New Oviraptorid Dinosaur (Dinosauria: Oviraptorosauria) from the Nemegt Formation of Southwestern Mongolia

Page 1

Introduction

Oviraptorosaurs are generally regarded as non-

avian theropod dinosaurs (Osborn, 1924; Bars-

bold, 1976, 1977, 1981, 1983, 1986, 1997; Gau-

thier, 1986; Smith, 1992; Sereno, 1999; Barsbold,

Currie et al., 2000; Barsbold, Osmólska et al.,

2000; Holtz, 2000, 2001; Norell et al., 1994;

Norell et al., 1995; Norell, Clark, and

Makovicky, 2001; Xu, Norell et al., 2002). They

are medium-sized theropods characterized by a

short, deep skull with toothless jaws in derived

forms, teeth present in primitive forms such as

Incisivosaurus gauthieri (Xu, Cheng et al., 2002)

and Caudipteryx zoui (Ji et al., 1998), pneu-

matilized caudal vertebrae present in derived

forms, anteriorly concave pubic shaft, and a pos-

teriorly curved ischium present in Oviraptoridae

(Barsbold and Osmólska, 1990; Barsbold,

Osmólska et al., 2000; Makovicky and Sues,

1998). Oviraptorosauria is here regarded to in-

clude three families: Oviraptoridae, Caenagnathi-

dae, and Caudipterygidae (Barsbold, 1976; Stern-

berg, 1940; Currie, 2000; Clark et al., 2001; Ji et

al., 1998; Zhou and Wang, 2000; Zhou et al.,

2000).

The Oviraptorosauria are mainly distributed in

Mongolia (Osborn, 1924; Barsbold, 1976, 1977,

1981, 1983, 1986, 1997; Osmólska, 1976; Norell

et al., 1994, 1995; Wester, 1996; Clark et al.,

1999, 2001, 2002), China (Dong and Currie,

1996; Ji et al., 1998; Lü et al., 2000; Xu, Cheng

et al., 2002; Lü, 2003) and North America

(Sternberg, 1940; Cracraft, 1971; Currie and

Russell, 1988; Sues, 1997). Additionally, ovirap-

torosaurs from the Southern Hemisphere have

been reported (Frey and Martill, 1995; Currie et

al., 1996; Frankfurt and Chiappe, 1999), al-

though validity of the taxonomic assignment is

doubtful.

In the summer of 1996, an incomplete ovirap-

torid skeleton with a nearly complete skull was

discovered in the Nemegt Formation of Nemegt

Basin, southwestern Mongolia by the Mongolian

New Oviraptorid Dinosaur (Dinosauria: Oviraptorosauria) from

the Nemegt Formation of Southwestern Mongolia

Junchang Lü1, Yukimitsu Tomida2, Yoichi Azuma3,

Zhiming Dong4 and Yuong-Nam Lee5

1 Institute of Geology, Chinese Academy of Geological Sciences, Beijing 100037, China

2 National Science Museum, 3–23–1 Hyakunincho, Shinjukuku, Tokyo 169–0073, Japan

3 Fukui Prefectural Dinosaur Museum, 51–11 Terao, Muroko, Katsuyama 911–8601, Japan

4 Institute of Paleontology and Paleoanthropology, Chinese Academy of Sciences, Beijing 100044, China

5 Korea Institute of Geoscience and Mineral Resources, Geology & Geoinformation Division,

30 Gajeong-dong, Yuseong-gu, Daejeon 305–350, South Korea

Abstract Nemegtia barsboldi gen. et sp. nov. here described is a new oviraptorid dinosaur from

the Late Cretaceous (mid-Maastrichtian) Nemegt Formation of southwestern Mongolia. It differs

from other oviraptorids in the skull having a well-developed crest, the anterior margin of which is

nearly vertical, and the dorsal margin of the skull and the anterior margin of the crest form nearly

90°; the nasal process of the premaxilla being less exposed on the dorsal surface of the skull than

those in other known oviraptorids; the length of the frontal being approximately one fourth that of

the parietal along the midline of the skull. Phylogenetic analysis shows that Nemegtia barsboldi is

more closely related to Citipati osmolskae than to any other oviraptorosaurs.

Key words: Nemegt Basin, Mongolia, Nemegt Formation, Late Cretaceous, Oviraptorosauria,

Nemegtia.

Bull. Natn. Sci. Mus., Tokyo, Ser. C, 30, pp. 95–130, December 22, 2004

Page 2

Highland International Dinosaur Project team. It

was initially described by Lü et al. (2002) as In-

genia sp. The skull morphology (vertical anterior

margin of the developed crest on the skull, less

exposed nasal process of the premaxilla, and a

process on the quadrate bearing a convex surface

projecting into the cotyla on the medial surface

of the quadratojugal) indicates that it is distinct

from named oviraptorosaurs. It represents a new

taxon from the Late Cretaceous.

Systematic Paleontology

Theropoda Marsh, 1881

Oviraptorosauria Barsbold, 1976

Family Oviraptoridae Barsbold, 1976

Nemegtia gen. nov.

Diagnosis: As for the only known species.

Etymology: Nemegtia refers to the fossil lo-

cality, the Nemegt Basin of southwestern Mon-

golia.

Nemegtia barsboldi sp. nov.

(Figs. 1, 3, 5, 6)

Ingenia sp. Lü et al., 2002

Etymology: The species is named in honor of

Dr. R. Barsbold, the Mongolian vertebrate pale-

ontologist, one of the leaders of the Mongolian

Highland International Dinosaur Project.

Type locality and horizon: The Nemegt lo-

cality of Gradzinski et al. (1968), Nemegt Basin,

southern Gobi Desert, Mongolia; Nemegt Forma-

tion, mid-Maastrichtian (Jerzykiewicz and Rus-

sell, 1991) of the Upper Cretaceous.

Holotype: GIN100/2112 (formerly PC100/

2112, see Lü et al., 2002): an incomplete skele-

ton with a nearly complete skull. Specimen

stored in the Paleontological Center of the Mon-

golian Academy of Sciences, Ulan Bator, Mon-

golia.

Diagnosis: An oviraptorid displaying the fol-

lowing derived characters. The skull with a naso-

premaxillary crest, anterior margin (the nasal

process of the premaxilla) nearly vertical; the

nasal process of the premaxilla, which forms part

of the crest, being barely exposed in dorsal view;

the anteroposterior length of the frontal is ap-

proximately one quarter of that of the parietals; a

presence of prefrontal; process on the quadrate

bears a convex surface projecting into the cotyla

on the medial surface of the quadratojugal.

Mandibular condyles of the quadrate situated

rostrally to the occipital condyle.

Description

1. Skull and mandible

The skull (Fig. 1) is deep, narrow and short,

and has a naso-premaxillary vertical crest. The

jaws are toothless, and the articular has a medial

process. These characters indicate that Nemegtia

barsboldi is a derived oviraptorosaur.

Skull: The premaxillae, vomers and maxillae

form a hard palate as in other oviraptorid di-

nosaurs (Fig. 1C), and the premaxillae form the

main part of the crest. Both premaxillae are

fused. The central palatal part of the joined pre-

maxillae is strongly concave. There is a cleft on

the central part of the palate. Anteriorly, small

foramina distributed on the lateral sides of the

suture between the premaxillae may represent

nutrient openings. The nasal processes of the pre-

maxillae are perpendicular to the palatal surface

of the upper jaw, and extend dorsally, connecting

with the nasals at the highest point of the crest

(Fig. 1A), and is barely exposed in dorsal view of

the skull. The crest extends posteriorly and ven-

trally, and forms a round arc at its highest point.

The antorbital cavity consists of a large inter-

nal antorbital fenestra posteriorly and a small

maxillary fenestra (see Witmer, 1997) anteroven-

trally. The internal antorbital fenestra is sub-oval,

and the long axis is vertical. The external surface

of the maxilla is smooth, and moderately con-

cave. Ventrally, there are two longitudinal, round-

ed ridges on each maxilla as in other ovirap-

torids. In addition to the lateral ridge, there is a

longitudinal groove. Posteriorly, there is a ven-

trally directed tooth-like process on the palatal

shelf of each maxilla as in other oviraptorids.

Lü, Tomida, Azuma, Dong and Lee

Page 3

New Oviraptorid Dinosaur, Mongolia

po p

stf

I n

nar

cav antorb

sof

lat

par

Fig. 1. The skull of Nemegtia barsboldi gen. et sp. nov. (GIN100/2112). In lateral (A), dorsal (B), and ventral

views (C). Scale bar5 cm. Abbreviations, as in appendix 1.

Page 4

Both nasals are completely fused posteriorly

with no trace of the suture. Anteriorly, the nasals

are not fused, forming a “” shaped cleft, which

receives the nasal processes of the premaxillae

(Fig. 1B). The dorsal portion of the lacrimal is

crescent shaped in lateral view. The shaft of the

lacrimal is stout, and forms the posterior margin

of the antorbital cavity. There is a large lacrimal

foramen in the middle of the anterior edge of the

orbit. The posterodorsal part of the lacrimal is

wedged between the prefrontal and frontal. The

lateral surface of the lacrimal is smooth. Pos-

terodorsal to the antorbital cavity, there is a large

recessed pneumatic pocket (6mm high, 4mm

wide), similar to that in Citipati osmolskae

(Clark et al., 2002).

Anterodorsal to the left orbit, there is a small

triangular bone, which is located between the

frontal and lacrimal. It should be a prefrontal, al-

though this bone has not been reported in other

oviraptorosaurs.

The jugal is a triradiate bone. The posterior

part of its anterior branch is stout and plate-like.

Anteriorly, it gradually becomes rod-like and

then ends as a thin plate (Fig. 2), where it inserts

between the maxilla, lacrimal and ectopterygoid.

The postorbital process of the jugal inclines pos-

terodorsally, and forms almost 2/3 of the height

of the postorbital bar, similar to other ovirap-

torosaurs (Barsbold et al., 1990). The process be-

comes widened, and slightly twisted, and its me-

dial surface is smooth, near the position where it

contacts the descending process of the postor-

bital. The dorsal part of the postorbital process of

the jugal is tightly attached to the posteroventral

surface of the postorbital. The posterior branch

of the jugal contacts the anterior process of the

quadratojugal. The angle of the postorbital

branch and posterior branch of the jugal is 90°,

which form the anterior and ventral margins of

the large lateral temporal fenestra, respectively.

The width of the frontal is much greater than

its length. In lateral view, the whole central part

of the frontal surface is convex, thus the trans-

verse cross-section of the frontal is arc-shaped.

The anteroposterior length of the frontal is ap-

proximately one quarter of that of the parietals.

Anteriorly, the suture of the frontal and the nasal

is relatively straight, while posteriorly, the suture

with the parietal is convex anteriorly (Fig. 1B).

There is a deep groove along the suture with the

nasal. When compared with other oviraptorids,

Lü, Tomida, Azuma, Dong and Lee

Medial view of the quadratojugal

cross-section of the jugal

Posterior view of

the quadrate

Fig. 2. Skull of Nemegtia barsboldi gen. et sp. nov. (GIN100/2112) showing the structure of the jugal and the

relationship between the quadrate and the quadratojugal.

Page 5

the frontal is very short medially and widens lat-

erally towards orbital margin.

The parietals are fused. Their dorsal surface is

smooth. A sagittal parietal ridge is present. This

ridge extends anteriorly and disappears at the

level of the anterior corner of the supratemporal

fossa, not reaching the frontoparietal suture.

The right squamosal is almost completely pre-

served. It curves posteriorly. There is a notch,

which holds the distal lateral head of the

quadrate on the middle of the lateral margin. The

anterior process of the squamosal forms the pos-

tolateral margin of the sub-oval supratemporal

fossa. The long axis of the suboval supratemporal

fossa points anterolaterally, making an angle with

the long axis of the skull of nearly 30°.

The postorbital is a triradiate bone. The frontal

process of the postorbital is upturned about 90

degrees. Its descending process occupies 2/3 of

the height of the postorbital bar. The anterior sur-

face of this process is smooth. The angle of the

anteromedial process and the posterior squamos-

al process is greater than 90°. The frontal process

covers the orbital process of the frontal. The

squamosal process extends posteriorly with a

sharp dorsal margin and overlies the anterior

process of the squamosal. This process does not

reach the posterolateral corner of the supratem-

poral fossa. The upper part of the postorbital bar

forms most of the anterolateral margin of the

supratemporal fossa.

A shallow groove separates the articular sur-

face into two parts in the quadrate. In ventral

view, the width of the mandibular articular sur-

face of the quadrate is greater than its length.

The mandibular condyles are situated rostrally to

the occipital condyle. The articular surface for

the quadratojugal is convex, while that on the

quadratojugal for the quadrate is concave. This

joint forms a mobile junction (Fig. 2), which is

similar to the condition in most birds and other

oviraptorosaurs (Maryanska and Osmólska,

1997). The convex quadrate and concave quadra-

tojugal joint is reversed from the condition in

other oviraptorosaurs (Maryanska and Osmólska,

1997).

In ventral view, the vomer has a sharp process,

which projects ventrally. This tooth-like process

is also formed in part by processes of the maxil-

lae. The vomers are completely fused anteriorly.

The posterior parts of the vomers are separated

from each other by the cultriform process of the

parasphenoid, which wedges rostrally between

the vomers as in all oviraptorids. The middle part

of the vomer is relatively long, its surface is

smooth, and its dorsal part is fused with the ante-

rior part of the cultriform process. The posterior

part of the vomer is covered by the anterior part

of the palatal ramus of the pterygoid. The fused

part of the vomer projects ventrally, and the mid-

dle part has a round ventral surface, while its

posterior part is relatively flat.

