracotherioides (Colbert, 1935; Farooq et al., 2007a).

Dorcabune from the Siwaliks of Pakistan were erected by Pilgrim (1915), named three

species: Dorcabune anthracotherioides, Dorcabune hyaemoschoides, Dorcabune nagrii,

and Dorcabune latidens. Dorcabune anthracotherioides and D. nagrii are considered as

valid species where as Dorcabune hyaemoschoides and Dorcabune latidens are known

by very poor record (Colbert, 1935; Farooq, 2006). According to Gentry (1978)

Dorcabune is most probably an anthracotheriid, however a number of collected dental

specimens from the Middle Siwaliks after Pilgrim (1910) evidently prove its inclusion in

Tragulina.

D. anthracotherioides

02468

1012141618

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34Length

Wid

th

Lower Second Molar Lower Third Molar

Figure 25: Scatter diagram showing dental proportions of D. cf. anthracotherioides’s

studied sample. Comparative data are taken from Colbert (1935) and Farooq et al.

(2007a).

115

Order Perissodactyla Owen, 1848

Family Equidae Gray, 1821

Subfamily Equinae Steinmann and Doderlein, 1890

Genus Hipparion de Chirstol, 1832

Type species: Hipparion prostylum Gervias.

Generic Diagnosis: Isolated and usually oval-elongated protocones in upper molars

during early to late stage of wear. Tridactyl feet with elongate slender third metapodials.

Prominent nasomaxillary fossa with a relatively well developed anterior rim.

Hipparionine horses with markded reduction of preorbital fossa in length, dorsoventral

height, medial depth and posterior pocketing. Maxillary cheek teeth show a tendency

towards simplification of prefossette and postfossette ornamentation with thinner enamel

bands. Plicaballins become simplified to a single morphology. Hypoglyph usually,

moderately to shallowly incised in middle adult wear (Colbert, 1935).

Known Distribution: North America, Asia, Africa (Pilbeam et al., 1997).

Hipparion theobaldi (Lydekker) Colbert, 1935

Type Specimen: GSI C153, a left maxilla with milk molars.

Type Locality: Keypar, Middle Siwaliks (Colbert, 1935).

Stratigraphic Range: Middle Siwaliks (Ghaffar, 2005; Naseem et al., 2009).

Diagnosis: A large Hipparion with tridactyl feet and deep preorbital facial fossa

separated from the orbit by a relatively wide preorbital bar, fossa deeply pocketed

posteriorly, medially deep and with a well defined continuous peripheral border including

116

the anterior rim, cheek teeth complexly ornamented with thickly banded fossette plication

and bifid to trifid plicaballins, protocones distinctly flattened lingually and rounded

buccally, hypoglyphs deeply incised, P2 anterostyle elongate, (Colbert 1935).

Studied Material: Upper dentition: PUPC 07/61, isolated left P2; PUPC 07/65, isolated

left M1; PUPC 07/66, isolated right M1; PUPC 07/57, isolated left M2; PUPC 07/58,

isolated left M3. Lower dentition: PUPC 07/60, isolated right P2; PUPC 07/59, isolated

right P3; PUPC 07/78, isolated left P4; PUPC 07/124, isolated right M3.

Description

Upper Dentition

PUPC 07/61 is a well preserved tooth and it is in early wear. The premolar is almost

triangular with characteristically well developed anterostyle. It is strongly elongated and

pillar like (Fig. 26(1)). All the major cusps are well developed and preserved. The

protocone is an isolated compressed pillar and elongated in shape. It is covered by a

moderately thick layer of cement. The hypoconal groove is well developed and placed

posteriorly. The styles are well preserved, strongly developed and prominent. The

mesostyle is pillar like structure and is similar to the parastyle in general appearance.

Both the styles are broad at the base and narrow at the apex. The metastyle is moderately

developed and straight in shape. The hypostyle is weak and not prominent like the other

styles. The protoloph, the metaloph, and the ectoloph are distinguished. The crown is

highly plicated (Fig. 26(1)). The prefossette and postfossette are plicated with maximum

enamel foldings. The plicaballin consists of bifurcated folds.

117

PUPC 07/65 and PUPC 07/66 (Figs. 26(2-3)) are the first upper molars of maxillary

series. The molars are hypsodont and well preserved. The enamel is moderately thick and

wrinkled. The layer of matrix is present all over the crown. The protocone, paracone,

metacone and hypocone are preserved. The protoloph, ectoloph and metaloph are

distinguished. The protocone is not pillar like but sub-ovate and isolated. The paracone

and the hypocone are not broad while the metacone is angulatory. The styles are well

preserved. The mesostyle is strongly developed while metastyle is weakly developed.

The pre-fossette and post-fossestte are richly plicated with maximum enamel foldings at

the posterior border of prefossette and at the anterior border of postfossette. The

pliprotoloph, plihypostyle of postfossette are well preserved. The hypoconal groove is

prominent and the fossettes are clear.

PUPC 07/57 is the well preserved second hypsodont molar (Fig. 26(4)). The enamel is

rugose, moderately thick and wrinkled. The protocone is pillar like, sub-ovate and

isolated which is the diagnostic character of Hipparion molar. The styles are well

preserved on the buccal side. The mesostyle is strongly developed and more prominent

than the parastyle and the metastyle. The parastyle is weak comparatively than the

mesostyle and the metastyle. The pre-fossette and post-fossette are richly plicated with

maximum enamel folding at the posterior borders of the post-fossette and at the anterior

border of the pre-fossettte. The ectoloph, protoloph and metaloph are well preserved. The

thick layer of the cement is present lingually. The pliprotoloph, pre-fossttes and plipost-

fossettes are clearly visible. The hypoconal groove is well developed, prominent and

present posteriorly of the molars.

118

The M3 is somewhat narrow posteriorly (Fig. 26(5)). The plicabillins are bifid and the

hypoglaph is deeply incised. The protocone is rounded lingually and buccally. The tooth

is in early wear. The pre- and postfossettes are moderately complex.

Lower Dentition

PUPC 07/60 comprises right isolated second premolar (Fig. 27(6)). The enamel is

moderately thick. The principal conids are well developed. The protoconid is completely

isolated from rest of the premolar and it is an oval shape. The buccal conids are spindle or

cresentric in shape i.e. both are broad in the middle and narrow anteroposteriorly. The

entoconid are triangular in shape and join with the metastylid through an isthamus, giving

the appearance of double knot. The hypoconid is broad. The stylids are well developed

and prominent like the conids. The ptycostylid is a fold like structure present at the

mesial border of the hypoconid. The mesostylid is triangular in its outline. It is joined

with metaconid by a narrow isthamus giving the appearance of double knot like structure.

There is no demarcation of the ptycostylid and the anterior end of the hypoconid.

PUPC 07/59 includes the P3 of the right mandible (Fig. 27(7)). It is in excellent state of

preservation and in an early stage of wear. The metaconid and metasylid are joined by a

narrow isthamus giving the appearance of double knot. The lingual side of the tooth is

vertically higher than the buccal one. The ectostylid and mesostylid are well preserved

and prominent. The ectosylid is prominent pillar like structure. The protostylid is absent.

There is a deep plication formed by the union of the posterior side of the parastylid and

the anterior side of the protoconid. There are two prominent invaginations lingually,

named the metaflexid anteriorly and the entoflexid posteriorly. The metaflexid is narrow

in the middle, while broad anteroposteriorly. The entoflexid is elongated, curved

119

anteriorly, while broad posteriorly. The wall of the entoflexid is wrinkled. The lower

fourh premolar PUPC 07/78 is in early wear (Fig. 27(8)). The enamel is thick and

wrinkled. The protoconid and the hypoconid are comparatively broad. The metaconid is

broad and triangular and united with narrow isthmus. The protostylid is bulky and

narrow, while the mesostylid and the entostylid are weak. The metaflexid is narrow while

the entoflexid is triangular in shape and narrow. The hypoconulid is absent but the

mesostylid is present on the lingual side.

Lingualflexids are shallow on all premolars. Entoflexids are elongate and have complex

anteriormost borders where they separate metaconid/metastylid. The premolars have

ectoflexids which do not separate the metaconids/metastylids. Metaconid is rounded on

the P2, having angular posterior borders on the P3-4. Metastylids have marked angular

posterior borders on all premolars. Protostylids are present on P3-4.

The ectoflexid in the lower third molar PUPC 07/124 (Fig. 27(9)) does not separate

metaconid and metaflexid. The molar has angular facing borders of metaconid and

metastylid. There are no plicaballinids. Lingualflexid is deeply incised and elongate.

Entoflexid is elongate and has a complex border. Metaflexid is square and has complex

border. The protostylid is present in the molar. The hypoconulid is present which is

concave buccally. The comparative dental measurements are provided in table 10.

120

Table 10: Comparative measurements of the cheek teeth of H. theobaldi in mm

(millimeters). * The studied specimens. Referred materials are taken from Colbert (1935)

and Ghaffar (2005).

Taxa Number Nature/Position Length WidthH. theobaldi PUPC 07/61* P2 37.0 26.0

PUPC 07/65* M1 26.0 27.0PUPC 07/66* M1 26.5 25.5PUPC 07/57* M2 27.5 27.5PUPC 07/58* M3 28.0 20.7PUPC 07/60* P2 33.0 15.6PUPC 07/59 * P3 30.5 20.0PUPC 07/78* P4 25.3 20.0PUPC 07/124* M3 33.0 13.0AMNH 19857 P2 32.0 26.5 PUPC 83/284 P2 39.5 21.5GSI C153 P2 38.5 26.0PUPC 83/498 P2 40.0 22.0AMNH 19466 M1 26.0 26.0AMNH 19857 M1 25.0 21.0AMNH 19836 M2 22.0 21.0AMNH 19492 M2 20.0 22.0AMNH 19711 M2 28.0 29.0AMNH 19466 M2 26.0 26.0AMNH 19466 M3 24.0 22.0AMNH 19857 M3 25.0 23.0PUPC 83/498 P2 31.0 12.0PUPC 83/285 P2 31.0 15.0PUPC 83/290 P2 30.0 12.0PUPC 83/786 P2 32.5 13.5PUPC 83/787 P2 33.0 13.0PUPC 86/183 P2 30.5 14.0PUPC 00/94 P2 31.0 18.5PUPC 00/94 P3 25.0 18.0PUPC 83/498 P3 26.5 13.0PUPC 00/94 P4 25.0 18.0PUPC 83/498 P4 25.0 24.0PUPC 87/309 P4 26.0 15.5PUPC 83/498 M3 28.0 11.5PUPC 87/255 M3 26.5 13.0PUPC 96/23 M3 28.5 13.5PUPC 00/94 M3 30.0 13.5

121

Figure 26: Hipparion theobaldi. 1. PUPC 07/61, lP2. 2. PUPC 07/65, lM1. 3. PUPC

07/66, rM1. 4. PUPC 07/57, lM2. 5. PUPC 07/58, lM3. a = occlusal view, b = lingual

view, c = buccal view. Scale bar equals 10 mm.

122

Figure 27: Hipparion theobaldi. 6. PUPC 07/60, rP2. 7. PUPC 07/59, rP3. 8. PUPC

07/78, lP4. 9. PUPC 07/124, rM3. a = occlusal view, b = lingual view, c = buccal view.

Scale bar equals 10 mm.

123

Comparison and Discussion

The general appearance of the studied specimens and strong pillar like isolated protocone

exclude the specimens from the genus Equus and favor their inclusion in the genus

Hipparion. The Middle Siwalik is represented by four Hipparion species H. nagriensis,

H. perimense, H. theobaldi and H. antelopinum (Colbert, 1935). Two of them are found

abundant in the Middle Siwaliks (Iqbal et al., 2009). Hipparion nagriensis is

comparatively small species (Mac Fadden and Woodburne, 1982; Naseem et al., 2009).

The protocone of H. perimense is flattened lingually and rounded buccally (Ghaffar,

2005). Hipparion antilopinum is large with complicated plications and oval shape

protocone. Hipparion theobaldi are large, having less complicated plications. Hipparion

theobaldi differs from H. antilopinum of having compressed protocone, as compared to

round oval shape protocone in H. antilopinum. Furthermore, the enamel borders of

cavities relatively simple in H. theobaldi and complicated in H. antilopinum (Lydekker,

1882; Colbert, 1935).

The morphology of the studied specimens reveals all the features of species H. theobaldi

as described by Lydekker (1882), Colbert (1935) and Ghaffar (2005) and the specimens

are assigned to H. theobaldi. This species is characterized by the isolated, compressed

and pillar like protocone, the molar size is greater than H. antilopinum and H. nagriensis,

and smaller than H. perimense. The enamel bordering of the cavities are relatively

simple. The specimens are extremely hypsodont and show less complicated plications.

The studied specimens show the same basic features of the species such as an anterostyle

in premolars and isolated protocone in cheek teeth, simple enamel bordering of central

cavities with large size. According to Colbert (1935), it is heavy and larger species as

124

compared to H. antilopinum and H. nagriensis. All the studied premolars reflect

hipparionine features i.e., they are longer than broad with longitudinally elongated

protocones and hypocones (Figs. 26-28). The enamel plications are moderately

complicated. These features are also mentioned by Lydekker (1882) with type specimen.

The genus Hipparion was erected by De Christol (1832) on the basis of fossil material

collected from the Turolian age locality of Mt-Luberon in the province of Vaucluse in

France. Christol (1832) characterized this Mt-Luberon horse with isolated protocones in

the upper molars and tridactyl feet. Christol did not designate a type species of the genus.

This was done by Gervais (1849) when he described a syntypic series from Mt-Luberon,

including H. prostylum, H. mesostylum and H. diplostylum.

