ARTICLE IN PRESS Fossil fruit of Cocos L. (Arecaceae) from Maastrichtian-Danian sediments of central India and its phytogeographical significance

Abstract

A fossilised palm fruit of Cocos L. (C. binoriensis sp. nov.) is reported from the Binori Reserve Forest, Ghansor, Seoni District, Madhya Pradesh, India. The fruit is a 3-dimensionally preserved drupe, ovoid with clearly visible longitudinal ridges. The husk is made up of a thin smooth exocarp and fibrous mesocarp, with vertical and horizontal fibres present on the inner surface of endocarp. The fruit is Maastrichtian-Danian in age and is the world's oldest fossil record of Cocos. The genus Cocos is now distributed in coastal areas of pantropical regions. The occurrence of Cocos along with coastal and mangrove remains such as Acrostichum, Barringtonia, Nypa, Sonneratia and marine algae Distichoplax and Peyssonellia previously recorded from Dec-can Intertrappean beds further confirms the proximity of sea in the area in central India and indicates warm and humid conditions. The presence of Cocos and previously recorded palaeoflora supports the existence of tropical wet evergreen to semi-evergreen forests at the time of deposition in the area, in contrast to the dry to moist deciduous forests existing today in central India. The probable reasons for the change in climatic conditions are withdrawal of an arm of the sea from central India, the change in latitude, and a significant uplift of the Western Ghats during post-trappean times.

Full text

ARTICLE IN PRESS
Acta Palaeobotanica 54(1), 2014
DOI: 10.2478/acpa-2014-000x
Fossil fruit of Cocos L. (Arecaceae) from
Maastrichtian-Danian sediments of central India
and its phytogeographical signicance
RASHMI SRIVASTAVA and GAURAV SRIVASTAVA
Birbal Sahni Institute of Palaeobotany, 53 University Road Lucknow 226 007, India;
e-mail: rashmi_bsip@yahoo.com, gaurav_jan10@yahoo.co.in
Received 20 October 2014; accepted for publication 14 March 2014
ABSTRACT. A fossilised palm fruit of Cocos L. (C. binoriensis sp. nov.) is reported from the Binori Reserve
Forest, Ghansor, Seoni District, Madhya Pradesh, India. The fruit is a 3-dimensionally preserved drupe, ovoid
with clearly visible longitudinal ridges. The husk is made up of a thin smooth exocarp and brous mesocarp,
with vertical and horizontal bres present on the inner surface of endocarp. The fruit is Maastrichtian-Danian
in age and is the world’s oldest fossil record of Cocos. The genus Cocos is now distributed in coastal areas of
pantropical regions. The occurrence of Cocos along with coastal and mangrove remains such as Acrostichum,
Barringtonia, Nypa, Sonneratia and marine algae Distichoplax and Peyssonellia previously recorded from Dec-
can Intertrappean beds further conrms the proximity of sea in the area in central India and indicates warm and
humid conditions. The presence of Cocos and previously recorded palaeoora supports the existence of tropical
wet evergreen to semi-evergreen forests at the time of deposition in the area, in contrast to the dry to moist
deciduous forests existing today in central India. The probable reasons for the change in climatic conditions are
withdrawal of an arm of the sea from central India, the change in latitude, and a signicant uplift of the Western
Ghats during post-trappean times.
KEYWORDS: Cocos, Arecaceae, Maastrichtian-Danian, coastal, climate, pantropical.
INTRODUCTION
The Deccan Volcanic Province of India is one
of the largest continental ood basalts in the
history of the Earth. It was formed by volcanic
eruption and the outpouring of lava in pen-
insular India, associated with the movement
of the Indian Plate over the Reunion Hotspot
(Chatterjee et al. 2013). Radiometric dating
and magnetostratigraphic studies (Keller et
al. 2009a, b, Chenet et al. 2009 and references
therein) indicate that volcanism extended from
ca 67.5 ±1 to 63 Ma, with a bulk eruption (ca
80% of the total volume of Deccan basalts) in
chron 29R (II phase) at 65 ±1 Ma (Chenet et
al. 2009). It has been observed that Deccan
volcanism and the accompanying global cli-
mate change at the K–T boundary led to the
extinction of dinosaurs and decline of planktic
foraminifera and other biota (Khosla & Sahni
2003, Keller et al. 2009a, b). According to
Cripps et al. (2005), however, Deccan volcan-
ism had hardly any effect on animal and oral
productivity. Couvreur et al. (2011) also sug-
gested constant diversication of palms (Yule
Process/Museum Model) in tropical rain forest
ecosystems until the Neogene. Our observa-
tions also suggest that palms along with other
terrestrial angiosperms continued and diver-
sied throughout the K–T mass extinction
event, as a large number of fossil palms along
with eudicots are reported from intertrappean
beds of central India (Srivastava 2011).