The palatine forms the posterior and lateral

margins of the elliptic choana. It connects with

the vomer medially, and with the maxilla antero-

laterally. The posterolateral part of the palatine

contacts the anteromedial side of the ectoptery-

goid, and its posteromedial part contacts the

palatal ramus of the pterygoid.

Ventrally, the pterygoids have elongated

palatal processes, which do not meet along the

middle line until they contact the vomers. Each

palatal process has a longitudinal groove, which

divides the palatal branch of the pterygoid into

two parts anteriorly: the lateral part connects

with the ectopterygoid, while the medial one con-

nects with the palatine and quadrate.

The ectopterygoid is located rostral to the

pterygoid. It extends vertically, similar to other

derived oviraptorids, whereas in other theropods,

it extends horizontally. The posterior end of the

dorsal part disappears near anterior end of the

jugal. There is an opening within ectopterygoid,

which is close to the maxilla. This opening may

be homologous to the dorsal ectopterygoid recess

of dromaeosaurids (Witmer, 1997).

The epipterygoid is present on both sides of

the skull. It is a thin plate-like bone with an ir-

regular triangular shape, whose free vertex angle

projects dorsally. It tightly ties to the pterygoid

branch of the quadrate. Its anterior margin is

sharp and free.

New Oviraptorid Dinosaur, Mongolia

Page 6

The basioccipital condyle is ball-shaped, with

a weak neck region. The articular surface of the

occipital condyle faces posteroventrally. The an-

terior part of the basioccipital projects ventrally

and laterally, forming the posterior part of the

basal tubera. The suture of the basioccipital and

basisphenoid is clear. Its middle part projects

posteriorly; therefore, in the ventral view, the su-

ture has a weak “W” shape.

The basisphenoid forms the posteroventral part

of the floor of the braincase. The sutures with the

pterygoid, and the parasphenoid are not clear. In

ventral view, it is narrow anteriorly and broad

posteriorly. The portion comprising the basal tu-

bera is strongly developed. Its middle portion is

strongly depressed, and has a large opening. This

opening is probably the Vidian canal for the pas-

sage of the internal carotid artery and the palatal

branch of nerve VII, which entered the pituitary

fossa through the basisphenoid. The basiptery-

goid process is absent, similar to other derived

oviraptorosaurs (Clark et al., 2002).

The tall parasphenoid process is delicate. It is

shallow anteriorly and deep posteriorly. Its depth

gradually lessens anteriorly, as it connects with

the vomer. Its posterodorsal part extends dorsally

between the epipterygoid and pterygoid, contact-

ing the laterosphenoid. In ventral view, it extends

forward into the large space called interpterygoid

vacuity. Its length is about 1/4 of that of the skull.

The laterosphenoid is a thin plate-like bone,

anterior to the floor of the braincase. The or-

bitosphenoid is also a thin plate like bone, anteri-

or to laterosphenoid portion of the floor of the

braincase.

The suture between the exoccipital and

supraoccipital is not clear, but the trend of the su-

ture shows that both exoccipitals connect above

the oval foramen magnum. Thus, they separate

the supraoccipital from the foramen magnum,

which is larger than the occipital condyle.

Mandible: The lower jaw is short and deep

(Fig. 3). The dorsal margin of the external

mandibular fenestra is formed by the surangular

and the dorsal arch of the dentary. The surangular

has a middle branch, which projects into the ex-

ternal mandibular fenestra, as in other ovirap-

torids. The ventral surface of the lower jaw is

convex, and the mandibular symphysis is short.

In dorsal view, the jaw slightly projects laterally

in the middle part, and both branches of the

lower jaw are parallel posteriorly from the level

of the posterior margin of the mandibular symph-

ysis (Fig. 3A). The dentaries have a dorsal pro-

trusion near the posterior margin of the mandibu-

lar symphysis. The large external mandibular

fenestra is located at the anterior part of the

lower jaw. The coronoid process is posterior to

the mandibular fenestra.

The mandibular symphysis is short and deep.

No evidence shows that a coronoid is present in

the present specimen. CT scanning shows that

the dorsal margin of the mandibular symphysis is

thin, while its lower margin (lower part) is stout,

the symphysis is highly pneumatized, and the

spaces are uniform in size and distribution (Figs.

4A, B). There is no bony construction between

the hollow spaces of the right and left dentaries.

The cross section near the mandibular symphysis

shows that both sides are steep, displaying a “V”

shape. It is “U” shaped in Caenagnathus and

there is a bar of bone between the dorsal and ven-

tral surface supporting the midline ridge on the

lingual surface of the symphysial shelf (Currie et

al., 1993; Figs. 4C, D). Around the anterior mar-

gin of the external mandibular fenestra, the den-

taries are deflected inwards, forming a shallowly

flattened surface, thus producing a thin and sharp

anterior edge on the external mandibular fenes-

tra. The ventral process of the denary is relatively

weak, and it tightly extends along the ventrolater-

al part of the anterior portion of the angular, and

disappears at the level of the coronoid eminence.

The dorsal process of the dentary is comparative-

ly massive, and dorsally, its distal end is divided

into lateral and medial prongs, with the anterior

branch of the articular-surangular-coronoid com-

plex (ASC) wedged between them. It disappears

anterior to the coronoid eminence. The ventral

process of the dentary extends more posteriorly

than that of the dorsal process. Only the external

surfaces of both processes of the dentaries are

100

Lü, Tomida, Azuma, Dong and Lee

Page 7

New Oviraptorid Dinosaur, Mongolia

101

emf

pre

asc

Fig. 3. Lower jaw of Nemegtia barsboldi gen. et sp. nov. (GIN100/2112). In dorsal (A), ventral (B), and lateral

Page 8

smooth. Anterior to the external mandibular fen-

estra and the lateral edges of the mandibular

symphysis, irregular small holes are densely dis-

tributed. A row of four small foramina is sym-

metrically distributed on both sides of the serrat-

ed suture of the mandibular symphysis. This row

of small foramina extends posterolaterally, form-

ing a nearly 30° angle with the serrated symphy-

seal suture. The lateral surface of the ventral part

of the mandibular symphysis is convex, and there

is a process centrally.

The surangular and articular are fused in the

posterior part of the lower jaw, with no clear su-

ture between them. The fused part occupies the

position, which corresponds to the position of the

articular, surangular, and coronoid in other

theropods. Currie et al. (1993) named it as the

articular-surangular-coronoid complex (ASC).

There is a longitudinal rounded ridge on the ar-

ticular surface. This ridge corresponds to the lon-

gitudinal sulcus on the articular surface of the

quadrate condyle. The ridge separates the articu-

lar surface into two parts: a medial one, called

the internal mandibular process by Sternberg

(1940), is large and semicircular, and a lateral

one, called the external mandibular process

(Sternberg, 1940), is smaller and sub-oval. The

surface of the lateral one slightly slants laterally.

The lengths of both processes are nearly equal.

The retroarticular process is slender, extending

posteroventrally from the articular surface. Its

ventral and medial surfaces are covered by the

prearticular. Its lateral surface is covered by artic-

ular-surangular-coronoid complex. The angular

does not cover the prearticular, and it projects

posteromedially. The articular-surangular-coro-

102

Lü, Tomida, Azuma, Dong and Lee

Fig. 4. Comparison CT scans of the corresponding positions of the lower jaws of Nemegtia (GIN100/2112) and

Caenagnathus. A, C: cross section through anterior part of the symphyseal region; B, D: cross section

through the posterior portion of the symphyseal region. Scale bar10 mm (C and D are chosen from Currie

et al., 1993, Fig. 7). Abbreviations as in appendix 1.

Page 9

noid complex occupies the major part, posterior

to the posterior margin of the external mandibu-

lar fenestra. There is a deep, longitudinal groove

on the lateral surface of the anteroventral part of

the articular bone. This depression should be the

insertion area for the external mandibular adduc-

tors. The coronoid process projects posteromedi-

ally.

The angular wedges anteriorly between the

splenial and the ventral process of the dentary,

and disappears anteriorly to the anterior margin

of the external mandibular fenestra. Medially, the

suture with the prearticular is clear, but, laterally,

the suture with the surangular is not clear.

The prearticular is located at the ventromedial

surface of the articular. It is a thin plate-like

bone, which covers a part of the medial surface

of the lower jaw.

The splenial is strap-like, wider anteriorly than

posteriorly. It contacts the dentary anteriorly, but

it does not take part in the formation of the

mandibular symphysis. Its medial surface is

smooth. It becomes thin and sharp posteriorly,

and disappears at the level of 1/3 length of the

lower jaw from the retroarticular process.

2. Postcranial skeleton

A nearly complete cervical series, part of the

dorsal vertebrae, a nearly complete sacrum, both

ilia, the proximal ends of the pubis and ischium,

the proximal end of the femur, parts of the

humerus, and the complete right radius are pre-

served.

Cervical series: The cervical series (Fig. 5)

includes 12 of 13 vertebrae naturally articulated

(including a small part of the axis). Although

sometimes, it is difficult to distinguish the poste-

rior cervical vertebrae from the anterior dorsal

vertebrae, the presence of a weak median ventral

keel and a hypapophysis on the last of the natu-

rally articulated 13 vertebrae indicate that this

last vertebra should be the first dorsal vertebra.

Thus, with the atlas, the number of the cervical

vertebrae is 13. The measurements of the pre-

New Oviraptorid Dinosaur, Mongolia

103

Fig. 5. The cervical vertebrae and the anterior dorsal vertebra (the last one) of Nemegtia barsboldi gen. et sp.

nov. (GIN100/2112). In lateral view (A), and dorsal view (B). Scale bar5 cm. Abbreviations as in appendix

Page 10

served 13 vertebrae are given in Table 1.

Only the posterior part of the axis is preserved,

and it is fused with the third cervical vertebra.

Two postzygapophyseal facets are preserved, and

the distal end of the postzygapophysis is relative-

ly sharp. There is no epipophysis on the dorsal

surfaces of postzygapophyses.

The anterior end of the 3rd cervical vertebra is

almost twice as wide as the posterior end. The

central part of the lateral surface is strongly de-

pressed with a well-developed elongated pleuro-

coel. In the lateral view, the centrum is wedge-

shaped. There is a weak epipophysis on the pos-

terodorsal surface of the postzygapophysis. The

posterior articular end of the centrum is concave.

The neural arch of the 4th cervical vertebra is

obviously larger than that of the 3rd centrum, but

the length of the centrum is slightly shorter than

the anterior one. From the lateral view, the anteri-

or articular end slants anterodorsally at 45°, cor-

responding to the posterior end of the anterior

connected centrum. The anterior margin of the

prezygapophysis slightly outruns the anterior

margin of the centrum, whereas the posterior

margin of the postzygapophysis is at the level of

the posterior end.

In dorsal view, the distance between the

postzygapophyses of the 5th cervical vertebra is

greater than that between the prezygapophyses.

The epipophysis is well developed. The anterior-

ly slanted spine is much lower than the previous

one. The anterior margin of the prezygapophysis

extends far beyond the margin of the anterior end

of the centrum. The posterior margin of the

postzygapophysis is located anterodorsally to the

posterior articular end of the centrum. Compared

with the 4th vertebra, the relative position of the

neural arch on the centrum is greatly changed,

the position of the neural arch is much more for-

ward.

The distance between the prezygapophyses

and that of the postzygapophyses of the 6th cer-

vical vertebra is nearly equal to the length of the

centra, thus forming a square outline. This is dif-

ferent from that of other centrum in that the ante-

rior and posterior widths are equal in dorsal view,

the diapophyseal facets are larger than that of the

anterior ones, but smaller than that of the posteri-

or ones.

In dorsal view, the distance between the prezy-

gapophyses of the 7th cervical vertebra is longer

than that of the postzygapophyses. The lateral

surface of the centrum is concave, with a well-

developed pleurocoel on the central part. The an-

terior margin of the pleurocoel is sharp, its pos-

teroventral margin is not clear. The size, position

and the morphology of epipophysis are similar to

that of the 6th cervical vertebra.

The neural arch of the 8th cervical vertebra is

obviously smaller than those of the anterior ones.

The distance between the prezygapophyses is

slightly greater than that of the postzygapophy-

ses. The epipophysis is relatively small and locat-

ed anteriorly. The facet of the postzygapophysis

faces posteroventrally.

In the 9th to 13th cervical vertebrae, the size

of each vertebra is nearly equal. The position of

pleurocoels on the lateral surface of the centra

gradually moves posteriorly from the 9th cen-

trum to the 13th centrum. It arrives at the central

part of the centrum in the 12th cervical vertebra,

and the pleurocoel is oval. The radius of the pos-

terior margin of the pleurocoel is larger than

those of the anterior ones, while the radius of the

anterior margin of the pleurocoel is larger than

the posterior one in the anterior vertebrae. There

104

Lü, Tomida, Azuma, Dong and Lee

Table 1. The measurements of the 12 (including part of axis but not the atlas) articulated cervical vertebrae and

the first dorsal vertebra (mm).

C10 C11 C12 C13

Length of the centrum

Width of the centrum

30.5 39

42*

52*

“*” represents estimation, “?” missing.

Page 11

are clear fossae posteroventral to the prezy-

gapophyses of the 11th and the 12th cervical ver-

tebrae. The infradiapophyseal fossa is reduced in

the posterior vertebrae. There is no infraprezy-

gapophyseal fossa. The distance between the

prezygapophyses of the five vertebrae is equal to

that of the postzygapophyses, but this distance is

shorter than the length of the centrum. Therefore,

their outlines are rectangle. The facet of the

postzygapophysis faces postolaterally. The poste-

rior part of the ventral surface of the centrum is

flat and slightly expanded.

