- Email: [email protected]
Note on pliocene small mammals from the Mirpur district, Azad Kashmir, Pakistan
I
NOTE ON PLIOCENE SMALL MAMMALS FROM THE M I R P U R DISTRICT, AZAD KASHMIR, PAKISTAN
IQBAL U. C H E E M A , S. MAHMOOD R A Z A , LAWRENCEJ. F L Y N N CHEEMA I.U., RAZA S.M. & FLYNN L.J. 1997. Note on Pliocene small mammals from the Mirpur district, Azad Kashmir, Pakistan. [Note sur de petits mammiferes plioc~nes du district de Mirpur, Azad Kashmir, PaMstan]. GEOBIOS, 30, 1: 115-119. Villeurbanne, le 28.02.1997. Manuscrit ddposd le 02.05.1994; acceptd ddfinitivement le 06.11.1995. ABSTRACT' - Small mammals from two mid-Pliocene localities in Azad Kashmir produce a modern shrew comparable to living Suncus and murids assignable to three genera. These are Golunda kelleri, a species described previously from an early Pleistocene locality in northern Pakistan, cf. Hadromys sp., and Mus sp. The murids date to about 3 Ma and thus include the oldest record of Golunda and of the Hadromys lineage. KEYWORDS: PLIOCENE SIWALIKS,FOSSIL SMALLMAMMALS,MURIDAE, SUNCUS. Rt~SUMt~ - Les petits mammiferes r~colt~s dans deux localit4s du Plioc~ne moyen du Kashmir Azad consistent en une musaraigne du genre Suncus et trois muridds: Golunda kelleri, esp~ce ddj/~dgcrite darts le Plgistocgne infdrieur du Nord Pakistan, cf. Hadromys sp. et Mus sp. Ces muridgs, datgs d'environ 3 Ma, constituent les plus anciens reprgsentants des genres Golunda et Hadromys. MOTS-CLES: PLIOCI~NE DES SIWALIKS,PETITS MAMMIFI)]RESFOSSILES, MURIDAE, SUNCUS.
INTRODUCTION An approximately one kilometer thickness of Upper Siwalik sub-Group rocks in the southeastern Potwar and adjacent southwestern Kashmir record the Plio-Pleistocene m a m m a l i a n and tectono-geomorphic history of the Himalayas (Opdyke et al. 1979; B u r b a n k & Beck 1989; Hussain et al. 1992). Careful stratigraphically controlled faunal collection integrated with local magnetic reversal s t r a t i g r a p h y (correlated with the Geomagnetic Time Scale) has provided a unique fine-tuned calibration for estimating species longevities, deciphering faunal diversity patterns, and for establishing a firm biostratigraphic zonation of the Neogene Siwalik sequence in the Potwar Plateau (see for example Barry et al. 1982, 1991; Flynn et al. 1990; Jacobs et al. 1989). Recent studies of the Upper Siwalik rocks have centered main]ty in the Rohtas, Pabbi Hills and ManglaSamwal area (Keller et al. 1977; Opdyke et al. 1979; Hussain et al. 1992). Thick alternations of grey and brown sandstone and red and brown mudstone with prominent conglomerate interbeds in the uppermost parts characterise the Upper Siwalik sequence. The lithofacies changes from multi-storey exten-
sire sheet sandbodies in the lower parts to the mudstone dominated upper part, and the varying composition and proportional increase of conglomeratic interbeds in the higher parts of the sequence reflect changing patterns of the fluvial system in the area. These phenomena have been interpreted to be controlled by appearances of tectonic elements in the Himalayas, such as the uplift of the Pir Panjal and the Margalla hills in the immediate northern vicinity of the area. The 850 meter thick Upper Siwalik rocks in the Mangla-Samwal anticline have been divided into a lower Samwal Formation, a middle Kakra Formation and an upper dominantly conglomeratic Mirpur Formation (Arif 1985). The entire sequence ranges in age from ca. 3.4 Ma to 1.0 Ma (Johnson et al. 1979): the Samwal-Kakra Formation boundary is around 1.6 Ma (Arif 1985). Hussain et al. (1992) have reported a rich mammalian fauna from the Samwal Formation including Crocuta, Stegodon, Elephas, Equus, Sivatherium, Hexaprotodon and Hemibos besides other bovines, antelopes, giraffes, and rhinoceroses. This assemblage corresponds with the Elephas planifrons Interval Zone of Barry et al. (1982). Hussain et al. (1992), however, suggested a subdivision into a lower E. planifrons Range Zone
116 c)
Islamabad%
flY. :. . . . .
/'
Chart
\
•'i::. •, • • . .. c . ...,.'
•
\.
.
.i ';. • • : ..j • ...
Kabal ,.
Colony
.:.','
..~
' . ' .
I FOSSIL LOCALITIES
" ~" '" ' ' .]:-'."' [""i '"( '!',...:..'.. ~ " " "".:' '/'~'"":"/":"-.""-"~ ). "K"~ ."' "'""~ , : .amwa~' "v .:" ~:/~" " ; ~ " : >
VIL~GE/TO~ JHELUN
RIVER
~"...
:.i.i)~! ":,!:. :.ii.i;!::?..:L
BED
"
" ' ~r"~'~
.~k-
' "" J
Okm
1
2
3
4
I
I
I
I
1
FIGURE 1 - Map of parts of eastern Potwar Plateau Southwestern Kashmir showing the ibssil sites (HGSP 8929 and PMNH 8802). Carte du Potwar E s t et du K a s h m i r Ouest montrant les localit~s fossilif~res.