The name Sivalhippus theobaldi was introduced by Lydekker (1877a). Shortly thereafter

Lydekker (1877b) synonymized Sivalhippus with Hippotherium retaining the name

Hippotherium. Lydekker used the combination Hippotherium theobaldi in two

subsequent publications (1882, 1883b) until finally synonymizing the genus with

Hipparion as H. theobaldi (Lydekker, 1885, 1886). Later on, the name H. theobaldi was

followed by Pilgrim (1913), Matthew (1929), Colbert (1935), Gromova (1952), Hussain

(1971) as referring smaller specimens to H. antelopinum and large to H. theobaldi.

Forsten (1968) however choose not to recognize H. antelopinum arguing that all Siwalik

hipparions belong to one polytypic species H. primigenium Skinner and Mac Fadden

(1977) recognize the distinctiveness of both these species but referred them

Cormohipparion. Mac Fadden (1984) similarly recognizing the species Cormohipparion

theobaldi. Phylogenetically Cormohipparion theobaldi appears to lie close to primitive

Eurasian Hipparion as morphologically idealized by H. primigenium. The specimen in

125

AMNH and yale-GSP collection referred to Cormohipparion theobaldi are more robust

then any of the metapodial material for Eurasian primitive hipparionines.

Skinner and Mac Fadden (1977) created new genus Cormohipparion and stated that all

the hipparionine should be included in this new genus. But later on Bernor and Hussain

(1985) stated that there is no need to erect new genus. Moreover, Mac Fadden and

Skinner (1977) have also suggested that it is highly probable that all the Siwalik

hipparionines should be included in the genus Cormohipparion with specific differences

based on dentition size and feet which are characteristics, classically employed in

hipparion taxonomy. Forsten (1968) was the most extreme in jumping all the Siwalik

Hipparion specimens into single highly variable population of H. primigenium.

According to him, the Siwalik Hipparion group is evolutionary conservative, and

compares closely with H. primigenium and other primitive horses in complex cheek teeth

fossette ornamentation, double or complex plicaballins, deeply incised hypoglyph and

elongate P2 anterostyle. He further argued that C. gracile and C. moldavicum are also

best derived from Eurasian horses similar to H. primigenium. The presence of an anterior

rim on facial fossa is also a primitive character shared by several Holarctic hipparionines.

The H. antilopium and H. theobaldi are significantly different in facial, dental and

possibly posteranial features from species of North American cormohipparions (Ghaffar,

2005). The North American cormohipparions and Eurasian hipparionines obscure the

plylogentic relationship of Holarctic cormohipparion. Subsequently, a long series of

investigations on old world hipparions, have incorporated these original morphological

characters in referring a vast array of species to this one genus. Consequently, a strong

central dogma has arisen in the systematics of Hipparion. Most paleontologists believed

126

that hipparion evolved in the New World from an unidentified species of Merychippus.

But some recent investigators have argued that several distinct lineages may be present in

the Old World and the Old World morphology evolutionary pattern may not be accurate.

As a result, the genus Hipparion has become more circumscribed in its morphological

characterization and the number of species included within the genus.

Hipparion first appeared in the Siwaliks by a single migration record in lithologic

boundary of the Nagri Formation (Hussain, 1971). The Nagri, type locality of the Nagri

Formation has yielded the richest record of the Siwalik hipparions. Hipparion species are

important markers in faunal correlations known as fossil record from middle Miocene to

Pleistocene. The oldest occurrence of the Siwalik hipparion is from the Nagri Formation,

ca 9.5 Ma (Hussain, 1971), while the old world occurrence of Equus is from Hasnot, ca

2.48 Ma (Barry et al., 1982) that is the oldest occurrence of Equus in Southern Asia.

Hussain (1971) made the first contemporary revision of the Siwalik hipparions. He

suggested that Hipparion first appeared in the Siwaliks by a single migration record in

lithologic boundary of the Nagri Formation and subsequently underwent autochthonous

evolution. He recognized three species of the Siwalik Hipparion: (1) H. nagriensis, (2)

H. theobaldi , and (3) H. antelopinum. More recently, Ghaffar et al. (2003)

considered the validity of the genus Hipparion with its four species H. antelopinum, H.

sivalensis, H. theobaldi and H. nagriensis.

127

A.

H. theobaldi

0

5

10

15

20

25

30

0 10 20 30 40 50Length

Wid

thUpper Second Premolar Upper First MolarUpper Second Molar Upper Third Molar

B.

H. theobaldi

0

5

10

15

20

25

30

0 5 10 15 20 25 30 35Length

Wid

th

Lower Second Premolar Lower Third Premolar Lower Fourth Premolar Lower Third Molar

Figure 28: Scatter diagram showing dental proportions of H. theobaldi’s studied sample.

The referred materials are taken from Colbert (1935) and Ghaffar (2005).

128

Family Rhinocerotidae Owen, 1848

Subfamily Rhinocerotinae Owen, 1845

Tribe Teleoceratini Hay, 1885

Genus Brachypotherium Roger, 1904

Type Species: Aceratherium perimense Falconer and Cautley, 1847.

Generic Diagnosis: A rhinoceros of gigantic size with hypsodont teeth. Skull rather short

and deep, with retracted nasals; zygomatic arch heavy; postglenoid separate from

posttympanie. Upper incisor present and well developed. Molars with moderately

developed crochet, weaker antecrochet and rudimentary crista. Protocone somewhat

pinehed off. Lower molars narrow and compressed. Mandibular symphysis narrow

(Heissig, 1972).

Known Distribution: Southern Asia, South-Eastern Asia and Western Asia (Heissig,

1972).

Brachypotherium perimense Falconer and Cautley, 1847

Cotypes: The specimens figured by Falconer and Cautley, (1847: pl. LXXV, figs. 13-16,

and LXXVI, figs. 14-17).

Type locality: Peram Island, India (Colbert, 1935).

Stratigraphic Range: Lower to Middle Siwaliks (Khan, A. M., 2010).

Diagnosis: Very large species of the genus Brachypotherium with relatively high cheek

teeth. All generic features are extremely developed. Nasals are shortened and hornless.

The upper molars have weak constrictions of the inner cusp; reduced antecrochet usually

129

present. Upper Premolars are molariform, usually with highly convex exterior. Lower

molars are almost without buccal fold; cingula usually reduced and short (Heissig, 1972).

Studied Specimens: Lower dentition: PUPC 07/52, a right mandibular ramus having P3 –

M3. PUPC 07/53, a left mandibular ramus having P3 – M2.

Description

Mandible

PUPC 07/52 and PUPC 07/53 is a well preserved right and left mandibular ramii

respectively (Figs. 29-30). The hemimandible PUPC 07/52 is moderately long having 390

mm length. PUPC 07/52 mandible depth at the P3 is 62 mm, at the P4 is 67 mm, at the M1

82 mm, at the M2 is 84 mm and at the M3 is 86 mm (Fig. 29). The horizontal ramii are

thick and their lower margins are slightly curved. The sagittal lingual groove is wide

posteriorly and absent anteriorly.

Lower Dentition

The molars are large and wide. All the cheek teeth are in early wear and show the distinct

morphology of rhinoceros; the valleys are open lingually (Figs. 29(1a), 30(1a)). The

enamel is fairly thick, uniform in thickness and rugose. The buccal and the lingual

cingula are absent. In P3 the paraconid has an anterior extension. The protoconid is a

round and has a buccal extension. The hypoconid is large and the entoconid is round. The

paralophid is long and the protolophid and the metalophid are broad. The hypolophid is

long. The anterior valley is not well developed. The posterior valley is U-shaped.

The paraconid is round in the P4. The metalophid is not distinct from the hypolophid at

the point of wear. The anterior valley is not well developed and V- shaped. The posterior

130

valley is deep and U-shaped. In M1 the paraconid is long and narrow but shorter than that

of the protoconid. The hypolophid is long and prominent but not distinct from the

metalophid at their junction due to wear. The paralophid is long and narrow than the

protolophid. The anterior and the posterior valleys are V – shaped. The damaged

paraconid is slightly round in the M2. The protoconid is long and broad. The metaconid is

short. The hypoconid is broad. The entoconid is long, little worn and have lingual

extension. The paralophid is short. The protolophid and the hypolophid are broad and

long. The anterior valley is slightly U-shaped. The posterior valley is deep and U-shaped.

The M3 hypolophid is long. The valleys are U-shaped.

PUPC 07/53 is a well-preserved left mandibular ramus with P3-M2 (Fig. 30). It is 280 mm

in length. All the cheek teeth are well preserved and show the distinct morphology. The

mandible is moderately long. The vertical height below P3 is 62 mm, below P4 is 66 mm,

below M1 is 82 mm, and below M2 is 85 mm. The P3 has anterior extension in the

paraconid. The protoconid is broad. The metaconid and the entoconid are round. The

hypoconid is wide. The metalophid and the protolophid are joined. The anterior valley is

not well developed. The posterior valley is V – shaped. The buccal and lingual cingula

are absent.

P4 has slightly broad paraconid. The entoconid, protoconid and metaconid are round but

the metaconid is not distinct from protoconid due to wear. The hypoconid is long while

the paralophid is short. The metaconid and the entoconid have lingual extension. The

protolophid is long. The metalophid is broad. The hypolophid is long and wide. The

protolophid and the hypolophid seem to be joining each other due to wear. The anterior

131

valley is not well developed while the posterior valley is U – shaped. There is V – shaped

trigonid.

The paraconid is short in M1. The protoconid is broad. The metaconid is not distinct due

to wear. The hypoconid is indistinct but long. The entoconid is pointed and has lingual

extension. The paralophid is narrow. The protolophid is long and broad. The hypolophid

is long and wide. The metalophid is broad. The posterior valley is U-shaped. The

paraconid is constricted in M2. The protoconid is broken but broad. The metaconid is

round. The hypoconid is pointed and the entoconid is round. The parolophid is short and

narrow. The protolophid is worn. The hypolophid is prominent and long. The anterior

valley is slightly V – shaped. The posterior valley is wide, deep, long and U – shaped.

The measurements are provided in table 11.

Table 11: Comparative dental measurements of the cheek teeth of Brachypotherium in

mm (millimeters). * The studied specimens. Referred data are taken from Colbert (1935),

Heissig (1972), and Cerdeño and Hussain (1997).

Taxon Number Nature/Position Length Width B. perimense PUPC 07/52* P3 41 31

P4 48 33 M1 56 32 M2 62 35 M3 55 35

PUPC 07/53* P3 40 31 P4 47 34 M1 56 33 M2 62 36

AMNH 19454 P3 40 26 P4 49 37 M1 53 36 M2 64 40 M3 72 35 M2 - 33 M2 55 30 M3 67 33

132

Figure 29: Brachypotherium perimense. 1. PUPC 07/52, a right mandibular ramus

having P3 – M3. a = occlusal view, b = lingual view, c = buccal view. Scale bar equals 50

mm.

133

Figure 30: Brachypotherium perimense. 2. PUPC 07/53, a left mandibular ramus having

P3 – M2. a = occlusal view, b = lingual view, c = buccal view. Scale bar equals 50 mm.

134

Comparison and Discussion

The studied lower dentition from the Nagri type area is identical to Brachypotherium

perminse in morphology to that described by Heissig (1972) from the Nagri Formation of

the Middle Siwaliks. PUPC 07/52 and PUPC 07/53 are comparable to AMNH 19454, as

identified of Aceratherium perimense (Colbert, 1935). PUPC 07/652 and PUPC 07/53 are

comparable to the sample described by Cerdeño and Hussain (1997) (Table 11; Fig. 31).

The dental measurements of PUPC 07/52 and PUPC 07/53 show that both specimens

probably belong to the same animal.

Colbert (1935) recognized Aceratherium permiense from the Lower and the Middle

Siwalik sediments while Heissig (1972) placed the species in Brachypotherium. Heissig

(1972) reported Brachypotherium in the Kamlial Formation of the Lower Siwaliks.

Cerdeño and Hussain (1997) described fossil remains of Brachypotherium permiense

from the Miocene Manchar Formation, Sind, Pakistan, whose morphology is similar to

those described by Heissig (1972) from the Siwlaiks of Pakistan; the P1 being wider, the

M2 narrower, and the lower teeth having closer dimensions. Other postcranial remains

from the Sind are smaller than those described by Heissig (1972), but this difference in

size may be due to older age of the Manchar Formation (Lower Chinji) with respect to

the latter specimens that belongs to the middle and upper Chinji, Nagri or Dhok Pathan

formations Cerdeño and Hussain, 1997). The difference of the dental remains size may

also be attributed to the age differences of animals. Gentry (1987) while describing the

Brachypotherium sp. from Miocene of Saudi Arabia considered the large size and

flatness of the buccal wall of upper molars and the small size of the paracone rib in

comparison with the large flat area, persistent internal cingula on its upper cheek teeth

135

and external cingula on its upper and lower molars as important characteristics for its

generic identity. Antoine et al. (2000) considered the European Brachypotherium

brachypus as an Asiatic migrant because closely related species have previous

occurrence in Pakistan and the surrounding areas. They (Antoine et al. 2000) suggested

that Aprotodon fatehjangense (Pilgrim, 1910) described from Asia has a very close

resemblance with Brachypotherium brachypus and must be regarded as recent synonym

of Brachypotherium Roger, 1904. Brachypotherium fatehjangense is senior synonym of

Aprotodon fatehjangense (Antoine and Welcomme, 2000).

The Brachypotherium is supposed to have a preference for soft diet and a more forested

environment (Andrew et al., 1996, 1997), which is comparable to the middle Miocene

Dhok Pathan Formation in the Siwlaiks. Brachypotherium perimense (Colbert 1935) is a

large species; lower teeth are charactersized by the smooth external groove, hardly

marked, as it is in other teleocertines (Cerdeño and Hussain, 1997). Heissig (2003)

indicated that Brachypotherium perimense is the most frequent species in times of

transition and rare during most humid and most arid times and this species point out in

the Nagri formation the beginnings of less humid conditions. Brachypotherium has often

been compared to hippos, and was certainly a marsh or lake dweller (Geraads and Sarac,

2003).