The neural spines of the cervical vertebrae are

short and centered on neural arch, giving neural

arches an “X” shaped appearance (Fig. 5B), as

that in Caenagnathidae, Microvenator celer, and

other oviraptorids (Makovicky and Sues, 1998).

The vertebrae in the middle of the cervical series

are the largest (5th, 6th and 7th). The facets of

the postzygapophysis from the 3rd to 8th cervical

vertebrae face posteroventrally, while those from

the 9th to 13th vertebrae face posterolaterally. An

infradiapophyseal fossa appears in the 11th and

12th cervical vertebrae. In dorsal view, the out-

line formed by the four zygapophyses in the sixth

cervical is nearly square, it is rectangle (the angle

formed by the right-left prezygapophysis with the

centered neural arch is greater than 90°) in the

4th and 5th cervical vertebrae, but the angle is

less than 90° after the seventh cervical vertebrae.

There are also variations in the pleurocoels. They

are relatively small in anterior vertebrae. The an-

terior margin of the oval pleurocoel is larger than

that of the posterior margin from the 3rd to 9th

cervical vertebrae, whereas the pleurocoel of the

10th is circular. From the 11th to the last cervical

vertebrae, the posterior margin of the pleurocoel

is larger than its anterior margin.

Almost all the cervical ribs are preserved. The

axis rib is slender, rod-like and it is the longest

one among the cervical ribs. It extends to the an-

terior end of the 4th cervical rib. The 3rd cervical

rib is also rod-like, longer than the length of the

3rd centrum.

The 5th and the 6th cervical ribs are very simi-

lar. The rib head is fused with the parapophysis,

while the rib tubercle is not fused with the cen-

trum. The rib tubercle of the 5th cervical rib is

smaller than that of the 6th cervical rib.

In the 7th to 12th cervical ribs, the rib tubercle

and rib head are completely fused with the di-

apophyses and parapophyses of the centra. The

tubercle of the 13th cervical rib is clearly not

fused with parapophysis, and its morphology is

similar to the anterior ones.

Dorsal vertebrae: Only the anterior one and

a half dorsal vertebrae, and three posterior dorsal

vertebrae and two neural arches, which articulat-

ed with the sacral vertebrae, are preserved. The

count of dorsal vertebrae is uncertain.

The pleurocoel in the 1st dorsal vertebra is

much larger than those of cervical vertebrae. The

distance between the postzygapophyses becomes

shorter. There is a weak ventral keel. A clear pro-

jection on the middle part of the ventral margin

of the anterior end of the vertebra represents the

hypapophysis. According to Sues (1997), the

presence of the hypapophysis generally appears

to be restricted to the first three or four dorsal

vertebrae in non-avialan theropods. Compared

with the cervical vertebrae, the first dorsal verte-

bra is short and high, and the posterior articular

end is flat. The spine is triangular in anterior and

posterior views. The anterior margin of the

prezygapophyseal facet approaches the middle of

the anterior vertebra, unlike that of the cervical

vertebrae. Only the anterior part of the 2nd dor-

sal vertebra is preserved, and the anterior articu-

lar surface is strongly concave.

On the five preserved posterior dorsal verte-

brae, only the neural arches of the anterior two

posterior dorsals are preserved. The base of the

transverse process is wider than on posterior ver-

tebra. The pleurocoel in the middle two is large

and circular. The neural spine is wide, thin and

plate-like. Its width is nearly 2/3 of the length of

the vertebra.

The last dorsal vertebra is located medial to

the anterior end of the ilium, and has a free rib,

but its transverse process contacts the ilium. The

ventral surface of the centrum is round. There is

a large pleurocoel in the middle of the lateral sur-

New Oviraptorid Dinosaur, Mongolia

105

Page 12

face of the centrum. The posterior articular end

is not fused to the first sacral vertebra and the ar-

ticular surface is slightly concave.

Sacral vertebrae: There are 8 sacral verte-

brae (Fig. 6). The anterior six are completely pre-

served, the 7th sacral vertebra is partly preserved,

and the 8th has preserved only a small anterior

part and its neural arch. Whether their spines are

fused together or not is uncertain, due to their

being covered by the matrix, but it is clear that

the distal ends of the spines do not rise above the

dorsal margins of the ilia. The measurements of

the sacral vertebrae are given in Table 2.

The width of the anterior end of the 1st sacral

106

Lü, Tomida, Azuma, Dong and Lee

Fig. 6. Sacrum and pelvic girdle of Nemegtia barsboldi gen. et sp. nov. (GIN100/2112).

Page 13

vertebra is greater than its posterior end. The an-

terior articular end is strongly concave. The ven-

tral surface of the centrum is smooth. The 2nd

sacral vertebra is similar to the 1st, except that

the pleurocoel is elongated and enlarged, and it

occupies nearly 1/2 the length of the centrum.

The width of the anterior end of the 3rd centrum

is larger than its posterior end, while in the 4th

and 5th centra, the width of both ends is equal.

There is a shallow groove on the anterior portion

of the ventral surface in the 4th centrum. This is

clearly different from other centra. The pleuro-

coel becomes smaller, and the sacral rib of the

4th vertebra is located at the same level as the is-

chiadic peduncle. The anterior articular end is

larger than the posterior in the 6th centrum, and

the contact area of the rib with the centrum is the

largest among the sacrals, occupying nearly the

whole length of the centrum. Only a small part of

its anterior end is fused with the posterior end of

the 5th centrum. The small pleurocoel is located

on the mid-posterior part of the centrum. The

distal end of the 7th sacral rib is divided into two

parts, which contact the ilium. Posterior to where

the sacral rib is fused with the sacral centrum,

there are two fenestrae, which are nearly parallel

to each other, and they are much larger than the

anterior pleurocoels. The right sacral rib of the

8th sacral vertebra is fused with the transverse

process of the centrum and is nearly rod-like.

The middle of the rib shaft is relatively slender,

but both ends are widened. Its distal end tightly

contacts the internal surface of the ilium. The

preserved part shows that the fusion with the

connected centra is weaker than that of other cen-

trum.

Caudal vertebrae: Only the neural arches of

the first two caudal vertebrae are preserved. The

nearly vertical facet of the prezygapophysis faces

inwards, while the postzygapophysis faces later-

ally.

Shoulder girdle and the fore limb: The proxi-

mal end of the left scapula is preserved. It is sim-

ilar to those of other oviraptorosaurs. Distal parts

of both humeri are preserved (Figs. 7A, B). The

preserved portions are similar to other ovirap-

torosaurs, but there is a fossa on the anterior sur-

face (Fig. 7A, fmb) occupying a similar position

to the fossa m. brachialis (Baumel and Witmer,

1993) in modern birds. This is not observed in

other oviraptorosaurs.

The right radius is completely preserved (Figs.

7C, D). It is straight, and the cross section of the

shaft is oval. Both ends are slightly expanded, but

the proximal is larger.

Pelvic girdle and the hind limb (Fig. 6): The

preserved length of the ilium is 280 mm. In later-

al view, the dorsal margin of the ilium is straight,

and the depth of the preacetabular process is the

same as that of the postacetabular process. The

anterodorsal part of the preacetabular process is

concave, while its ventral part is convex. The

pubic peduncle is slender and projects down-

wards and its articular end is triangular. The is-

chiadic peduncle is relatively stout with a round

articular end. Dorsally, the ilia nearly meet, but

are not fused. The same condition is also found

in other oviraptorosaurs, such as Nomingia gobi-

ensis (Barsbold, Osmólska et al., 2000).

Only the proximal part of both ischia is pre-

served. The process that contacts the pubis is

smaller than that which articulates with the ilium.

The proximal ends of both pubes are preserved.

Only the proximal end of the femur is pre-

served. There is a clear neck between the femoral

head and the shaft of the femur. The angle of the

neck and the shaft is about 90°. There is a shal-

low groove between the greater trochanter and

New Oviraptorid Dinosaur, Mongolia

107

Table 2. The measurements of the sacral vertebrae (mm).

Length of the centrum

Width of the centrum

“?” missing.

Page 14

the femoral head. The lesser trochanter is smaller

and fingerlike. It connects tightly with the greater

trochanter. The broken surface shows that many

bony struts are distributed near the lateral wall of

the shaft.

Comparison and Discussion

The high, narrow and short skull with tooth-

less jaws and the medial process of the articular

indicate that Nemegtia barsboldi is a derived ovi-

raptorosaur (Barsbold and Osmólska, 1990;

Barsbold, 1997). Nemegtia is different from Ovi-

raptor in the skull crest. The anterior margin of

the skull crest is vertical and the highest point of

the crest is located between the nasals and pre-

maxillae in Nemegtia barsboldi. The dorsal ante-

rior margin of the premaxilla projects forward

and the ventral anterior margin of the crest is

concave posteriorly in the specimen GIN 100/42,

which was described as Oviraptor philoceratops

by Barsbold et al. (1990). The highest point of

the crest is located above between the lacrimals

in O. mongoliensis, although the anterior margin

of the crest is nearly vertical (Barsbold and

Osmólska, 1990). Dorsally, the premaxillae are

barely exposed, and the anteroposterior length of

the parietal is greater than the frontal in Nemeg-

tia barsboldi, but the exposed nasal process of

the premaxilla is much larger than the nasals, and

the lengths of the parietal and the frontal are

nearly equal in O. philoceratops (Barsbold and

Osmólska, 1990). An additional opening appear-

ing on the wall within the antorbital cavity in Ne-

megtia barsboldi is smaller than that of the O.

philoceratops. The posterior margin of the

quadrate condyle is distinctly posterior to the an-

teroventral margin of the occipital condyle in O.

philoceratops. The shape of the ilium, and the

number of the sacral vertebrae are also different

between Nemegtia and Oviraptor.

Nemegtia barsboldi has eight sacral vertebrae

whereas Ingenia yanshini has six or seven. Ne-

megtia barsboldi has a distinct crest in contrast

108

Lü, Tomida, Azuma, Dong and Lee

fmb

Fig. 7. The left humerus in anterior view (A) and posterior view (B) and the right radius in medial view (C) and

lateral view (D) of Nemegtia barsboldi gen. et sp. nov. (GIN100/2112). Scale bar4 cm. Abbreviations as in

appendix 1.

Page 15

to Ingenia yanshini. The ischiadic peduncle of

the ilium is relatively stout, and the pubic pedun-

cle is relatively weak in Nemegtia barsboldi, as

opposed to the condition in Ingenia yanshini.

Nemegtia barsboldi is different from Con-

choraptor gracilis in having a developed crest,

while the skull has no crest in Conchoraptor gra-

cilis.

Nemegtia barsboldi is different from Citipati

osmolskae, in that the convex crest is higher than

the dorsal surface of the skull, and the occiput

and the quadrate are vertical, while the top of the

skull crest is below the dorsal surface of the

skull, the occiput and the quadrate sloped an-

terodorsally in Citipati osmolskae (Clark et al.,

2001; 2002), the width and the length of the cer-

vical vertebrae are not so different in Nemegtia

barsboldi, but the length of the cervical vertebrae

is about two times longer than their width in Citi-

pati osmolskae. The phylogenetic analysis shows

that Nemegtia barsboldi is more closely related

to Citipati osmolskae than to any other ovirap-

torosaurs (Fig. 8).

Nemegtia barsboldi differs from Khaan

mckennai (Clark et al., 2001) in having both a

distinct skull crest and a parietal crest, which are

New Oviraptorid Dinosaur, Mongolia

109

Allosauroidea

Ornithomimidae

Dromaeosauridae

Troodontidae

Archaeopteryx lithographica

Avimimus portentosus

Chirostenotes pergracilis

Nomingia gobiensis

Ingenia yanshini

Conchoraptor gracilis

Heyuannia huangi

Nemegtia barsboldi

Citipati osmolskae

GIN 100/42

"Oviraptor" mongoliensis

Oviraptor philoceratops

Khaan mckennai

Microvenator celer

Caudipteryx zoui

Incisivosaurus gauthieri

Fig. 8. One most parsimonious tree found by PAUP.

Page 16

lacking in Khaan. The dorsal process of the pre-

maxilla is vertical in Nemegtia barsboldi, but it

extends posterodorsally in Khaan mckennai, the

jugal extends further anteriorly in Nemegtia bars-

boldi, but it extends further posteriorly in Khaan

mckennai.

Nemegtia barsboldi is different from Nomingia

gobiensis in that having eight sacral vertebrae,

compared to five in Nomingia gobiensis; the dor-

sal margin of the ilium is straight in Nemegtia

barsboldi, but convex in Nomingia gobiensis; the

pubic peduncle is more slender and weaker in

Nemegtia barsboldi, compared to Nomingia gob-

iensis. Nomingia has a pygostyle (Barsbold, Cur-

rie et al., 2000).

Nemegtia barsboldi differs Microvenator celer

in having one pleurocoel on the cervicals and hy-

pophysis on the ventral surface of dorsal verte-

brae in Nemegtia, compared with two pleuro-

coels and no hypophysis in Microvenator celer;

the number of sacral vertebrae is at least 8 in Ne-

megtia barsboldi, it is less than 6 in Microvena-

tor celer.

Nemegtia barsboldi differs from Caudipteryx

zoui in that Nemegtia barsboldi has no premaxil-

lary teeth, while Caudipteryx zoui has premaxil-

lary teeth. Nemegtia has a well-developed crest

in the skull, but the skull of Caudipteryx zoui

lacks a crest. Additionally, the pubic peduncle is

slender and weaker than ischiadic peduncle in

Nemegtia barsboldi, but it is stronger in

Caudipteryx zoui.