Deccan Intertrappean sediments were depos-
ited in lacustrine and uviatile environments
during quiescent phases between intervals of

2
ARTICLE IN PRESS
volcanic activity while the Indian Plate was
still an isolated land mass moving northwards
toward the Eurasian tectonic plate. The inter-
trappean beds sandwiched between successive
lava ows are highly fossiliferous and contain
diverse plant and animal fossils, mostly terres-
trial and usually not age-diagnostic. Earlier the
Deccan Intertrappean beds were considered to
be early Tertiary due to the predominance of
angiosperm remains (Sahni 1934, Bande 1992),
but studies of microoral and faunal (especially
dinosaur) assemblages from several intertrap-
pean localities have led to general acceptance of
Maastrichtian age for most of the intertrappean
exposures, though there are also a few Danian
indicators (Sahni 1983, Kar & Srinivasan 1988,
Khosla 1999, Khosla & Sahni 2003, Bajpai
2009, Keller et al. 2009a, Samant & Mohabey
2009, Srivastava 2011).
In the present communication we describe
a fossil fruit resembling Cocos L. from Maas-
trichtian-Danian sediments of Deccan Inter-
trappean exposures of Ghansor, Seoni Dis-
trict, Madhya Pradesh, central India. During
the period of deposition of the fossil described
here, India was already separated from the
rest of the Gondwanan continents but had not
yet collided with Asia. The fossil locality was
situated at ca 17°S palaeolatitude (ODSN) and
is now at 22°40′52″N (Fig. 1A, B).
Arecaceae/Palmae, the family of Cocos, is
monophyletic in origin and has been placed
within the commelinid clade of the monocoty-
ledons (Chase et al. 2006, Davis et al. 2006).
Palms are an important and characteristic
component of tropical rainforest ecosystems
having a pantropical distribution (Couvreur et
al. 2011). In temperate regions their diversity
Fig. 1. A. World map showing the modern distribution of Cocos nucifera (dashed line) (after Shukla et al. 2012) and present
fossil locality (red dot); B. Palaeocontinental map showing the area of the fossil locality (red dot) at 65 Ma (ODSN)

3
ARTICLE IN PRESS
is much lower and they have very limited frost
tolerance because of their architecture (a large
crown of evergreen leaves) (Tomlinson 1990,
Jones 1995, Lötschert 2006). On the basis of
the oldest reliable megafossil records from
Europe and North America (late Coniacian to
early Santonian – Berry 1914, Turonian-Cam-
panian – Kvaček & Herman 2004, Manchester
et al. 2010) and molecular phylogenetic stud-
ies (Couvreur et al. 2011), palms have been
suggested to be of Laurasian origin. However,
palm pollen are recorded from the Campanian
of Japan (Takahashi 1964) and from the Maas-
trichtian onwards all over the globe (Harley
2006). The family consists of 201 genera and
2650 species distributed in ve sub-families:
Arecoideae, Calamoideae, Ceroxyloideae, Cor-
yphoideae, and Nypoideae (Dranseld et al.
2005, Govaerts & Dranseld 2005, Mabberley
2005, Dranseld et al. 2008). In India, palms
are represented today by 20 genera and 96
species (Kulkarni & Mulani 2004).
The fruit of Cocos nucifera L., commonly
known as the coconut, is a very important
food plant having both domestic and commer-
cial importance. In view of the importance of
coconut in culture, the environment and agri-
culture, the origin and dispersal of the genus
is a much-discussed topic amongst biogeog-
raphers and palaeobotanists (Harries 1992,
Gunn et al. 2011). Earlier, on the basis of fossil
records, the genus was hypothesized to have
originated in South America (Colombia) during the
middle-late Palaeocene (Gomez-Navaro et al. 2009).
Later, Couvreur et al. (2011) used the Colombian fos-
sil record (Gomez-Navarro et al. 2009) in conjunction
with molecular phylogenetic studies to infer an age of
54.8 Ma for the stem node of sub-tribe Attaleineae,
tribe Cocoseae, to which Cocos belongs.