(between 3.3 Ma to 2.7 Ma) and an upper E. hysudricus Range Zone (between 2.7 Ma to 1.6 Ma). The small mammals reported here come from two localities, P M N H 8802 and HGSP 8929, in the Samwal Formation (Fig. 1). The PMNH 8802 site includes a few fragmentary molluscan shells and fish and reptilian bones, in a yellowish grey sandstone bed. HGSP 8929, however, has also produced some larger mammals. The stratigraphic position of these localities vis-a-vis the Johnson et al. (1979) magnetic polarity stratigraphy of the ManglaSamwal area suggests an age for PMNH 8802 of ca. 3.0 Ma and for HGSP 8929 an age of 2.6 Ma. This is the first report of a small mammal fauna from the Mangla-Samwal area and also for the late Pliocene of the Potwar-Kashmir area. Paleontologists in India are actively investigating late Pliocene rocks of the Karewas (Kotlia 1992) and around Saketi Park (Rajeev Patnaik, unpublished dissertation).
SYSTEMATIC PAT,AEONTOLOGY
D e s c r i p t i o n - This specimen h a s a distinctive preservation. It is a lustrous red-brown, thin bone being translucent; t h e r e is no indication of original tooth pigmentation. The caronoid and condyle are broken, so identification is b a s e d on the molar. The tooth is relatively u n r e d u c e d , b u t v e r y narrow. Its well developed trigonid blade b e t w e e n protoconid and p a r a c o n i d curves gently anterolingually. The m u c h s h o r t e r protoconid- metaconid blade is t r a n s v e r s e in orientation and the m e t a c o n i d is the s m a l l e s t of t h e s e t h r e e cusps. The m o d e r a t e l y s h o r t talonid is modified in structure. Its large single cusp, posterior in position, is n e a r l y on a level w i t h the metaconid.A major crest dips from it to the base of the metaconid. A second low crest descends anteriorly down the lingual side of the talonid cusp. There is a posterolabial cingulum on the talonid, none lingually, and a h e a v y anterolabial cingulum continuous w i t h a distinct ridge anterior to the paraconid.
Order INSECTIVORA Family SORICIDAE Genus S u n c u s EHRENBERG,1833
, ,.,/~t~
i!~
cf. Suncus sp. Fig. 2 m a t e r i a l - PMNH 5001, right dentary fragment bearing M/3. Referred
L o c a l i t y - HGSP 8929.
FIGURE 2 - cf. Suncus sp. (A-B) PMNH 5001, right M/3. Bar represents 1 mm.
117 FIGURE 3 - Golunda kelleri P M N H 5002, left M/1 ; P M N H 5003~ left M/2 ; P M N H 5004, left M/3 ; P M N H 5005, r i g h t M \ I . barre repr~sente 1 mm.
(A) (B) (C) (D) La
JAIl\ •..~
~f ~
.~
C B
Order RODENTIA Family MURIDAE Genus Golunda GRAY, 1837
Golunda kelleri JACOBS, 1978 Fig. 3 R e f e r r e d m : ~ t e r i a l - Four isolated teeth, P M N H 5002, 5003, 5004 and 5005. left for M/l, M/2 a n d M/3, and r i g h t M \ I respectively. L o c a l i t y - P M N H 8802.
D e s c r i p t i o n - The teeth range in color from light to dark chocolate brown. The M/1 and M/2 are similar in light brown coloration and moderately advanced wear. Root ends are broken. Cusps are arcuate to subcircular and positioned as Jacobs (1978) describes. The labial anteroconid of M/1 is smaller and posterior to the lingual anteroconid, and there is no ante M romedial cusp between them. The metaconid touches the lingual anteroconid. The hypoconid and entoconid converge in a remnant of the medial mure. A labial bulge in the hypoconid may represent a worn C1. The posterior cingulum is a small oval. The labial cingulum is greatly reduced, but there are small cuspules between labial anteroconid and protoconid, protoconid and hypoconid, and posterior to the hypoconid. There are six roots, a large anterior and two posterior roots, and three across the middle of the tooth. The protoconid and metaconid of M/2 join anteriorly, but the hypoconid, the smallest cusp, does not touch the entoconid. There is a small labial anteroconid and weak labial cingulum. There is a low anterior cingulum and small, oval posterior cingulure. A large root occurs at each corner of the square tooth, and a fifth small root is anteromedial in position. The triangular M/3 is lightly worn but corroded. There are three cusps, unconnected protoconid and metaconid, and medially placed entoconid. The anterior cingulum is clear but a posterior cingulum is not so evident. A small, labial cuspule is considered the hypoconid (see Musser 1987).
Although damaged, M/3 appears to have h a d two large anterior roots and one or two posterior roots. The upper first molar is rounded in outline with high, strongly inclined cusps. They are weakly joined in early wear. Medial cusps are the largest and most arcuate. The anterostyle and lingual and labial anterocones compose the first chevron; both anterostyle and labial anterocone are slightly posterior to the lingual anterocone. In the second chevron, the enterostyle and paracone are both slightly posterior to the protocone. The third chevron includes only two cusps, a large hypocone and very small metacone joined by a narrow posterior connection. The metacone is lateral to the hypocone and abuts the back of the paracone. There is no posterior cingulum. The base of this specimen is damaged and slightly corroded but at least five major roots were present. Genus Hadromys
THOMAS,1911
cf. Hadromys s p . Fig. 4 R e f e r r e d m a t e r i a l - P M N H 5006, left M \ 1 f r a g m e n t . L o c a l i t y - P M N H 8802.