Nemegtia barsboldi differs from Heyuannia

huangi (Lü, 2003) in that the articular surface for

the quadratojugal is convex, while it is more

groove-like in Heyuannia huangi. Nemegtia

barsboldi has a developed skull crest, whereas

Heyuannia huangi does not. In the lateral view,

the dorsal margin of the ilium is straight through-

out its length, while it is convex in Heyuannia

huangi.

The relative position of the external nasal

opening and the antorbital cavity varies in differ-

ent oviraptorosaurs. The apparent primitive state

is with the dorsal margin of the external nasal

opening below or near the level of the dorsal

margin of the antorbital cavity (Fig. 9), such as in

Incisivosaurus gauthieri. In Caudipteryx sp.

(IVPP V 12430, Zhou et al., 2000), although the

skull is heavily crushed, it still can be inferred

that the dorsal margin of the external nasal open-

ing is below the dorsal margin of the antorbital

cavity. Derived forms have the dorsal margin of

the external nasal opening above the dorsal mar-

gin of the antorbital cavity, a larger distance

being more derived. This condition in Nemegtia

barsboldi shows that Nemegtia barsboldi is a

more derived oviraptorosaur.

The morphology of the lower jaw of Nemegtia

shows that it is similar to those of derived ovirap-

torosaurs, Oviraptor philoceratops and O. mon-

goliensis, Ingenia yanshini (Barsbold et al.,

1990), Citipati osmolskae (Clark et al., 2002),

Khaan mckennai (Clark et al., 2001), and

Heyuannia huangi (Lü, 2003), in which the lower

jaw is short and deep, the upper margin of the ex-

ternal mandibular fenestra is formed by the

surangular and the dorsal branch of the dentary,

the surangular has a middle branch, which pro-

jects into the external mandibular fenestra. The

ventral surface of the lower jaw is convex, and

the mandibular symphysis is short. This is differ-

ent from that of Caenagnathus collinsi (Stern-

berg, 1940; Cracraft, 1971; Currie et al., 1993),

where the lower jaw is slender and shallow, the

mandibular symphysis is also longer, the external

mandibular fenestra is elongate with a dorsal

margin, which is formed by the surangular, and

the surangular had no middle branch. Inci-

sivosaurus gauthieri (Xu, Cheng et al., 2002)

also has a longer lower jaw, similar to that of

Caenagnathus collinsi, but it bears teeth. It

may represent a primitive caenagnathid ovirap-

torosaur.

Eight sacral vertebrae are present in Nemegtia

barsboldi, similar to some derived oviraptors but

more than most known non-avian theropod

dinosaurs. More basal theropods, such as

Compsognathus longipes (Ostrom, 1978), and

Dilophosaurus wetherilli (Welles, 1984) have

four sacrals, five in the relatively advanced

theropods-Ceratosaurus (Gilmore, 1920), Al-

110

Lü, Tomida, Azuma, Dong and Lee

Page 17

New Oviraptorid Dinosaur, Mongolia

111

4CM

5CM

Fig. 9. Skulls of (A) Incisivosaurus gauthieri, (B) GIN 100/2112, (C) Khaan mckennai (from Clark et al.,

2001), (D) Oviraptor philoceratops (reversed, from Clark et al., 2002), (E) Conchoraptor gracilis, (F) “Ovi-

raptor” mongoliensis, (G) Citipati osmolskae (from Clark et al., 2002), and (H) Oviraptor philoceratops

(GIN 100/42) (from Barsbold, 1986). Scale bar5 cm in A, B, E and F, 2 cm in C and D, 4 cm in H.

Page 18

losaurus fragilis (Madsen, 1976), Nomingia gob-

iensis (Barsbold, Osmólska et al., 2000), six or

seven in Oviraptor (Barsbold et al., 1990), and

eight in Heyuannia huangi (Lü, 2003).

Nemegtia barsboldi independently acquired

the avian characters such as the fused premaxil-

lae, as in Confuciusornis sanctus (Chiappe et al.,

1999), toothless jaws, and the presence of nutri-

ent openings on the premaxilla and the maxilla as

in other oviraptorids.

Sharp ventrolateral margins of the premaxillae

indicate that a keratinous structure probably cov-

ered the end of the rostrum, as in Erlicosaurus

andrewsi (Clark et al., 1994), ornithomimids

(Norell, Makovicky, and Currie, 2001; Kobayashi

and Lü, 2003), and birds. The rod-like jugal is

similar to that of Confuciusornis sanctus and

other birds (Elzanowski, 1999), the mobile con-

dition between the quadrate and the quadratoju-

gal is similar to the condition of most birds, and

so are the increased cervical and sacral vertebral

counts.

Phylogenetic Analysis

In order to determine the phylogenetic status

of Nemegtia barsboldi among oviraptorosaurs,

20 taxa and 200 characters (107 cranial and 93

postcranial characters) (Appendices 2 and 3) are

used for this analysis. All characters are equally

weighted and unrooted. The character/matrix are

modified from Maryanska et al. (2002). Five new

genera of Oviraptorosauria are added to the ma-

trix; these include Khaan mckennai (Clark et al.,

2001), Citipati osmolskae (Clark et al., 2001,

2002), Incisivosaurus gauthieri (Xu, Cheng et

al., 2002), Nemegtia barsboldi, and Heyuannia

huangi (Lü, 2003). The aim of this analysis is to

determine the phylogenetic position of Nemegtia

barsboldi among oviraptorosaurs, so the most

primitive forms such as Herrerasaurus is-

chigualastensis, Coelophysis bauri, and the spe-

cialized forms such as Tyrannosauridae and Al-

varezsauridae, which were used by Maryanska et

al. (2002) as the outgroup in their analysis, are

excluded from the present analysis. Although Mi-

crovenator celer and Troodontidae were exclud-

ed in Maryanska et al.’s analysis due to the large

amount of missing data they thought (Maryanska

et al., 2002), taxa should not be excluded a priori

from phylogenetic analysis based only on the

number of preserved characters (Kearney and

Clark, 2003). Therefore, in the present analysis,

these taxa are still employed. Most characters are

from Maryanska et al. (2002). New characters

and their sources in appendix 2 are in bold.

Phylogenetic analysis was performed using

MacClade 3.08 (Maddison and Maddison, 1992)

and PAUP 4.0b (Swofford, 1998). Because of the

large data set (20 taxa) and many missing charac-

ter states, a Heuristic Search was used (Swofford

and Begle, 1993), with branch-swapping options

of the TBR swapping algorithm method. The

analysis resulted in one most parsimonious tree

(tree length481; consistency index0.5073; re-

tention index0.6269) (Fig. 8). This tree shows

that Oviraptorosauria forms a monophyletic

group. Nineteen unambiguous synapomorphies

support this clade. These characters are 1(1),

9(1), 23(1), 37(1), 40(1), 41(2), 43(1), 46(1),

47(1), 55(1), 58(1), 69(1), 71(1), 76(1), 77(1),

81(1), 84(1), 86(1) and 88(1). The primitive

forms, such as Incisivosaurus gauthieri,

Caudipteryx zoui, Microvenator celer, and Avim-

imus portentosus form an ascending sequence

from most basal to more derived in the tree.

Avimimus portentosus is a basal form of Caenag-

nathoidea. Following Barsbold (1981; 1983),

Maryanska et al. (2002) used GIN 100/42 (IGM

100/42) as the representative of Oviraptor philo-

ceratops in their phylogenetic analysis. The pre-

sent analysis shows that GIN100/42 and Ovirap-

tor philoceratops are closely related. GIN100/42

either belongs to a different species of the same

genus as O. philoceratops or the same species,

the latter differs from Clark et al.’s judgment

(Clark et al., 2002). Nemegtia barsboldi is close-

ly related to Citipati osmolskae.

Conclusion

Nemegtia barsboldi is distinguished by at least

112

Lü, Tomida, Azuma, Dong and Lee

Page 19

five autapomorphies from other known ovirap-

torosaurs (vertical skull crest, anteroposterior

length of the frontal approximately 25% of that

of the parietals in dorsal view, less exposed nasal

process of the premaxilla on the dorsal surface of

the skull, a process on the quadrate projecting

into the cotyla of the quadatojugal, mandibular

condyles of the quadrate situated rostrally to the

occipital condyle). Phylogenetic analysis shows

that Nemegtia barsboldi is closer to Citipati os-

molskae than to other oviraptorosaurs.

Acknowledgments

The authors would like to thank Drs. Louis L.

Jacobs, Dale A. Winkler (SMU, USA), H.

Osmólska (Poland), R. E. Molnar (USA), R.

Barsbold (Mongolia), and Y. Kobayashi (Japan)

for their valuable comments on the early versions

of the manuscript, and to acknowledge the mem-

bers of the Mongol Highland International Di-

nosaur Project. Special thanks go to Y.

Kobayashi, who found this beautiful specimen in

1996. Thanks also go to Drs. Zhou, Z. and Xu,

X. (IVPP of Chinese Academy of Sciences, Bei-

jing, China), for access of the specimens in their

care. This project is supported by Chunichi Shin-

bun Co. Ltd., Kyoto Kagaku Co. Ltd., Chukyo

TV Broadcasting Co. Ltd., and Tokai Bank Ltd.;

plus Institute for the Study of Earth and Man at

Southern Methodist University, the Jurassic

Foundation, and the Chang Ying-Chien Science

Grant for USA-China Collaborative Field Re-

search to Junchang Lü.

References

Barsbold, R., 1976. [On a new Late Cretaceous family of

small theropods (Oviraptoridae fam. n.) of Mongolia].

Doklady Akademia Nauk SSSR, 226(3): 685–688 (in

Russian).

Barsbold, R., 1977. Kinetism and peculiarity of the jaw

apparatus of oviraptors (Theropoda, Saurischia). Trans-

actions, Joint Soviet-Mongolia Geological Expedition,

4: 37–47 (in Russian).

Barsbold, R., 1981. Toothless carnivorous dinosaurs of

Mongolia. Trudy Sovmestnoi Sovetsko-Mongol’skoi Pa-

leontologicheskoi Ekspeditsii, 15: 28–39.

Barsbold, R., 1983. Carnivorous dinosaurs from the Cre-

taceous of Mongolia. Trudy Sovmestnoi Sovetsko-Mon-

gol’skoi Paleontologicheskoi Ekspeditsii, 19: 5–120.

Barsbold, R., 1986. [Raubdinosaurier Oviraptoren]. In

Vorobyeva, E. I. (ed). Herpetologische Untersuchungen

in der Mongolischen Volksrepublik. Pp. 210–223. Akad.

Nauk. S. S. S. R. Inst. Evolyucionnoy Morfologii i

Ekologii Zhivotnykh im. A. M. Severtsova, Moskva (in

Russian with German summary).

Barsbold, R., 1997. Oviraptorosauria. In Currie, P. J. and

Padian, K. (eds.). Encyclopedia of Dinosaurs. Pp.

505–509. Academic Press, San Diego.

Barsbold, R., Currie, P. J., Myhrvold, N. P., Osmólska, H.,

Tsogtbaatar, K. & Watabe, M., 2000. A pygostyle from

a non-avian theropod. Nature, 403: 155–156.

Barsbold, R., Maryanska, T. & Osmólska, H., 1990. Ovi-

raptorosauria. In Weishampel, D. B., Dodson, P. and

Osmólska, H. (eds.). The Dinosauria. Pp. 249–258.

University of California Press, Berkeley.

Barsbold, R. & Osmólska, H., 1999. The skull of Veloci-

raptor (Theropoda) from the Late Cretaceous of Mon-

golia. Acta Palaeontologica Polonica, 44(2): 189–219.

Barsbold, R., Osmólska, H., Watabe, M., Currie, P. J. &

Tsogtbaatar, K., 2000. A new oviraptorosaur (dinosaur,

theropod) from Mongolia: the first dinosaur with a py-

gostyle. Acta Palaeontologica Polonica, 45(2): 97–106.

Baumel, J. J. & Witmer, M., 1993. Osteologia. In Baumel,

J. J., King, A. S. et al. (eds.). Handbook of avian anato-

my: nomina anatomica avium, second edition. Pp.

45–132. Cambridge, Massachusetts.

Bonaparte, J. F., Novas, F. E. & Coria, R. A., 1990.

Carnotaurus sastrei Bonaparte, the horned, lightly built

carnosaur from the Middle Cretaceous of Patagonia.

Contributions in Science Nat. History Museum Los An-

geles County, 416: 1–42.

Chiappe, L. M., 1996. Late Cretaceous birds of southern

South America: Anatomy and systematics of enantior-

nithes and Patagopteryx deferrariisi. Munchner Ge-

owiss. Abh. (A), 30: 203–224.

Chiappe, L. M., 2001. Phylogenetic relationships among

basal birds. In Gauthier, J. and Gall, L. F. (eds.). New

perspectives on the origin and early evolution of birds,

Proceedings of the International Symposium in Honor

of John H. Ostrom. Pp. 125–139. Yale Peabody Muse-

um.

Chiappe, L. M., 2002. Basal bird phylogeny: Problems

and solutions. In Chiappe L. M. and Witmer L. M.

(eds.). Mesozoic birds above the heads of dinosaurs.

Pp. 448–472. University of California Press. Berkeley.

Chiappe, L. M., Ji, S-A., Ji, Q. & Norell, M. A.,1999.

Anatomy and systematics of the Confuciusornithidae

(Theropoda: Aves) from the late Mesozoic of north-

eastern China. Bulletin of the American Museum of

New Oviraptorid Dinosaur, Mongolia

113

Page 20

Natural History, 242: 1–89.

