Content uploaded by Haiyan Tong
Author content
All content in this area was uploaded by Haiyan Tong on Sep 27, 2015
Content may be subject to copyright.
An early ‘ostrich dinosaur’ (Theropoda: Ornithomimosauria) from
the Early Cretaceous Sao Khua Formation of NE Thailand
ERIC BUFFETAUT
1
*, VARAVUDH SUTEETHORN
2
& HAIYAN TONG
1
1
CNRS (UMR 8538, Laboratoire de Ge
´
ologie de l’Ecole Normale Supe
´
rieure), 24 rue
Lhomond 75231 Paris Cedex 05, France
2
Bureau of Fossil Research and Museum, Department of Mineral Resources, Rama VI Road,
Bangkok 10400, Thailand
*Corresponding author (e-mail: eric.buffetaut@wanadoo.fr)
Abstract: Postcranial remains of a small theropod dinosaur, including vertebrae, incomplete
pubes, tibiae, an incomplete fibula, metatarsals and phalanges, from the Early Cretaceous Sao
Khua Formation of Phu Wiang, Khon Kaen Province, NE Thailand, are described as a new
taxon of ornithomimosaur, Kinnareemimus khonkaenensis, gen. et sp. nov. This early ‘ostrich dino-
saur’ is characterized by a fairly advanced metatarsus, in which metatarsal III, although still visible
proximally between metatarsals II and IV in cranial view, is markedly ‘pinched’ more distally and
becomes triangular in cross-section. The condition of its metatarsus shows that Kinnareemimus
khonkaenensis is more derived than the geologically younger primitive ornithomimosaurs Harpy-
mimus and Garudimimus, but less derived than Archaeornithomimus. Its occurrence in the Early
Cretaceous of Thailand suggests that advanced ornithomimosaurs may have originated in Asia.
The fossil vertebrate locality of Phu Wiang 5, in red
clays of the Sao Khua Formation in the Phu Wiang
hills near the town of Phu Wiang (Phu Wiang
District, Khon Kaen Province, NE Thailand), has
yielded a curious dinosaur assemblage consisting
mainly of disarticulated bones of large and small
individuals of the sauropod Phuwiangosaurus
sirindhornae Martin, Buffetaut & Suteethorn 1994
(Martin 1994; Martin et al. 1994, 1999), and a
small ornithomimosaur theropod (Buffetaut et al.
1995; Buffetaut & Suteethorn 1998, 1999). The
latter represents a new taxon of Ornithomimosauria
that, although geologically ancient, shows some
derived characters, and provides new evidence
about the acquisition of some peculiar characters
of the hind limb in the so-called ‘ostrich dinosaurs’.
Geographical and geological setting
The Phu Wiang hills are a vast synclinal structure in
the western part of the Khorat Plateau of NE
Thailand, located just NE of the town of Phu
Wiang, and about 100 km NNE of the provincial
capital Khon Kaen. Several of the non-marine for-
mations of the Mesozoic Khorat Group crop out in
the Phu Wiang syncline, with the Aptian Khok
Kruat Formation occupying the centre, followed
towards the periphery (and therefore of increasing
geological age) by the Phu Phan, Sao Khua and
Phra Wihan Formations. Dinosaur footprints are
known from the Phra Wihan Formation at Phu
Wiang, but the most productive formation in terms
of vertebrate remains is the Sao Khua Formation.
The first dinosaur bone identified in Thailand
(Ingavat et al. 1978) came from the Sao Khua For-
mation at Phu Wiang, which also yielded the type
skeletons of the sauropod Phuwiangosaurus sirind-
hornae Martin, Buffetaut & Suteethorn 1994, and
the theropod Siamotyrannus isanensis, as well as
abundant remains of fishes, turtles and crocodilians.
A number of vertebrate localities are known in the
Sao Khua Formation at Phu Wiang (distinguished
by numbers; see map published by Martin 1994)
and have different fossil contents, although all are
part of the same general assemblage, the differences
being largely due to taphonomic conditions.
The Sao Khua Formation is one of the most
fossiliferous formations of the Khorat Group. Its
age is not well constrained, although it can firmly
be placed in the Early Cretaceous because it overlies
the Phra Wihan Formation, which has yielded Early
Cretaceous palynomorphs (Racey et al. 1996). It is
overlain by the Phu Phan Formation, which has
yielded very few fossils (except for dinosaur foot-
prints and rare sauropod bones) and is itself overlain
by the Khok Kruat Formation, which is referred to
the Aptian on the basis of its vertebrates and paly-
nomorphs (Cappetta et al. 1990; Racey et al.
1996). The Sao Khua Formation can thus be con-
sidered as later than earliest Cretaceous and earlier
than Aptian. A Valanginian–Hauterivian age was
suggested by Buffetaut & Suteethorn (1999), but
the Sao Khua Formation may be somewhat
younger (Barremian?).
From:BUFFETAUT, E., CUNY, G., LE LOEUFF,J.&SUTEETHORN, V. (eds) Late Palaeozoic and Mesozoic
Continental Ecosystems in SE Asia. The Geological Society, London, Special Publications, 315, 229 –243.
DOI: 10.1144/SP315.16 0305-8719/09/$15.00 # The Geological Society of London 2009.
by guest on September 15, 2015http://sp.lyellcollection.org/Downloaded from
The depositional environment of the Sao Khua
Formation was discussed at some length by
Mouret et al. (1993) and Racey et al. (1996); it is
considered as having been deposited on an extensive
floodplain of a low-energy meandering river system.
At Phu Wiang 5, the fossil-bearing layer,
exposed in a small creek, consists of red clays in
which dinosaur bones were found scattered. None
of the theropod bones described below were found
in articulation, but they almost all come from a
small excavation (Phu Wiang 5A), not more than
2m
2
, where they were mingled with bones of very
young specimens of Phuwiangosaurus (described
by Martin 1994; Martin et al. 1999). There is
every reason to believe that all the ornithomimosaur
bones from that locality are from a few individuals
of a single taxon. Other vertebrate fossils found at
Phu Wiang 5 include a tooth of the spinosaurid ther-
opod Siamosaurus, crocodile teeth, turtle plates and
bivalves (Martin et al. 1999).
Systematic description
Dinosauria Owen 1842
Theropoda Marsh 1881
Coelurosauria von Huene 1914
Ornithomimosauria Barsbold 1976
Kinnareemimus gen. nov.
Etymology. From Kinnaree, graceful beings of Thai
mythology, with the body of a woman and the legs
of a bird, said to inhabit the depths of the legendary
Himmapan Forest, by allusion to the bird-like feet of
this dinosaur.
Diagnosis. As for type species.
Kinnareemimus khonkaenensis sp. nov.
Etymology. From Khon Kaen Province in NE
Thailand, where the type material was found.
Diagnosis. An ornithomimid dinosaur in which
metatarsal III is visible in cranial view between
the proximal ends of metatarsals II and IV (unlike
the condition in Archaeornithomimus and more
derived ornithomimids), but becomes rod-like dis-
tally and expands again, with a triangular cross-
section, closer to the distal end (unlike the condition
in Garudimimus and Harpymimus).
Locality and horizon. Locality Phu Wiang 5, Phu
Wiang district, Khon Kaen Province, Thailand;
Sao Khua Formation, Early Cretaceous.
Holotype. An incomplete left third metatarsal, with
the distal end and part of the shaft (PW5A-100).