D e s c r i p t i o n - This dark brown tooth fragment shows light wear. Its cusps are tall and slightly inclined. The first chevron is broken away. Cusps of the second chevron are strongly joined by high connections. The protocone is the biggest cusp in the second chevron, followed by enterostyle, then paracone. Both enterostyle and paracone are
A
B
FIGURE 4 - cf. Hadromys sp. (A-B) P M N H 5006, left M \ I . B a r r e p r e s e n t s 1 mm.
118 slightly posterior to the protocone. The conical hypocone dominates the third chevron, which is well separated by a deep cleft from the second. There is a small metacone oppressed to the labial side of the hypocone; there is no lingual Style. There is no posterior cingulum. Genus Mus
LINNAEUS,
1758
Mus sp. Fig. 5 R e f e r r e d m a t e r i a l - PMNH 5007, left M \ I fragment. L o c a l i t y - PMNH 8802.
D e s c r i p t i o n - This small specimen lacks its first chevron. The paracone is firmly joined to the protocone. The enterostyle is more separated and more posterior than paracone and a posterolingual spur. The hypocone is large and weakly joined to the enterostyle spur. The posterolabial metacone is unreduced. There is no posterior cingulum. D i s c u s s i o n - The specimens described here come from two localities in the Mirpur area and indicate considerable potential to build a Pliocene biostratigraphy for small mammals of northern Pakistan. Only one identifiable specimen, the shrew, cf. Suncus sp. comes from locality 8929. The Mirpur shrew M/3 is very narrow (Table 1) with a simple talonid dominated by a single cusp and high crest. The trigonid arms diverge broadly and the protoconid is high. The anterolabial cingulum is prominent. The specimen resembles some of the smaller species of Suncus, but its M/3 is distinctive in its narrowness and broad trigonid. The six murid specimens indicate three species, all from locality P M N H 8802 (Table 1). This locality correlates roughly to the top of the Mammoth
event in the section of H u s s a i n et al. (1992), ca. 3 Ma. Four specimens are assigned to Golunda kelleri, a species recorded in the younger locality DP 24 (Pabbi Hills area, Fig. 1; Jacobs 1978), which occurs m a g n e t o - s t r a t i g r a p h i c a l l y above the Olduvai event. Whether slight morphological differences in molars of the two samples reflect individual variation or time remains to be seen. The Mirpur M/1-2 have a slightly stronger labial cingulum and are possibly lower crowned, although wear stage makes this uncertain. This sample shows the diagnostic feature of G. kelleri, a centrally placed entoconid on M/3. The Mirpur Golunda adds the upper first molar, which was lacking in the original hypodigm of G. kelleri described by Jacobs (1978). This tooth differs from that of extant G. ellioti in that it is more elongated anteroposteriorly, with chevrons less crowned, and its metacone is less reduced. It is similar in length/width proportions to Golunda gurai SABATIER(1982), but at the small end of the size range for that species. Central cusps on M \ 1 are more crescentic in G. kelleri, labial cusps are smaller and rounderounded, and the metacone is smaller than in G. gurai. Musser (1987) rejected the latter as a member of Golunda. The cf. Hadromys sp. is smaller, lower crowned than Hadromys loujacobsi. Similarly, the chevrons are less crowned and slightly more curved. Whether the Mirpur taxon is generically distinct from Hadromys awaits more fossil finds. Whether it precedes or is a c o n t e m p o r a r y of early Hadromys awaits more field work. In any case, it indicates that the Hadromys lineage was present by 3 Ma. The third murid from P M N H 8802 represents a species of Mus based on cusp morphology and lack of a posterior cingulum. It is slightly smaller than Mus auctor and Mus sp. of DP 24. With reference to murids from localities DP 13
Element
Length
Width u
cf. S u n c u s sp.
M/3
1.15
Golunda kelleri
M\I M/1 M/2 M/3
2.80 2.60 1.95
cf. Hadromys sp.
M\ 1 (2nd chevron)
1.70
M u s sp.
M\ 1 (and chevron)
1.03
o
•
0.45 20.5 1.90 2.00 ca.1.70 ca.170
TABLE 1 - Measurements (mm) of molars described from localities HGSP 8929 and PMNH 8802. Dimensions (ram) des molaires ddcrites dans les deux loealitds.
FIGURE 5 - M u s sp. PMNH 5007, left M \ I . Bar represents 1 m m .