MATERIAL AND METHODS
The fossil fruit was collected from Ghansor village
(22°40′28″N; 80°02′E) situated in the Binori Reserve Forest
(Block No. 444–445), Seoni District, Madhya Pradesh, India
(Fig. 2). The fossiliferous Intertrappean bed, 1.5 m thick, is
associated with basal ow of the Dhuma Formation of the
Amarkantak Group of Deccan Traps comprised of chert,
cherty limestone and thin shaley fragments between cherts
(GSI, 2002). Petried palm stems and leaf remains (Sahni
1934, Guleria & Mehrotra 1999) and dicotyledonous woods
(Srivastava 2008, 2010) were earlier reported from the area.
The associated sediments attached to the fruit specimens were
removed with a ne chisel and the cleaned fruit was photo-
graphed under low-angle sunlight using a 10 mpx digital cam-
era. The holotype is housed in the museum of the Birbal Sahni
Institute of Palaeobotany. Another specimen from the same site
was sectioned for examination of anatomical structure but no
internal anatomy was preserved, so it seems that these speci-
mens represent molds and casts of successive layers within the
fruit, rather than having been permineralised.
Fig. 2. Map showing the fossil locality

4
ARTICLE IN PRESS
RESULTS
SYSTEMATIC PALAEOBOTANY
Class Monocotyledons (Liliopsida)
Family Arecaceae Schultz Schultzenst.
(nom. altern.)
Subfamily Arecoideae Grifth
Tribe Cocoseae Mart.
Subtribe Attaleineae Drude
Genus Cocos L.
Cocos binoriensis Srivastava & Srivastava
sp. nov.
(Pl. 1, gs 1, 3–4)
Etymology. The specic name is derived
from the Binori Reserve Forest.
H o l ot yp e. Museum of Birbal Sahni Institute
of Palaeobotany, No. BSIP 40107.
Diagnosis. Fruit a 3-dimensionally pre-
served drupe, ovoid, somewhat triangular,
longitudinal ridges present, apex obtuse with
depression in centre, base broad. Husk made up
of exocarp and mesocarp. Exocarp preserved in
places, smooth, thin; mesocarp brous, vertical
and horizontal bres present on inner surface
of endocarp.
T y p e l oc al i t y. Ghansor, Binori Reserve
Forest, Seoni District, Madhya Pradesh, India.
Type horizon. Deccan Intertrappean Beds.
A g e. Maastrichtian-Danian.
Description. Fruit a 3-dimensionally preser-
ved drupe, shape ovoid, somewhat triangular,
slightly broader at base, ca 11.7 cm long and
10.0 cm wide; apex obtuse with depression in
centre, two longitudinal ridges clearly visible;
base obtuse, longitudinal bre scars converging
at base (Pl. 1, gs 1, 2). Husk made up of exo-
carp and mesocarp. Exocarp preserved in places,
smooth, thin; mesocarp brous, marks of longi-
tudinal bres seen (Pl. 1, gs 3, 4); endocarp
2–4 mm thick, vertical and longitudinal bres
present on inner surface (Pl. 1, g. 4).
Botanical affinity. The diagnostic cha-
racters of the fossil fruit – ovoid, somewhat
triangular, the presence of longitudinal ridges,
a brous mesocarp surrounding a smooth endo-
carp, and relatively large size – can only be
found in Arecaceae and more specically sub-
tribe Attaleinae. Among the modern genera of
Attaleinae, only Cocos is similar to the present
fossil in terms of the aforementioned characters.
However, due to the presence of the husk (exo-
carp and mesocarp) the endocarp was not fully
exposed, so it was not possible to determine
whether the three germination pores characte-
ristic of Cocos were present in the specimen.
Fossil records of Cocos have been reported
from all of the Gondwana continents (e.g.
India, Australia, New Zealand, South America)
except Africa and Antarctica. Fossils of Cocos-
like fruits, fossilised endocarps especially, are
numerous. They are recorded from the Pliocene
of Australia as Cocos nucifera (Rigby 1995) and
from Miocene and Pliocene sediments of New
Zealand, for example C. zeylandica Berry (1926)
and Cocos fruit (Ballance et al. 1981). A mid-
dle-late Palaeocene occurrence was recently
reported from northern South America as cf.