Chiappe, L. M., Norell, M. A. & Clark, J. M., 1996. Phy-

logenetic position of Mononykus (Aves: Alvarezsauri-

dae) from the Late Cretaceous of the Gobi Desert.

Memoirs of the Queensland Museum, 39(3): 557–582.

Chiappe, L. P., Norell, M. A. & Clark, J. M., 1998. The

skull of a relative of the stem-group bird Mononykus.

Nature, 392: 275–278.

Clark, J. M., Altangerel, P. & Norell, M. A., 1994. The

skull of Erlicosaurus andrewsi, a Late Cretaceous seg-

nosaur (Theropoda: Therizinosauridea) from Mongolia.

American Museum Novitates, 3115: 1–39.

Clark, J. M., Norell, M. A. & Barsbold, R., 2001. Two

new oviraptorids (Theropoda: Oviraptorosauria), Upper

Cretaceous Djadokhta Formation, Ukhaa Tolgod, Mon-

golia. Journal of Vertebrate Paleontology, 21(2): 209–

213.

Clark, J. M., Norell, M. A. & Chiappe, L.M., 1999. An

oviraptorid skeleton from the late Cretaceous of Ukhaa

Tolgod, Mongolia, preserved in an avian-like brooding

position over an oviraptorid nest. American Museum

Novitates, 3265: 1–36.

Clark, J. M., Norell, M. A. & Rowe, T., 2002. Cranial

anatomy of Citipati osmolskae (Theropoda, Ovirap-

torosauria), and a reinterpretation of the holotype of

Oviraptor philoceratops. American Museum Novitates,

3364: 1–24.

Colbert, E. H. & Russell, D. A., 1969. The small Creta-

ceous dinosaur Dromaeosaurus. American Museum

Novitates, 2380: 1–49.

Cracraft, J., 1971. Caenagnathiformes: Cretaceous birds

convergent in jaw mechanism to dicynodont reptiles.

Journal of Paleontology, 45(2): 805–809.

Cracraft, J. 1986. The origin and early diversification of

birds. Paleobiology, 12: 383–399.

Currie, P. J., 1985. Cranial anatomy of Stenonychosaurus

inequalis (Saurischia, Theropoda) and its bearing on

the origin of birds. Canadian Journal of Earth

Sciences, 22: 1643–1658.

Currie, P. J., 1995. New information on the anatomy and

relationships of Dromaeosaurus albertensis (Di-

nosauria: Theropoda). Journal of Vertebrate Paleontol-

ogy, 15: 576–591.

Currie, P. J., 2000. Theropods from the Cretaceous of

Mongolia. In Benton, M. J., Shishkin, M. A., Unwin, D.

M. and Kurochkin, E. N. (eds.). The age of dinosaurs in

Russia and Mongolia. Pp. 434–455. Cambridge Uni-

versity Press, Cambridge.

Currie, P. J., Godfrey, S. J. & Nessov, L., 1993. New cae-

nagnathid (Dinosauria: Theropoda) specimens from the

Upper Cretaceous of North America and Asia. Canadi-

an Journal of Earth Sciences, 30: 2255–2272.

Currie, P. J. & Russell, D. A., 1988. Osteology and rela-

tionships of Chirostenotes (Saurischia, Theropoda)

from the Judith River (Oldman) Formation of Alberta,

Canada. Canadian Journal of Earth Sciences, 25: 972–

986.

Currie, P. J., Vickers-Rich, P. & Rich, T. H., 1996. Possi-

ble oviraptorosaur (Theropoda, Dinosauria) specimens

from the Early Cretaceous Otway Group of Dinosaur

Cove, Australia. Alcheringa, 20: 73–79.

Currie, P. J. and Zhao, X.-J., 1993a: A new carnosaur (Di-

nosauria: Theropoda) from the Jurassic of Xinjiang,

People’s Republic of China. Canadian Journal of Earth

Sciences, 30: 2037–2081.

Currie, P. J. & Zhao, X.-J., 1993b. A new troodontid (Di-

nosauria, Theropoda) braincase from the Dinosaur Park

Formation (Campanian) of Alberta. Canadian Journal

of Earth Sciences, 30: 2231–2247.

Dong, Z.-M. & Currie, P. J., 1996. On the discovery of an

oviraptorid skeleton on a nest of eggs at Byan Man-

dahu, Inner Mongolia, People’s Republic of China.

Canadian Journal of Earth Sciences, 33: 631–636.

Elzanowski, A., 1999. A comparison of the jaw skeleton

in theropods and birds, with a description of the palate

in the Oviraptoridae. In Olson, S. L. (ed.). Avian Pale-

ontology at the Close of the 20th Century: Proceedings

of the 4th International Meeting of the Society of Avian

Paleontology and Evolution, Washington, D.C., 4–7

June 1996. Smithsonian Contributions to Paleobiology,

89: 311–323.

Frankfurt, N. G. & Chiappe, L. M., 1999. A possible ovi-

raptorosaur from the Late Cretaceous of northwestern

Argentina. Journal of Vertebrate Paleontology, 19(1):

101–105.

Frey, E. & Martill, D. M., 1995. A possible ovirap-

torosaurid theropod from the Santana Formation

(Lower Cretaceous, ?Albian) of Brazil. Neues Jahrbuch

für Geologie und Paläontologie Monatshefte, 7: 397–

412.

Gauthier, J., 1986. Saurischian monophyly and the origin

of birds. In Padian K. (ed.). The origin of birds and the

evolution of flight. Memoirs of the California Academy

of Sciences 8. San Francisco, Pp. 1–55.

Gilmore, C. W., 1920. Osteology of the carnivorous Di-

nosauria in the United States National Museum,

with special reference to the genera Antrodemus (Al-

losaurus) and Ceratosaurus. United States National

Museum, Bulletin, 110: 1–154.

Gradzinski, R., Kazmierczak, J. & Lefeld, J., 1968. Geo-

graphical and geological data from the Polish-Mongo-

lian Palaeontological Expeditions. Palaeontologia

Polonica, 19: 33–92.

Holmgren, N., 1955. Studies on the phylogeny of birds.

Acta Zoologica, 36: 243–328.

Holtz, T. R. Jr., 1994. The arctometatarsalian pes, an un-

usual structure of the metatarsus of Cretaceous

Theropoda (Dinosauria: Saurischia). Journal of Verte-

114

Lü, Tomida, Azuma, Dong and Lee

Page 21

brate Paleontology, 14(4): 480–519.

Holtz, T. R. Jr., 2000. A new phylogeny of the carnivorous

dinosaurs. Gaia, 15: 5–61.

Holtz, T. R. Jr., 2001. Arctometatarsalia revised: The

problem of homoplasy in reconstructing theropod phy-

logeny. In Gauthier, J. and Gall, L. F. (eds.). New per-

spectives on the origin and Early evolution of birds:

Proceedings of the International Symposium in Honor

of John H. Ostrom. Pp. 99–122. Peabody Museum of

Natural History. Yale University, New Haven.

Hwang, S.-H., Norell, M. A., Ji, Q. & Gao, K.-Q., 2002.

New specimens of Microraptor zhaoianus (Theropoda:

Dromaeosauridae) from Northeastern China. American

Museum Novitates, 3381: 1–44

Hwang, S.-H., Norell, M. A., Ji, Q. & Gao, K.-Q., 2004.

A large compognathid from the Early Cretaceous Yixi-

an Formation of China. Journal of Systematic Palaeon-

tology, 2(1): 13–30.

Jerzykiewicz, T. & Russell, D. A., 1991. Late Mesozoic

stratigraphy and vertebrates of the Gobi Basin. Creta-

ceous Research, 12: 345–377.

Ji, Q., Currie, P. J., Norell, M. A. & Ji, S-A., 1998. Two

feathered dinosaurs from northeastern China. Nature,

393: 753–761.

Ji, Q., Ji, S.-A., You, H.-L., Zhang, J.-P., Yuan, C.-X., Ji,

X.-X., Li, J.-L. & Li, Y.-X., 2002. Discovery of an

avialae bird Shenzhouraptor sinensis gen. et sp. nov. —

from China, Geological Bulletin of China 21(7): 363–

369.

Ji, Q., Ji, S.-A., You, H.-L., Zhang, J.-P., Zhang, H.-B.,

Zhang, N.-J., Yuan, C.-X. & Ji, X.-X., 2003. An Early

Cretaceous avialan bird, Shenzhouraptor sinensis from

western Liaoning, China. Acta Geologica Sinica, 77(1):

21–26.

Kearney, M. & Clark, J. M., 2003. Problems due to miss-

ing data in phylogenetic analyses including fossils: a

critical review. Journal of Vertebrate Paleontology,

23(2): 263–274.

Kobayashi, Y. & Lü, J.-C., 2003. A new ornithomimid di-

nosaur with gregarious habits from the Late Cretaceous

of China. Acta Palaeontologica Polonica, 48(2): 235–

259.

Lü, J.-C., 2003. A new oviraptorosaurid (Theropoda: Ovi-

raptorosauria) from the Late Cretaceous of southern

China. Journal of Vertebrate Paleontology, 22(4):

871–875.

Lü, J.-C., Dong, Z.-M., Azuma,Y., Barsbold, R. & Tomi-

da, Y., 2002. Oviraptorosaurs compared to birds. In

Zhou, Z.-H., and Zhang, F.-C. (eds.). Proceedings of

the 5th Symposium of the Society of Avian Paleontol-

ogy and Evolution, Science Press, Beijing China, Pp.

175–189.

Lü, J.-C., Huang, D. & Zhang, S., 2000. The first discov-

ery of oviraptorosaur from Guangdong Province. Verte-

brata PalAsiatica, 38(2): 99.

Maddison, W. P. & Maddison, D. R., 1992. MacClade:

analysis of phylogeny and character evolution, version

3. Sinauer Associates, Inc. Publishers, Sunderland,

Massachusetts.

Madsen, J. H. Jr., 1976. Allosaurus fragilis: a revised oste-

ology. Utah Geological and Mineral Survey, Bulletin,

109: 1–163.

Makovicky, P. J. & Sues, H.-D., 1998. Anatomy and phy-

logenetic relationships of the theropod dinosaur Mi-

crovenator celer from the Lower Cretaceous of Mon-

tana. American Museum Novitates, 3240: 1–27.

Marsh, O. C., 1881. Classification of the Dinosauria. The

American Journal of Science, Third series, 23: 81–86.

Maryanska, T. & Osmólska, H., 1997. The quadrate of

oviraptorid dinosaurs. Acta Palaeontologica Polonica,

42(3): 361–371.

Maryanska, T., Osmólska, H. & Wolsan, M., 2002.

Avialan status for Oviraptorosauria. Acta Palaeontolog-

ica Polonica, 47(1): 97–116.

Molnar, R. E., 1973. The cranial morphology and me-

chanics of Tyrannosaurus rex (Reptilia: Saurischia).

PhD. Diss., University of California at Los Angeles.

Molnar, R. E., 1991. The cranial morphology of Tyran-

nosaurus rex. Palaeontographica, 217(4–6): 137–176.

Norell, M. A., Clark, J. M., Chiappe, L. M. & Dashzeveg,

D., 1995. A nesting dinosaur. Nature, 378: 774–776.

Norell, M. A., Clark, J. M., Dashzeveg, D., Barsbold, R.,

Chiappe, L. M., Davidson, A. K., Mckenna M. C., Alt-

tanerel, P. & Novacek, M. J., 1994. A theropod dinosaur

embryo and the affinities of the Flaming Cliffs Di-

nosaur eggs. Science, 266: 779–782.

Norell, M. A., Clark, J. M. & Makovicky, P. J., 2001.

Phylogenetic relationships among coelurosaurian

theropods. In Gauthier, J. and Gall, L. F. (eds.). New

perspectives on the origin and Early evolution of birds:

Proceedings of the International Symposium in Honor

of John H. Ostrom. Pp. 49–67. Peabody Mus. Nat. Hist.

Yale University, New Haven.

Norell, M. A., Makovicky, P. J. & Currie, P. J., 2001. The

beaks of ostrich dinosaurs. Nature, 412: 873–874.

Novas, F. E., 1993. New information on the systematics

and postcranial skeleton of Herrerasaurus ischigualast-

ensis (Theropoda: Herrerasauridae) from the Is-

chigualasto Formation (Upper Triassic) of Argentina.

Journal of Vertebrate Paleontology, 13: 400–423.

Osborn, H. F., 1924. Three new theropoda, protoceratops

zone, central Mongolia. American Museum Novitates,

144(7): 1–12.

Osmólska, H., 1976. New light on the skull anatomy and

systematic position of Oviraptor. Nature, 262: 683–

684.

Osmólska, H., Roniewicz, E. & Barsbold, R., 1972. A

new dinosaur Gallimimus bullatus, n. gen. n. sp. (Or-

New Oviraptorid Dinosaur, Mongolia

115

Page 22

nithomimidae) from the Upper Cretaceous of Mongo-

lia. Palaeontol. Polonica, 27: 103–143.

Ostrom, J. H., 1978. The osteology of Compsognathus

longipes Wagner. Zitteliana, 4: 73–118.

Rauhut, O. W. M., 2003. The interrelationships and evolu-

tion of basal theropod dinosaurs. Special Papers in

Palaeontology, 69: 1–213.

Russell, D. A. & Dong, Z.-M., 1993. The affinities of a

new theropod from the Alxa Desert, Inner Mongolia,

People’s Republic of China. Canadian Journal of Earth

Sciences, 30: 2107–2127.