Referred material. The distal end of a right third
metatarsal (PW5A-107); the middle part of a left
third metatarsal (PW5A-103); the proximal end of
a left third metatarsal (PW5A-104); the proximal
end of a right third metatarsal (PW5A-131); a com-
plete left second metatarsal (PW5A-101); a nearly-
complete left second metatarsal (PW5A-105); the
proximal end of a left fourth metatarsal (PW5A-
102); a complete right fourth metatarsal (PW5A-
106); two proximal ends of fourth metatarsals
(PW5A-108, 109); seven more or less complete
pedal phalanges (PW5A-115, 116, 117, 118, 119,
120, 121); an incomplete pedal ungual phalanx
(PW5A-122); a complete right tibia (PW5A-110);
a complete left tibia (PW5A-111); the proximal
end of a left fibula (PW5A-112); the proximal end
of a left pubis (PW5A-114); the proximal end of a
right pubis (PW5A-113); a dorsal centrum
(PW5A-123); an incomplete caudal vertebra
(PW5A-124); an anterior caudal centrum (PW5A-
130); a middle caudal centrum (PW5A-125); four
distal caudal centra (PW5A-126, 127, 128, 129).
Description of the material
Vertebrae (Fig. 1). One comparatively large, weakly
amphicoelous, hourglass-shaped centrum (PW5A-
123) may be from a posterior dorsal vertebra, as it
lacks chevron facets. The neural arch is missing,
having detached at the level of the unfused neurocen-
tral suture, suggestive of an immature individual.
Several caudal vertebrae are present. One of
the largest (PW5A-124) shows the neural arch
completely fused to the centrum, without any clear
trace of a suture, which may suggest an adult indi-
vidual. It is incomplete, only the posterior half
being preserved. The centrum is hollow, a condition
previously reported for Archaeornithomimus asiati-
cus (Gilmore 1933). There are no transverse
processes, but a longitudinal ridge is present on
each lateral face of the centrum. The neural spine
appears to have been very low, and the postzygapo-
physes are broken. The specimen generally
resembles the 15th to 17th caudals of Gallimimus
bullatus (Osmo
´
lska et al. 1972). Another similar-
sized caudal vertebra (PW5A-130) is reduced to
its amphicoelous centrum, with an open neurocen-
tral suture. The hourglass-shaped centrum, with
chevron facets at the posterior end, is relatively
short and high, and its anterior articular surface is
wider than the posterior facet. It may be a first
caudal. A small specimen (PW5A-128), which
also shows longitudinal ridges, has a low neural
arch firmly fused to the centrum. The remaining
caudal vertebrae (PW5A-126, 127) are small and
reduced to their centra, with no sign of fusion
along the neurocentral centure. The amphicoelous
centra are elongate, hourglass-shaped, with
chevron facets at both the anterior and posterior
end, linked by paired longitudinal ridges. A larger
E. BUFFETAUT ET AL.230
by guest on September 15, 2015http://sp.lyellcollection.org/Downloaded from
caudal vertebra (PW5A-125) shows chevrons only
at the posterior end and a single ventral ridge. All
the vertebral elements from Phu Wiang 5 are gener-
ally similar to the vertebrae described in other
ornithomimosaurs (Barsbold & Osmo
´
lska 1990).
Pubis (Fig. 2). The only pelvic elements present are
the proximal ends of a right (PW5A-114) and a left
(PW5A-113) pubis, broken at the beginning of the
shaft. A well-marked obturator notch is present.
The sutural area for the ilium is large and crescent-
shaped. A smooth area corresponding to the rim of
the acetabulum is short with a trapezoidal outline.
The contact area for the ischium is small and
smooth, without rugosities. By comparison with
the pubes of other ornithomimosaurs, this bone
Fig. 1. Kinnareemimus khonkaenensis gen. et sp. nov. Vertebrae in dorsal (a, c, e, g, i, k, m, o) and lateral (b, d, f, h, j, l,
n, p) views; (a, b) PW5A-123; (c, d) PW5A-130; (e, f) PW5A-124; (g, h) PW5A-125; (i, j) PW5A-126; (k, l)
PW5A-127; (m, n) PW5A-128; (o, p) PW5A-129. Scale bar: 50 mm.
Fig. 2. Kinnareemimus khonkaenensis gen. et sp. nov. Proximal ends of pubes. Left pubis PW5A-113 in lateral (a) and
proximal (b) views; right pubis PW5A-114 in lateral view (c). Scale bar: 10 mm.
EARLY OSTRICH DINOSAUR FROM THAILAND 231
by guest on September 15, 2015http://sp.lyellcollection.org/Downloaded from
appears very slender, but this may be due to imma-
turity of the specimens.
Tibia (Fig. 3). The tibia is represented by two com-
plete specimens, one from the right side (PW5A-
111) and one from the left (PW5A-110). Although
they are nearly the same length, the left tibia is
more robust than the right one and probably
belongs to a slightly older individual, although
sexual dimorphism may also be involved. The
distal articular area shows a well-defined cnemial
crest which is thickened cranially. In proximal
view, the lateral margin of the cnemial crest is mark-
edly concave, forming a distinct notch between the
fibular condyle and the cnemial crest, whereas the
medial margin is convex. The condyles for articula-
tion with the femur are separated caudally by a dis-
tinct incisura.
On the cranial surface of the shaft, the cnemial
crest decreases quickly in height, being continued
along the shaft by a faint ridge, which is more
pronounced on the more robust left tibia; a similar
condition was described by Sullivan (1997) in a
tibia from the Kirtland Formation of New Mexico
referred to Ornithomimus antiquus. The fibular
crest is prominent, extending to the proximal third
of the length of the shaft, and is more strongly devel-
oped on the left tibia, in accordance with general
greater robustness. The shaft is slender and rather
straight, with only a faint concavity of the medial
margin and convexity of the lateral margin. At
mid-shaft, the cross-section is D-shaped. The distal
end is expanded mediolaterally and compressed cra-
niocaudally. The cranial face shows a flat surface for
reception of the ascending process of the astragalus,
which is bounded medially by the raised medial rim.
The distal articular surface is approximately tri-
angular in outline, the medial margin being dis-
tinctly longer than the lateral one. The caudal face
bears a blunt ridge that is offset medially. Its
lateral rim is raised into a distinct ridge, correspond-
ing to the articulation with the fibula.
The tibiae from Phu Wiang 5 are generally
similar to those of Late Cretaceous ornithomimids.
In Archaeornithomimus asiaticus, however, the
proximal articular region shows a proximal protru-
sion that is not present in Kinnareemimus khonkae-
nensis. Comparisons with more basal forms are
limited because the tibia is unknown in Pelecanimi-
mus polyodon and Shenzhousaurus orientalis, and
very incompletely preserved in the holotype of
Harpymimus okladnikovi. Garudimimus brevipes
has a more robust tibia than Kinnareemimus khon-
kaenensis, which may be the result of differences
in individual age, but otherwise few morphological
differences can be seen.
Fig. 3. Kinnareemimus khonkaenensis gen. et sp. nov. tibiae and fibula. Left tibia PW5A-110 in cranial (a), medial (c),
caudal (e), lateral (g) and proximal (i) views. Right tibia PW5A-111 in cranial (b), medial (d), caudal (f), lateral (h) and
proximal (j) views. Proximal end of left fibula PW5A-112 in medial view (k). Scale bar: 50 mm.
E. BUFFETAUT ET AL.232
by guest on September 15, 2015http://sp.lyellcollection.org/Downloaded from
In the material from Phu Wiang 5, the astragalus
and calcaneum are not attached to the tibia, reflect-
ing the juvenile status of the specimens.
Fibula (Fig. 3). The fibula is represented only by a
left proximal end (PW5A-112), which is expanded
craniocaudally and comma-shaped in proximal
view, with the sharp end in caudal position. The
lateral face is convex and the medial face concave,
showing the inception of a broad and deep groove
that apparently extended for some distance along
the shaft. A similar groove is present in Gallimimus
bullatus (Osmo
´
lska et al. 1972).
Metatarsus. Several more or less complete metatar-
sal bones, belonging to several individuals, have
been found at Phu Wiang 5 and allow a fairly accu-
rate reconstruction of the metatarsus, on which the
diagnosis of Kinnareemimus khonkaenensis is
based.