136

B. perimense

0

10

20

30

40

50

0 10 20 30 40 50 60 70 80 90Length

Wid

thLower Third Premolar Lower Fourth Premolar Lower First Molar

Lower Second Molar Lower Third Molar

Figure 31: Scatter diagram showing dental proportions of B. perimense’s studied sample.

Referred data are taken from Colbert (1935), Heissig (1972), and Cerdeño and Hussain

(1997).

137

DISCUSSION

Faunal Correlation

The fauna from the Nagri includes at present the following species: Listriodon

pentapotamiae, Selenoportax cf. vexillarius, Pachyportax cf. latidens, Tragoportax

punjabicus, Miotragocerus cf. gluten, Gazella cf. lydekkeri, Giraffokeryx punjabiensis,

Giraffa cf. priscilla, Dorcatherium cf. minus, Dorcatherium cf. majus, Dorcabune cf.

anthracotherioides, Hipparion theobaldi, Brachypotherium perimense. The thirteen taxa

of the true ungulates have been recognized from the Late Miocene deposits exposed in

the Nagri type locality of the Nagri Formation, northern Pakistan. This faunal list may be

compared with that of other Late Miocene Siwalik localities, Dhok Pathan and Hasnot

and the middle Miocene Siwalik locality, Chinji. This list obviously indicates a faunal

spectrum of only orders Artiodactyla and Perissodactyla and several orders such as

Proboscidea, Carnivora are not included in this study. However, well documented Late

Miocene mammalian faunal successions are known from the Siwaliks (Lydekker, 1876,

1878; Pilgrim, 1937, 1939; Matthew, 1929; Colbert, 1935; Thomas, 1984; Akhtar, 1992,

1996; Bhatti, 2005; Farooq, 2006; Khan, 2007, 2008; Khan, A. M., 2010). Among the

common taxa, the Nagri Dorcatherium seems more primitive than the Dhok Pathan one

because of its heavy styles and more bunodont teeth. The Nagri Hipparion also shows a

simplification in plicaballins. The Nagri stratigraphically is situated below the Dhok

Pathan Formation and the list is characteristic of the early Late Miocene in the

chronologic successions of faunas in the Siwaliks (Barry et al., 2002).

138

This faunal list is biased towards large mammals, because many fossils were recovered

during quarry work, and excavations have been of very limited extent. Still, this faunal

association contains enough significant elements to allow comparison with some other

upper Miocene faunas from Europe and the Near Siwaliks. Although the latest Miocene

(7.5 – 5.3 Ma) faunas of the Siwaliks are well known (Pilgrim, 1937, 1939; Akhtar, 1992;

Khan, 2007, 2008; Bibi, 2007; Khan et al., 2009a; Khan, A. M., 2010), the early Late

Miocene (11 – 10 Ma) fauna are poorly known in the Siwaliks. However, the Nagri fauna

is one of the best representatives for this age.

The Nagri fauna indicates strong resemblance to the late Miocene fauna of the Dhok

Pathan and the Hasnot (Akhtar, 1992; Khan, 2007) and the early Late Miocene faunas of

East Africa (Pickford, 1981). However, the Nagri fauna indicates weak resemblance to

the Late middle Miocene of the Chinji Formation (Thomas, 1984). The Nagri fauna is

similar to the faunas from Samos and Pikermi of Greece, Maragheh of Iran, and Dhok

Pathan Formation of the Siwaliks (Solounias, 1981; Thomas, 1984; Akhtar, 1992; Khan,

2007; Kostopoulos, 2009). This fauna is particularly significant in providing evidence

regarding the Late Miocene faunal interchange between African and Eurasia (Pickford,

1981; Thomas, 1984; Pilbeam et al., 1997; Barry et al., 2002; Bibi and Gulec, 2008;

Kostopoulos, 2009). The occurrence of several bovids (Selenoportax cf. vexillarius,

Pachyportax cf. latidens, Tragoportax punjabicus, Miotragocerus cf. gluten, Gazella cf.

lydekkeri), tragulids (Dorcatherium cf. minus, Dorcatherium cf. majus, Dorcabune cf.

anthracotherioides), hipparionines (Hipparion theobaldi) and rhinos (Brachypotherium

perimense) was also mentioned in the Late Miocene of the Siwalik, Greco-Iranian

province and Eurasia (Sen et al., 1997; Bibi et al., 2009; Kostopoulos, 2009). The Nagri

139

fauna is different from that of the Dhok Pathan and the Hasnot by not having cervids,

Bramatherium, and large suids.

Gazella is abundantly recorded from Samos, Pikermi, Maragheh, Dhok Pathan and

Hasnot (Pilgrim, 1939; Akhtar, 1992; Khan, 2007; Bibi and Gulec, 2008; Kostopoulos,

2009). The Siwalik Gazella lydekkeri most closely approximates the morphology and

perhaps the evolutionary stage of the Eurasian Gazella capricornis based on primarily on

the degree of premolar row reduction (Bibi and Gulec, 2008). Gazella lydekkeri is known

primarily from the Nagri Formation and the known age range of Gazella lydekkeri upto

the Mio-Pliocene boundary or beyond (Thomas, 1984). Selenoportax vexillarius and

Pachyportax latidens are recorded from the late Miocene and Pliocene faunas at the

Siwaliks, the main bone beds of the Dhok Pathan and the Hasnot (Khan et al., 2009a).

The Dhok Pathan and the Hasnot faunas are contemporaneous (10.1-3.4 Ma) but the

Nagri faunas are older (ca 11.2-10.1 Ma). An age range for these species that includes all

these localities would be quite broad, potentially as old as the base of the Nagri up until

3.4 Ma, the upper limit to the Dhok Pathan Formation (Barry et al., 2002). Tragoportax is

present in the Nagri Formation and also in the Dhok Pathan Formation (Pilgrim, 1937,

1939; Bibi et al., 2009). Tragoportax is recorded also from the late Miocene of Pikermi,

Molayan, Samos and Mytilini (Bibi and Gulec, 2008; Bibi et al., 2009).

Listriodon pentapotamiae have been found in the Siwalik localities ranging in age from

Middle Miocene to early Late Miocene (Pickford, 1988; Van der Made, 1996; in this

thesis). In Europe a few specimens have been found in sites which have also yielded the

equid Hipparion as in the Nagri type section but by the end of MN9 listriodonts were

extinct everywhere. It is noted that listriodonts became extinct in Europe, China, India

140

and Africa over a short period at the end of the Middle Miocene and the beginning of the

Late Miocene. Thus, the available evidence about the broad tendencies of listriodont

evolution in the Indian subcontinent accords with that from Europe and China, which

supports the view that for much of the middle Miocene and the early Late Miocene,

Europe, Asia and the Indian subcontinent were all part of a single biogeographic region

(Pickford and Morales, 2003).

In summary then, the Nagri faunas are similar to those from Dhok Pathan, Pikermi,

Samos, Maragheh and Sivas, though also some elements to older site like Chinji

(Pickford, 1981; Thomas, 1984; Akhtar, 1992; Pilbeam et al., 1997; Sen et al., 1997;

Barry et al., 2002; Bibi and Gulec, 2008; Bibi et al., 2009; Kostopoulos, 2009).

Biostratigraphy

The stratigraphical ranges of the bovids of the Nagri are restricted to Late Miocene (Bibi,

2007; Khan, 2007; Khan et al., 2009a). The suid Listriodon pentapotamiae is restricted to

the earliest Late Miocene of the Siwaliks (Pickford, 1988; Van der Made, 1996). More

recently, the cervids have been recorded in the latest Late Miocene of the Siwaliks: Dhok

Pathan and Hasnot (Ghaffar, 2005). Prior to it the first appearance of the cervids are

recorded from the late Pliocene of the Siwaliks (Barry et al., 2002). The cervids are

absent in the earliest Late Miocene of the Siwaliks. The presence of Tragoportax

confirms Late Miocene age but allows no further refinement. Nevertheless,

Miotragocerus indicates a more primitive bovid assemblage.

Listriodon pentapotamiae is recorded in the Siwaliks from as early as the middle

Miocene until the earliest Late Miocene (Welcomme et al., 2001). In the Chinji strata

141

Listriodon pentapotamiae is the most common suid (Pickford, 1988). Several Middle

Siwalik localities of the Nagri Formation exhibit the presence of Listriodon

pentapotamiae and however, the species is widely distributed in the Middle Miocene

(Pickford, 1988; Pickford and Morales, 2003). Brachyopotherium perimense is a

common species of the Siwalik Late Miocene (Heissig, 1972) and its stratigraphic range

is the earliest Late Miocene to the latest Late Miocene (Heissig, 1972).

The Palaeotraginae is documented sporadically in the Chinji Formation of the Lower

Siwaliks (Colbert, 1933; Bhatti et al., 2007). Giraffokeryx punjabiensis has already been

mentioned several localities of the Late Middle Miocene age (Bhatti, 2005), occupying a

wide territory from Western Europe to India (Bohlin, 1926; Gentry et al., 1999; NOW

database, 2003). A rare occurrence of the species is found in the early Late Miocene and

now from the Nagri type section confirms its presence in the early Late Miocene. The

persistence of Giraffokeryx punjabiensis and Listriodon pentapotamiae in the Nagri type

section seems to contradict their time range in the Siwaliks, where they are considered to

be typical Lower Siwalik elements. Giraffokeryx punjabiensis and Listriodon

pentapotamiae are here rather in favour of an early Late Miocene age. The Nagri

specimens attributed to Dorcatherium and Dorcobune lack any stratigraphic indication.

The specimens have the typical fossilization status seen in specimens from the Nagri

ravine, suggesting an early Late Miocene age.

Accordingly, an estimated age of early Late Miocene age is reasonable for the Nagri

fauna, equating to somewhere between about 11 to 10.1 Ma. This estimate, based solely

on the true ungulates, is in agreement with previous estimates of 11.2-10.1 Ma for the

Nagri Formation sites of the Siwaliks using radiometric dating data (Pilbeam et al., 1997;

142

Barry et al., 1980, 1982, 2002). Pilbeam et al. (1997) gave the chronology of the top

Ghabir sites of the type section and suggested 10.75 and 10.84 Ma for this section. Barry

et al. (2002) suggested early Late Miocene age for the type section. Their arguments were

mainly magnetostratigraphic (using ages from the Berggren et al., 1985, and Cande and

Kent, 1995 time scale ) and also palaeontological (to get possible ages for the Siwalik

hipparionines). The fossiliferous level of the Nagri type section was radiometrically

bracketed between 10.1 and 11.2 Ma (Pilbeam et al., 1997; Barry et al., 2002) and should

consequently be correlated to the earliest Vallesian, MN9 (Bernor et al., 1996).

In general, the Nagri true ungulates are strongly indicative of the Middle Siwaliks, which

have a combined absolute range of 10.1-11.2 Ma (Barry et al., 2002) and relate well to

other Late Miocene Eurasian faunas. On the whole, the faunas suggest that the Nagri is

older than the Dhok Pathan and the Hasnot. The biostratigraphical position of the Nagri

locality relative to the Dhok Pathan and the differences in appearance and disappearance

of fauna suggest that the Nagri fauna is of early Late Miocene age.

Paleoecology and Paleoenvironment

Recent work on the geology of the Nagri (Barry et al., 2002) has documented that the

deposits were formed as part of a large river system by coexisting emergent and interfan

river system, with the larger emergent Nagri system beginning at 10.1 Ma (Willis and

Behrensmeyer, 1995). Mammal fossils at the Nagri are recovered from fluvial channel

deposits. The interpretation of the Nagri paleoenvironment based on geological

information is supported by paleontological evidence. The Nagri fauna has an aquatic

component that includes a mix of freshwater animals that can tolerate slow-moving water

143

with a high sediment contents (e.g. crocodiles, turtles). The land mammal fauna is

dominated by a number of artiodactyl and perissodactyl species that were likely preferred

an aquatic-margin habitat, as they are found most commonly in fluviatile deposits (Barry

et al., 2002). Recovery of a fair number of well preserved terrestrial vertebrates supports

previous interpretations of the Nagri paleoenvironment, they appear to be preferentially

preserved in swampy and fluviatile settings. As for paleoenvironmental indications, the

Nagri faunas are consistently found in deposits that accumulated in or near swamps and

shallow lakes (Badgley and Behrensmeyer, 1980; Badgley, 1986; Badgley and Tauxe,

1990; Barry et al., 1980, 1982, 2002) and suggest the presence of forest in the Nagri

region at the time of deposition.

The presence of the tragulids and the giraffids, which were certainly browsers, definitely

speaks in favour of wet forested environments. Hipparion fauna suggests a

sclerophyllous evergreen woodland environment, similar to today’s mixed monsoon

forest and grassland glades of north central India (Solounias, 1999). Fortelius et al.

(1996) studied the body mass diversity of the West Eurasian suoids in relationship to

environmental conditions, and suggested that the loss of species of small size was

correlated to a progressive development of increasingly open and seasonal habitats.

During the Late Miocene, suid diversity in western Eurasia was very low, and suids in

MN12 and MN13 are only recorded in the largest size class (201-1000 kg). Microstonyx

major has been recorded from the type section (Van der Made and Hussain, 1989). The

presence of Microstonyx major and Listriodon pentapotamiae in the type area of the

Nagri Formation confirm that the environment appears to have been more humid than the

latest Late Miocene (Pickford, 1988; Van der Made and Hussain, 1989; Pickford et al.,

144

2004). It should be noted that the Nagri has some hypsodont members (Thomas, 1977;

Akhtar, 1992). This provides evidence for an open environment.

Paleoenvironmental interpretations using vertebrate fossils are often based largely on

paleodietary reconstructions using dento-gnathic and postcranial functional analyses in

conjunction with the relative abundances of ecologically differentiated taxa (Bibi and

Gulec, 2008). Tragoportax from the sites of Pikermi and Samos (Solounias and Hayek,

1993) as well as Molayan (Merceron et al., 2004) concluded that Tragoportax species

were variable mixed feeders with strong grazing habits. An analysis of maxillary and

zygomatic morphology among living and fossil bovids indicated that Tragoportax were

mixed feeders while Gazella was a mixed feeder or browser (Solounias et al., 1995).