Sanz, J. L. & Bonaparte, J. F. 1992. Iberomesornis

romerali, a fossil small bird articulated skeleton from

the Early Cretaceous of Spain. Proceedings of the II In-

ernational Symposium of Avialian Paleontology, 1988,

Pp. 39–49.

Sereno, P. C., 1999. The evolution of dinosaurs. Science,

284: 2137–2147.

Sereno, P. C., 2000. Iberomesornis romerali (Aves, Or-

nithothoraces) reevaluated as an Early Cretaceous enan-

tiornithine. Neues Jahrbuch für Geologie und

Paläontologie Abhandlungen, 215(3): 365–395.

Sereno, P. C. & Novas, F. E., 1993. The skull and neck of

the basal theropod Herrerasaurus ischigualastensis.

Journal of Vertebrate Paleontology, 13(4): 451–476.

Sereno, P. C. & Rao, C.-G., 1992. Early evolution of avian

flight and perching: new evidence from the Lower Cre-

taceous of China. Science, 255: 845–848.

Smith, D., 1992. The type specimen of Oviraptor philo-

ceratops, a theropod dinosaur from the Upper Creta-

ceous of Mongolia. Neues Jahrbuch für Geologie und

Paläontologie Abhandlungen, 186(3): 365–388.

Sternberg, R. M., 1940. A toothless bird from the Creta-

ceous of Alberta. Journal of Paleontology, 14(1): 81–

85.

Sues, H.-D., 1997. On Chirostenotes, a Late Cretaceous

oviraptorosaur (Dinosauria: Theropoda) from western

North America. Journal of Vertebrate Paleontology, 17:

698–716.

Swofford, D. L., 1998. PAUP*: phylogenetic analysis

using parsimony (and other methods). Version 4.0. Sin-

auer, Sunderland, Mass.

Swofford, D. L. & Begle, D. P., 1993. PAUP: phylogenetic

analysis using parsimony. Version 3.1. User’s manual.

Illinois Natural History Survey, Champaign, Illinois.

Welles, S. P., 1984. Dilophosaurus wetherilli (Dinosauria,

Theropoda). Osteology and comparisons. Palaeontolo-

graphica A, 185: 85–180.

Wester, D., 1996. Dinosaurs of the Gobi. National Geo-

graphic, 190(1): 70–89.

Witmer, L. M., 1990. The craniofacial air sac system of

Mesozoic birds (Aves). Zoological Journal of the Lin-

nean Society of London, 100: 327–378.

Witmer, L. M., 1997. The evolution of the antorbital cavi-

ty of archosaurs: a study in soft-tissue reconstruction in

the fossil record with an analysis of the function of

pneumaticity. Society of Vertebrate Paleontology mem-

oir 3. Journal of Vertebrate Paleontology, 17: supple-

ment to number 1.

Xu, X., Cheng, Y.-N., Wang, X.-L. & Chang, C.-X., 2002.

An unusual oviraptorosaurian dinosaur from China.

Nature, 419: 291–293.

Xu, X., Norell, M. A., Wang, X.-L., Makovicky, P. J. &

Wu, X.-C., 2002. A basal troodontid from the Early

Cretaceous of China. Nature, 415: 780–784.

Zhou, Z.-H., & Wang, X.-L., 2000. A new species of

Caudipteryx from the Yixian Formation of Liaoning,

Northwest China. Vertebrata PalAsiatica, 38(2): 111–

127.

Zhou, Z.-H., Wang, X.-L., Zhang, F.-C., & Xu, X., 2000.

Important features of Caudipteryx-evidence from

two nearly complete new specimens. Vertebrata PalAsi-

atica, 38(4): 241–254.

Zhou, Z.-H. & Zhang, F.-C., 2002. A long-tailed, seed-

eating bird from the Early Cretaceous of China. Nature,

418: 405–409.

Appendix 1

Abbreviations: An: angular; ar: articular;

asc: articular-surangular-coronoid complex; bs:

basisphenoid; cav antorb: antorbital cavity; ch:

choana; CH: chamber; d: dentary; ec: ectoptery-

goid; emf: external mandibular fenestra; eo: ex-

occipital; f: frontal; fmb: fossa m. brachialis; j:

jugal; l: lacrimal; lat: laterosphenoid; m: maxilla;

mf: maxillary fenestra; na: nasal; nar: narial

opening; oc: occipital condyle; o, orbit; or: or-

bitosphenoid; SP: supporting pillar; p: parietal;

pa: palatine; par: parasphenoid; pm: premaxilla;

po: postorbital pre: prearticular; pt: pterygoid; q:

quadrate; qj: quadratojugal; sof: suborbital fenes-

tra; sq: squamosal; stf: supratemporal fenestra; t:

tooth-like process formed by maxillae and

vomers; v: vomer.

Appendix 2.

Characters for phylogenetic analysis of the re-

lationships among Oviraptorosauria (modified

from Maryanska et al., 2002 with additional taxa

and recoded and additional characters as noted in

bold) and data matrix: 0plesiomorphic charac-

ter state; 1, 2, 3derived character states;

?missing data; pcharacter not applicable.

116

Lü, Tomida, Azuma, Dong and Lee

Page 23

1. Preorbital skull length to basal skull

length ratio: 0.6 or more (0); 0.5 or less

(1).

2. Longitudinal pneumatized crest-like

prominence on the skull roof: absent (0),

present (1).

3. Premaxilla main body length (ventral) to

height (below the naris) ratio: 1.0–1.4(0);

more than 1.7(1); 0.7 or less (2).

4. Otosphenoidal crest vertical on ba-

sisphenoid and prootic, and does not

border an enlarged pneumatic recess

(0) or well developed, crescent shaped,

thin crest forms anterior edge of en-

larged pneumatic recess (1). This struc-

ture forms the anterior, and most distinct,

border of the “lateral depression” of the

middle ear region (see Currie, 1985; Cur-

rie and Zhao, 1993) of troodontids and

some extant avians (Hwang et al., 2004).

5. Subnarial (maxillary) process of the pre-

maxilla: contacts the nasal, the maxilla ex-

cluded from the narial border (0); does not

contact the nasal, the maxilla participates

in formation of the narial border (1).

6. Premaxillae in adult: unfused (0); fused

(1) (Chiappe, 1996).

7. Pneumatization of the premaxilla: absent

(0); present (1).

8. Basisphenoid recess present between

basisphenoid and basioccipital (0) or

entirely within basisphenoid (1) or ab-

sent (2) (Hwang et al., 2004).

9. Subantorbital portion of the maxilla: not

inset medially (0); inset medially (1).

10. Palatal shelf of the maxilla with two longi-

tudinal ridges and a tooth-like process: ab-

sent (0); present (1).

11. Rim around the antorbital fossa: well pro-

nounced (0); poorly delimited (1).

12. Antorbital fossa: not bordered rostrally by

the premaxilla (0); bordered rostrally by

the premaxilla (1).

13. Preorbital region of the skull in post-

hatchling individuals: elongate, nasals

considerably longer than frontals, max-

illa at least twice the length of the pre-

maxilla (0); shortened, nasals subequal

in length to frontals or shorter, maxil-

lary length less than twice the length of

the premaxillary body (1) (Rauhut,

2003). In the majority of theropods, the

preorbital part of the skull forms an elon-

gate snout, with maxilla and nasal being

two of the longest bones of the skull roof.

In the basal oviraptorosaur Incisivosaurus

gauthieri and in Nemegtia barsboldi, the

nasal is longer than the frontal (Xu et al.,

2002). In Caudipteryx zoui (Ji et al.,

1998) and most derived oviraptorosaurs,

the nasal is shorter than the frontal.

14. Maxillary process of premaxilla con-

tacts nasal to form posterior border of

nares (0) or maxillary process reduced

so that maxilla participates broadly in

external nares (1) or maxillary process

of premaxilla extends posteriorly to sep-

arate maxilla from nasal posterior to

nares (2) (Hwang et al., 2004).

15. Nasal recesses: absent (0); present (1).

16. Caudal margin of the naris: rostral to the

rostral border of the antorbital fossa (0);

nearly reaching or overlapping the rostral

part of the antorbital fossa (1); overlap-

ping most of the antorbital fossa (2).

17. Ventral margin of the external naris: at the

level of the maxilla (0); dorsal to the max-

illa (1).

18. Prefrontal: present (0); absent or fused

with the lacrimal (1). In the majority of

theropods, the prefrontal is large. Pre-

frontal may be absent in all ovirap-

torosaurs, including the primitive Inci-

sivosaurus gauthieri.

19. Lacrimal recess: absent (0); present (1).

20. Premaxillary symphysis acute, V-

shaped (0); or rounded, U-shaped (1)

(Hwang et al., 2004).

21. Pronounced, round accessory antorbital

fenestra absent (0); or present (1). A

small fenestra, variously termed the acces-

sory antorbital fenestra or maxillary fen-

New Oviraptorid Dinosaur, Mongolia

117

Page 24

estra, penetrates the medial wall of the an-

torbital fossa anterior to the antorbital fen-

estra in a variety of coelurosaurs and other

theropods (Hwang et al., 2004).

22. Parietal length to frontal length ratio: 0.6

or less (0); 1.0 or more (1).

23. Narial region apneumatic or poorly

pneumatized (0); or with extensive

pneumatic fossae, especially along pos-

terodorsal rim of fossa (1) (Hwang et

al., 2004).

24. Sagittal crest on the parietals: absent (0);

present (1).

25. Jugal pneumatic recess in posteroven-

tral corner of antorbital fossa present

(0); or absent (1) (Hwang et al., 2004).

26. Infratemporal fenestra: ventrally nearly as

long as high rostrally (0); shorter ventrally

than (1); large, square (2); not separate

from the orbit (3).

27. Descending (prequadratic) process of the

squamosal: constricting the dorsal part of

the infratemporal fenestra (0); not con-

stricting the infratemporal fenestra (1).

28. Jugal and quadratojugal separate (0);

or quadratojugal and jugal fused and

not distinguishable from one another

(1) (Hwang et al., 2004).

29. Suborbital part of the jugal: deep

dorsoventrally and flattened lateromedially

(0); shallow dorsoventrally or rod-shaped

(1).

30. Jugal-postorbital contact: present (0); ab-

sent (1).

31. Quadratojugal process of the jugal in lat-

eral view: forked (0); tapering (1); fused

with the quadratojugal (2).

32. Quadratojugal-squamosal contact: tips

of the bones widely separated (0); the

contact present (1) (modified).

33. Ascending (squamosal) process of the

quadratojugal: massive, bordering about

the ventral half of the infratemporal fenes-

tra (0); slender, bordering the ventral half

or less of the infratemporal fenestra (1);

slender, bordering the ventral two-thirds

or more of the infratemporal fenestra (2);

absent (3).

34. Dorsal part of the quadrate: erect (0); di-

rected backwards (1).

35. Otic process of the quadrate: articulating

only with the squamosal (0); articulating

with the squamosal and the lateral wall of

the braincase (1).

36. Pneumatization of the quadrate: absent

(0); present (1). In many theropods, the

quadrate is a solid bone and it is not in-

vaded by a diverticulum of the tympanic

pneumatic system, but the quadrate is

pneumatized by the quadrate diverticulum

in most modern birds. The diverticulum

enters the bone ventromedially, dorsome-

dially or caudally (Witmer, 1990). In dro-

maeosaurids (Velociraptor mongoliensis)

and other non-avian theropods, the

quadrate lacks pneumatic foramina (Col-

bert and Russell, 1969; Madsen, 1976;

Bonaparte et al., 1990; Barsbold and

Osmólska, 1999). Among non-avian

theropods, troodontids (Currie and Zhao,

1993b) and tyrannosaurids (Molnar, 1991)

are the exception, in which the quadrate is

pneumatic. A large pneumatic foramen is

placed anteromedially and somewhat

below the mid-height in some oviraptorids

(Maryanska and Osmólska, 1997). Some

birds, such as Archaeopteryx lithographi-

ca, Hesperornis regalis and Parahesperor-

nis alexi (Chiappe, 1996) also lack pneu-

matic foramina. The quadrate diverticu-

lum perforates the lateral surface of the

quadrate in Patagopteryx deferrariisi (Chi-

appe, 1996), and a similar condition is

present in Heyuannia huangi (Lü, 2003),

but it is absent in “Oviraptor” mongolien-

sis. The lateral position of the entrance of

the quadrate diverticulum is regarded as

an autapomorphy of Patagopteryx defer-

rariisi (Chiappe, 1996), but the similar po-

sition in Heyuannia indicates an indepen-

dent acquisition. Witmer (1990) consid-

ered the presence of quadrate pneumatici-

118

Lü, Tomida, Azuma, Dong and Lee

Page 25

ty as a synapomorphy of the Carinatae

[IchthyornithiformesAves], but Chiappe

(1996) thought that a pneumatic quadrate

was present at least in the common ance-

stor of Patagopteryx deferrariisi and the

Ornithurae and it might have evolved as

early as at the origin of the ornithotho-

races. The pneumatic quadrate in ovirap-

torosaurs is considered here as indepen-

dently acquired.

37. Accessory process for a contact with the

quadratojugal on the distal end of the

quadrate: absent (0); present (1).

38. Quadratojugal sutured to the quadrate

(0); or joined through a ligamentary ar-

ticulation (1) (Chiappe, 2001, 2002). In

Archaeopteryx and most non-avian thero-

pod dinosaurs, the quadratojugal is su-

tured to the quadrate, in almost all de-

rived oviraptorosaurs, a cotyle is present

either on the quadrate (“Oviraptor” mon-

goliensis, GIN 100/42, and Conchoraptor

gracilis, Maryanska et al., 2002) or on the

medial surface of the quadratojugal GIN

100/2112 (Lü et al., 2002). Heyuannia

huangi has a groove-like structure on the

quadrate (Lü, 2003). The presence of a

cotyle or a groove-like surface on the

quadrate or quadratojugal is interpreted

here as a ligamentary articulation between

the quadrate and quadratojugal. In Citipati

osmolskae, the quadrate is tightly sutured

to the quadratojugal (Clark et al., 2002).