Metatarsal II (Fig. 4) is represented by two com-
plete (PW5A-101) or nearly complete (PW5A-105)
specimens from the left side and two proximal ends
from the right side. The proximal articular region is
more expanded craniocaudally than mediolaterally.
The proximal articular surface is approximately tri-
angular in outline, with a rounded anterolateral
corner that overhangs the shaft. The caudal angle
is elongate and overhangs the posterior face of the
shaft, forming a well-marked rounded process that
markedly projects caudally. This process does not
project to the same extent in advanced ornithomi-
mids such as Struthiomimus altus (Osborn 1916),
Gallimimus bullatus (Osmo
´
lska et al. 1972) or
Sinornithomimus dongi (Kobayashi & Lu
¨
2003).
In less derived forms such as Archaeornithomimus
asiaticus (Smith & Galton 1990), Garudimimus bre-
vipes (Kobayashi & Barsbold 2005b) and Harpymi-
mus okladnikovi (Kobayashi & Barsbold 2005a),
the process is more developed, and therefore
resembles the condition in Kinnareemimus khon-
kaenensis. A first digit is present in Garudimimus
brevipes (Barsbold 1981), and its presence is uncer-
tain in Harpymimus okladnikovi (Kobayashi &
Barsbold 2005a). The shaft of metatarsal II is rela-
tively straight in its proximal half, with a D-
shaped cross-section, the lateral face being convex
and the medial face flat and smooth (this flat area
may have served for the contact with a first metatar-
sal, of which no remains have been found). More
distally, the condition becomes reversed, with a
flat lateral face (to accommodate the flat surface of
metatarsal III) and a convex medial face. In its
distal third, the shaft curves medially, so that the
distal articular end is directed somewhat laterally
rather than distally. The lateral condyle is much
larger and more bulbous than the medial one,
which is very narrow mediolaterally. The condyles
Fig. 4. Kinnareemimus khonkaenensis gen. et sp. nov. Left metatarsals II PW5A-101 (a –c) and PW5A-105 (d –f)in
lateral (a, b), cranial (c, d) and medial (e, f) views. Scale bar: 50 mm.
EARLY OSTRICH DINOSAUR FROM THAILAND 233
by guest on September 15, 2015http://sp.lyellcollection.org/Downloaded from
are separated by a deep caudal groove. There are
deep fossae both medially and laterally.
Metatarsal III (Fig. 5) is represented by two
proximal ends, one from the left side (PW5A-104)
and one from the right (PW5A-131), the middle
part of a shaft from the right side (PW5A-103), a
distal segment from the left side with the articular
end and a part of the shaft (PW5A-100, the holotype
of Kinnareemimus khonkaenensis), and a left distal
articular end (PW5A-107). On the basis of these
various elements, which complement one another,
a fairly accurate reconstruction of metatarsal III
can be obtained. In proximal view, the proximal
articular surface is much compressed mediolaterally
and roughly triangular in outline, with the apex
located cranially. Both the lateral and medial faces
are flat. Unlike the condition in advanced ornithomi-
mids, such as Ornithomimus velox (Marsh 1890,
1896), Struthiomimus altus (Lambe 1902; Osborn
1916), Gallimimus bullatus (Osmo
´
lska et al.
1972), Anserimimus planinychus (Barsbold 1988)
or Sinornithomimus dongi (Kobayashi & Lu
¨
2003), and even Archaeornithomimus asiaticus
(Smith & Galton 1990), the craniocaudal extent of
the proximal end of metatarsal III in Kinnareemimus
khonkaenensis was such that it was still visible in
cranial view, between the proximal ends of metatar-
sals II and IV, as in Garudimimus brevipes and Har-
pymimus okladnikovi. More distally, the shaft
becomes extremely narrow both mediolaterally
Fig. 5. Kinnareemimus khonkaenensis gen. et sp. nov. metatarsals III. Proximal end of metatarsal III PW5A-104 in
(a) cranial and (b) medial views. Middle part of shaft of right metatarsal III PW5A-103 in (c) cranial and (d) medial
views. Distal end of left metatarsal III PW5A-107 in cranial (e) and caudal (f) views. Distal end of left metatarsal III
PW5A-100 in cranial (g) and caudal (h) views. Scale bar: 50 mm.
E. BUFFETAUT ET AL.234
by guest on September 15, 2015http://sp.lyellcollection.org/Downloaded from
and craniocaudally, forming a thin bony rod with a
rectangular cross-section, a condition similar to
that seen in Archaeornithomimus asiaticus and
more advanced ornithomimids (the extreme slen-
derness of the bony rod explaining why specimens
are often broken or incomplete at that level). In
both Garudimimus brevipes and Harpymimus
okladnikovi, on the other hand, the corresponding
part of the shaft, although compressed mediolater-
ally, never becomes so thin and rod-like. Yet more
distally, the shaft of metatarsal III in Kinnareemi-
mus khonkaenensis becomes triangular in cross-
section, with a flat cranial face and oblique lateral
and medial faces, which meet at a sharp angle caud-
ally. The bone simultaneously become much
broader cranially, thus overlapping to some extent
metatarsals II and III. This is the usual condition
in advanced ornithomimids (including Archaeor-
nithomimus asiaticus), which differs from that
seen in Garudimimus brevipes and Harpymimus
okladnikovi, in which the cross-section of metatarsal
III at this level is more D-shaped than triangular.
The ‘medial expansion’ of the shaft (where it
slightly overlaps metatarsal II: see Kobayashi &
Barsbold 2005b, fig. 20) is apparently closer to the
distal end in Kinnareemimus khonkaenensis than
in Garudimimus brevipes and Harpymimus okladni-
kovi, and in this Kinnareemimus approximates more
closely the arctometatarsalian condition seen in
derived ornithomimids. In the more distal part,
close to the articulation, the shaft of metatarsal III
in Kinnareemimus khonkaenensis becomes
D-shaped in cross-section and slightly narrower
cranially. The articular head shows two subequal
Fig. 6. Kinnareemimus khonkaenensis gen. et sp. nov. Right metatarsal IV PW5A-106 in cranial (a), medial (b) and
caudal (c) views. Left proximal parts of metatarsals PW5A-102 (d, e) and PW5A-108 (f, g) in medial (d, f) and lateral
(e, g) views. Scale bar: 50 mm.
EARLY OSTRICH DINOSAUR FROM THAILAND 235
by guest on September 15, 2015http://sp.lyellcollection.org/Downloaded from
condyles separated by a shallow groove caudally.
There are deep ligamentous fossae on both sides.
Metatarsal IV (Fig. 6) is represented by a com-
plete right specimen (PW5A-106, in which part of
the distal half was poorly preserved and was recon-
structed), a proximal fragment comprising the
articular region and part of the shaft from the left
side (PW5A-102), and two left proximal ends with
small portions of the shaft (PW5A-108, 109). This
bone is longer than metatarsal II and more slender.
The proximal articular surface is somewhat
teardrop-shaped in outline, with a rounded cranial
margin and a pointed caudal end. The medial face
of the bone, in its proximal part, is flat, to accommo-
date metatarsal III. This flat facet extends from the
cranial to the caudal margin, showing that there
was no cranial contact with metatarsal II and that
the proximal part of metatarsal III was visible in
cranial view. The cranial edge of the articular
head overhangs the shaft. Caudally, there is a
process that projects less markedly than the
similar process on metatarsal II. A distinct concav-
ity on the lateral face of this process probably corre-
sponds to a contact area for metatarsal V (of which
no remains have been found). The proximal half of
the shaft is straight, with a cross-section that is
rounded cranially and angled caudally. More dis-
tally, the medial face of the shaft becomes flatter,
where it could have accommodated the flat
oblique lateral surface of metatarsal III. In its
distal third, the shaft curves laterally, so that the
distal articular region is oriented somewhat later-
ally, at an angle to the axis of the metatarsus. The
distal articulation is ginglymoid, with a medial
condyle that is more developed than the narrow
lateral condyle. The condyles are separated caudally
by a deep and narrow groove. There is a deep fossa
on the medial side, and a much smaller one on the
lateral side.