119
and DP 24 Fig. 1; Jacobs (197S), 5.7 and 1.6 Ma, respectively, the f a u n a of PMNH 8802 is consistent with its intermediate age. All three sites contain Mus. DP 13 is characterized by the presence of Parapelomys robertsi and the last record of Karnimata. DP 24 contains Hadromys and Golunda. The P M N H 8802 site contains a Hydromys-like form more primitive t h a n H. loujacobsi, and the first record of Golunda. That Golunda kelleri is known from localities at 3.0 and 1.6 Ma shows t h a t this species survived the end-Pliocene interval of climatic change t h a t effected much of the globe. It also establishes an earlier Asian record of Golunda, consistent with postulated origin of the genus on t h a t continent (Jacobs 1978). A c k n o w l e d g m e n t - We thank Dr. Farhat Hussain, Director General, Geological Survey of Pakistan, Dr. Shahzad A. Mufti, Director General, Pakistan Museum of Natural History, and Dr. Taseer Hussain, Howard University, for their encouragement and permission to undertake this study. We are also indebted to Mr. Mohammad Arif and (Late) M. Asif Jah, of Geological Survey of Pakistan, for their assistance in the field. Dr. Nayyer Iqbal has very kindly provided the French translation. We would also thank S. Jaffer Hussain Naqvi for typing the manuscript and Mrs Samyia Haroon for drawing the figures. Geobios reviewers, Dr. Hans de Bruijn and Dr. Pierre Mein, are thanked for their many useful comments. This research was supported by the Geological Survey of Pakistan, the Pakistan Museum of Natural History and by Smithsonian Institute's grants to Dr. Taseer Hussain and to Dr. David Pilbeam.
REFERENCES ARIF M. 1985 - S t r a t i g r a p h y a n d chronology of verteb r a t e fossil localities i n the M i r p u r area, Azad Kashmir. Geological Survey of Pakistan Information Release, 264: 1-13. BARRYJ.C., LINDSAYE.H. & JACOBS L.L. 1982. A biostratigraphic zonation of the middle a n d upper Siwaliks of t h e P o t w a r P l a t e a u of n o r t h e r n P a k i s t a n .
Palaeogeography, Palaeoclimatology, Palaeoecology, 37: 95-130.
FLYNN L.J., PILBEAM D., JACOBS L.L., BARRY J.C., BEHRENSMEYERA.K. & KAPPELMAN J. 1990 - The Siwaliks time and f a u n a s i n a Miocene terrestrial setting. Journal of Geology, Chicago, 98: 589-604. HUSSAIN S.W., VAN DEN BERGH G.D., STEENSMA K.J., DE VISSER J.A., DE VOS J., ARIF M., vAN DAM J., SONDAAR P.Y. & MALIK S.B. 1992 - Biostratigraphy of the PlioPleistocene c o n t i n e n t a l s e d i m e n t s (Upper Siwaliks) of the M a n g l a - S a m w a l Anticline, Azad Kashmir,
Pakistan. Proceedings of the Koninklijke nederlandse Akademie van wetenschappen, 95: 65-80. JACOBS L.L. 1978 - Fossil Rodents (Rhizomyidae a n d Muridae) from Neogene Siwalik deposits, Pakistan.
Museum Northern Arizona, Flagstaff, Bulletin Series, 52, 103 p. JACOBS L.L., FLYNN L.J. & DOWNS 1989 - Neogene rodents of s o u t h e r n Asia. In C.C. Black & M.R. Dawson (eds.), Papers on fossil rodents in honor of Albert E l m e r Wood. Natural History Museum, Los Angeles, Sciences Series, 33: 157-177.
JOHNSON G.D., JOHNSON N.M., OPDYKE, N.D. & TAHIRKHELI R.A.K. 1979 - Magnetic reversal stratigraphy and s e d i m e n t a r y tectonic history of the U p p e r Siwalik Group, e a s t e r n Salt R a n g e a n d s o u t h w e s t e r n Kashmir. In A. FARAH & K.A. DE JONQ (eds), Geody-
namics of Pakistan Geological Survey of Pakistan: 149-165.
KELLER H.M., TAHIRKHELI R.A.K., MIRZA M.A., JOHNSON G.D., JOHNSON N.M. & OPDYKE N.D. 1977 - Magnetic polarity stratigraphy of the upper Siwalik deposits, Pabbi Hills, Pakistan. Earth and Planetary Science
Letters, 36: 187-201. Kotlia B.S. 1992 Pliocene m u r i d s (Rodentia, Mammalia) from K a s h m i r Basin, W e s t e r n India.
Neues Jahrbuch fiir Geologic Paliiontologie, Abhandlungen, 183: 339-357. M~SSER G.G. 1987 - The occurrence of Hadromys (Rodentia: Muridae) in early Pleistocene Siwalik s t r a t a in n o r t h e r n P a k i s t a n and its b e a r i n g on biogeographic affinities b e t w e e n I n d i a n a n d northeastern African r o u t i n e rodents. American Museum of Natural History, Novitates, 2883, 36 p.
OPDYKE N.D., LINDSAY E.H., JOHNSON G.D., JOHNSON N.M., TAHIRKHELI R.A.K. & MIRZA M.A. 1979 Magnetic polarity stratigraphy and vertebrate paleontology of the Upper Siwalik subgroup of northern Pakistan. Palaeogeography, Palaeoclimatology,
Palaeoecology, 27, 34 p. SABATIER M. 1982 - Les rongeurs du site plioc~ne h hominid~s de H a d a r (Ethiopie). Palaeovertebrata, 12 (1), 56 p. I.U. C H E E M A
Earth Sciences Division Pakistan Museum of Natural History Garden Avenue Shakar Parian, Islamabad, Pakistan S.M. RAZA Oil and Gas Development Corporation Islamabad, Pakistan L.Jo FLYNN Peabody Museum Harvard University Cambridge, MA 02138, U.S.A.