Cocos sp. (Gomez-Navaro et al. 2009). A few
fruits resembling coconut have been described
from various Deccan Intertrappean localities of
central India under various morphotaxa such
as Palmocarpon cocoides (Mehrotra 1987),
Cocos intertrappeansis (Patil & Upadhye 1984),
Cocos nucifera-like fruit (Tripathi et al. 1999)
and Cocos pantii (Mishra 2004), while Cocos
sahnii was reported from the early Eocene of
Rajasthan (Kaul 1951, Shukla et al. 2012).
Besides fruits, Sahni (1946) reported a petri-
ed palm stem of Palmoxylon (Cocos) sunda-
ram, and Bonde et al. (2004) reported a basal
portion of a stem with roots closely resembling
modern Cocos from the Deccan Intertrappean
sediments of central India.
The fossil fruit Palmocarpon cocoides
(Mehrotra 1987) cannot be compared with the
present specimen because of its different shape
and insufcient details, while Cocos intertrap-
peansis (3.0 × 5.0 cm, Patil & Upadhye 1984)
and C. zeylandica (2.5 × 5.0 cm; Berry 1926)
differ in being smaller. Cocos pantii is larger
(10.0–15.0 × 8.0–13.0 cm, Mishra 2004), and
the Cocos nucifera-like fruit and the above spe-
cies are based on anatomical characters. Cocos
sahnii from the Eocene of Rajasthan is based
on impressions of the endocarp and mesocarp
and differs in shape from the present fos-
sil. The compressed fruit cf. Cocos sp. from

5
ARTICLE IN PRESS
Colombia shows similar external morphology
but is larger (15 × 25 cm). Therefore a new
species, Cocos binoriensis sp. nov., has been
established. The specic epithet refers to the
locality of the fossil.
DISCUSSION
The origin and dispersal of coconut palms
are still a matter of debate and interest in
view of its dispersal by sea currents (Mahabale
1978, Ward & Brookeld 1992) and its ability
to germinate even after oating in sea water
for 110 days (Edmondson 1941). Some authors
believe that it originated in western Pacic
islands of tropical Asia, Polynesia or Melane-
sia (Beccari 1963, Corner 1966, Moore 1973,
Harries 1978), from where it was dispersed,
mainly via oceanic currents, to sandy and cor-
alline tropical coasts, but this is not supported
by the fossil record. Others suggest a South
American origin (Gomez-Navaro et al. 2009)
and a later range extension to the Indo-Pacic
region (Guppy 1906, Cook 1910, Gunn 2004).
The available records of Cocos known
from Australia, New Zealand and South
America indicate its distribution in different
stratigraphic successions of Cenozoic sedi-
ments (from Palaeocene to Miocene-Pliocene),
whereas records of Cocos fruit from Maastrich-
tian-Danian sediments of Deccan Intertrap-
pean beds indicate the earliest occurrence of
Cocos on the Indian Peninsula. The oldest fos-
sils of Cocos may well be of late Cretaceous
origin in the Deccan Intertrappean period,
later dispersing into Southeast Asia and other
parts of the world, rafting northwards on the
Indian plate, supporting the ‘Out of India’ dis-
persal hypothesis. Three well-supported sub-
tribes (Bactridinae, Elaeidinae, Attaleinae)
are recognised, with the groups Bactridinae
and Elaeidinae resolved as sister clades on the
basis of molecular phylogenetic analysis of the
tribe Cocoseae (Dranseld et al. 2005, 2008,
Govaerts & Dranseld 2005). Gunn (2004)
suggests that sub-tribes Bactridinae-Elaeidi-
nae diverged from Attaleinae between 50 and
60 Ma ago. The present fossil record of Cocos
from a well-dated horizon (65.5–61.7 Ma;
Gradstein et al. 2004) of Deccan Intertrappean
sediments suggests that sub-tribe Attaleinae
must have diverged earlier than 60 Ma. Phy-
logenetic analyses of palms from Colombia,
South America, also support the hypothesis
(Gomez-Navarro 2009, Futey et al. 2012).
The coconut palm is now cultivated in coastal
areas of Southeast Asia and Melanesia and has
a wide pantropical distribution (Fig. 1 A). The
occurrence of Cocos along with known coastal
and mangrove fossils like Acrostichum (Bonde
& Kumaran 2002), Barringtonia (Srivastava et
al. 2009), Nypa (Chitaley & Nambudiri 1995),
Sonneratia (Srivastava 2008) and marine algae
Distichoplax and Peyssonellia (Bande et al.