Phalanges (Fig. 7 ). Eight more or less complete
pedal phalanges are referable to Kinnareemimus
khonkaenensis. They include a right proximal
phalanx of digit II (PW5A-116) closely resembling
a specimen referred by Gilmore (1920, fig. 73) to
Ornithomimus affinis, two right second phalanges
of digit III (PW5A-113, -119) also resembling a
specimen illustrated by Gilmore (1920, fig. 74), a
right proximal phalanx of digit III (PW5A-115),
which is remarkably long and slender, two possible
Fig. 7. Kinnareemimus khonkaenensis gen. et sp. nov. Phalanges in dorsal (a, c, e, g, i, k, m, o) and lateral (b, d, f, h, j, l,
n, p) views; (a, b) PW5A-116; (c, d) PW5A-115; (e, f) PW5A-120; (g, h) PW5A-119; (i, j) PW5A-118; (k, l)
PW5A-123; (m, n) PW5A-121; (o, p) PW5A-122. Scale bar: 50 mm.
E. BUFFETAUT ET AL.236
by guest on September 15, 2015http://sp.lyellcollection.org/Downloaded from
right second phalanges of digit III (PW5A-117), and
a right third phalanx of digit IV (PW5A-121) similar
to the corresponding element in Gallimimus bulla-
tus (Osmo
´
lska et al. 1972). All these elements
resemble the corresponding bones of other ornitho-
mimosaurs but appear slightly more slender and
elongate than the phalanges of Harpymimus oklad-
nikovi and Garudimimus brevipes (Kobayashi &
Barsbold 2005a, b), and even Gallimimus bullatus
(Osmo
´
lska et al. 1972). A single ungual phalanx
(PW5A-122), lacking it tip, is present. It shows a
moderate curvature, has a groove on the medial
side, lacks a flexor tubercle on the ventral face,
and closely resembles the ungual phalanx of the
third digit of Struthiomimus altus, as illustrated by
Osborn (1916). The ventral surface is bounded
medially by a ridge that ends posteriorly in a small
spur (not visible on the damaged lateral margin).
All these characters occur in other ornithomimids
(Osmo
´
lska et al. 1972).
Relationships of Kinnareemimus
khonkaenensis
Although many elements of its skeleton remain
unknown, Kinnareemimus khonkaenensis can be
distinguished from other ornithomimosaurs by the
structure of its metatarsus (Fig. 8). Ornithomimo-
saurs as a group exhibit various degrees of constric-
tion of metatarsal III (Makovicky et al. 2004); but
the so-called ‘arctometatarsalian’ condition (Holtz
1994, 2001) is not developed to the same extent in
all members of the group. In basal forms in which
the metatarsus is known (which excludes Pelecani-
mimus polyodon and Shenzhousaurus orientalis),
metatarsal III is only moderately compressed med-
iolaterally and its proximal end is still visible
between those of metatarsals II and IV in cranial
view; it does not show a triangular cross-section in
its distal half. Both Garudimimus brevipes and Har-
pymimus okladnikovi show this type of metatarsal
III (the reconstruction of the tarsus of Garudimimus
brevipes in Currie 2000, fig. 22.6 J, which shows
metatarsal III hidden proximally by metatarsals II
and IV, is incorrect: see Kobayashi & Barsbold
2005b), the former being slightly more advanced
in the direction of the arctometatarsalian condition
than the latter (Kobayashi & Barsbold 2005b). In
contrast, more advanced forms (Ornithomimidae
sensu Makovicky et al. 2004), including Archaeor-
nithomimus asiaticus and more derived forms, have
the proximal part of metatarsal III hidden from
cranial view by metatarsals II and IV, which meet
cranially in their proximal region; in addition,
metatarsal III, which is very thin in its proximal
half, becomes broader more distally, with a tri-
angular cross-section, and spreads medially and
Fig. 8. Kinnareemimus khonkaenensis gen. et sp. nov.
Reconstruction of the metatarsus in proximal (a) and
cranial (b) views.
EARLY OSTRICH DINOSAUR FROM THAILAND 237
by guest on September 15, 2015http://sp.lyellcollection.org/Downloaded from
laterally over metatarsals II and IV. Kinnareemimus
khonkaenensis shows an intermediate condition
between that of the more basal forms and that of
the Ornithomimidae: its metatarsal III is still
visible, as a thin sliver of bone, between metatarsals
II and IV in cranial view, but more distally becomes
a thin rod of bone before it expands again, with a tri-
angular cross-section, in its distal half. Although the
metatarsal III of Kinnareemimus khonkaenensis still
shows a relatively basal condition in its proximal
end, it is extremely similar to the metatarsal III of
advanced ornithomimids more distally. Neverthe-
less, a fully arctometatarsalian condition, with the
proximal part of metatarsus III hidden from
cranial view by metatarsals II and IV, is considered
as one of the main synapomorphies of Ornithomimi-
dae within Ornithomimosauria (Makovicky et al.
2004). From this point of view, Kinnareemimus
khonkaenensis cannot be placed among the Ornitho-
mimidae, but can be considered as the sister-group
of that clade, occupying an intermediate position
between Garudimimus brevipes and Archaeornitho-
mimus asiaticus, currently the most basal ornithomi-
mid. Garudimimus brevipes and Harpymimus
okladnikovi have been placed in families of their
own (Garudimimidae Barsbold 1981 and Harpy-
mimidae Barsbold & Perle 1984), which have
been accepted by various workers (Barsbold &
Osmo
´
lska 1990; Kobayashi & Barsbold 2005a, b),
although they are not recognized by Makovicky
et al. 2004). Because its metatarsus is more
advanced than that of both Garudimimus brevipes
and Harpymimus okladnikovi, there seems to be
no reason for referring Kinnareemimus khonkaenen-
sis to either of these families. Pending the discovery
of more complete material, it seems preferable not
to erect a new family for the Thai form.
As noted above, the age of the Sao Khua For-
mation is not perfectly well constrained, but it
clearly belongs to the Early Cretaceous and is
older than Aptian. This makes Kinnareemimus
khonkaenensis one of the geologically oldest
known ornithomimosaurs. The purported Late
Triassic ornithomimosaur Shuvosaurus inexpecta-
tus from Texas (Chatterjee 1993), is not an ornitho-
mimosaur according to Rauhut (1997, 2003); the
recent description of Effigia okeeffeae, a Late Trias-
sic suchian from New Mexico showing conver-
gently acquired ornithomimosaur-like features,
shows that Shuvosaurus is a member of a Late Trias-
sic non-dinosaurian group of ornithomimosaur
mimics (Nesbitt & Norell 2006; Nesbitt 2007).
Elaphrosaurus bambergi, from the Late Jurassic of
Tanzania (Janensch 1925), once considered as an
ornithomimosaur (Nopcsa 1928; Russell 1972;
Galton 1982) is now placed among the Ceratosauria
(Rauhut 2003; Tykoski & Rowe 2004). Other
reports of pre-Cretaceous ornithomimosaurs
include two phalanges from the Kimmeridgian of
southern England (Brokenshire & Clarke 1993),
which do resemble the corresponding elements of
ornithomimosaurs, but are too fragmentary for a
more accurate assessment.
Among Early Cretaceous ornithomimosaurs,
Kinnareemimus khonkaenensis is either coeval
with or slightly older than the two most basal
forms known from that group, Pelecanimimus poly-
odon from the Barremian of Spain (Pe
´
rez-Moreno
et al. 1994) and Shenzhousaurus orientalis from
the Barremian of NE China (Ji et al. 2003). Both
taxa show various primitive characters, including
the presence of teeth. Unfortunately, Kinnareemi-
mus khonkaenensis cannot be compared with these
forms, because of the lack of skeletal elements in
common. In particular, the tarsus is unknown in
both Pelecanimimus polyodon and Shenzhousaurus
orientalis, and nothing is known of the jaws of Kin-
nareemimus khonkaenensis. It cannot be excluded
that Pelecanimimus polyodon or Shenzhousaurus
orientalis had a Kinnareemimus-like tarsus.