NOTE ON PLIOCENE SMALL MAMMALS FROM THE M I R P U R DISTRICT, AZAD KASHMIR, PAKISTAN
IQBAL U. C H E E M A , S. MAHMOOD R A Z A , LAWRENCEJ. F L Y N N CHEEMA I.U., RAZA S.M. & FLYNN L.J. 1997. Note on Pliocene small mammals from the Mirpur district, Azad Kashmir, Pakistan. [Note sur de petits mammiferes plioc~nes du district de Mirpur, Azad Kashmir, PaMstan]. GEOBIOS, 30, 1: 115-119. Villeurbanne, le 28.02.1997. Manuscrit ddposd le 02.05.1994; acceptd ddfinitivement le 06.11.1995. ABSTRACT' - Small mammals from two mid-Pliocene localities in Azad Kashmir produce a modern shrew comparable to living Suncus and murids assignable to three genera. These are Golunda kelleri, a species described previously from an early Pleistocene locality in northern Pakistan, cf. Hadromys sp., and Mus sp. The murids date to about 3 Ma and thus include the oldest record of Golunda and of the Hadromys lineage. KEYWORDS: PLIOCENE SIWALIKS,FOSSIL SMALLMAMMALS,MURIDAE, SUNCUS. Rt~SUMt~ - Les petits mammiferes r~colt~s dans deux localit4s du Plioc~ne moyen du Kashmir Azad consistent en une musaraigne du genre Suncus et trois muridds: Golunda kelleri, esp~ce ddj/~dgcrite darts le Plgistocgne infdrieur du Nord Pakistan, cf. Hadromys sp. et Mus sp. Ces muridgs, datgs d'environ 3 Ma, constituent les plus anciens reprgsentants des genres Golunda et Hadromys. MOTS-CLES: PLIOCI~NE DES SIWALIKS,PETITS MAMMIFI)]RESFOSSILES, MURIDAE, SUNCUS.
INTRODUCTION An approximately one kilometer thickness of Upper Siwalik sub-Group rocks in the southeastern Potwar and adjacent southwestern Kashmir record the Plio-Pleistocene m a m m a l i a n and tectono-geomorphic history of the Himalayas (Opdyke et al. 1979; B u r b a n k & Beck 1989; Hussain et al. 1992). Careful stratigraphically controlled faunal collection integrated with local magnetic reversal s t r a t i g r a p h y (correlated with the Geomagnetic Time Scale) has provided a unique fine-tuned calibration for estimating species longevities, deciphering faunal diversity patterns, and for establishing a firm biostratigraphic zonation of the Neogene Siwalik sequence in the Potwar Plateau (see for example Barry et al. 1982, 1991; Flynn et al. 1990; Jacobs et al. 1989). Recent studies of the Upper Siwalik rocks have centered main]ty in the Rohtas, Pabbi Hills and ManglaSamwal area (Keller et al. 1977; Opdyke et al. 1979; Hussain et al. 1992). Thick alternations of grey and brown sandstone and red and brown mudstone with prominent conglomerate interbeds in the uppermost parts characterise the Upper Siwalik sequence. The lithofacies changes from multi-storey exten-
sire sheet sandbodies in the lower parts to the mudstone dominated upper part, and the varying composition and proportional increase of conglomeratic interbeds in the higher parts of the sequence reflect changing patterns of the fluvial system in the area. These phenomena have been interpreted to be controlled by appearances of tectonic elements in the Himalayas, such as the uplift of the Pir Panjal and the Margalla hills in the immediate northern vicinity of the area. The 850 meter thick Upper Siwalik rocks in the Mangla-Samwal anticline have been divided into a lower Samwal Formation, a middle Kakra Formation and an upper dominantly conglomeratic Mirpur Formation (Arif 1985). The entire sequence ranges in age from ca. 3.4 Ma to 1.0 Ma (Johnson et al. 1979): the Samwal-Kakra Formation boundary is around 1.6 Ma (Arif 1985). Hussain et al. (1992) have reported a rich mammalian fauna from the Samwal Formation including Crocuta, Stegodon, Elephas, Equus, Sivatherium, Hexaprotodon and Hemibos besides other bovines, antelopes, giraffes, and rhinoceroses. This assemblage corresponds with the Elephas planifrons Interval Zone of Barry et al. (1982). Hussain et al. (1992), however, suggested a subdivision into a lower E. planifrons Range Zone
116 c)
Islamabad%
flY. :. . . . .
/'
Chart
\
•'i::. •, • • . .. c . ...,.'
•
\.
.
.i ';. • • : ..j • ...
Kabal ,.
Colony
.:.','
..~
' . ' .
I FOSSIL LOCALITIES
" ~" '" ' ' .]:-'."' [""i '"( '!',...:..'.. ~ " " "".:' '/'~'"":"/":"-.""-"~ ). "K"~ ."' "'""~ , : .amwa~' "v .:" ~:/~" " ; ~ " : >
VIL~GE/TO~ JHELUN
RIVER
~"...
:.i.i)~! ":,!:. :.ii.i;!::?..:L
BED
"
" ' ~r"~'~
.~k-
' "" J
Okm
1
2
3
4
I
I
I
I
1
FIGURE 1 - Map of parts of eastern Potwar Plateau Southwestern Kashmir showing the ibssil sites (HGSP 8929 and PMNH 8802). Carte du Potwar E s t et du K a s h m i r Ouest montrant les localit~s fossilif~res.