More balanced mixed-feeding was interpreted for microwear of Gazella (Merceron et al.,

2004). A comprehensive study of the postcranial morphology of extant and extinct bovids

by Köhler (1993) found similarities between the build and proportions of Tragoportax

amalthea and certain deer (Cervus) and found a leafy diet and light woodland habitat as

most likely for this species. Köhler also reconstructed Gazella as a browser inhabiting

more open country. Miotragocerus was interpreted as a leaf and herb-eater inhabiting

shrubland to light woodland.

In terms of number of specimens, the Nagri true ungulate assemblage is dominated by

bovids, giraffids and tragulids with Hipparion and Listriodon, Brachypotherium

constituting much smaller percentages (Appendix 1). The high proportion of

boselaphines may be representative of the presence of drier and more open habitats than

would be expected for the remaining bovids, particularly if the fossil gazelle resembled

the living species in their ability to inhabit semi-arid to arid environments. The specific

145

richness of the boselaphins in the Nagri type section suggests open areas biomass

(Thomas, 1984) as the living boselaphins (Boselaphus tragocamelus and Tetracerus

quadricornis) prefer open areas.

Dorcatherium and Miotragocerus are familiar for more or less closed and humid habitats

(Kohler, 1993; Gentry, 2005; Eronen and Rössner, 2007). This supports the assumption

of an earliest Late Miocene Siwalik humid habitat with abundant cover. The taxonomic

faunal composition suggests a humid habitat pocket with abundant cover indicating the

dominance of forested landscapes during the early Late Miocene times of the Siwaliks.

The presence of Dorcatherium in the type section can be assumed a strong attachment to

wet, forested habitats with dense understory, where the animals could hide in vegetation

or water from predators (Rössner, 2010).

In summary, the dominance of boselaphins in the Nagri reflects an environment of

mainly open forest. A significant representation of tragulids with adaptations to ecotonal

wet and swampy habitats indicates humid conditions. The alternation of dry and flood

seasons would have caused a highly differentiated mosaic ecotone environment, which

would have offered an outstanding number of habitats and niches and consequently an

exceptionally large number of species. Because of alternation of ground conditions

seasonal migration events might have occurred. Finally, the Nagri fauna suggest the

existence of a vast wetland environment with alternating dry and flood seasons which

forced a mosaic of ecotonal habitats with many niches and corresponding adaptations.

146

CONCLUSIONS

The study of the new true ungulate material collected from the Nagri type area, the

Middle Siwaliks, Pakistan, allows recognizing 13 species from fifteen fossil sites, SN1-

15. This includes a species of Listriodon, a species of Hipparion, a species of

Selenoportax, a species of Pachyportax, a great abundance of the bovid Tragoportax, a

Gazella, a Miotragocerus, a species of Giraffokeryx, and only rare representation of

Dorcatherium, Dorcabune and Giraffa. The early Late Miocene fossil ungulates from the

Nagri constitute an assemblage that is fairly typical of the Siwalik region during this

time.

From biochronological point of view, the co-existence of Listriodon pentapotamiae,

Hipparion theolbaldi, Brachypotherium perimense, Selenoportax and Pachyportax

indicate close relationships to the early Late Miocene of the Siwaliks. The true ungulate

fauna allows a biochronological estimate of early Late Miocene for the Nagri assemblage

based on similarities with faunas from Dhok Pathan, Hasnot, Maragheh, Pikermi and

Samos, corresponding well to previous estimates (Thomas, 1984; Barry et al., 2002; Bibi

and Gulec, 2008; Kostopoulos, 2009; Khan et al., 2009a). In conclusion, the true

ungulate assemblage of the Nagri type section rather indicates an early Late Miocene age

(earliest Vallesian, MN 9) which is in agreement with the radiometric dating data

provided by Barry et al. (2002).

The fossil ungulates indicate the ancient environment at Nagri comprised woodland to

shrubland. The type area may have comprised relatively more humid, closed habitats

based on the ecomorphology of their taxa, particularly Miotragocerus and Dorcatherium,

which may have been an inhabitant of wetlands (Köhler, 1993). The Nagri occurrence

increases again the similarities between Indo-Siwalik faunas and those of the Greco-

Iranian province, the early Late Miocene ungulates common to both realms.

147

REFERENCES

ACHARYYA, S. K., 1994. The Cenozoic Foreland basin and tectonics of the Eastern

Sub-Himalaya: Problems and prospects. In Kumar, R. and Ghosh S. K. (eds.)

Siwalik Foreland Basin of Himalaya. Himal. Geol., 15: 1-415.

AKHTAR, M., 1992. Taxonomy and Distribution of the Siwalik Bovids. Ph. D. Diss.,

University of the Punjab, Lahore, Pakistan.

AKHTAR, M., 1995. Pachyportax giganteus, new species (Mammalia, Artiodactyla,

Bovidae) from the Dhok Pathan, district Chakwal, Punjab, Pakistan. Pak. J.

Zool., 27 (4): 337-340.

AKHTAR, M., 1996. A new species of the genus Selenoportax (Mammalia, Artiodactyla,

Bovidae) from Dhok Pathan, district Chakwal, Punjab, Pakistan. Proc. Pak.

Cong. Zool., 16: 91-96.

AKHTAR, M., HAMEED, Z.B., AMIN, M. AND NAZIR, M., 1997. An evidence on the

validity of the species Pachyportax nagrii Pilgrim (mammalia, Artiodactyla,

Bovidae). Pak. J. Geol., 4: 1-3.

ANDERSON, R. V., 1927. Tertiary stratigraphy and orogeny of the Northern Punjab.

Geol. Soc. Am. Bull., 38: 665-720.

ANDREWS, P. AND CRONIN, J. E., 1982. The relationships of Sivapithecus and

Ramapithecus and the evolution of the orangutan, Nat., 297: 541-546.

ANDREWS, P., HARRISON, T., DELSON, E., BERNOR, R. L. AND MARTIN, L.,

1996. Distribution and biochronology of European and southwest Asian

Miocene catarrhines. In (R.L. Bernor, V. Fahlbusch and H.W. Mittman, Eds),

The Evoluton of Western Eurasian Neogene Mammal Founas, New York:

Columbia University Press: pp. 168-207.

ANDREWS, P., LORD, J. AND EVANS, E. M., 1997. Patterns of ecological diversity in

fossil and modern mammalian faunas. Biol. J. Linn. Soc., 11: 177-205.

ANTOINE, P. O. AND WELCOMME, J. L., 2000. A new Rhinoceros from the Lower

Miocene of the Bugti Hills, Baluchistan, Pakistan: The earliest

Elasmotheriine. Palaeont., 43(5): 795-816.

148

ANTOINE, P. O., BULOT, C. AND GINSBURG, L., 2000. Une faune rare de

rhinocérotidés (Mammalia, Perissodactyla) dans leMiocène inférieur de

Pellecahus (Gers, France), Geob., 33: 249–255.

ARAMBOURG, C. AND PIVETEAU, J., 1929. Dorcatherium puyhauberti sp. N.

Pontian nr. Salonica. Ann. Paleont., 18(2-3): 34.

ARAMBOURG, C., 1933. Mammiferes Miocenes du Tarkana (Afrique Orientale). Ann.

Paleont. Paris, 22: 121-148.

BADGLEY, C. AND GINGERICH, P. D., 1988. Sampling and faunal turnover in early

Eocene manmals. Palaeog., Palaeocl., Palaeoec., 63: 141-157.

BADGLEY, C. E, AND L. TAUXE., 1990. Paleomagnetic stratigraphy and time in

sediments: studies in alluvial Siwalik rocks of Pakistan. J. Geol., 98: 457-477.

BADGLEY, C. E. AND BEHRENSMEYER, A. K., 1980. Paleoecology of Middle

Siwalik sediments and faunas, northern Pakistan. Palaeog., Palaeocl.,

Palaeoec., 30: 133-155.

BADGLEY, C. E., BARTELS, W. S., MORGAN, M. E., BEHRENSMEYER, A. K.

AND RAZA, S. M., 1995. Taphonomy of vertebrate assemblages from the

Paleogene of the northwestern Wyoming and the Neogene of the northern

Pakistan. Palaeog., Palaeocl., Palaeoec., 115: 157-180.

BADGLEY, C., 1986. Taphonomy of mammalian fossil remains from Siwalik rocks of

Pakistan. Paleob., 12: 119-142.

BADGLEY, C., WILL, D. AND LAWRENCE, F., 2008. Taphonomy of Small-Mammal

Fossil Assemblages from the Middle Miocene Chinji Formation, Siwalik

Group, Pakistan. Nat. Sc. Mus. monographs, 14: 145-166.

BAGATI, T. N. AND KUMAR, R., 1994. Clay mineralogy of Middle Siwalik sequence

in Mohand area. Dhera Dun: implication for climate and source area. In:

Kumar, R., Ghosh, S.K., Phadtare, N.R. (Eds.), Siwalik Foreland Basin of

Himalaya. Himal. Geol., 15: 219-228.

BARNDT, J., 1977. The magnetic polarity stratigraphy of the type locality of the Dhok

Pathan Faunal Stage, Potwar Plateau, Pakistan. Master Diss. Dartmouth

College Hanover, N.H.

149

BARRY, J. C. AND FLYNN, L. J., 1989, Key biostratigrapic events in the Siwalik

sequence – In: Lindsay, E. H., Fahlbusch V. & Mein, P. (eds.): European

Neogene Mammal Chronology. NATO ASI Series, (A) 180: 557-571; New

York, Plenum.

BARRY, J. C. MORGAN, M. E., FLYNN, L. J., PILBEAM, D., BEHRENSMEYER, A.

K., MAHMOOD, S. R., KHAN, I. A., BADGLEY, C., HICKS, J. AND

KELLEY, J., 2002. Faunal and environmental change in the late Miocene

Siwaliks of northern Pakistan. Paleob., 28 (2): 1-71.

BARRY, J. C., BEHRENSMEYER, A. K. AND MORGAN, M., 1980. A geologic and

biostratigraphic framework for Miocene sediments near Khaur Village,

northern Pakistan. Postil., 183: 1-19.

BARRY, J. C., JHONSON, N. M., RAZA, S. M. AND JACOBS, L. L., 1985. Neogene

mammalian faunal change in southern Asia: correlations with climatic,

tectonic and eustatic events. Geol., 13: 637-640.

BARRY, J. C., LINDSAY, E. H. AND JACOBS, L. L., 1982. A biostratigraphic

zonation of the Middle and Upper Siwaliks of the Potwar Platue of northern

Pakistan. Palaeog., Palaeocl., Palaeoec., 37: 95-130.

BARRY, J. C., MORGAN, M. E., FLYNN, L. J. PILBEAM, D., JACOBS, L. L.,

LINDSAY, E. H., RAZA, S. M. AND SOLOUNIAZ, N., 1995. Patterns of

faunal turnover and diversity in the Neogene Siwaliks of northern Pakistan.

Palaeog., Palaeocl., Palaeoec., 115: 209-226.

BARRY, J. C., MORGAN, M. E., WRINKLER, A. J., FLYNN, L. J, LINDSAY, L. H.,

JACOBS, L. L. AND PILBEAM, D., 1991. Faunal interchange and Miocene

terrestrial vertebrates of southern Asia. Paleob., 17: 231-245.

BARRY, J. C. AND FLYNN, L. J., 1990. Key biostratigraphic events in the Siwalik

Sequence. In: European Neogene Mamma Chronology (ed. E.H. Lindsay et

al.), pp. 557-571. New York.

BARRY. J. C., COTE, S., MACLATCHY, L., LINDSAY, E. H., KITYO, R. AND

RAJPAR, A. R., 2005. Oligocene and Early Miocene Ruminants (Mammalia,

Artiodactyla) from Pakistan and Uganda. Palaeont. Electr., 8 (1): 1- 29:

885MB.

150

BASU, P. K., 2004. Siwalik mammals of the Jammu Sub-Himalaya, India: an appraisal

of their diversity and habitas. Quater. Int., 117: 105-118.

BECK, R. A. AND BURBANK, D. W., 1990. Continental – scale diversion of rivers: a

control of alluvial stratigraphy. Geological Society of America; Abstract with

Programs, 22:238.

BEHRENSMEYER, A. K, WILLIS, B. J. AND QUADE, J., 1995. Floodplains and

paleosols of Pakistan Neogene and Wyoming, Paleogene deposits: a

comparative study. Palaeog., Palaeocl., Palaeoec., 115: 37-60.

BEHRENSMEYER, A. K. AND TAUXE, L., 1982. Isochronous fluvial systems in

Miocene deposits of northern Pakistan. Sediment., 29, 331-352.

BEHRENSMEYER, A. K., 1987. Miocene fluvial facies and vertebrate and vertebrate

taphhonomy in northern Pakistan. In recent developments in fluvial

sedimentology (ed. F. G. Ethridge. R. M Flores, and M. D. Harvey). Society

of Economic Paleontologists and Mineralogists. Spec. Pub., 39: 169-176.

BEHRENSMEYER, A. K., 1988. Vertebrate preservation in fluvial channels. Palaeog.,

Palaeocl., Palaeoec., 63: 183-199.

BEHRENSMEYER, A. AND BARRY, J., 2005. Biostratigraphic Surveys in the Siwaliks

of Pakistan. A Method for Standardized Surface Sampling of the Vertebrate

Fossil Record. Palaeont. Electr., 8 (1): 1-24.

BERGGREN, W. A., KENT, D. V., FLYNN, J. J. AND VAN COUVERING, J. A.,

1985. Cenozoic geochronology. Geol. Soc. Am. Bull., 96: 1407-1418.

BERNOR R. L., SOLOUNIAS, N., SWISHER III C. C. AND VAN COUVERING J. A.