However, in the only other known toothed ornitho-
mimosaur, Garudimimus brevipes, metatarsal III is
less derived than in Kinnareemimus khonkaenensis.
It may be mentioned that Kinnareemimus
khonkaenensis is older than the purported ornitho-
mimosaur Timimus hermani from the late Aptian –
early Albian Otway Group of Victoria, Australia
(Rich & Vickers-Rich 1994). Whatever the exact
systematic position of this taxon, it postdates the
earliest well-established ornithomimosaurs from
Asia (and Europe) and therefore cannot be used to
support a Gondwanan origin of ornithomimosaurs.
Both of the primitive ornithomimosaurs that can
be compared with Kinnareemimus khonkaenensis
are from Mongolia and probably of later geological
age than the Thai form. Harpymimus okladnikovi is
from the Shinekhudag Svita, which, although
regarded as Hauterivian –Barremian (Shuvalov
2000) or Aptian – Albian (Currie 2000) by some
workers, has been referred to the late Albian on
the basis of magnetostratigraphy and palynology
by Hicks et al. (1999), an age estimate confirmed
by Nichols et al. (2006) on the basis of palyno-
morphs. Garudimimus brevipes is from the Bayn
Shireh Svita, which is considered as Cenomanian–
Turonian (Currie 2000) or Cenomanian –Santonian
(Hicks et al. 1999; Shuvalov 2000). Kinnareemimus
khonkaenensis is thus in all likelihood geologically
older than Harpymimus okladnikovi and certainly
older than Garudimimus brevipes. In the structure
of its metatarsus (which is the only feature that
allows significant comparisons), the Thai form is
clearly more advanced than the geologically
younger Mongolian taxa. This suggests more or
less separate evolution of the Central Asian and
SE Asian ornithomimosaurs during the Early
E. BUFFETAUT ET AL.238
by guest on September 15, 2015http://sp.lyellcollection.org/Downloaded from
Cretaceous, with the Mongolian forms (or at least
some of them) remaining more basal in foot struc-
ture than those from SE Asia. In this connection, it
is worth mentioning that the dinosaur fauna from
the Sao Khua Formation shows some peculiar fea-
tures (such as the apparent absence of
ornithischians) that may point to isolation of SE
Asia at the time of its deposition (Buffetaut et al.
2006), although the geographical and/or environ-
mental reasons for such an isolation remain
obscure (the presence of mountains chains resulting
from Mesozoic tectonic activity in Asia may be an
explanation). The dinosaur assemblage from the
Sao Khua Formation appears to be rather different
from the more or less coeval assemblages of the
‘Jehol Biota’ of NE China, for instance, and resem-
blances with Mongolian faunas are also limited. It
should be mentioned, however, that an ornithomi-
mosaur pes reminiscent of Kinnareemimus khon-
kaenensis has been described from the Middle
Grey Unit of the Xinminbao Group of the Mazong-
shan area, Gansu, China, by Shapiro et al. (2003);
metarsals II and III of this form are generally
similar to those of the Thai taxon and suggest that
the metatarsus was close to the arctometatarsalian
condition (however, the proximal end of metatarsal
III is missing). According to Tang et al. (2001), the
Middle Grey Unit of the Xinminbao Group may be
Albian. The specimen from Mazongshan thus indi-
cates that an ornithomimosaur more advanced than
the roughly coeval Harpymimus okladnikovi was
present in what is now NW China at the end of the
Early Cretaceous. Although this distribution
pattern may suggest that ornithomimosaurs with
an advanced type of metatarsus appeared in SE
Asia somewhat earlier than in other parts of the
world, the fact that metatarsus structure is
unknown in important Early Cretaceous taxa such
as Pelecanimimus polyodon and Shenzhousaurus
orientalis makes this assumption difficult to test
on the basis of the available fossils.
Archaeornithomimus asiaticus is currently the
most basal ornithomimid, showing a fully arcto-
metatarsalian condition. It is geologically much
younger than Kinnareemimus khonkaenensis,
coming as it does from the Iren Dabasu Formation
of Inner Mongolia, which was once thought to be
Cenomanian, but may in fact be as young as Seno-
nian, possibly Campanian (Currie & Eberth 1993)
or even early Maastrichtian (Van Itterbeeck et al.
2005). The question of when the arctometatarsalian
condition was achieved in ornithomimosaurs cur-
rently remains unanswered because of insufficient
evidence concerning that group during the early
stages of the Late Cretaceous. According to
Alifanov & Averianov (2006), the ornithomimosaur
bones from the Santonian of Kansai, Tajikistan, do
not differ from those of other late Cretaceous
ornithomimids; however, but this material does
not include metatarsals. Similarly, the ornithomi-
mosaur material reported from the Turonian of
Uzbekistan (Nessov 1995; Averianov 2006) and
the Cenomanian of Kyrgyzstan (Averianov 2006)
provides no information about the condition of the
metatarsus. However, the fact that Kinnareemimus
khonkaenensis was very close to the arctometatarsa-
lian condition suggests that it may have been
attained much earlier (i.e. in the Early Cretaceous)
than could be expected on the basis of the fossil
record as it was known before the Thai form was
discovered. This reasoning could be reversed,
however: the fact that Kinnareemimus khonkaenen-
sis antedates the less derived (as far as the
metatarsus is concerned) Harpymimus okladnikovi
and Garudimimus brevipes could be considered as
being in accordance with Holtz’s opinion that
these non-arctometatarsalian ornithimosaurs ‘rep-
resent reversal to a more primitive state’ (Holtz
2001, p. 113). In that case, the condition in
Kinnareemimus khonkaenensis could be interpreted
as an incipient stage in this reversal from a fully
arctometatarsalian condition; although this is not
how Holtz (2001) interpreted the (at the time still
unnamed) Thai form on the basis of the preliminary
description given by Buffetaut & Suteethorn (1998).
However, all recent studies of ornithomimosaur
phylogeny (Ji et al.
2003; Kobayashi & Lu
¨
2003;
Makovicky et al. 2004; Kobayashi & Barsbold
2005a, b) place Harpymimus and Garudimimus in
a basal position relative to the fully arctometatarsa-
lian Ornithomimidae, except that of Kobayashi &
Barsbold (2006, fig. 8B), in which Archaeornitho-
mimus is more basal than Garudimimus. It therefore
seems more likely that Kinnareemimus in fact
occupies an intermediate position between these
basal forms and the more derived ones, including
Archaeornithomimus.
An attempt has been made to assess the cursorial
abilities of Kinnareemimus khonkaenensis by com-
parison with other ornithomimosaurs, using the
length ratio of the tibia to metatarsal III, under the
assumption that forms with a longer metatarsus
were more advanced in this respect than those
with a shorter metatarsus. The length of metatarsus
III in Kinnareemimus khonkaenensis had to be esti-
mated because no complete specimen is available.
By articulating together metatarsal II PW5A-101
and incomplete metatarsal III PW5A-100, which
are of similar robustness and may be from the
same individual, a total length of 145 mm was esti-
mated for PW5A-100. This was compared with the
length of tibia PWA-1, which may be from the same
individual, although that cannot be demonstrated.