1981) indicate marine incursions in central
India. The idea of a marine seaway in central
India was originally proposed by Lakhanpal
(1970) and later substantiated by palaeontologi-
cal evidence (Sahni 1983, Keller et al. 2009a,
Bajpai 2009). Presently the area is dominated
by dry to moist deciduous forest because of the
change in the climate, which may be due to
withdrawal of an arm of the sea, the change in
latitude, and a signicant uplift of the Western
Ghats during post-trappean times (Bande &
Prakash 1982, Gunell et al. 2003).
ACKNOWLEDGMENTS
We are grateful to Prof. Sunil Bajpai (Director,
Birbal Sahni Institute of Palaeobotany, Lucknow) for
constructive suggestions and permission to publish
the paper (Ref. BSIP/RPCC/2012-114), Prof. Steven
R. Manchester (Florida Museum Natural History) for
valuable comments and suggestions, Dr. Dhananjay
M. Mohabey (Deputy Director General, Retd., Geologi-
cal Survey of India) for valuable geological informa-
tion including the age of the area, Dr. T.L.P. Couvreur
(France) for sending a valuable palm reprints and Mr.
L.K. Chowdhury (Chief Forest Conservator) permit-
ting us to collect plant fossils from the Binori Reserve
Forest, Ghansor, Seoni Circle, Madhya Pradesh. R.S.
thanks Dr. Dasharath K. Kapgate (J.M. Patel College,
Bhandara, Maharastra) for accompanying her during
the eld work.
REFERENCES
BAJPAI S. 2009. Biotic perspective of the Deccan vol-
canism and India–Asia collision: Recent advances.
In: Mukunda N. (ed.), Current Trends in Science
– Platinum Jubilee Spec. Publ., Indian Acad. Sci.,
505–516.
BALLANCE P.F., GREGORY M.R. & GIBSON G.W.
1981. Coconuts in Miocene turbidities in New Zea-
land: possible evidence for tsunami origin of same
turbidity currents. Geology, 9: 592–598.
BANDE M.B. 1992. The Palaeogene vegetation of Pen-
insular India (megafossil evidences). Palaeobota-
nist, 40: 275–284.

6
ARTICLE IN PRESS
BANDE M.B. & PRAKASH U. 1982. Palaeoclimate
and palaeogeography of central India during Early
Tertiary. Geophytology, 12(2): 152–165.
BANDE M.B., PRAKASH U. & BONDE S.D. 1981.
Occurrence of Peyssonnelia and Distichoplax in the
Deccan Intertrappeans with remarks on the age of
Chhindwara traps and palaeoecology of the region.
Geophytology, 11(2): 182–188.
BECCARI O. 1963. The origin and dispersion of Cocos
nucifera. Principes, 7: 57–69.
BERRY E.W. 1914. The Upper Cretaceous and Eocene
oras of South Carolina and Georgia. US Geol.
Surv. Prof. Paper, 84: 5–200.
BERRY E.W. 1926. Cocos and Phymatocaryon in the Pli-
ocene of New Zealand. Am. Jour. Sci., 12: 181–184.
BONDE S.D., GAMRE P.G. & MAHABALE T.S. 2004.
Further contribution to Palmoxylon (Cocos) sunda-
ram Sahni: Structure of the rooting base and its
afnities: 229–235. In: Srivastava P.C. (ed.), Vistas
in Palaeobotany and Plant Morphology: Evolution-
ary and Environmental Perspectives. Prof. D.D.
Pant Commemoration Vol.
BONDE S.D. & KUMARAN K.P.N. 2002. A perminer-
alised species of mangrove fern Achrostichum from
the Deccan Intertrappean beds of India. Rev. Pal-
aeobot. Palynol., 120: 285–299.
CHASE M.W., FAY M.F., DEVEY D.S., MAURIN O.,
RONSTED N., DAVIES J. & PILLON Y. 2006.
Multigene analyses of monocot relationships: A
summary. Aliso, 22: 63–75.
CHATTERJEE S., GOSWAMI A. & SCOTESE C.R.
2013. The longest voyage: Tectonic, magmatic, and
palaeoclimatic evolution of Indian plate during its
northward ight from Gondwana to Asia. Gond.
Res., 23: 238–267.