(between 3.3 Ma to 2.7 Ma) and an upper E. hysudricus Range Zone (between 2.7 Ma to 1.6 Ma). The small mammals reported here come from two localities, P M N H 8802 and HGSP 8929, in the Samwal Formation (Fig. 1). The PMNH 8802 site includes a few fragmentary molluscan shells and fish and reptilian bones, in a yellowish grey sandstone bed. HGSP 8929, however, has also produced some larger mammals. The stratigraphic position of these localities vis-a-vis the Johnson et al. (1979) magnetic polarity stratigraphy of the ManglaSamwal area suggests an age for PMNH 8802 of ca. 3.0 Ma and for HGSP 8929 an age of 2.6 Ma. This is the first report of a small mammal fauna from the Mangla-Samwal area and also for the late Pliocene of the Potwar-Kashmir area. Paleontologists in India are actively investigating late Pliocene rocks of the Karewas (Kotlia 1992) and around Saketi Park (Rajeev Patnaik, unpublished dissertation).
SYSTEMATIC PAT,AEONTOLOGY
D e s c r i p t i o n - This specimen h a s a distinctive preservation. It is a lustrous red-brown, thin bone being translucent; t h e r e is no indication of original tooth pigmentation. The caronoid and condyle are broken, so identification is b a s e d on the molar. The tooth is relatively u n r e d u c e d , b u t v e r y narrow. Its well developed trigonid blade b e t w e e n protoconid and p a r a c o n i d curves gently anterolingually. The m u c h s h o r t e r protoconid- metaconid blade is t r a n s v e r s e in orientation and the m e t a c o n i d is the s m a l l e s t of t h e s e t h r e e cusps. The m o d e r a t e l y s h o r t talonid is modified in structure. Its large single cusp, posterior in position, is n e a r l y on a level w i t h the metaconid.A major crest dips from it to the base of the metaconid. A second low crest descends anteriorly down the lingual side of the talonid cusp. There is a posterolabial cingulum on the talonid, none lingually, and a h e a v y anterolabial cingulum continuous w i t h a distinct ridge anterior to the paraconid.
Order INSECTIVORA Family SORICIDAE Genus S u n c u s EHRENBERG,1833
, ,.,/~t~
i!~
cf. Suncus sp. Fig. 2 m a t e r i a l - PMNH 5001, right dentary fragment bearing M/3. Referred
L o c a l i t y - HGSP 8929.
FIGURE 2 - cf. Suncus sp. (A-B) PMNH 5001, right M/3. Bar represents 1 mm.
117 FIGURE 3 - Golunda kelleri P M N H 5002, left M/1 ; P M N H 5003~ left M/2 ; P M N H 5004, left M/3 ; P M N H 5005, r i g h t M \ I . barre repr~sente 1 mm.
(A) (B) (C) (D) La
JAIl\ •..~
~f ~
.~
C B
Order RODENTIA Family MURIDAE Genus Golunda GRAY, 1837
Golunda kelleri JACOBS, 1978 Fig. 3 R e f e r r e d m : ~ t e r i a l - Four isolated teeth, P M N H 5002, 5003, 5004 and 5005. left for M/l, M/2 a n d M/3, and r i g h t M \ I respectively. L o c a l i t y - P M N H 8802.
D e s c r i p t i o n - The teeth range in color from light to dark chocolate brown. The M/1 and M/2 are similar in light brown coloration and moderately advanced wear. Root ends are broken. Cusps are arcuate to subcircular and positioned as Jacobs (1978) describes. The labial anteroconid of M/1 is smaller and posterior to the lingual anteroconid, and there is no ante M romedial cusp between them. The metaconid touches the lingual anteroconid. The hypoconid and entoconid converge in a remnant of the medial mure. A labial bulge in the hypoconid may represent a worn C1. The posterior cingulum is a small oval. The labial cingulum is greatly reduced, but there are small cuspules between labial anteroconid and protoconid, protoconid and hypoconid, and posterior to the hypoconid. There are six roots, a large anterior and two posterior roots, and three across the middle of the tooth. The protoconid and metaconid of M/2 join anteriorly, but the hypoconid, the smallest cusp, does not touch the entoconid. There is a small labial anteroconid and weak labial cingulum. There is a low anterior cingulum and small, oval posterior cingulure. A large root occurs at each corner of the square tooth, and a fifth small root is anteromedial in position. The triangular M/3 is lightly worn but corroded. There are three cusps, unconnected protoconid and metaconid, and medially placed entoconid. The anterior cingulum is clear but a posterior cingulum is not so evident. A small, labial cuspule is considered the hypoconid (see Musser 1987).
Although damaged, M/3 appears to have h a d two large anterior roots and one or two posterior roots. The upper first molar is rounded in outline with high, strongly inclined cusps. They are weakly joined in early wear. Medial cusps are the largest and most arcuate. The anterostyle and lingual and labial anterocones compose the first chevron; both anterostyle and labial anterocone are slightly posterior to the lingual anterocone. In the second chevron, the enterostyle and paracone are both slightly posterior to the protocone. The third chevron includes only two cusps, a large hypocone and very small metacone joined by a narrow posterior connection. The metacone is lateral to the hypocone and abuts the back of the paracone. There is no posterior cingulum. The base of this specimen is damaged and slightly corroded but at least five major roots were present. Genus Hadromys
THOMAS,1911
cf. Hadromys s p . Fig. 4 R e f e r r e d m a t e r i a l - P M N H 5006, left M \ 1 f r a g m e n t . L o c a l i t y - P M N H 8802.