A shallow depression is present on the lat-

eral condyle of the quadrate in Confuciu-

sornis sanctus (Chiappe et al., 1999),

therefore it is interpreted as having a liga-

mentary articulation between the quadrate

and quadratojugal.

39. Mandibular condyles of the quadrate situ-

ated: caudal to the occipital condyle (0);

in the same vertical plane as the occipital

condyle (1); rostral to the occipital

condyle (2). In Herrerasaurus is-

chigualastensis (Sereno and Novas, 1993),

Allosaurus and sinraptorids (Currie and

Zhao, 1993a), the mandibular condyle of

the quadrate is caudal to the occipital

condyle. In Archaeopteryx lithographica,

Incisivosaurus gauthieri and Caudipteryx

sp. (IVPP V 12430), the mandibular

condyle of the quadrate is rostral to the

occipital condyle. Although there are no

completely three-dimensionally preserved

skulls of Caudipteryx, the preserved skull

is little disturbed in Caudipteryx sp. (IVPP

V 12430). It shows clearly relative posi-

tion between the mandibular condyles of

the quadrate and the occipital condyle. In

Nemegtia barsboldi and “Oviraptor” mon-

goliensis and Heyuannia huangi, the

mandibular condyles of the quadrate are

in the same vertical plane as the occipital

condyle.

40. Enlarged foramen or foramina opening

laterally at the angle of the lacrimal, ab-

sent (0); or present (1) (Hwang et al.,

2002).

41. Paroccipital process directed: laterad (0);

lateroventrad (1), ventrad (2). In both Her-

rerasaurus ischigualastensis and Al-

losaurus fragilis, the paroccipital process

is directed lateroventrally, but it is directed

much more strongly so in A. fragilis than

in H. ischigualastensis. The paroccipital

process directed ventrad in Incisivosaurus

gauthieri, Chirostenotes pergracilis, and

derived oviraptorosaurs.

42. Dorsal surface of parietals flat, lateral

ridge borders supratemporal fenestra

(0); or parietals dorsally convex with

very low sagittal crest along midline (1);

or dorsally convex with a well-devel-

oped sagittal crest (2) (Hwang et al.,

2004).

43. Foramen magnum: smaller than or equal

to the occipital condyle (0); larger than the

occipital condyle (1).

44. Basal tubera: modestly pronounced (0);

well pronounced, widely separated (1).

45. Pneumatization of the basisphenoid: weak

or absent (0); extensive (1).

New Oviraptorid Dinosaur, Mongolia

119

Page 26

46. Basipterygoid process: well developed

(0); strongly reduced (1); absent (2).

47. Posterior end of dentary without pos-

terodorsal process dorsal to mandibular

fenestra (0); or with dorsal process

above anterior end of mandibular fenes-

tra (1); or with elongate dorsal process

extending over most of fenestra (2)

(Hwang et al., 2004).

48. Parasphenoid rostrum: horizontal or di-

rected rostrodorsad (0); slanting rostro-

ventrad (1).

49. Tibiofibular crest in the lateral condyle

of femur: absent (0); present (1) (Ji et

al., 1998).

50. Parietals separate (0) or fused (1)

(Hwang et al., 2004).

51. Foramen magnum subcircular, slightly

wider than tall (0); or oval, taller than

wide (1) (Hwang et al., 2004).

52. Medially extended pterygoids meeting

each other along the midline and ventrally

underlying the basisphenoid and parasphe-

noid: absent (0); present (1).

53. Quadrate wing of the pterygoid: distinct

from the braincase wall (0); overlapping

the braincase (1).

54. Pterygoid basal process for contact with

the basisphenoid: absent (0); present (1).

55. Ectopterygoid situated: lateral to the

pterygoid (0); rostral to the pterygoid (1).

56. Ectopterygoid contacts with the maxilla

and lacrimal: absent (0); present (1).

57. Hook-like jugal process on the ectoptery-

goid: present (0); absent (1).

58. Massive pterygoid-ectopterygoid longitu-

dinal bar: absent (0); present (1).

59. Palate extending below the cheek margin:

absent (0); present (1).

60. Palatine: tetraradiate or trapezoid (0); tri-

radiate, without a jugal process (1); devel-

oped in three plane perpendicular to each

other (2). Triradiate palatine is present In-

cisivosaurus gauthieri (Xu, Cheng et al.,

2002), Citipati osmolskae (Clark et al.,

2002) and GIN100\2112. So Hwang et al.

(2004), in which IGM100\42 was coded as

having a tetraradiate palatine, may be

wrong.

61. Pterygoid wing of the palatine situated:

dorsal to the pterygoid (0); ventral to the

pterygoid (1).

62. Maxillary process of the palatine: shorter

than vomeral process (0); longer than the

vomeral process (1).

63. Vomer: distant from the parasphenoid ros-

trum (0); approaching or in contact with

the rostrum (1).

64. Suborbital (ectopterygoid-palatine) fenes-

tra: well-developed (0); closed or reduced

(1).

65. Pterygopalatine fenestra: absent (0); pre-

sent (1).

66. Jaw joint: distant from the skull midline

(0); close to the skull midline (1).

67. Occipital condyle without constricted

neck (0); or subspherical with constrict-

ed neck (1) (Hwang et al., 2004).

68. Mandibular symphysis: loose (0); tightly

sutured (1); fused (2).

69. Extended symphyseal shelf at the

mandibular symphysis: absent (0); present

(1).

70. Downturned symphyseal portion of the

dentary: absent (0); present (1).

71. U-shaped mandibular symphysis: absent

(0); present (1).

72. Retroarticular process’ length to total

mandibular length ratio: less than 0.05 or

the process absent (0); about 0.10 (1).

73. Mandible maximum height to length ratio:

about 0.2(0); about 0.1 (1); 0.3–0.4(2).

74. External mandibular fenestra’s height to

length ratio: 0.2–0.5(0); 0.7–1.0 (1); the

fenestra absent (2). In Caenagnathus

collinsi and Incisivosaurus gauthieri, the

ratio is less than 0.25, but in the derived

oviraptorosaurs such as Ingenia yanshini,

Heyuannia huangi, Oviraptor philocer-

atops, “Oviraptor” mongoliensis, Con-

choraptor gracilis, Citipati osmolskae,

this ratio is greater than 0.25.

120

Lü, Tomida, Azuma, Dong and Lee

Page 27

75. External mandibular fenestra’s length

to total mandibular length ratio:

0.15–0.2(0), 0.1– or less (1); 0.21 or

more (2); the fenestra absent (3) (modi-

fied).

76. Coossification of the articular with the

surangular: absent (0); present (1).

77. Mandibular rami in dorsal view: straight

(0); bowed laterad at the mid-length (1).

78. Rostrodorsal margin of the dentary:

straight or weakly convex (0); deeply

concave (1) (modified). In most theropods

and Oviraptor philoceratops, the ros-

trodorsal margin of the dentary is straight.

In the basal oviraptorosaur Incisivosaurus

gauthieri, it is slightly convex. It is deeply

concave in Oviraptor philoceratops, Cae-

nagnathus collinsi, “Oviraptor” mon-

goliensis, Heyuannia huangi, Citipati os-

molskae, Conchoraptor gracilis, Nemegtia

barsboldi, and Ingenia yanshini.

79. Caudal margin of the dentary: incised,

producing two caudal processes (0);

oblique (1). The posterior end of the den-

tary is strongly forked to form the anterior

margin of the external mandibular fenestra

in oviraptorosaurs. In several theropods,

such as Allosaurus, dromaeosaurids, or-

nithomimosaurs, therizinosauroids and

tyrannosaurs, and Archaeopteryx, the cau-

dal margin of the dentary is oblique.

80. Long and shallow caudodorsal process of

the dentary: present (0); absent (1).

81. Long and shallow caudoventral process of

the dentary, extending caudad at least to

the caudal border of the external mandibu-

lar fenestra: absent (0); present (1).

82. Pronounced coronoid eminence: absent

(0); present (1) (modified). The coronoid

eminence on the surangular is absent in

most theropods. In Incisivosaurus gau-

thieri the coronoid eminence is weakly de-

veloped, so it is coded here as (0). The

pronounced coronoid eminence is present

in Avimimus portentosus and derived ovi-

raptorosaurs.

83. External mandibular fenestra oval (0);

or subdivided by a spinous rostral

process of the surangular (1) (Hwang et

al., 2002). The external mandibular fenes-

tra is elongate, and no spinous rostral

process of the surangular is present in In-

cisivosaurus gauthieri (Xu, Cheng et al.,

2002) and Caudipteryx sp. (IVPP V

12430, pers. observation). In derived ovi-

raptorosaurs, the external mandibular fen-

estra is subdivided by a spinous rostral

process of the surangular.

84. Mandibular articular facet for the

quadrate: formed of the surangular and ar-

ticular (0); formed exclusively of the artic-

ular (1).

85. Mandibular articular facet for the

quadrate: with one or two cotylae (0); con-

vex in lateral view, transversely wide (1).

86. Articular facet for the mandibular joint

positioned: below the dorsal margin of the

caudal part of the mandibular ramus (0);

above this margin (1).

87. Rostral part of the prearticular: deep, ap-

proaching the dorsal margin of the

mandible (0); shallow, strap-like, not ap-

proaching the dorsal margin of the

mandible (1).

88. Splenial: subtriangular, approaching the

dorsal margin of the mandible (0); strap-

like, shallow, not approaching the dorsal

margin of the mandible (1).

89. Mandibular adductor fossa: rostrally de-

limited, occupying the caudal part of the

mandible (0); large, rostrally and dorsally

extended, not delimited rostrally (1).

90. Coronoid bone: well developed (0); re-

duced (1) or absent (2) (modified). The

coronoid bone is present in most

theropods, but it is weakly developed and

strap-like in Incisivosaurus gauthieri (Xu

et al., 2002b), Caudipteryx sp. (IVPP V

12430; Pers. observation) and Citipati os-

molskae (Clark et al., 2002). It is also pre-

sent in therizinosauroids (Clark et al.,

1994). It is absent in birds (Archaeo-

New Oviraptorid Dinosaur, Mongolia

121

Page 28

pteryx), ornithomimosaurs (Osmólska et

al., 1972) and other derived ovirap-

torosaurs including Nemegtia barsboldi,

Conchoraptor gracilis, Ingenia and Ovi-

raptor.

91. Premaxillary teeth: present (0); absent (1).

Premaxillary teeth are present in Inci-

sivosaurus, Caudipteryx, Archaeopteryx

and most theropods. But premaxillary

teeth are absent in therizinosauroids

(Clark et al., 1994), derived ovirap-

torosaurs and advanced birds.

92. Maxillary tooth row: extends at least to

the level of the preorbital bar (0); does not

reach the level of the preorbital bar (1);

maxillary teeth absent (2). Maxillary teeth

are present in Incisivosaurus gauthieri,

and the majority of theropods. Maxillary

teeth are absent in Caudipteryx zoui and

derived oviraptorosaurs.

93. Dentary teeth: present (0); absent (1).

Dentary teeth are present in Inci-

sivosaurus gauthieri, but absent in de-

rived oviraptorosaurs, Microvenator celer

(Makovicky and Sues, 1998), Caudipteryx

zoui (Ji et al., 1998), Avimimus portento-

sus, advanced ornithomimosaurs and

birds.

94. Number of the cervicals (excluding the

cervicodorsal): not more than 10 (0); more

10 (1).

95. Cranial articular facets of the centra in the

anterior postaxial cervicals: not inclined

or only slightly inclined (0); strongly in-

clined ventrocaudal, almost continuous

with the ventral surface of the centra (1);

ball-shaped (2).

96. Anterior cervical centra: not extending

posteriorly beyond the respective neural

arches (0); extending posteriorly beyond

the respective neural arches (1).

97. Epiphyses on the postaxial cervicals: in

form of a low crest or rugosity (0); prong-

shaped (1). The epiphyses on the postaxial

cervicals are in form of a low crest in Mi-

crovenator celer, a rugosity in Coeloph-

ysis bauri, Avimimus portentosus,

Archaeopteryx lithographica, Ornithomi-

mosauria, Therizinosauria, Oviraptor

philoceratops, Caudipteryx sp., Nemegtia

barsboldi, Heyuannia huangi, Conchorap-

tor gracilis, Ingenia yanshini. The epiphy-

ses are only present in the third and fourth

cervicals of “Oviraptor” mongoliensis.

They are very developed, prong-shaped,

and usually extending over the distal end

of the postzygopophyses in Herrerasaurus

ischigualastensis, Allosaurus fragilis,

Dromaeosauridae, Troodontidae and

Tyrannosauridae.

98. Cervical ribs: loosely attached to verte-

brae in adults (0); firmly attached (1).

99. Shafts of the cervical ribs: longer than the

respective centra (0); not longer than the

respective centra (1).

100. Pleurocoels or lateral excavations on the

dorsal centra: absent (0); present (1).

101. Postzygapophyses on the dorsals: not ex-

tending beyond the respective centra (0);

markedly extending beyond the centra (1).

102. Number of the vertebrae included in the

sacrum in adults: not more than five (0);

more than five (1) (the term sacrum is

used here instead of synsacrum in

Maryanska et al.’s original description).

103. Sacral spine in adults: unfused (0); fused

(1).