CHENET A., COURTILLOT V., FLUTEAU F.,
GÉRARD M., QUIDELLEUR X., KHADRI S.F.R.,
SUBBARAO K.V. & THORDARSON T. 2009. Deter-
mination of rapid Deccan eruptions across the Cre-
taceous-Tertiary boundary using palaeomagnetic
secular variation: 2. Constraints from analysis of
eight new sections and synthesis for a 3500-m-thick
composite section. J. Geophysical Res., 114: B06103.
CHITALEY S. & NAMBUDIRI E.M.V. 1995. Anatomy
of Nypa fruits reviewed from new specimens from
the Deccan Intertrappean ora of India. 83–94. In:
Pant D.D. (ed.), Proc. Internatn. Conf. on Global
Environment and Diversication of Plants through
Geological Time. Allahabad.
COOK O.F. 1910. History of the coconut palm in
America. Contributions from the U. S. National
Herbarium, 14: 271–342.
CORNER E.J.H. 1966. The natural history of palms.
London, UK: Weidenfeld and Nicholson.
COUVREUR T.L.P., FOREST F. & BAKER W.J. 2011.
Origin and global diversication patterns of tropi-
cal rain forests: Inferences from a complete genus-
level phylogeny of palms. BMC Biology, 9: 44.
CRIPPS J.A., WIDDOWSON M., SPICER R.A. &
JOLLY D.W. 2005. Coastal ecosystem response to
late stage Deccan Trap volcanism: the post K-T
boundary (Danian) palynofacies of Mumbai (Bom-
bay), west India. Palaeogeogr. Palaeoclimat. Pal-
aeoecol., 216: 303–332.
DAVIS J.I., PETERSEN G., SEBERG O., STE-
VENSON D.W., HARDY C.R., SIMMONS M.P.
MICHELANGELI F.A., GOLDMAN D.H., CAMP-
BELL L.M., SPECHT C.D. & COHEN J.I. 2006.
Are mitochondrial genes useful for the analysis of
monocot relationships? Taxon, 55(4): 857–870.
DRANSFIELD J., UHL N.W., ASMUSSEN C.B.,
BAKER W.J., HARLEY M.M. & LEWIS C.E. 2005.
A new phylogenetic classication of the palm fam-
ily, Arecaceae. Kew Bull., 60: 559–569.
DRANSFIELD J., UHL N.W., ASMUSSEN C.B.,
BAKER W.J., HARLEY M.M. & LEWIS C.E. 2008.
Genera palmarum: The evolution and classication
of palms. Royal Botanic Gardens, Kew, UK.
EDMONDSON C.H. 1941. Viability of coconut seed
after oating in sea. Occasional papers of Bermice
P. Bishop Museum Honolulu, Hawaii, 16: 293–304.
FUTEY M.K., GANDOLFO M.A., ZAMALOA M.C.,
CÚNEO R. & CLADERA G. 2012. Arecaceae fossil
fruits from the Paleocene of Patagonia, Argentina.
Bot. Rev., 78: 205–234.
GOMEZ-NAVARRO C., JARAMILLO C., HERRERA F.,
WING S.L. & CALLJAS R. 2009. Palms (Arecaceae)
from a Palaeocene rainforest of northern Colombia.
Am. J. Bot., 96: 1300–1312.
GOVAERTS R. & DRANSFIELD J. 2005. World Check-
list of Palms. Royal Botanic Garden, Kew, UK.
GRADSTEIN F.M., OGG J.G. & SMITH A. 2004.
A Geologic Time Scale. Cambridge, UK: Cambridge
University Press.
G.S.I. 2002. District resource map of Seoni District,
Madhya Pradesh.
GULERIA J.S. & MEHROTRA R.C. 1999. On some
plant remains from Deccan intertrappean localities
of Seoni and Mandla districts of Madhya Pradesh,
India. Palaeobotanist, 47: 68–87.
GUNELL Y., GALLAGHER K., CARTER A., WID-
DOWSON M. & HURFORD A.J. 2003. Denudation
history of the continental margin of western pen-
insular India since the early Mesozoic-reconciling
apatite ssion-track data with geomorphology.
Earth Planet. Sci. Lett., 215: 187–201.
GUNN B.F. 2004. The phylogeny of the Cocoseae (Are-
caceae) with emphasis on Cocos nucifera. Ann. Mis-
souri Bot. Gard., 91: 505–522.