D e s c r i p t i o n - This dark brown tooth fragment shows light wear. Its cusps are tall and slightly inclined. The first chevron is broken away. Cusps of the second chevron are strongly joined by high connections. The protocone is the biggest cusp in the second chevron, followed by enterostyle, then paracone. Both enterostyle and paracone are
A
B
FIGURE 4 - cf. Hadromys sp. (A-B) P M N H 5006, left M \ I . B a r r e p r e s e n t s 1 mm.
118 slightly posterior to the protocone. The conical hypocone dominates the third chevron, which is well separated by a deep cleft from the second. There is a small metacone oppressed to the labial side of the hypocone; there is no lingual Style. There is no posterior cingulum. Genus Mus
LINNAEUS,
1758
Mus sp. Fig. 5 R e f e r r e d m a t e r i a l - PMNH 5007, left M \ I fragment. L o c a l i t y - PMNH 8802.
D e s c r i p t i o n - This small specimen lacks its first chevron. The paracone is firmly joined to the protocone. The enterostyle is more separated and more posterior than paracone and a posterolingual spur. The hypocone is large and weakly joined to the enterostyle spur. The posterolabial metacone is unreduced. There is no posterior cingulum. D i s c u s s i o n - The specimens described here come from two localities in the Mirpur area and indicate considerable potential to build a Pliocene biostratigraphy for small mammals of northern Pakistan. Only one identifiable specimen, the shrew, cf. Suncus sp. comes from locality 8929. The Mirpur shrew M/3 is very narrow (Table 1) with a simple talonid dominated by a single cusp and high crest. The trigonid arms diverge broadly and the protoconid is high. The anterolabial cingulum is prominent. The specimen resembles some of the smaller species of Suncus, but its M/3 is distinctive in its narrowness and broad trigonid. The six murid specimens indicate three species, all from locality P M N H 8802 (Table 1). This locality correlates roughly to the top of the Mammoth
event in the section of H u s s a i n et al. (1992), ca. 3 Ma. Four specimens are assigned to Golunda kelleri, a species recorded in the younger locality DP 24 (Pabbi Hills area, Fig. 1; Jacobs 1978), which occurs m a g n e t o - s t r a t i g r a p h i c a l l y above the Olduvai event. Whether slight morphological differences in molars of the two samples reflect individual variation or time remains to be seen. The Mirpur M/1-2 have a slightly stronger labial cingulum and are possibly lower crowned, although wear stage makes this uncertain. This sample shows the diagnostic feature of G. kelleri, a centrally placed entoconid on M/3. The Mirpur Golunda adds the upper first molar, which was lacking in the original hypodigm of G. kelleri described by Jacobs (1978). This tooth differs from that of extant G. ellioti in that it is more elongated anteroposteriorly, with chevrons less crowned, and its metacone is less reduced. It is similar in length/width proportions to Golunda gurai SABATIER(1982), but at the small end of the size range for that species. Central cusps on M \ 1 are more crescentic in G. kelleri, labial cusps are smaller and rounderounded, and the metacone is smaller than in G. gurai. Musser (1987) rejected the latter as a member of Golunda. The cf. Hadromys sp. is smaller, lower crowned than Hadromys loujacobsi. Similarly, the chevrons are less crowned and slightly more curved. Whether the Mirpur taxon is generically distinct from Hadromys awaits more fossil finds. Whether it precedes or is a c o n t e m p o r a r y of early Hadromys awaits more field work. In any case, it indicates that the Hadromys lineage was present by 3 Ma. The third murid from P M N H 8802 represents a species of Mus based on cusp morphology and lack of a posterior cingulum. It is slightly smaller than Mus auctor and Mus sp. of DP 24. With reference to murids from localities DP 13
Element
Length
Width u
cf. S u n c u s sp.
M/3
1.15
Golunda kelleri
M\I M/1 M/2 M/3
2.80 2.60 1.95
cf. Hadromys sp.
M\ 1 (2nd chevron)
1.70
M u s sp.
M\ 1 (and chevron)
1.03
o
•
0.45 20.5 1.90 2.00 ca.1.70 ca.170
TABLE 1 - Measurements (mm) of molars described from localities HGSP 8929 and PMNH 8802. Dimensions (ram) des molaires ddcrites dans les deux loealitds.
FIGURE 5 - M u s sp. PMNH 5007, left M \ I . Bar represents 1 m m .