1996. The correlation of three classical Pikermian mammal faunas. Maragheh,

Samos and Pikermi-with the European MN unit system. In BERNOR R. L.,

FAHLBUSCH V. and MITTMANN H. W. (eds): The Evolution of Western

Eurasian Neogene Mammal Faunas. Columbia University Press: 137-156.

BERNOR, R. L., 1984. A zoogeographic theater and biochronologic play: the time

biofacies phenomena of Eurasian and African Miocne mammal provinces.

Paleobio. Continent., 14: 121-142.

151

BERNOR, R. L., 1986. Mammalian biostrtigraphy, geochronology, and zoogeographic

relationships of the Late Miocene Maragheh fauna, Iran. J Vert. Palaeont., 6:

76-95.

BERNOR, R. L., SCOTT, R. S., FORTELIUS, M., KAPPELMAN. J. AND SEN, S.,

2003. Systematics and Evolution of the Late Miocene Hipparions from Sinap,

Turkey. In: Fortelius, M., Kappelman, J., Sen, S. and Bernor, R. L. (Eds). The

Geology and Paleontology of the Miocene Sinap Formation, Turkey: New

York (Columbia University Press).

BERNOR, R. L. AND HUSSAIN, S. T., 1985. An assessment of the systematics

phylogenetic and biogeographic relationships of Siwalik Hipparionine Horses.

J. V. P., 5(1): 32-87.

BHATTACHARYA, N. AND MISRA, S. S., 1963. Petrology and sedimentation of

Middle Siwalik clays at Dhokhand, Saharanpur district U.P. Idia. Beit. Zur

Miner. Und Petgr., 9: 139-147.

BHATTACHARYA, N., 1970. Clay mineralogy and trace element geochemistry of

Subathu, Dharamsala and Siwalik sediments in Himalayan foothills of

northwest India. J. Geol. Soc. India, 11(4): 309-332.

BHATTI, Z. H., 2005. Taxonomy, Evolutionary History & Biogeography of the Siwalik

Giraffids. Ph. D. Diss., University of the Punjab, Pakistan.

BHATTI, Z. H., QURESHI, M. A., KHAN, M. A., AKHTAR, M., GHAFFAR, A. AND

EJAZ, M. 2007. Individual variations in some premolars of species

Giraffokeryx punjabiensis (Mammalia, Giraffidae) from Lower Siwalik

(Chinji Formation) of Pakistan. Contribution to Geology of Pakistan; Proc.

Pak. Geol. Cong. 2004, 5: 261-272.

BIBI, F. AND GULEC, E. S., 2008. Bovidae (Mammalia: Artiodactyla) from the Late

Miocene of Sivas, Turkey. J. V.P., 28(2): 501-519.

BIBI, F., 2007. Origin, paleoecology and paleobiogeography of early Bovini. Palaeog.,

Palaeocl., Palaeoec., 248(1-2): 60-72.

BIBI, F., BUKHSIANIDZE, M., GENTRY, A. W., GERAADS, D., KOSTOPOULOS,

D. S. AND VRBA, E. S., 2009. The Fossil Record and Evolution of Bovidae:

State of the Field. Palaeont. Elect., 12(3): 1-11.152

BISWAS, S. K., 1994. Status of exploration for hydrocarbons in Siwalik Basin of India

and future trends. In Symposium on Siwalik Bsin, 1991. Geol. Soc. India,

Dehra: 283-300.

BLAINVILLE, H. M., 1816. Sur plusieurs espèces d animaux mammifères de 1’ order

des Ruminants. Bull. Soc. Phil. Paris, 73-82.

BOHLIN, B., 1926. Die familie Giraffidae. Palaeotologia Sinica Pekin. C4(I): 1-78.

BURBANK, D. W., DERRY, L. A. AND FRANCE, L. C., 1993. Reduced Himalayan

sediment production 8 Myrago despite an intensified monsoon. Nature, 364:

48-50.

CANDE, S. C. AND KENT, D. V., 1995. Revised calibration of the geomagnetic polarity

time scale for the Cretaceous and Cenozoic. J. Geoph.l Res., B100: 6093-

6095.

CERDEÑO, E. AND HUSSAIN, T., 1997. On the Rhinocerotidae from the Miocene

Manchar Formation, Sind, Pakistan. Pak. J. Zool., 29(3):263-276.

CERDEÑO, E., 1995. Cladistic analysis of the Family Rhinocerotidae (Perissodactyla).

Am. Mus. Novit., 3143: 1-25.

CERLING, T. E., WANG, Y., AND QUADE, J., 1993. Expansion of C, ecosystems as an

indicator of global ecological change in the late Miocene. Natur., 361: 344-

345.

CHAMELY, H., 1989. Clay Sedimentology. Springer Verlag, New York, pp. 623.

CHAUDHRI, R. S. AND GILL. G. T. S., 1983. Clay mineralogy of the Siwalik Group of

Simla Hills, northwestern Himalaya. . Geol. Soc. India, 24: 159-165.

CHAUHAN, P. R., 2003. An overview of the Siwalik Acheulian & Recosidering its

chronological relationship with the Soanian. A theoretical perspective. The

Sheffield graduate J. Arch., 7: 1-17.

CHEEMA, M. R., RAZA, S. M. AND AHMAD, H., 1977. Cainozoic, in Shah, S. M. I.

(ed).”Stratigraphy of Pakistan” Geol. Sur. Pak. Memoirs, 12: 56-98.

CHRISTOL, J. DE, 1832. [No title, Description of Hipparion]. Annales des Science et de

I’ Industrie du Midi de la France, 1: 180-181.

COLBERT, E. H., 1933. A skull and mandible of Giraffokeryx punjabiensis (Pilgrim).

Am. Mus. Novit., 632: 1–14.

153

COLBERT, E. H., 1935. Siwalik mammals in the American Museum of Natural History.

Trans. Am. Phil. Soc., N. S., 26: 1-401.

COTTER, G. P., 1933. The geology of the part of the Attock district, West of Longitude

72° 45; India Geol. Surv. Mem., 55: 63-161.

DANILCHIK, W. AND SHAH, S. M. L, 1967. Stratigraphic nomenclature of formation

in Trans-Indus mountains Mianwali distt. West Pakistnan, U. S. Geol. Surv.

Proj. Report (F1), PK: 33-45.

DENNELL, R. W., COARD, R., TURNER, A., 2008. Predators and Scavengers in Early

Pleistocene southern Asia. Quater. Int., 192: 78-88.

DENNELL, R., COARD, R. AND TURNER, A. 2006: The biostratigraphy and magnetic

polarity zonation of the Pabbi Hills, northern Pakistan: An Upper Siwalik

(Pinjor Stage) Upper Pliocene-Lower Pleistocene fluvial sequence. Palaeog.,

Palaeocl., Palaeoec., 234: 168-185.

EINSELE, G., RATSCHBACHER, L. AND WETZEL, A., 1996. The Himalaya Bengal

fan denudation-accumulation system during the past 20 Ma. J. Geol., 104:

163-184.

ERONEN, J. T. AND RÖSSNER, G. E., 2007. Wetland paradise lost: Miocene

community dynamics in large herbivorous mammals from the German

Molasse Basin. Evolut. Ecol. Res., 9: 471– 494.

FALCONER, H., 1868. Palaentological memoirs. Eds by R.I. Murchison, Lond., 1: 157-

169.

FAROOQ, U., 2006. Studies of Evolutionary Trends in Dentition of the Siwalik

Tragulids. Ph. D. Diss., University of the Punjab, Pakistan.

FAROOQ, U., KHAN, M. A., AKHTAR, M., KHAN, A. M., 2008. Lower Dentition of

Dorcatherium majus (Tragulidae, Mammalia) in the Lower and Middle

Siwaliks (Miocene) of Pakistan. Turk. J. Zool., 32: 91-98.

FAROOQ, U., KHAN, M. A. AND AKHTAR, M., 2007a. Dorcabune nagrii

(Ruminantia, Tragulidae) from the Upper part of the Middle Siwaliks. J. Appl.

Sc., 7(10):1428-1431.

154

FAROOQ, U., KHAN, M. A., AKHTAR, M. AND KHAN, A. M., 2007b. Dorcatherium

majus, a study of upper dentition from the Lower and Middle Siwaliks of

Pakistan. J. Appl. Sc., 7(9): 1299-1303.

FATMI, A. N., 1973. Lithostratigraphic units of the Kohat Potwar Province, Indus Basin,

Pakistan. Pak. Geol. Surv. Mem., 10: 80.

FLYNN, L. J., 1986. Species longevity, stasis, and stairsteps in rhizomyid rodents, in

Flanagan, K. M. and Lillegraven, J. A. (ed.), “Vertebrates, Phylogeny, and

Philosophy.” Contr. Geol., Special Paper, 3: 273-285.

FLYNN, L. J., 2003. Small mammal indicators of forest paleo-environment in the

Siwalik deposits of the Potwar Plateau, Pakistan, Deins., 10: 183-196.

FLYNN, L., J., BARRY, J. C., MORGAN, M. E., PIBEAM, D., JACOBS, L. L. AND

LINDSAY, E. H., 1995. Neogene Siwalik Mammalian Lineages: Species

longevities, rates of change, and modes of speciation. Palaeog., Palaeocl.,

Palaeoec., 115: 249-264.

FLYNN. L. J., PILBEAM, D., JACOBS, L. L., BARRY, J. C., BEHRENSMEYER, A.

K., AND KAPPELMAN, J. W., 1990. The Siwaliks of Pakistan: Time and

faunas in a Miocene terrestrial setting. J. Geol., 98: 589-604.

FORSTEN, A. M., 1968. Revision of the Palearctic Hipparion. Act. Zool. Fennico, 119:

1-134.

FORTELIUS, M., MADE VAN DER, J., BERNOR, R., 1996. Middle and Late Miocene

Suoidea of central Europe and the eastern Mediterranean: evolution,

biogeography and paleoecology. In: Bernor, R., et al. (Eds.), the evolution of

Western Eurasian Neogene mammal faunas. Columbia University Press,

NewYork, pp. 348–377.

FRANZ-ODENDAAL, T. A., KAISER, T. M. AND BERNOR, R. L., 2003. Systematics

and dietary evaluation of a fossil equid from South Africa-implications for

habitat conditions during the Early Pliocene. S. Afric. J. Sc., 99: 453-459.

FREEMAN, K. H. AND HAYS, J. M., 1992. Fractionation of carbon isotopes by

phytoplankton and estimates of ancient GO, levels. Glob. Biogeoch. Cycles, 6:

185-198.

GAUDRY, A., 1865. Animaux fossils at geologie de I. Attique, Paris, pp. 264-308.

155

GENTRY, A. W., ROSSNER, G. E. AND HEIZMAN, E. P. S., 1999. Suborder

Ruminantia, in Rossner, G. E., and Heissig, K. (eds.). The Miocene land

mammals of Europe: Munchen, Verlag Dr. Friedrich Pfeil, 225-258.

GENTRY, A. W. AND HOOKER, J. J., 1988. The phylogeny of Artiodactyla. In: The

Phylogeny and Classification of the Tetrapods. Vol. 2: Mammals, pp. 235-272

(ed. M. J. Benton), Systematics Association Special Volume No. 35B

Clarendon, Oxford.

GENTRY, A. W., 1966. Fossil Antilopini of East Africa. Bull. Brit. Mus. Nat. Hist.

(Geol). 12: 45-106.

GENTRY, A. W., 1970. The Bovidae (Mammalia) of the Fort Ternan fossil fauna: pp.

243-323 in L. S. B. Leakey and R. J. G. Savage (eds.). Fossil Vertebrates of

Africa, Vol. 2. Academic Press, London.

GENTRY, A. W., 1974. A new genus and species of Pliocene boselaphine (Bovidae,

Mammalia) from South Africa, Ann. S. Afr. Mus., 65(5): 145-188.

GENTRY, A. W., 1978. Bovidae; pp. 540-572 in V. J. Maglio and H. B. S. Cooke (eds.),

Evolution of African Mammals. Harvard University Press, Cambridge,

Massachusetts.

GENTRY, A. W., 1987. Rhinoceros from the Miocene of Saudi Arabia. Bull. Br. Mus.

Nat. Hist. (Geol.), 41(4): 409-432.

GENTRY, A. W., 1994. The Miocene differentiation of Old World Pecora (Mammalia).

Hist. Biol., 7: 115-158.

GENTRY, A. W., 1997. Fossil ruminants (Mammalia) from the Manonga Valley.

Tanania. In Neogene Paleontlogy of the Manonga Valley, Tanzania, pp. 107-

135 (ed. T. J. Harrison). Plenum Press, New York.

GENTRY, A. W., 1999. Fossil Pecorans from the Baynunah Formation, Emirate of Abu

Dhabi, United Arab Emirates. Chap. 22 in Abu Dhabi pecorans, Yale

University Press, New Haven, 290-316.

GENTRY, A.W., 2005. Ruminants of Rudabanya. Palaeontograph. Italic., 90: 283-302.

GERAADS, D. AND SARAC, G., 2003. Rhinocerotidae from the Middle Miocene

Hominoid locality of Candir (Turkey). Cour. Forsch. Inst. Senckenberg, 240:

217-231.156

GERAADS, D., 1986. Remarques sur la systrematique et la phylogenic des Girffidae

(Artiodactyla. Mammalia). Geob., 19: 465-477.

GERAADS, D., ALEMSEGED, Z. AND BELLON, H., 2002. The late Miocene

mammalian fauna of Chorora. Awash basin, Ethiopia: systematics,

biochronology and the 40K-40Ar ages of the associated volcanics. Tert. Res.,

21 (1-4): 113-122.

GERAADS, D., GULUC, E. AND SARAC, G., 1995. Middle Miocene Ruminants from

Inonu, Central Turkey. N. Jb. Geol. Palaont. Mh. Stuttgart, 8: 462- 474.