GUNN B.F., BAUDOUIN L. & OLSEN K.M. 2011.
Independent Origins of Cultivated Coconut (Cocos
nucifera L.) in the Old World Tropics. PLOS ONE,
6(6): e21143. doi:10.1371/journal.pone.0021143
GUPPY H.B. 1906. Observations of a Naturalist in the
Pacic between 1896 and 1899. Plant dispersal, 2.
London, U.K.: Macmillan.

7
ARTICLE IN PRESS
HARLEY M.M. 2006. A summary of fossil records for
Arecaceae, Botanical Journal of the Linnean Soci-
ety, 151: 39–67.
HARRIES H.C. 1978. The evolution, dissemination,
and classication of Cocos nucifera L. Bot. Rev.,
44: 265–319.
HARRIES H.C. 1992. Biogeography of the Coconut
Cocos nucifera L. Principes, 36(3): 155–162.
JONES D.L. 1995. Palms throughout the world.
Chatswood: Reed Books.
KAR R.K. & SRINIVASAN S. 1988. Late Cretaceous
palynofossils from the Deccan Intertrappean beds
of Mohgaon-Kalan, Chhindwara District, Madhya
Pradesh. Geophytology, 27: 17–22.
KAUL K.N. 1951. A palm fruit from Kapurdi (Jodh-
pur, Rajasthan Desert) Cocos sahnii sp. nov. Curr.
Sci., 20: 138.
KELLER G., ADATTE T., BAJPAI S., MOHABEY
D.M., WIDDOWSON M., KHOSLA A., SHARMA
R., KHOSLA S.C., GETSCH B., FLEITMANN D. &
SAHNI A. 2009a. K-T transition in Deccan Traps
of central India marks major marine seaway across
India. Earth Planet. Sci. Lett., 282: 10–23.
KELLER G., SAHNI A. & BAJPAI S. 2009b. Deccan
Volcanism, the KT Mass Extinction and Dinosaurs.
J. Biosci., 34 (5): 709–728.
KHOSLA A. & SAHNI A. 2003. Biodiversity during
the Deccan volcanic eruptive episode. J. Asian
Earth Sci., 21: 895–908.
KHOSLA S.C. 1999. Costabuntonia, a new genus of
Ostracoda from the Intertrappean beds (Paleo-
cene) of east coast of India. Micropaleontology, 45:
319–323.
KVAČEK J. & HERMAN A.B. 2004. Monocotyledons
from Early Campanian (Cretaceous) of Grunbach,
Lower Austria. Rev. Palaeobot. Palynol., 128: 323–
353.
KULKARNI A.R. & MULANI R.M. 2004. Indigenous
palms of India. Curr. Sci., 86(12): 1598–1603.
LAKHANPAL R.N. 1970. Tertiary oras of India and
their bearing on the historical geology of the region.
Taxon, 19(5): 675–694.
LÖTSCHERT W. 2006. Palmen: Botanik, Kultur, Nut-
zung. Stuttgart: Ulmer.
MABBERLEY D.J. 2005. The plant book, a portable
dictionary of the vascular plants. Cambridge: Cam-
bridge University Press.
MAHABALE T.S. 1978. The origin of Coconut.
Palaeobotanist, 25: 238–248.
MANCHESTER S.R., LEHMAN T.M. & WHEELER
E.A. 2010. Fossil palms (Arecaceae, Coryphoideae)
associated with juvenile herbivorous dinosaurs in
the upper Cretaceous Aguja Formation, Big Bend
National Park, Texas. Int. J. Plant Sci., 171(6):
679–689.
MEHROTRA R.C. 1987. Some new palm fruits from
the Deccan Intertrappean beds of Mandla District,
Madhya Pradesh. Geophytology, 17: 204–208.
MISHRA S.N. 2004. Cocos pantii sp. nov. The Tertiary
counterpart of modern coconut fruit from Amar-
kantak, India. 237–239. In: Srivastava P.C. (ed.),
Vistas in Palaeobotany and Plant Morphology:
Evolutionary and Environmental Perspectives.
U.P. Offset: India.
MOORE H.E. 1973. The major groups of palms and
their distribution. Gentry Herbarium, 11: 27–141.