Because of these various assumptions, the tibia/
metatarsal III ratio thus obtained for Kinnareemi-
mus khonkaenensis must be considered as only an
EARLY OSTRICH DINOSAUR FROM THAILAND 239
by guest on September 15, 2015http://sp.lyellcollection.org/Downloaded from
approximation. A comparison with the ratios
obtained for some other ornithomimosaurs is
shown below:
length of metatarsal III/length of tibia:
Kinnareemimus khonkaenensis (PW5A-100 and
PW5A-101) 0.56
Garudimimus brevipes (GIN 100, data from
Kobayashi & Barsbold 2005b) 0.59
Struthiomimus altus (AMNH 5339, data from
Osborn 1916) 0.68
Gallimimus bullatus (GIN 100/10, data from
Osmo
´
lska et al. 1972) 0.72
This comparison shows that Kinnareemimus
khonkaenensis had a relatively short tarsus, shorter
than that of ornithomimids, and even than that of a
relatively primitive taxon such as Garudimimus
brevipes, which in metatarsus structure is less
advanced than in the Thai form. This may be
explained by the greater geological age of Kinnaree-
mimus khonkaenensis and suggests that metatarsus
elongation and the acquisition of an arctometatarsa-
lian condition were not directly linked and did not
evolve at the same rate in all ornithomimosaur
lineages.
Russell (1972) noted that Archaeornithomimus
asiaticus has relatively stouter metatarsals than
more advanced ornithomimids, as shown by the
ratio between the midshaft circumference of meta-
tarsal IV and its length. From this point of view, Kin-
nareemimus khonkaenensis has remarkably slender
metatarsals, as shown by the following comparison:
circumference of metatarsal IV at midlength of
shaft/Length of metatarsal IV:
Archaeornithomimus asiaticus (AMNH 6565, after
Russell 1972) 25.9%
Struthiomimus altus (AMNH 5375, after Russell
1972) 20.6%
Kinnareemimus khonkaenensis (PW5-106) 18.5%
This shows an interesting contrast to the metatar-
sal III/tibia length ratios given above, as it suggests
that in the slenderness of its metatarsals Kinnaree-
mimus khonkaenensis was more derived than some
Late Cretaceous ornithomimids (if slenderness of
the metatarsus is considered as a derived feature).
However, the slenderness of metatarsal IV in Kin-
nareemimus khonkaenensis may be linked to the
juvenile character of most of the ornithomimosaur
material from Phu Wiang.
Conclusions
Kinnareemimus khonkaenensis is one of the geolo-
gically oldest well-attested representatives of the
Ornithomimosauria. Its occurrence in SE Asia
may be significant from a biogeographical point of
view, as an addition to the growing list of Early
Cretaceous ornithomimosaurs from Asia, which
includes Harpymimus okladnikovi from the
Aptian–Albian of Mongolia, Shenzhousaurus
orientalis from the Barremian of China, and a
possible ornithomimosaur vertebra from the late
Barremian to Aptian Sebayashi Formation of
Japan (Hasegawa et al. 1999). However, ornithomi-
mosaurs were clearly not restricted to Asia in
the Early Cretaceous, as attested by the presence
of Pelecanimimus polyodon in the Barremian of
Spain and possible remains from the Jurassic of
England. Their presence in North America in the
Early Cretaceous is questionable. Very fragmentary
remains from the Aptian Arundel Formation of
Maryland were described as Ornithomimus affinis
by Gilmore (1920), and placed in the genus
Archaeornithomimus by Russell (1972), but Smith
& Galton (1990, p. 264) concluded that most
of the material could not be identified beyond
theropod incertae sedis, and two incomplete meta-
tarsals could only be referred to ‘a coelurosaurian
grade’. It should be noted, however, that, as
already stated by Gilmore (1920, 1921), the pedal
phalanges described as Ornithomimus affinis are
remarkably similar to those of ornithomimosaurs
(including Kinnareemimus). Of special importance
is a fragmentary metatarsal III (USNM 5684;
Gilmore 1920, fig. 72; 1921, plate 113), which
apparently becomes triangular in cross-section a
short distance proximally to the distal head,
suggesting a form at least as advanced as Kinnaree-
mimus in this respect. However, the proximal part of
the specimen is not preserved. Although some
additional specimens have been reported
(Weishampel & Young 1996), more material is
clearly necessary to better assess the systematic
position of this material, but it does seem to
suggest that early ornithomimosaurs may have
been present in North America in the Early Cretac-
eous. Makovicky et al. (2004) and Weishampel
(2006) considered the material from the Arundel
Formation as belonging to indeterminate ornithomi-
mosaurs. These Early Cretaceous records of
ornithomimosaurs from Europe and possibly North
America, together with the enigmatic Timimus
hermani from the Aptian–Albian of Australia,
show that an Asian origin for ornithomimosaurs
cannot be taken for granted. Nevertheless, the
advanced features of the tarsus of Kinnareemimus
khonkaenensis, indicative of a trend toward the con-
dition seen in Late Cretaceous ornithomimids, may
suggest that derived ornithomimosaurs evolved in
Asia, although it should be admitted that there is
no evidence about metatarsus structure in important
non-Asian Early Cretaceous forms such as Peleca-
nimimus polyodon.
E. BUFFETAUT ET AL.240
by guest on September 15, 2015http://sp.lyellcollection.org/Downloaded from
What the condition of the metatarsus in Kinnar-
eemimus khonkaenensis does suggest is that the
acquisition of the arctometatarsalian condition in
ornithomimosaurs was not a simple linear process.
Although it is one of the earliest well-attested
ornithomimosaurs, Kinnareemimus khonkaenensis
is more advanced in this respect than some geologi-
cally younger forms such as Harpymimus okladni-
kovi and Garudimimus brevipes.
Our field work at Phu Wiang was funded by the Depart-
ment of Mineral Resources (Bangkok) and the Centre
National de la Recherche Scientifique (CNRS, Paris). Sub-
sequent research on the fossil vertebrates from the Sao
Khua Formation has been supported by the ECLIPSE Pro-
gramme of CNRS, a Thailand Research Fund–CNRS joint
project, and a ‘Partenariat Hubert Curien’ French–Thai
project. We thank E. S. Gaffney (New York), A. Milner
(London), and R. Barsbold (Ulaan Baatar) for access to
specimens in their care. This paper is dedicated to the
memory of the late Halszka Osmo
´
lska (Warsaw), a
leading expert on ornithomimosaurs, who helped us in
many ways in our study of the Thai material.
References
ALIFANOV,V.R.&AVERIANOV, A. O. 2006. On the
finding of ornithomimid dinosaurs (Saurischia,
Ornithomimosauria) in the Upper Cretaceous Beds of
Tajikistan. Paleontological Journal, 1, 103– 108.
A
VERIANOV, A. O. 2006. On an ornithomimid dinosaur
(Saurischia, Ornithomimosauria) from the Cenoma-
nian of Fergana. Paleontological Journal, 3, 323–327.
B
ARSBOLD, R. 1976. On the evolution and systematics
of the late Mesozoic dinosaurs. The Joint Soviet–
Mongolian Paleontological Expedition, Transactions,
3, 68–75.
B
ARSBOLD, R. 1981. Toothless carnivorous dinosaurs of
Mongolia. The Joint Soviet – Mongolian Paleontologi-
cal Expedition, Transactions, 15, 28– 39.
B
ARSBOLD, R. 1988. A new Late Cretaceous ornithomi-
mid from the MPR. Paleontologicheskii Zhurnal, 1,
122– 124.
B
ARSBOLD,R.&OSMO
´
LSKA, H. 1990. Ornithomimo-
sauria. In:W
EISHAMPEL , D. B., DODSON,P.&
O
SMO
´
LSKA, H. (eds) The Dinosauria. University of
California Press, Berkeley, 225–244.
B
ARSBOLD,R.&PERLE, A. 1984. On the first new find of
a primitive ornithomimosaur from the Cretaceous of
the MPR. Paleontologicheskii Zhurnal, 2, 121– 123.
B
ROKENSHIRE,A.J.&CLARKE, J. B. 1993. Important
recently collected dinosaurian remains from the
Lower Kimmeridge Clay at Weymouth. Proceedings
of the Dorset Natural History and Archaeological
Society, 115, 177 –178.