GERVAIS, P., 1849. Note sur la multiplicite des especes d’ hipparions (genre de

Chevaux a trios droits) qui sont enfouis a cucuron (Vaucluse). Comptes

Rendus, Academie des Sc. de Paris, 29: 284-286.

GHAFFAR, A., 2005. Studies on Equids, Cervids and Carnivora from the Siwalik Hills

of Pakistan. Ph. D. Diss., University of the Punjab, Lahore, Pakistan.

GHAFFAR, A., NAYYER, A. Q., BHATTI, Z. H. AND AKHTAR, M., 2003. Critical

analysis of Siwalik equids. Sci. Int. (Lahore), 15(2): 177-178.

GHOSH. S. K., KUMAR, R. AND SURESH, N., 2003. Influence of Mio-Pliocene

drainage reorganization in the detrital modes of sandstone, Subathu sub-basin,

Himalayan foreland basin. J. Him. Geol., 24: 35-46.

GILL, W.D., 1951. The stratigraphy of the Siwalik series in the Northern Potwar, Punjab,

Pakistan. Q. J. G. Soc., 107: 325-421.

GINSBURG, L., 1974. Les Tayassuidés des phosphorites du Quercy. Palaeovert., 6: 55-

85.

GROMOVA, V., 1952. Lc genre Hipparion (translated from Russian by St. Aubin). Bur.

Res. Min. Geol., 12: 1-288.

HAMILTON, W. R., 1973. The lower Miocene ruminants of Gebel Zelten, Libya.

Bulletin of the British Museum (Natural History) London. Geol., 21 (3): 75-

150.

HAN, DE-FEN., 1974. First discovery of Dorcabune in China. Vertebrate Palasiat.,

12(3): 217-221.

HEISSIG, K., 1972. Palaontolgische and geologische Untersuchungen im Tertia..r Vo

Pakistan, 5. Rhinocerotidae (Mamm.) aus den unteren und mittleren Siwalik-

157

Schichten. Bayerische Akademie der Wissenschaften Mathematisch-

Naturwissenschaftliche Klasse, Abhandlungen, Neue Folge, 152: 1-112.

HEISSIG, K., 2003. Change and Continuity in Rhinoceros faunas of Western Eurasia

from the Middle to the Upper Miocene. Eeden, Stara Lensa, pp. 35-37.

HILL, A., DRAKE, R., TAUXE, L., MONAGHAN, M., BARRY, J. C.,

BEHRENSMEYER, A. K., CURTIS, G., JACOBS, B. F., JACOBS, L.,

JOHNSON, N., PILBEAM, D., 1985. Neogene paleontology and

geochronology of the Baringo Basin, Kenya. J. Human Evol., 14: 759–773.

HOLLAND, T. H., 1926. Indian Geological Terminology, Mem. Geol. Surv. India, LI,

Pt. I.

HOOIJER, D. A., 1958. Fossil Bovidae from the Malay Archipelago and the Punjab.

Zoolog. Verhandlungen, 38: 1-110.

HUSSAIN, S. T., VAN DEN BERGH, G. D., STEENSMA, K. J., DE VISSER, J. A., DE

VOS, J., ARIF, M., FAN DAM, J., SONDAAR, P. Y. AND MALIK, S, M.,

1992. Biostratigraphy of the Plio-Pleistocene continental sediments (Upper

Siwaliks) of the Mangla-Samwal anticline, Azad Kashmir, Pakistan. Proc.

Koninklyke Nederlandse Akademie van Wetenschappen. Ser. B, 95: 65-80.

HUSSAIN, S. T., 1971. Revision of Hipparion (Equidae, Mammals) from the Siwalik

Hills of Pakistan and India. Bayerische Akademic der Wissenschaffen,

Abhandlungen, 147: 1-68.

IACOBELLIS, S. F. AND SOMERVILLE, R. C. J., 1991a. Diagnostic modeling of the

Indian monsoon onset, I: Model description and validation. J. Atmos. Sc., 48:

1948-1959.

IACOBELLIS, S. F. AND SOMERVILLE, R. C. J., 1991b. Diagnostic modeling of the

Indian monsoon onset, II: Budget and sensitivity studies: J. Atmos. Sc., 48:

1960-1971.

IQBAL, M., LIAQAT, A., KHAN, M.A. AND AKHTAR, M., 2009. Some new remains

of Hipparion from the Dhok Pathan type locality, Pakistan. J. A. P. S., 19(3):

154-157.

JACOBS, L. L., FLYNN, L. J. AND DOWNS, W. R., 1989. Neogene rodents of southern

Asia. In: C. C. Black and M. R. Dawson (Editors). Papers on Fossil Rodents

158

in Honor of Albert, F. Wood. Natural History Museum, Los Angeles

Company. Sc. Ser., 33: 157-177.

JACOBS, L. L., FLYNN, L. J., DOWNS, W. R. AND BARRY, J. C., 1990. Quo vadis

Antemus? The Siwalik muroid record. In: E. H. Lindsay, V. Fahlbusch and P.

Mein (Editors). European Neogene Mammal Biochronology. Plenum Press,

New York, pp. 573-586.

JANIS, C. M. AND SCOTT, K. M., 1987a. Grades and clades in hornless ruminant

evolution: the reality of Gelocidae and the systematic position of

Lophiomeryx and Bachitherium. J. V.P., 7: 200-216.

JANIS, C. M. AND SCOTT, K. M., 1987b. The inter-relationship of Higher Ruminant

families with special emphasis on the members of the Cervoidea. Am. Mus.

Novit., 2893: 1-85.

JOHNSON, G. D., ZETTLER, P., NAESER, C. W., JOHNOSN, N. M., SUMMERS, D.

M. FROST, C. D., OPDYKE, N. D. AND TAHIRKHELI, R. A. K., 1982.

The occurrence and fission-track ages of Late Neogene and Quatrnary

volcanic seiments, Siwalik group, northern Pakistan. Palaeog., Palaeocl.,

Palaeoec., 37: 63-93.

JOHNSON, N. M. AND MCGEE, V. E., 1983. Magnetic polarity, stratigraphy,:

stochastic properties of data, sampling problems and the evaluation of

interpretations. J. Geoph. Res., 88B: 1213-1221.

JOHNSON, N. M., OPDYKE, N. D., JOHNSON, G. D., LINDSAY, E. H. AND

TAHIRKHELI, R. A. K., 1982. Magnetic polarity, stratigraphy and ages of

Siwalik group rocks of the Potwar Plateau, Pakistan. Palaeog., Palaeocl.,

Palaeoec., 37: 17-42.

JOHNSON, N. M., SHEIKH, K. A., DAWSON, S. E. AND MCRAE, L. E., 1988. The

use of magnetic-reversal time lines in stratigraphic analysis: a case study in

measuring variability in sedimentation rates. In K. L. Kleinspehn and C.

Paola. (Eds.), New perspectives in basin analysis. Springer, New York, pp.

189-200.

159

JOHNSON, N. M., STIX, J., TAUXE, L., CERVENY, P. F. AND TAHIRKHELI, R. A.

K., 1985. Paleomagnetic chronology, fluvial processes and tectonic

implications of the Siwalik deposits near Chinji Village, Pakistan. J. Geol.,

93: 27-40.

JORGENSEN, D. W., HARVEY, M. D., SCHUMM, S. A. AND FLAM, L., 1993.

Morphology and dynamics of the Indus River: implicatins for the Mohen Jo

Daro site. In: Himalaya to the sea: geology, geomorphology and the

Quaternary (Ed. By J.F. Shroder). Routledge, New York, pp. 288-326.

KAISER, T. M. AND FORTELIUS, M., 2003. Differential mesowear including upper

and lower molars opening mesowear analysis for lower molars and premolars

in hypsodont equids. J. Morp., 258: 67-83.

KAISER, T. M., 2003. The dietary regimes of two contemporaneous populations of

Hippotherium primigenium (Perissodactyla, Equidae) from the Vallesian

(upper Miocene) of Southern Germany. Palaeog., Palaeocl., Polaeoec., 198:

381-102.

KAISER, T. M., BERNOR, R. L., FRANZEN, J. L., SCOTT, R. AND SOLOUNIAS, N.,

2003. New Interpretations of the Systmatics and Palaeoceology of the Dorn-

Durkheim I Hipparions (Late Miocene, Turolian Age [MNII]), Rheinhessen,

Germany. Senckenbergiana lethaea molars – opening mesowear analysis for,

lower molars and premolars in hypsodont equids. J. Morph., 258: 67-83.

KALEEM ULLAH, ARIF, M. AND SHAH, M. T., 2006. Petrography of Sandstones

from the Kamlial and Chinji Formations, Southwestern Kohat Plateau, NW,

Pakistan: Implications for Source Lithology and Paleoclimate. J. Himal. Earth

Sc., 39: 1-13.

KAPPELMAN, J., DUNCAN, A., FESEHA, M., LUNKKA, J. P., EKART, D.,

MCDOWELL, F., RYAN, T. AND SWISHER III, C. C., 2003. Chronology

of the Sinap Formation. In: Fortelius, M. Kappelman, J., Sen, S. and Bernor,

R.L. (Eds): The Geology and Paleontology of the Miocene Sinap Formation.

Turkey: 41-68; New York (Columbia University Press).

KAPPELMAN, J., KELLEY, J., PILBEAM, D., SHEIKH, K. A., WARD, S., ANWAR,

M., BARRY, J. C., BROWN, B., HAKE, P., JOHNSON, N. M., RAZA, S. M.

160

AND SHAH, M. L., 1991. The earliest occurrence of Sivapithecus from the

Middle Miocene Chinji Formation of Pakistan. J. Human Evol., 21: 61-73.

KAUP, J. J., 1833. Der Krallen-Phalanz Von Eppelsheim, nach welchem Hr. V. Cuvier

seinen Riesen-Phalanz Von Eppelsheim, nach welchem Hr. V. Cuvier seinen

Riesen-Pangolin. Manis gigante, aufstallte, gehort zu Dinotherium. Neues

Jahrd. Min., 2: 172-176.

KELLER, H. M. R., TAHIRKHELI, A. K., MIRZA, M. A., JOHNSON, G. D.,

JOHNSON, N. M. AND OPDYKE, N. D., 1977. Magnetic polarity

stratigraphy of the Upper Swialik deposits, Pabbi Hills, Pakistan. Ear. Planet.

Sc. Letters, 36: 187-201.

KHAN, A. M., 2010. Taxonomy and Distribution of Rhinoceroses from the Siwalik Hills

of Pakistan Ph. D. Diss. (unpublished), University of the Punjab, Lahore,

Pakistan.

KHAN, I. A., BRIDGE, J. S., KAPPELMAN, J. AND WILSON, R., 2006. Evolution of

Miocene fluvial environments, eastern Potwar Plateau, northern Pakistan. J.

Int. Assoc. Sedimentologists, 44 (2): 221-225.

KHAN, I. A., BRIDGE, J. S., KAPPELMANT, J. AND WILSON, T. R., 1997.

Evolution of Miocene fluvial environments, eastern Potwar plateau, northern

Pakistan. Sediment., 44: 221-251.

KHAN, M. A., 2007. Description of Selenoportax vexillarius Molars from Dhok Pathan

Village (Middle Siwaliks), Pakistan. Pak. J.Biol.Sc., 10(18): 3166-3169.

KHAN, M. A., 2008. Fossil bovids from the late Miocene of Padri, Jhelum, Pakistan.

Pak. J. Zool., 40(1): 25-29.

KHAN, M. A., AKHTAR, M. AND QURESHI, M. A., 2005a. Discovery of a

Bramatherium (Giraffid) horn core from the Dhok Pathan Formation (Middle

Siwaliks) of Hasnot, Potwar Plateau, Pakistan. Geol. Bull. Punjab, 40-41(6):

21-25.

KHAN, M. A., GHAFAR, A., ALI, Z., FAROOQ, U., HAMEED, Z. B. AND AKHTAR,

M., 2005b. Report on Mammalian Fossils of Chinji Formation, Dhulian,

Pakistan. Am. J. Appl. Sc., 2(12): 1619-1628.

161

KHAN, M. A., KOSTOPOULOS, D. S., AKHTAR, M. AND NAZIR, M., 2010a. Bison

remains from the Upper Siwaliks of Pakistan. Neu. Jahrb. für Geol. und

Paläont., 258(1): 121-128.

KHAN, M. A., BUTT, S. S., KHAN, A. M. AND AKHTAR, M., 2010b. A New

Collection of Giraffokeryx punjabiensis (Giraffidae, Ruminantia,

Artiodactyla) from the Lehri Outcrops, Jhelum, Northern Pakistan. Pak. J. Sc.,

62(2): 120-123.

KHAN, M. A. AND FAROOQ, M. U., 2006. Paleobiogeography of the

Siwalik Ruminants. Int. J. Zool. Res., 2(2): 100-109.

KHAN, M. A., LIOPOULOS, G. AND AKHTAR, M., 2009a. Boselaphines

(Artiodactyla, Ruminantia, Bovidae) from the Middle Siwaliks of Hasnot.

Pakistan. Geob., 42: 739-753.

KHAN, M. A., MALIK, M., KHAN, A. M., IQBAL, M. AND AKHTAR, M., 2009b.

Mammalian remains in the Chinji type locality of the Chinji Formation: A

new collection. J. A. P. S., 19(4): 224-229.

KIDWELL, S. M., 1988. Taphonomic comparison of passive and active continental

margins: Neogene shell beds of the Atlantic coastal plain and sothern Gulf of

Calitornia. Palaeg., Palaec., Palaeoec., 63: 201-223.

KINGSTON, J. D., JACOBS, B. F., HILL, A. AND DEINO, A., 2002. Stratigraphy, age

and environments of the late Miocene Mpesida Beds, Tugen Hills, Kenya. J.