ODSN – www.odsn.de/odsn/index.html
PATIL G.V. & UPADHYE E.V. 1984. Cocos-like fruit
from Mohgaonkalan and its signicance towards
the stratigraphy of Mohgaonkalan Intertrap-
pean beds: 541–554. In: Sharma A.K., Mitra G.C.,
Banerjee M. (eds), Proc. Symp. Evolutionary Bot-
any and Biostratigraphy, New Delhi: Today and
Tomorrow’s Printers and Publishers.
RIGBY J.F. 1995. A fossil Cocos nucifera L. fruit
from the latest Pliocene of Queensland, Australia.
379–381. In: Pant D.D., Nautiyal D.D., Bhatnagar
A.N., Surange K.R., Bose M.N. & Khare P.K. (eds),
Birbal Sahni Centenary Volume, South Asian Pub-
lisher: Allahabad, India.
SAHNI A. 1983. Upper Cretaceous palaeobiogeography
of peninsular India and the Cretaceous-Paleocene
transition: The vertebrate evidence. 128–140. In:
Maheshwari H.K. (ed), Cretaceous of India. Indian
Association of Palynostratigraphers. Lucknow.
SAHNI B. 1934. The Deccan Traps: Are they Creta-
ceous or Tertiary? Curr. Sci., 3: 134–136.
SAHNI B. 1946. A silicied Cocos-like palm stem, Pal-
moxylon (Cocos) sundaram, from the Deccan Inter-
trappean Beds. J. Indian Bot. Soc., 26: 361–374.
SAMANT B. & MOHABEY D.M. 2009. Palynoora
from Deccan volcano-sedimentary sequence (Cre-
taceous-Palaeogene transition) of central India:
implications for spatio-temporal correlation. J.
Biosci., 34(5): 811–823.
SHUKLA A., MEHROTRA R.C. & GULERIA J.S.
2012. Cocos sahnii Kaul: A Cocos nucifera L.-like
fruit from the Early Eocene rainforest of Rajast-
han, western India. J. Biosci., 37(4): 769–776.
SRIVASTAVA R. 2008. Fossil woods resembling Son-
neratia with fungal infection from Deccan Inter-
trappean sediments of Seoni District, Madhya
Pradesh. Geophytology, 37: 87–92.
SRIVASTAVA R. 2010. Fossil dicotyledonous woods
from Deccan Intertrappean sediments of Ghansor,
Seoni District, Madhya Pradesh, India. Palaeobot-
anist, 59: 129–138.
SRIVASTAVA R. 2011. Indian upper Cretaceous-Ter-
tiary ora before collision of Indian Plate: A reap-
praisal of Central and Western Indian Flora. Mem.
Geol. Soc. India, 77: 281–292.
SRIVASTAVA R., KAPGATE D.K. & CHATERJEE S.
2009. Permineralised fungal remains in the fossil
wood of Barringtonia from the Deccan Intertrap-
pean sediments of Yavatmal District, Maharash-
tra, India. Palaeobotanist, 58: 11–19.

8
ARTICLE IN PRESS
TAKAHASHI K. 1964. Sporen und Pollen der Ober-
kretaz-eischen Hakobuchi-Schichtengruppe, Hok-
kaido. Mem. Fac. Sci. Kyushu Univ., Series D,
Geology, 14: 159–271.
TOMLINSON P.B. 1990. The structural biology of
palms. Oxford: Clarendon Press.
TRIPATHI R.P., MISHRA S.N. & SHARMA B.D.
1999. Cocos nucifera like petried fruit from the
Tertiary of Amarkantak, M.P., India. Palaeobota-
nist, 48: 251–255.
WARD R.G. & BROOKFIELD M.N. 1992. The disper-
sal of the coconut: Did it oat or was it carried to
Panama. J. Biogeogra., 19: 467–480.
Plate 1
Cocos binoriensis sp. nov.
1 Fossil fruit showing shape, size and two longitudinal ridges (red arrows).
2. Modern fruit of Cocos nucifera showing its shape, size and longitudinal ridges, similar to the fossil (red and
yellow arrows).
3. Longitudinally broken part of the fossil showing brous mesocarp (red arrows) and endocarp with horizontal
and longitudinal bres (black arrows).
4. Counterpart of the same fossil fruit showing inner surface of the endocarp having horizontal and longitudinal
bres (red arrows), brous mesocarp (black arrows) and basal portion (blue arrow).

Plate 1 9
R. Srivastava & G. Srivastava
Acta Palaeobot. 54(1)
ARTICLE IN PRESS