119
and DP 24 Fig. 1; Jacobs (197S), 5.7 and 1.6 Ma, respectively, the f a u n a of PMNH 8802 is consistent with its intermediate age. All three sites contain Mus. DP 13 is characterized by the presence of Parapelomys robertsi and the last record of Karnimata. DP 24 contains Hadromys and Golunda. The P M N H 8802 site contains a Hydromys-like form more primitive t h a n H. loujacobsi, and the first record of Golunda. That Golunda kelleri is known from localities at 3.0 and 1.6 Ma shows t h a t this species survived the end-Pliocene interval of climatic change t h a t effected much of the globe. It also establishes an earlier Asian record of Golunda, consistent with postulated origin of the genus on t h a t continent (Jacobs 1978). A c k n o w l e d g m e n t - We thank Dr. Farhat Hussain, Director General, Geological Survey of Pakistan, Dr. Shahzad A. Mufti, Director General, Pakistan Museum of Natural History, and Dr. Taseer Hussain, Howard University, for their encouragement and permission to undertake this study. We are also indebted to Mr. Mohammad Arif and (Late) M. Asif Jah, of Geological Survey of Pakistan, for their assistance in the field. Dr. Nayyer Iqbal has very kindly provided the French translation. We would also thank S. Jaffer Hussain Naqvi for typing the manuscript and Mrs Samyia Haroon for drawing the figures. Geobios reviewers, Dr. Hans de Bruijn and Dr. Pierre Mein, are thanked for their many useful comments. This research was supported by the Geological Survey of Pakistan, the Pakistan Museum of Natural History and by Smithsonian Institute's grants to Dr. Taseer Hussain and to Dr. David Pilbeam.
REFERENCES ARIF M. 1985 - S t r a t i g r a p h y a n d chronology of verteb r a t e fossil localities i n the M i r p u r area, Azad Kashmir. Geological Survey of Pakistan Information Release, 264: 1-13. BARRYJ.C., LINDSAYE.H. & JACOBS L.L. 1982. A biostratigraphic zonation of the middle a n d upper Siwaliks of t h e P o t w a r P l a t e a u of n o r t h e r n P a k i s t a n .
Palaeogeography, Palaeoclimatology, Palaeoecology, 37: 95-130.
FLYNN L.J., PILBEAM D., JACOBS L.L., BARRY J.C., BEHRENSMEYERA.K. & KAPPELMAN J. 1990 - The Siwaliks time and f a u n a s i n a Miocene terrestrial setting. Journal of Geology, Chicago, 98: 589-604. HUSSAIN S.W., VAN DEN BERGH G.D., STEENSMA K.J., DE VISSER J.A., DE VOS J., ARIF M., vAN DAM J., SONDAAR P.Y. & MALIK S.B. 1992 - Biostratigraphy of the PlioPleistocene c o n t i n e n t a l s e d i m e n t s (Upper Siwaliks) of the M a n g l a - S a m w a l Anticline, Azad Kashmir,
Pakistan. Proceedings of the Koninklijke nederlandse Akademie van wetenschappen, 95: 65-80. JACOBS L.L. 1978 - Fossil Rodents (Rhizomyidae a n d Muridae) from Neogene Siwalik deposits, Pakistan.
Museum Northern Arizona, Flagstaff, Bulletin Series, 52, 103 p. JACOBS L.L., FLYNN L.J. & DOWNS 1989 - Neogene rodents of s o u t h e r n Asia. In C.C. Black & M.R. Dawson (eds.), Papers on fossil rodents in honor of Albert E l m e r Wood. Natural History Museum, Los Angeles, Sciences Series, 33: 157-177.
JOHNSON G.D., JOHNSON N.M., OPDYKE, N.D. & TAHIRKHELI R.A.K. 1979 - Magnetic reversal stratigraphy and s e d i m e n t a r y tectonic history of the U p p e r Siwalik Group, e a s t e r n Salt R a n g e a n d s o u t h w e s t e r n Kashmir. In A. FARAH & K.A. DE JONQ (eds), Geody-
namics of Pakistan Geological Survey of Pakistan: 149-165.
KELLER H.M., TAHIRKHELI R.A.K., MIRZA M.A., JOHNSON G.D., JOHNSON N.M. & OPDYKE N.D. 1977 - Magnetic polarity stratigraphy of the upper Siwalik deposits, Pabbi Hills, Pakistan. Earth and Planetary Science
Letters, 36: 187-201. Kotlia B.S. 1992 Pliocene m u r i d s (Rodentia, Mammalia) from K a s h m i r Basin, W e s t e r n India.
Neues Jahrbuch fiir Geologic Paliiontologie, Abhandlungen, 183: 339-357. M~SSER G.G. 1987 - The occurrence of Hadromys (Rodentia: Muridae) in early Pleistocene Siwalik s t r a t a in n o r t h e r n P a k i s t a n and its b e a r i n g on biogeographic affinities b e t w e e n I n d i a n a n d northeastern African r o u t i n e rodents. American Museum of Natural History, Novitates, 2883, 36 p.
OPDYKE N.D., LINDSAY E.H., JOHNSON G.D., JOHNSON N.M., TAHIRKHELI R.A.K. & MIRZA M.A. 1979 Magnetic polarity stratigraphy and vertebrate paleontology of the Upper Siwalik subgroup of northern Pakistan. Palaeogeography, Palaeoclimatology,
Palaeoecology, 27, 34 p. SABATIER M. 1982 - Les rongeurs du site plioc~ne h hominid~s de H a d a r (Ethiopie). Palaeovertebrata, 12 (1), 56 p. I.U. C H E E M A
Earth Sciences Division Pakistan Museum of Natural History Garden Avenue Shakar Parian, Islamabad, Pakistan S.M. RAZA Oil and Gas Development Corporation Islamabad, Pakistan L.Jo FLYNN Peabody Museum Harvard University Cambridge, MA 02138, U.S.A.