Human Evol., 42: 95–116.

KLOOTWIJK, C. T., GEE, J. S., PEIRCE, J. W. AND SMITH, G. M., 1992. Neogene

evolution of the Himalayan-Tibetan region: constraints from ODP Site 758,

northern Ninetyeast Rdige: bearing on climatic change. Palaeog., Palaeocl.

Palaeoec., 95: 95-110.

KOCH, C. F., 1987. Prediction of sample size effects on the measured temporal and

geographic distribution patterns of species, Paleob., 13: 100-187.

KOHLER, M. 1993. Skeleton and habitat of recent and fossil ruminants. Münch.

geowissenschaftliche Abh. (A), 25: 1–88.

162

KOSTOPOULOS, D. S., 2009. The late Miocene mammal faunas of the Mytilinii basin,

Samos island, Greece: New collection, 14. Bovidae. Beitr. Palaont., 31: 339-

383.

KRAVTCHENKO, K. N., 1964. Soan Formation upper unit of Siwalik group in Potwar,

Sc. Indus., 2(3): 230-233.

KROON, D., STEENS, T. AND TROELSTRA, S. R., 1991. Onset of monsoonal related

upwellings in the western Arabian Sea as revealed by planktonic foraminifers.

Proc. Ocean Drill. Prog., Sci. Results, 117: 257-263.

KUMAR, R., GHOSH, S. K. AND SANGODE, S. J., 1999. Evolution of a fluvial system

in a Himalayan foreland basin, India, In: Macfarlane, A., Sorkhabi, R.B.,

Quade, J. (Eds.), Himalayan and Tibet: Mountain Roots to Mountain Tops.

Geol. Soc. Am. Spec. Pap., 328: 239-256.

KUMARAVEL, V., SANGODE, S. J., KUMAR, R. AND SIDDAIAH N. S., 2005.

Magnetic Polarity stratigraphy of Plio-Pleistocene Pinjor Formation (type

locality), Siwalik Group, NW Himalaya, India. Wadia Instit. Himal. Geol.,

88(9) 1453-1461.

KUMARAVEL, V., SANGODE, S. J., SIVA SIDDAIAH, N. AND KUMAR, R., 2009.

Major Element Geochemical Variations in a Miocene-PlioceneSiwalik

Paleosol Sequence: Implications to Soil Forming Processes in the Himalayan

Foreland Basin. J. Geol. Soc. India, 73: 759-772.

KUTZBACH, J. E., GUETTER, P. J., RUDDIMAN, W. F. AND PRELL, W. L., 1989.

Senstivity of climatic uplift in southern Asia and in American West:

Numerical experiments. J. Geophys. Res., 94: 18393-18407.

LARTET, E., 1837. Ser les debris fossils trouves a Sansan et sur les animaux antediluvins

en general. C. R. Ac. Sc., 5: 158-159.

LEWIS, G. E., 1937. A new Siwalik correlation (India), Amercan Jour. Sci. Ser., 5(195).

33: 191-204.

LIPSON, S. AND PILBEAM, D., 1982. Ramapithecus and hominoid evolution. Human

Evol., 11: 545-548.

LYDEKKER, R., 1876. Molar teeth and other remains of Mammalia from the India

Tertiaries. Pal. Ind., 10(2): 19-87.

163

LYDEKKER, R., 1877a. Notices of new and other vertebrata from Indian Tertiary and

Secondary rocks. Rec. Geol. Sur. Ind., 10: 30-42.

LYDEKKER, R., 1877b. Notices of new or rare mammals from the Siwaliks. Rec. Geol.

Sur. Ind., 10: 76-83.

LYDEKKER, R., 1878. Indian Tertiary and Post-Tertiary vertebrate. 3. Crania of

Ruminants. Pal. Ind., 10: 88-181.

LYDEKKER, R., 1879. Further Notices of Siwalik Mammalia. Rec. Geol. Surv. Ind.,

12(1): 33-52.

LYDEKKER, R., 1882. Siwalik and Narbada Equidae. Paleont. Indic. (X), Part 3, 2: 67-

98.

LYDEKKER, R., 1883a. Indian Tertiary and post-tertiary Vertebrata: Siwalik selenodont

Suina, etc. Memoirs of the Geological Survey of India, Palaent. Indic., 5(10):

143-177.

LYDEKKER, R., 1883b. Synopsis of the Fossil Vertebrata of India. Rec. Geol. Surv.

Ind., XVI: 61-93.

LYDEKKER, R., 1884. Additional Siwalik Perissodactyla and Proboscidea. Memoirs of

the Geological Survey of India, Palaeont. Indic., 3: (10): 1-34.

LYDEKKER, R., 1885. Catalogue of Siwalik Vertebrata in the Indian Museum. Part-1,

Mammalia, Calcutta.

LYDEKKER, R., 1886. Catalogue of Fossil Mammalia in the British Museum. Part 3,

London, pp. 186.

MAC FADDEN, B. J. AND WOODBURNE, M. O., 1982. Systematics of the Neogene

Siwalik hipparions (Mammalia, Equidae) based on cranial and dental

morphology. J. V. P., 2(2): 185-218.

MACFADDEN, B. J., 1984. Systematics and phylogeny of Hipparion, Neohipparion,

Nannhippus and Cormohipparion (Mammalia Equidae) from the Miocene and

Pliocene of the New World. Am. Mus. Bull., 179: 1–196.

MADE VAN DER, J. AND HUSSAIN, S. T., 1989. “Microstonyx” major (Suidae,

Artiodactyla) from the Type Area of the Nagri Formation, Siwalik Group,

Pakistan. Estudios Geol., 45: 409-416.

164

MADE VAN DER, J., 1994. Suoidea from the Lower Miocene of Cetina de Aragón,

Spain. Rev. Esp. Paleont., 9(1): 1-23.

MADE VAN DER, J., 1996. Listriodontinae (Suidae, Mammalia), their evolution,

systematics and distribution in time and space. Contr. Tert. Quater. Geol., 33:

3-254.

MANKINEN, E. A. AND DALRYMPLE, G. B., 1979. Revised geomagnetic polarity

time scale for the interval 0-5 m.y. B.P. J. Geoph. Res., 84B: 615-626.

MATTHEW, W. D., 1929. Critical observations upon Siwalik mammals (exclusive of

Proboscidea). Am. Mus. Nat. Hist. Bull., 56: 437-560.

MEDLICOTT, H. B., 1864. On the geological structure and relations of the Southern

portion of the Himalayan range between the Rivers Ganges and Rauce. Geol.

Sur. Ind., Mem. III.

MERCERON, G., BLONDEL, C., BRUNET, M., SEN, S., SOLOUNIAS, N., VIRIOT,

L. AND HEINTZ, E., 2004. The Late Miocene paleoenvironment of

Afghanistan as inferred from dental microwear in artiodactyls. Palaeog.,

Palaeocl., Palaeoec., 207: 143–163.

METAIS, G., BENAMMI, M., CHAIMANEE, Y., JAEGER, J. J., THAN TUN, TIN

THEIN AND DUCROCQ, S., 2000. Discovery of new ruminant dental

remains from the Middle Eocene Pondaung formation (Myanmar):

reassessment of the phylogenetic position of Indomeryx. Comptes Rendus de

I’Academie des Sc. de Paris, Serie Ha., 330: 1-7.

METAIS, G., CHAIMANEE, Y., JAEGER, J. J. AND DUCROCQ, S., 2001. New

remains of primitive ruminants from Thailand: evidence of the early evolution

of the Ruminantia in Asia. Zool. Script., 30: 231-248.

METAIS, G., GUO, J. AND BEARD, K. C., 2004. A New Small Dichobunid Artiodactyl

from Shanghuang (Middle Eocene, Eastern China): Implications for the Early

Evolution of Proto-Setenodonts in Asia. Bull. Carn. Mus. Nat. Hist., 36: 177-

197.

MITCHELL, G. AND SKINNER, J. D., 2003. On the origin, evolution and phylogeny of

giraffes Giraffa camelopardalis. Trans. Roy. Soc. S. Afric., 58: 51-73.

165

MORGAN, M. E., KINGSTON, J. D. AND MARINO, B. D., 1994. Carbon isotopic

evidence for the emergence of C4 plants in the Neogene from Pakistan and

Kenya. Natur., 267: 162-165.

MURPHY, M. A., 1977. On time-stratigraphic units. J. P., 51: 213-219.

NAKAYA, H., 1994. Faunal change of Late Miocene Africa and Eurasia: mammalian

fauna from the Namurungule Formation, Samburu Hills, northern Kenya.

African Study Monographs, Supp. Iss., 20: 1–112.

NANDA, A. C., 2002. Upper Siwalik mammalian faunas of India and associated events.

J. Asian Ear. Sc., 21: 47-58.

NANDA, A. C., 2008. Comments on the Pinjor Mammalian Fauna of the Siwalik Group

in relation to the Post-Siwalik Faunas of Peninsular India and Indo-Gangetic

Plain. Quater. Int., 192: 6-13.

NANDA, C. AND SHANI, A., 1990. Oligocene Vertebrates from the Ladakh Molasse

Group, Ladakh Himalaya: Paleobiogeographic Implications. J. Himal. Geol.,

I: 1-10.

NASEEM, L., KHAN, M. A, AKHTAR, M., IQBAL, M., KHAN, A. M. AND

FAROOQ, U., 2009. Hipparion from the Nagri Type Locality of the Nagri

Formation, Middle Siwaliks, Pakistan: Systematics. J. Nat. Sc., 7(1-2): 18-29.

NOW DATABASE 2003. Neogene of the Old World. Database of fossil mammals:

www.helsinki.fi/science/now

OLDHAM, R. D., 1893. Geology of India. Calcutta.

OPDYKE, N. D., LINDSAY, E., JOHNSON, G. D., JOHNSON, N., TAHIRKHELI, R.

A. K. AND MIRZA, M. A., 1979. Magnetic polarity, stratigraphy and

vertebrate paleontology of the Upper Siwalik Subgroup of northern Pakistan.

Palaeog., Palaeocl., Palaeoec., 27: 1-34.

PASCOE, E.H. 1964. A manual of the Geology of India and Burma, vol 111. Indian

Government Press, Calcutta, 111: 886.

PEARSON, H.S., 1932. Some Skulls of Perchoerus [Thynohyus] from the White River

and John Day Formations. Bull. Am. Mus. Nat. Hist., 48(3): 61-96.

166

PICKFORD, M. AND MORALES, J., 2003. New Listriodontinae (Mammalia, Suidae)

from Europe and a review of listriodont evolution, biostratigraphy and

biogeography. Geodiver., 25(2): 347-404.

PICKFORD, M., 1981. Preliminary Miocene Mammalian Biostratigraphy for Westem

Kenya. J. Human Evol., 10: 73-97.

PICKFORD, M., 1988. Revision of the Miocene Suidae of the Indian Subcontinent.

Münchner Geowissenschaften Abh., 12: 1-91.

PICKFORD, M., 2001. Africa’s smallest ruminant: A new tragulid from the Miocene of

Kenya and the biostratigraphy of East African Tragulidae. Geob., 34(4): 437–

447.

PICKFORD, M., SENUT, B., MOURER-CHAUVIRE, C., 2004. Early Pliocene

Tragulidae and peafowls in the Rift Valley, Kenya: Evidence for rainforest in

East Africa. Comptes Rendus Palevol., 3: 179–189.

PILBEAM, D. R., BEHRENSMEYER, A. K., BARRY, J. C. AND SHAH, S. M. I.,

1979. Miocene sediments and faunas of Pakistan. Postill., 179: 1-45.

PILBEAM, D., 1982. New hominoid skull material from the Miocene of Pakistan.

Natur., 295: 232-234.

PILBEAM, D., BARRY, J., MEYER, G. E., SHAH, S. M. I., PICKFORD, M. H. L.,

BISHOP, W. W., THOMAS, H. AND JACOBS, L. L., 1977. Geology and

Palaeontology of Neogene strata of Pakistan. Natur., London, 270: 684-689.

PILBEAM, D., MORGAN, M., BARRY, J. C. AND FLYNN, L., 1997. Eurpean MN

units and the Siwalik faunal sequence of Pakistan. Pp. 96-105 in R. L. Bernor,

V. Fahlbusch, and H-W. Mittmann, eds. The evolution of Western Eurasian

Neogene mammal faunas. Columbia University Press, New York.

PILGRIM, G. E. AND HOPWOOD, A. T., 1928. Catalogue of the Pontian Bovidae of

Europe. Brit. Mus. Nat. Hist. London.

PILGRIM, G. E., 1910. Notices of new Mammalian genera and species from the

Tertieries of India-Calcutta. Rec. Geol. Sur. Ind., 40: 63-71.

PILGRIM, G. E., 1911. The fossil Giraffidae of India. Mem. Geol. Sur. Ind. Palaeont.

Indic. N. S., 4 (I): 1-29.

167

PILGRIM, G. E., 1913. Correlation of the Siwaliks with Mammal Horizons of Europe.

Rec. Geol. Sur. Ind., 43(4): 264-326.

PILGRIM, G. E., 1915. The dentition of the Tragulid genus Dorcabune. Rec. Geol. Sur.

Ind., 45: 226–238.

PILGRIM, G. E., 1926. The Tertiary formations of India and the interrelation of marine

and terrestrial deposits. Proc. Pan-Pacific Sci. Cong. Austalia, 896-931.

PILGRIM, G. E., 1937. Siwalik antelopes and oxen in the American Museum of Natural

History. Bull. Amer. Mus. Nat. Hist., 72: 729-874.

PILGRIM, G. E., 1939. The fossil Bovidae of India. Pal. Ind., N. S., 26(1): 1-356.

PILGRIM. G. E., 1912. The vertebrate fauna of the Gaj Series in the Bugti Hills and the

Punjab. Paleont. Indic., N. S., 4: 1-83.