B
UFFETAUT,E.&SUTEETHORN, V. 1998. Early Cretac-
eous dinosaurs from Thailand and their bearing on the
early evolution and biogeographical history of some
groups of Cretaceous dinosaurs. New Mexico
Museum of Natural History and Science Bulletin, 14,
205– 210.
B
UFFETAUT,E.&SUTEETHORN, V. 1999. The dinosaur
fauna of the Sao Khua Formation of Thailand and the
beginning of the Cretaceous radiation of dinosaurs in
Asia. Palaeogeography, Palaeoclimatology, Palaeoe-
cology, 150, 13 –23.
B
UFFETAUT, E., SUTEETHORN, V., MARTIN, V., TONG,
H., C
HAIMANEE,Y.&TRIAMWICHANON, S. 1995.
New dinosaur discoveries in Thailand. In:
W
ANNAKAO, L., SRISUK, K., YOUNGME,W.&
L
ERTSIRIVORAKUL, R. (eds) Proceedings of the Inter-
national Conference on Geology, Geotechnology and
Mineral Resources of Indochina (GEOINDO 2005).
Khon Kaen University, Khon Kaen, 157–161.
B
UFFETAUT, E., SUTEETHORN,V.&TONG, H. 2006.
Dinosaur assemblages from Thailand: A comparison
with Chinese faunas. In:L
U
¨
, J., KOBAYASHI, Y.,
H
UANG,D.&LEE, Y. N. (eds) Papers from the
2005 Heyuan International Dinosaur Symposium.
Geological Publishing House, Beijing, 19– 37.
C
APPETTA, H., BUFFETAUT,E.&SUTEETHORN,V.
1990. A new hybodont shark from the Lower Cretac-
eous of Thailand. Neues Jahrbuch fu
¨
r Geologie und
Pala
¨
ontologie, Monatshefte, 11, 659– 666.
C
HATTERJEE, S. 1993. Shuvosaurus—a new theropod.
National Geographic Research and Exploration, 9,
274– 285.
C
URRIE, P. J. 2000. Theropods from the Cretaceous of
Mongolia. In:B
ENTON, M. J., SHISHKIN, M. A.,
U
NWIN,D.M.&KUROCHKIN, E. N. (eds) The Age
of Dinosaurs in Russia and Mongolia. Cambridge
University Press, Cambridge, 434–455.
C
URRIE,P.J.&EBERTH, D. A. 1993. Palaeontology,
sedimentology and palaeoecology of the Iren Dabasu
Formation (Upper Cretaceous), Inner Mongolia,
People’s Republic of China. Cretaceous Research,
14, 127–144.
G
ALTON, P. M. 1982. Elaphrosaurus, an ornithomimid
dinosaur from the Upper Jurassic of North America
and Africa. Pala
¨
ontologische Zeitschrift, 56, 265 –275.
G
ILMORE, C. W. 1920. Osteology of the carnivorous
Dinosauria in the United States National Museum,
with special reference to the genera Antrodemus (Allo-
saurus) and Ceratosaurus. United States National
Museum Bulletin, 110, 1–159.
G
ILMORE, C. W. 1921. The fauna of the Arundel For-
mation of Maryland. Proceedings of the US National
Museum, 59, 581 –594.
G
ILMORE, C. W. 1933. On the dinosaurian fauna of the
Iren Dabasu Formation. Bulletin of the American
Museum of Natural History, 67, 23– 78.
H
ASEGAWA, Y., MANABE, M., KASE, T., NAKAJIMA,S.
&T
AKAKUWA, Y. 1999. An ornithomimid vertebra
from the Early Cretaceous Sebayashi Formation,
Sanchu Terrane, Gunma Prefecture, Japan. Bulletin
of Gunma Museum of Natural History, 3, 1–6.
H
ICKS, J. F., BRINKMAN, D. L., NICHOLS,D.J.&
W
ATABE, M. 1999. Paleomagnetic and palynologic
analyses of Albian to Santonian strata at Bayn
Shireh, Burkhant, and Khuren Dukh, eastern Gobi
Desert, Mongolia. Cretaceous Research, 20, 829–850.
H
OLTZ, T. R. 1994. The arctometatarsalian pes, an unusual
structure of the metatarsus of Cretaceous Theropoda
(Dinosauria: Saurischia). Journal of Vertebrate
Paleontology, 14, 480–519.
EARLY OSTRICH DINOSAUR FROM THAILAND 241
by guest on September 15, 2015http://sp.lyellcollection.org/Downloaded from
HOLTZ, T. R. 2001. Arctometatarsalia revisited: The
problem of homoplasy in reconstructing theropod phy-
logeny. In:G
AUTHIER,J.&GALL, L. F. (eds) New
Perspectives on the Origin and Early Evolution of
Birds. Peabody Museum of Natural History, New
Haven, CT, 99–121.
I
NGAVAT, R., JANVIER,P.&TAQUET, P. 1978. De
´
cou-
verte en Thaı
¨
lande d’une portion de fe
´
mur de dinosaure
sauropode (Saurischia, Reptilia). Comptes Rendus
Sommaires de la Socie
´
te
´
Ge
´
ologique de France, 3,
140– 141.
J
ANENSCH, W. 1925. Die Coelurosaurier und Theropo-
den der Tendagura-Schichten Deutsch-Ostafrikas.
Palaeonto graphica, Supplement VII, 1. Reihe, 1,
1–99.
J
I, Q., NORELL, M. A., MAKOVICKY, P. J., GAO, K., JI,S.
&Y
UAN, C. 2003. An early ostrich dinosaur
and implications for ornithomimosaur phylogeny.
American Museum Novitates, 3420, 1–19.
K
OBAYASHI,Y.&BARSBOLD, R. 2005a. Anatomy of
Harpymimus okladnikovi Barsbold and Perle 1984
(Dinosauria: Theropoda) of Mongolia. In:
C
ARPENTER, K. (ed.) The Carnivorous Dinosaurs.
Indiana University Press, Bloomington, 97 –126.
K
OBAYASHI,Y.&BARSBOLD, R. 2005b. Reexamination
of a primitive ornithomimosaur, Garudimimus bre-
vipes Barsbold, 1981 (Dinosauria: Theropoda), from
the Late Cretaceous of Mongolia. Canadian Journal
of Earth Sciences, 42, 1501–1521.
K
OBAYASHI,Y.&BAR SBOLD, R. 2006. Ornithomimids
from the Nemegt Formation of Mongolia. Journal of
the Paleontological Society of Korea, 22, 195 –207.
K
OBAYASHI,Y.&LU
¨
, J. 2003. A new ornithomimid
dinosaur with gregarious habits from the Late Cretac-
eous of China. Acta Palaeontologica Polonica, 48,
235– 259.
L
AMBE, L. M. 1902. New genera and species from the
Belly River Series (Mid-Cretaceous). Contributions
to Canadian Palaeontology, 3, 23– 81.
M
AKOVICKY, P. J., KOBAYASHI,Y.&CURRIE,P.J.
2004. Ornithomimosauria. In:W
EISHAMPEL, D. B.,
D
ODSON,P.&OSMO
´
LSKA, H. (eds) The Dinosauria.
2nd edn. University of California Press, Berkeley,
137– 150.
M
ARSH, O. C. 1881. Principal characters of American
Jurassic dinosaurs. Part V. American Journal of
Science, 21, 417–423.
M
ARSH, O. C. 1890. Description of new dinosaurian
reptiles. American Journal of Science, 39, 81–86.
M
ARSH, O. C. 1896. The dinosaurs of North America. US
Geological Survey Annual Report, 16, 133–244.
M
ARTIN, V. 1994. Baby sauropods from the Sao Khua
Formation (Lower Cretaceous) in northeastern
Thailand. Gaia, 10, 147 –153.
M
ARTIN, V., BUFFETAUT,E.&SUTEETHORN, V. 1994.