PINFOLD, E. S., 1918. Notes on structure and stratigraphy in the north-west Punjab,

India. Rec. Geol. Sur. Ind., 49: 137-160.

PRASAD, K. N., 1968. The vertebrate fauna from the Siwalik beds of Haritayangar,

Himachal Pradesh, India. Palaeont. Ind., N. S., 39: 1-55.

PRELL, W. L. AND KUTZBACH, J. E., 1992. Sensitivity of the Indian monsoon to

forcing parameters and implications for evolution. Natur., 360: 647-652.

QUADE, J., CERLING, T. E. AND BOWMAN, J. R., 1989. Dramatic ecologic shift in

the latest Miocene of northern Pakistan, and its significance to the

development of the Asian Monsoon, Natur., 342: 163-166.

QUADE, J. AND CERLING, T. E., 1995. Expansion of C4 grasses in the Late Miocene

of Northern Pakistan: evidence from stable isotopes in paleosoles. Palaeog.,

Palaeocl., Palaeoec., 115: 91-116.

RAIVERMAN, V. AND SURESH, N., 1997. Clay mineral distribution in the Cenozoic

sequence of the western Himalayan Foothills. J. Indian Assoc. Sediment, 16:

63-75.

RAIVERMAN, V., 2002. Foreland Sedimentation in Himalayan Tectonic Regime a

relook at the orogenic process. Bishen Sing Mahendra Pal Singh, Dehra Dun,

India, p. 378.

RAYMO, M. E. AND RUDDIMAN, W. F., 1992. Tectonic forcing of late Cenozoic

climate. Natur., 359: 117-122.

168

RAYMOND, L. B., THOMAS, M. K. AND SHERY, V. N., 2004. The Oldest Ethiopian

Hipparion (Equinae, Perissodactyla) from Chorora. Forch Inst. Senkckenberg,

246: 213-226.

RAZA, S. M., 1983. Taphonomy and paleoecology of middle Miocene vertebrate

assemblages, southern Potwar Plateau, Pakistan. Thesis. Yale Univ., New

Haven.

RETALLACK, G. J., 1991. Miocene paleosols and ape habitats of Pakistan and Keny.

Oxford University Press, Oxford.

ROMER, A. S., 1974. Vertebrate palaeontology, III. The University of Chicago Press,

Chicago Hlinois. Pp. 1-687.

RÖSSNER, G. E., 2010. Systematics and palaeoecology of Ruminantia (Artiodactyla,

Mammalia) from the Miocene of Sandelzhausen (southern Germany, Northern

Alpine Foreland Basin). Palaont. Z., 1-40. (DOI 10.1007/s12542-010-0052-

2).

RUDDIMAN, W. F., RAYMO, M. E., MARTINSON, D. G., CLEMENT, B. M. AND

BACKMAN, J., 1989. Mid-Pleistocene evolution of Northern Hemisphere

climate. Paleoceanog., 4: 353-412.

SAHNI, A. AND MITRA, H. C., 1980. Neogene palaeobiography of the Indian

subcontintent with special reference to fossil vertebrates. Palaeog., Palaeocl.,

Palaeoec., 31: 39-62.

SAHNI, A., TIWARI, B. N. AND KUMAR, K., 1980. An Additional Lower Siwalik

Vertebrate Fauna from the Kalagarh Area, District Pauri Garhwal, Uttar

Pradesh, Proc. Ind. Geol. Cong. Poona, 3: 81-90.

SARAGE, D. E. AND RUSSELL, D. E., 1983. Mammalian Paleofaunas of the World,

London.

SARWAR, M. AND AKHTAR, M., 1987. A new Sivatherine giraffe from Pabbi Hills of

Potwar, Pakistan. Kashmir J. Geol., 5: 95-99.

SARWAR, M., 1977, Taxonomy and distribution of the Siwalik Proboscidea. Bull. Dept.

Zool. Univ. Punjab, N. S., 10: 1-172.

SAVAGE, R. J. G. AND LONG M. R., 1986. Mammal Evolution – An Illustrated Guide.

Giraffids. British Museum (Natural History), London, pp. 1-228.

169

SCOTT, R. S., KAPPELMAN, J. AND KELLEY, J., 1999. The Paleoenvironment of

Sivapithecus parvada. J. Human Evol., 36: 245-274.

SEN, S., BLIECK, A., BOUVRAIN, G., BRUNET, M., GERAADS, D., HEINTZ, E.

AND KOUFOS, G. D., 1997. Late Miocene mammals from Taghar,

Khurdkabul Basin, Afghanistan. Ann. de Paleont., 83: 233–266.

SHAH, S. M. I., 1980. Stratigraphy and economic geology of Central Salt Range Rec

Geol. Sur. Pak., 52: 1-104.

SHEIKH, I. M., PASHA, M. K., WILLIAMS, V. S., RAZA, S. Q. AND KHAN, K. S.,

2008. Environmental Geology of the Islamabad-Rawalpindi Area, Northern

Pakistan. In: Regional Studies of the Potwar-Plateau Area, Northern Pakistan.

Bull., 2078G.

SKINNER, M. F. AND MACFADDEN, B. J., 1977. Cormohipparion N. Gen.

(Mammalia, Equidae) from the North American Miocene (Barstovian-

Clarendonian). J. Palaeont., 51(5): 912-926.

SOLOUNIAS, N., 1981. The turolian fauna from the Island of Samos, Greece. Cont.

Vert. Evol., 6: 99-232.

SOLOUNIAS, N., 1999. The Paleoecology of the Pikermian Biome and the savanna

myth. In: Agustí, J., et al. (Eds.), Hominoid evolution and climatic change in

Europe I: the evolution of Neogene terrestrial ecosystems in Europe.

Cambridge University Press, Edinburgh, pp. 436–453.

SOLOUNIAS, N. AND HAYEK, L. A. C., 1993. New methods of tooth microwear

analysis and application to dietary determination of two extinct antelopes. J.

Zool., 229: 421–445.

SOLOUNIAS, N., MOELLEKEN, S. M. C. AND PLAVCAN, J. M., 1995. Predicting

the diet of extinct bovids using masseteric morphology. J. V. P., 15: 795–805.

SPASSOV, N. AND GERAADS, D., 2004. Tragoportax Pilgrim, 1937 and

Miotragocerus Stromer, 1928 (Mammalia, Bovidae) from the Turolian of

Hadjidimovo, Bulgaria, and a revision of the Late Miocene Mediterranean

Boselaphini. Geodiv., 26(2): 339-370.

TAUXE, I., 1979. A new date for Ramapithecus. Natur., 282: 399-401.

170

TAUXE, L. AND BADGLLY, C., 1988. Stratigraphy and remanence acquisition of a

paleomagnetic reversal in alluvial Siwalik rocks of Pakistan. Sediment., 35:

697-715.

TAUXE, L. AND OPDYKE, N. D., 1982. A time framework based on

magnetostratigraphy for the Siwalik sediments of the Khaur area, northern

Pakistan. Palaeog., Palaeocl., Palaeoec., 37: 43-61.

THOMAS H., 1977. Unn nouveau Bovide dans les couches a Hominoidea du Nagri

(Siwaliks moyens), Plateau de Potwar, Pakistan: Elachistoceras

khauristanensis gen. et sp. nov. (Bovidae, Artiodactyla, Mammalia). Bull.

Soc. Geol. France, (7), XIX, (2), 375-383. (eds.), The Evolution of Western

Eurasian Neogene Mammal Faunas. Columbia University Press, New York.

THOMAS, H., 1984. Les bovides ante-hipparions des Siwaliks inferieurs (Plateau du

Potwar), Pakistan. Memoires de la Societe Geologique de France, Paris, 145:

1-68.

TONG, YONGSHENG AND ZHONGRU, 1986. Odoichoerus, a new Suoid

(Artiodactyla, Mammalia) from the early Tertiary of Guangxi. Vert.

PalAsiatica, 24(2): 129-138, pl. l.

VON MEYER, H., 1846. Mitteilungenan Prof. Bronn gerichtet (Brief). Neues Jahrbuch

fur Minereralogie, Geologie und Palaontologie, 462–476.

WELCOMME, J. L., BENAMMI, M., CROCHET, J. Y., MARIVAUX, L., METAIS,

G., ANTOINE, P. O. AND BALOCH, I., 2001. Himalayan Forelands:

palaeontological evidence for Oligocene detrital deposits in the Bugti Hills

(Balochistan, Pakistan). Geol. Magaz., 138 (4): 397–405.

WEST, R. M., 1980 A minute new species of Dorcatherium (Tragulidae, Mammalia)

from the Chinji Formation near Daud Khel, Mianwali district, Pakistan,

MILW. Public Mus. Contrib, Biol. Geol., 3(33): 1-6.

WEST, R. M., HUTCHISON, J. H. AND MUNTHE, J., 2010. Miocene vertebrates from

the Siwalik group, Western Nepal. J. V. P., 11(1): 108-129.

WHITWORTH, T., 1958. Miocene ruminants of East Africa. Fossil Mammals of Africa.

Bull. Brit. Mus. Nat. Hist., 15: 1-50.

171

WILLIS, B. J. AND BEHRENSMEYER A. K., 1995. Fluvial systems in the Siwalik

Neogene and Wyoming Paleogene. Palaeog., Palaeocl., Palaeoc., 114: 13-35.

WILLIS, B., 1993a. Ancient river systems in the Himalayan foredeep, Chinji village area,

northern Pakistan. Sediment. Geol., 88: 1-76.

WILLIS, B., 1993b. Evolution of Miocene fluvial systems in the Himalayan foredeep

through a two kilometer thick succession in northern Pakistan. Sediment.

Geol., 88: 77-121.

WRIGHT, J. D. AND MILLER, K. G., 1993. Southern Ocean influences on Late Eocene

to Miocene deepwater circulation. In The Antarctic paleoenviroment: a

perspective on global change (eds. J. P. Kennet and D. A. Warnke), pp. 1-25.

American Geophysical Union, Washington.

ZACHOS, J., PAGANI, M., SLOAN, L., THOMAS, E. AND BILLUPS, K., 2001.

Trends, rhythms, and aberrations in global climate 65 Ma to present. Sc., 292:

686-693.

ZALEHA, M. J., 1994. Alluvial Deposition and Pedogenesis within the Miocene Indo-

Gangetic Foreland, Khaur Area, Northern Pakistan. Ph. D. Diss., State

University of New York at Binghamton.

ZALEHA, M. J., 1997. Intra- and extrabasinal controls on fluvial deposition in the

Miocene Indo-Gangetic forland basin, northern Pakistan. Sediment., 44: 369-

390.

ZALEHA, M. J., 2006. Fluvial and Lacustrine Palaeoenvironments of the Miocene

Siwalik Group, Khaur area, Northern Pakistan. Sediment., 44: 349-368.

ZITTEL, K. A., 1925. Text Book of Palaeontology. MacMillan & Co. Ltd. London, pp.

310.

172

APPENDICES

Appendix 1 - Studied Material

Listriodon pentapotamiae

Upper dentition: PC-GCUF 10/04, left first upper incisor (I1); PUPC 07/73, right

maxillary ramus with M1-2. Lower dentition: PC-GCUF 10/05, isolated left P4; PUPC

07/72, almost complete mandible with the partial canines, the right hemimendible with

M1-3 and the left hemimendible with M2-3.

Selenoportax cf. vexillarius

Upper dentition: PC-GCUF 10/07, isolated left M1. Lower dentition: PC-GCUF 10/06,

isolated left incisor (I1); PUPC 09/117, isolated right M1; PUPC 07/135, a fragment of

right mandible having M1-3.

Pachyportax cf. latidens

Upper dentition: PUPC 09/46, isolated right P3; PUPC 09/69, isolated left M2.

Tragoportax punjabicus

Upper dentition: PC-GCUF 10/08, isolated left P3; PUPC 09/66, isolated right M1; PC-

GCUF 10/09, partial tooth probably M1. Lower dentition: PC-GCUF 10/11, isolated right

P3; PUPC 09/70, isolated left P4; PUPC 07/77, isolated left M1; PUPC 07/86, isolated left

M1; PUPC 07/138, a mandibular ramus with P4-M2.

Miotragocerus cf. gluten

Upper dentition: PUPC 07/138, isolated left M1. Lower dentition: PC-GCUF 10/12,

isolated right P3.

Gazella cf. lydekkeri

Lower dentition: PC-GCUF 09/02, a right mandibular ramus with M1-3; PUPC 07/71,

isolated left M3.

Giraffokeryx punjabiensis

Upper dentition: PUPC 07/88, isolated left P3; PUPC 09/67, partially preserved isolated

right P3; PUPC 07/133, isolated right M2. Lower dentition: PUPC 09/43, left

hemimandible with M2-3, broken canine and alveoli of P3-M1; PUPC 07/90, isolated right

M3.

173

Giraffa cf. priscilla

Upper dentition: PUPC 07/131, isolated left M1; PUPC 07/89, isolated right M1.

Dorcatherium cf. minus

Upper dentition: PC-GCUF 10/10, isolated left dP4. Lower dentition: PUPC 07/69, a right

mandibular ramus with partial M1 and complete M2.

Dorcatherium cf. majus

Upper dentition: PC-GCUF 09/46, isolated right M2.

Dorcabune cf. anthracotherioides

Lower dentition: PUPC 07/87, isolated left M2.

Hipparion theobaldi

Upper dentition: PUPC 07/61, isolated left P2; PUPC 07/65, isolated left M1; PUPC

07/66, isolated right M1; PUPC 07/57, isolated left M2; PUPC 07/58, isolated left M3.

Lower dentition: PUPC 07/60, isolated right P2; PUPC 07/59, isolated right P3; PUPC

07/78, isolated left P4; PUPC 07/124, isolated right M3.

Brachypotherium perimense

Lower dentition: PUPC 07/52, a right mandibular ramus having P3 – M3. PUPC 07/53, a

left mandibular ramus having P3 – M2.

Appendix 2 - Reprint

174