A new genus of sauropod dinosaur from the Sao Khua
Formation (Late Jurassic or Early Cretaceous) of north-
eastern Thailand. Comptes Rendus de l’Acade
´
mie des
Sciences, Se
´
rie II, 319, 1085–1092.
M
ARTIN, V., SUTEETHOR N,V.&BUFFETAUT , E. 1999.
Description of the type and referred material of
Phuwiangosaurus sirindhornae Martin, Buffetaut &
Suteethorn, 1994, a sauropod from the Lower Cretac-
eous of Thailand. Oryctos, 2, 39–91.
M
OURET, C., HEGGEMANN, H., GOUADAIN,J.&
K
RISADASIMA, S. 1993. Geological history of the sili-
ciclastic strata of the Khorat Group in the Phu Phan
Range area, northeastern Thailand. In:T
HANASUTHI-
PITAK
, T. (ed.) Proceedings of the International Sym-
posium on Biostratigraphy of Mainland Southeast
Asia: Facies and Paleontology, 1. Department of Geo-
logical Sciences, Chiang Mai University, 23–49.
N
ESBITT, S. J. 2007. The anatomy of Effigia okeeffeae
(Archosauria, Suchia), theropod-like convergence,
and the distribution of related taxa. Bulletin of the
American Museum of Natural History, 302, 1–84.
N
ESBITT,S.J.&NORELL, M. A. 2006. Extreme conver-
gence in the body plans of an early suchian (Archo-
sauria) and ornithomimid dinosaurs (Theropoda).
Proceedings of the Royal Society of London, Series
B, 273, 1045–1048.
N
ESSOV, L. A. 1995. Dinosaurs of Northern Eurasia: New
data about assemblages, ecology and palaeobiogeo-
graphy. University of Saint-Petersburg, Institute of
Earth Crust.
N
ICHOLS, D. J., MATSUKAWA,M.&ITO, M. 2006. Paly-
nology and age of some Cretaceous nonmarine depos-
its in Mongolia and China. Cretaceous Research, 27,
241– 251.
N
OPCSA, F. 1928. The genera of reptiles. Palaeobiologica,
1, 163–188.
O
SBORN, H. F. 1916. Skeletal adaptations of Ornitho-
lestes, Struthiomimus and Tyrannosaurus. Bulletin of
the American Museum of Natural History, 35,
733– 771.
O
SMO
´
LSKA, H., RONIEWICZ,E.&BARSBOLD, R. 1972.
A new dinosaur, Gallimimus bullatus n.gen., n.sp.
(Ornithomimidae) from the Upper Cretaceous of Mon-
golia. Palaeontologia Polonica, 27, 103– 143.
O
WEN, R. 1842. Report on British fossil reptiles. Part II.
Report of the British Association for the Advancement
of Science, 11, 60–204.
P
E
´
REZ-MORENO, B. P., SANZ, J. L., BUSCALIONI, A. D.,
M
ORATALLA, J. J., ORTEGA,F.&RASSKIN -
G
UTMAN, D. 1994. A unique multitoothed ornithomi-
mosaur dinosaur from the Lower Cretaceous of Spain.
Nature, 370, 363 –367.
R
ACEY, A., LOVE, M. A., CANHAM , A. C., GOODALL,
J. G. S., P
OLACHAN,S.&JONES, P. D. 1996. Strati-
graphy and reservoir potential of the Mesozoic
Khorat Group, NE Thailand. Part 1: Stratigraphy and
sedimentary evolution. Journal of Petroleum
Geology, 19, 5 –40.
R
AUHUT, O. W. M. 1997. Zur Scha
¨
delanatomie von
Shuvosaurus inexpectatus (Dinosauria, Theropoda).
Terra Nostra, 7, 17–21.
R
AUHUT, O. W. M. 2003. The interrelationships and evol-
ution of basal theropod dinosaurs. Special Papers in
Palaeontology, 69, 1 –213.
R
ICH,T.H.&VICKERS-RICH, P. 1994. Neoceratopsians
and ornithomimosaurs: Dinosaurs of Gondwana
origin? National Geographic Research and Explora-
tion, 10, 129–131.
R
USSELL, D. A. 1972. Ostrich dinosaurs from the Late
Cretaceous of western Canada. Canadian Journal of
Earth Sciences, 9, 375– 402.
S
HAPIRO, M. D., YOU, H., SHUBIN, N. H., L UO,Z.&
D
OWNS, J. P. 2003. A large ornithomimid pes from
E. BUFFETAUT ET AL.242
by guest on September 15, 2015http://sp.lyellcollection.org/Downloaded from
the Lower Cretaceous of the Mazongshan area, north-
ern Gansu Province, People’s Republic of China.
Journal of Vertebrate Paleontology, 23, 695 –698.
S
HUVALOV, V. F. 2000. The Cretaceous stratigraphy
and palaeobiogeography of Mongolia. In:B
ENTON,
M. J., S
HISHKIN, M. A., UNWIN,D.M.&
K
UROCHKIN, E. N. (eds) The Age of Dinosaurs in
Russia and Mongolia. Cambridge University Press,
Cambridge, 256–278.
S
MITH,D.&GALTON, P. 1990. Osteology of Archaeor-
nithomimus asiaticus (Upper Cretaceous, Iren Dabasu
Formation, People’s Republic of China). Journal of
Vertebrate Paleontology, 10, 255–265.
S
ULLIVAN, R. M. 1997. A juvenile Ornithomimus anti-
quus (Dinosauria: Theropoda: Ornithomimosauria),
from the Upper Cretaceous Kirtland Formation
(De-Na-Zin Member), San Juan Basin, New Mexico.
In:A
NDERSON, O., KUES ,B.&LUCAS, S. (eds)
Mesozoic Geology and Paleontology of the Four
Corners Area. New Mexico Geological Society Guide-
book, 48
th
field conference, New Mexico Geological
Society, Socorro, 249–254.
T
ANG, F., LUO, Z., ZHOU, Z., YOU , H., GEORGI, J. A.,
T
ANG,Z.&WANG, X. 2001. Biostratigraphy and
palaeoenvironment of the dinosaur-bearing sediments
in Lower Cretaceous of Mazongshan area, Gansu
Province, China. Cretaceous Research, 22, 115–129.
T
YKOSKI,R.S.&ROWE, T. 2004. Ceratosauria. In:
W
EISHAMPEL, D. B., DODSON,P.&OSMO
´
LSKA,
H. (eds) The Dinosauria. 2nd edn. University of
California Press, Berkeley, 47– 70.
V
AN ITTERBEECK, J., HORNE, D. J., BULTY NCK,P.&
V
ANDENBERGHE, N. 2005. Stratigraphy and palaeo-
environment of the Upper Cretaceous dinosaur-
bearing Iren Dabasu Formation (Inner Mongolia,
People’s Republic of China). Cretaceous Research,
26, 699–725.
VON HUENE, F. 1914. Das natu
¨
rliche System der Saur-
ischia. Centralblatt fu
¨
r Mineralogie, Geologie und
Pala
¨
ontologie, B, 1914, 154– 158.
W
EISHAMPEL, D. B. 2006. Another look at the dinosaurs
of the East Coast of North America. In:T
ORCIDA
FERNA
´
NDEZ-BALDOR,F.&HUERTA HURTADO,P.
(eds) Actas de las III Jornadas Internacionales sobre
Paleontologia de Dinosaurios y su Entorno. Colectivo
Arqueolo
´
gico– Paleontolo
´
gico de Salas, Salas de Los
Infantes, 129–168.
W
EISHAMPEL,D.B.&YOUNG, L. 1996. Dinosaurs of
the East Coast. Johns Hopkins University Press,
Baltimore, MD.
EARLY OSTRICH DINOSAUR FROM THAILAND 243
by guest on September 15, 2015http://sp.lyellcollection.org/Downloaded from