104. Continuous sulcus along the ventral side

of the mid-sacral centra: absent (0); pre-

sent (1).

105. Pleurocoels on the sacral centra: absent

(0); present (1).

106. Scapula and coracoid separate (0), or

fused into scapulacoracoid (1) (Hwang

et al., 2004). The scapula and coracoid are

separate in most theropods, including Al-

losaurus fragilis, Tyrannosaurus rex,

Deinonychus antirrhopus, Chirostenotes

pergracilis, and Caudipteryx sp. They are

fused in Velociraptor mongoliensis, Ar-

chaeopteryx lithographica, Confuciusor-

nis sanctus, Ingenia yanshini, “Oviraptor”

122

Lü, Tomida, Azuma, Dong and Lee

Page 29

mongoliensis, Oviraptor philoceratops,

Conchoraptor gracilis, and Heyuannia

huangi. Because both character states

are present in dromaeosaurids and or-

nithomimids, this character is coded as 0/1

in them.

107. Hyposphene-hypantrum articulations

in trunk vertebrae present (0); or ab-

sent (1) (contrary to the code of Hwang

et al., 2004). These articulations are pre-

sent in Herrerasaurus ischigualastensis

(Novas, 1993), Allosaurus fragilis (Mad-

sen, 1976), Avimimus portentosus, Mi-

crovenator celer, Chirostenotes pergra-

cilis, Conchoraptor gracilis, ornitho-

mimids, troodontids, tyrannosaurids. They

are absent in Confuciusornis sanctus and

advanced birds. But it is not clear in

Archaeopteryx lithographica. In “Ovirap-

tor” mongoliensis, and Heyuannia huangi,

these articulations are very weak, so it was

coded here as absent (0).

108. Pleurocoels on the caudal centra: absent

(0); present at least in the proximal part of

the tail (1).

109. Number of caudals: more than 35 (0);

35–25 (1); fewer than 25 (2) (modified

from Lü et al., 2002). Most theropods

have more than 35 caudals. 22 caudals are

present in Caudipteryx sp. (Zhou et al.,

2000), 24 caudals in Nomingia gobiensis

(Barsbold, Osmólska et al., 2000), about

33 in “Oviraptor” mongoliensis. There are

about 26 caudals in Microraptor zhaoianus

and 25 in Shenzhouraptor (Jeholornis,

Zhou and Zhang, 2002; Ji et al., 2002;

2003).

110. Pygostyle: absent or rudimentary (fewer

than three elements) (0), present (1). The

last three to six caudal vertebrae fuse to

form a pygostyle in modern birds (Baumel

and Raikow, 1993) that support the tail

feathers. A pygostyle is present in the or-

nithurines, hesperorniforms and ichthyor-

nithiforms, enantiornithines (Iberomesor-

nis, Sinornis, Sereno, 2000; Sereno and

Rao, 1992), but is absent in non-avian

theropod dinosaurs, Archaeopteryx litho-

graphica (Chiappe, 1996) and the basal

bird Shenzhouraptor (Zhou and Zhang,

2002; Ji et al., 2003). The presence of a

pygostyle in Nomingia gobiensis is of

great interest in the study of the relation-

ship between birds and non-avian

theropods, although it is most parsimo-

nious to consider it a convergence

(Osmólska, pers. comm., 2003). Some

have even regarded the pygostyle in mod-

ern ratite and carinate birds, to be not ho-

mologous (Holmgren, 1955). The struc-

ture of pygostyle in Nomingia gobiensis

(Barsbold, Osmólska et al., 2000) and the

ostrich (Holmgren, 1955) are very similar

(fused dorsally in the vertebrae rather than

ventrally in carinate birds). The complete

distal tail of Heyuannia huangi shows that

there is no pygostyle. It is also absent in

Caudipteryx zoui (Ji et al., 1998; Zhou et

al., 2000). The development of a py-

gostyle was regarded as a synapomorphy

of the clade composed of all birds more

advanced than Archaeopteryx lithographi-

ca (Gauthier, 1986). Cracraft (1986) con-

sidered it as synapomorphy of Ornithurae,

whereas Sanz and Bonaparte (1992) con-

sidered it as synapomorphic for Iberome-

sornis plus Ornithurae. In agreement with

Sanz and Bonaparte (1992), Chiappe

(1996) considered it as a synapomorphy

of the Ornithothoraces. The discoveries of

Jeholornis and Rahonavis further supports

Sanz and Buscalioni (1992) and Chiappe

(1996)’s hypotheses (Chiappe, 2002).

111. Scapular caudal end: blunt and much

expanded (0); tapered to a sharp point

or slightly expanded (1) (modified from

Chiappe, 2002).

112. Scapular and coracoid: nearly in the

same plane (0); forming a distinct angle

(1) at the level of the glenoid cavity

(Chiappe et al., 1998).

113. Distal caudal prezygapophyses: overlap-

New Oviraptorid Dinosaur, Mongolia

123

Page 30

ping less than a half of the centrum of the

preceding vertebra (0); overlapping at

least a half of the preceding vertebra (1).

114. Hypapophyses in the cervicodorsal verte-

bral region: absent (0); small (1); promi-

nent (2). The hypapophyses are absent in

Coelophysis bauri, Herrerasaurus is-

chigualastensisi, Allosaurus fragilis, Mi-

crovenator celer, ornithomimids, tyran-

nosauridae. Because it is absent in Micro-

raptor, present in Velociraptor mongolien-

sis, this character is coded here as 0/1 in

Dromaeosauridae. It is not prominent in

Heyuannia huangi, Caudipteryx sp., and

Nemegtia barsboldi. It is well developed

in “Oviraptor” mongoliensis, Conchorap-

tor gracilis, Chirostenotes pergracilis,

Avimimus portentosus, and troodontids.

115. Distal chevrons: deeper than long (0);

longer than deep (1).

116. Sternum: unossified or small (0); ossified,

large (1).

117. Scapula length to humerus length ratio:

0.8–1.1 (0); 1.2 or more (1); 0.7 or less

(2).

118. Acromion: projecting dorsad (0); everted

laterad (1); projecting cranial (2).

119. Caudoventral process on the coracoid: ab-

sent (0); short, not extending beyond the

glenoid diameter (1); long, caudoventrally

extending beyond the glenoid (2).

120. Orientation of the glenoid on the pectoral

girdle: caudoventral (0); lateral (1).

121. Deltopectoral crest: low, with the width

equal to or smaller than the shaft diameter

(0); expanded, wider than the shaft diame-

ter (1).

122. Internal tuberosity on humerus: weakly

pronounced or absent (0); well pro-

nounced but low (1); subtriangular, dis-

tinctly extended medially (2); in form of a

longitudinally short, tuber-like extension,

sharply delimited from the shaft and usu-

ally also from the humeral head (3).

123. Deltopectoral crest (measured from the

humeral head to the apex) extending for:

about the proximal third of the humerus

length or less (0); about 40–50% of the

humerus length (1).

124. Epicondyles on the humerus: absent or

poorly developed (0); the ectepicondyle

more prominent than the entepicondyle

(1); the entepicondyle more prominent

than the ectepicondyle (2); the ectepi-

condyle and entepicondyle about equally

prominent (3).

125. Shaft of the ulna: straight (0); bowed, con-

vex caudally (1); bowed, concave caudally

(2).

126. Radius length to humerus length ratio:

0.80 or less (0); 0.85 or more (1).

127. Distal carpals: flat, mostly separate (0);

carpals I and II separate, carpal I with

the proximal trochlea (1); carpals I and

II fused, half-moon-shaped, with the

trochlea on the proximal surface, cover-

ing metacarpals I and II (2). The ex-

panded semilunate (half-moon-shaped)

carpal block completely caps the proximal

surface of metacarpals I and II in ovirap-

torosaurs, dromaeosaurids and avialians. It

is regarded as diagnostic of Maniraptora.

The absence in ornithomimosaurs and

adult tyrannosaurids are treated as rever-

sals (modified from Holtz, 2001).

128. Combined lengths of manual phalanges

III-I and III-2: greater than the length of

phalanx III-3 (0); less than or equal to the

length of phalanx III-3 (1).

129. Metacarpal I length to metacarpal II

length ratio: 0.5 or more (0); less than 0.5

(1).

130. Proximal margin of the metacarpal I in

dorsal view: straight, horizontal (0); an-

gled due to the medial extent of the carpal

trochlea (1).

131. Metacarpal II relative to metacarpal III:

shorter (0); subequal (1); longer (2). In

Heyuannia huangi, and Ingenia yanshini,

the length of metacarpal III is shorter than

that of metacarpal II; In Conchoraptor

gracilis, the length of metacarpal III is

124

Lü, Tomida, Azuma, Dong and Lee

Page 31

longer than that of metacarpal II; In

Caudipteryx, metacarpal II is subequal to

metacarpal III.

132. Metacarpal II length to humerus length

ratio: 0.4 or less (0); more than 0.4 (1).

133. Metacarpal III: unmodified (0); very slen-

der (1).

134. Lip or nubbin on the proximodorsal edge

of the manual unguals: absent (0); present

(1).

135. Manus length to humerus length plus ra-

dius length ratio: 0.50–0.65 (0); more than

0.65 (1); less than 0.50 (2).

136. Manus length to femur length ratio: 0.3–

0.6 (0); more than 0.7(1); less than 0.2 (2).

137. Humerus length to femur length ratio:

0.5–0.6 (0); less than 0.4 (1); 0.7 or more

(2).

138. Dorsal margins of the opposite iliac

blades: well separated from each other (0);

close to or contacting each other along

their medial sections (1).

139. Dorsal margin of the ilium along the cen-

tral portion of the blade: straight (0);

arched (1).

140. Preacetabular process relative to the

postacetabular process (the lengths mea-

sured from center of the acetabulum):

shorter or equal (0); longer (1).

141. Preacetabular process: not expanded or

weakly expanded ventrally below the level

of the dorsal acetabular margin (0); ex-

panded ventrally well below the level of

the dorsal acetabular margin (1).

142. Morphology of the ventral margin of the

preacetabular process: the cuppedicus

fossa absent, the margin transversely nar-

row (0); the cuppedicus fossa or wide

shelf present (1); the margin flat, wide at

least the base of the pubic peduncle (2).

143. Cranioventral process on the preacetabular

blade: absent (0); rounded (1); hook-like

(2).

144. Distal end of the postacetabular process:

truncated or broadly rounded (0); nar-

rowed or acuminate (1).

145. Supracetabular crest: well developed (0);

reduced or absent (1).

146. Craniocaudal length of the pubic pe-

duncle: about as long as the ischiadic

peduncle (0); distinctly longer than the

ischiadic peduncle (1); distinctly short-

er than the ischiadic peduncle (2) (mod-

ified).

147. Dorsoventral extension of the pubic pe-

duncle: level with the ischiadic peduncle

(0); deeper than the ischiadic peduncle

(1).

148. Brevis fossa: absent or small (0); large (1).

149. Antitrochanter on the ilium: present (0);

absent (1).

150. Ilium length to femur length ratio: 0.5–0.7

(0); 0.8 or more (1).

151. Hypocleidium on furcula absent (0); or

present (1). The hypocleidium is a

process extending from the ventral mid-

line of the furcula, and is attached to

the sternum by a ligament in extant

birds (Hwang et al., 2004).

152. Pubic shaft: straight (0); concave cranially

(1).

153. Pubic foot: with the cranial and caudal

processes being about equally long (0);

with the cranial process being longer

than the caudal process (1); with the

cranial process being shorter than the

caudal process (2); absent (3) (state 2 is

modified).

154. Dorsoventral length of the pubic apron:

longer than half total length of the pubis

(0); not longer than the half of total length

of the pubis (1).

155. Caudal margin of the ischiadic shaft:

straight or almost straight (0); strongly

concave (1).

156. Position of the obturator process on the is-

chium: proximal (0); at about mid-length

(1); distal (2); obturator process lacking

(3).

157. Distal end of the ischium: not expanded

(0); expanded (1).

158. Ischium length to pubis length ratio: 0.75

New Oviraptorid Dinosaur, Mongolia

125

Page 32

or more (0); 0.70 or less (1).

159. Posterior (greater) trochanter: weakly sep-

arated or not separated from the femoral

head (0); distinctly separated from the

femoral head (1).

160. Craniocaudal extent of the posterior

trochanter of femur: short (0); long (1).

161. Pleurocoels absent on sacral vertebrae

(0); or present on anterior sacrals only

(1); or present on all sacrals (2). A pleu-

rocoel may be present on the first sacral in

Alxasaurus elesitaiensis, although this

area is badly crushed (Russell and Dong,

1993) (Hwang et al., 2004).

162. Anterior and posterior trochanters: well

separated (0); contacting (1); fused (2).

163. Dorsal extremity of the anterior trochanter

of femur: well below the posterior

trochanter (0); about level with the poste-

rior trochanter (1).

164. Fourth trochanter of femur: well devel-

oped (0); weakly developed or absent (1).

165. Adductor fossa and associated craniome-

dial crest on the distal femur: weak or ab-

sent (0); well developed (1).

166. Strong distal projection of the fibular

condyle on the femur: absent (0); present

(1).

167. Medial surface of the fibular head: flat or

shallowly concave (0); with a deep fossa

(1).

168. Contacts of the distal end of the fibula

with tarsus: present (0); absent (1). In

most theropod dinosaurs, and Heyuannia

huangi, the distal end of the fibula con-

tacts the tarsus, but in Ingenia yanshini,

this contact is absent, similar to the case in

most birds.

169. Ascending process of the astragalus: as

tall as wide across the base (0); taller than

wide (1).

170. Distal tarsals: not fused with metatarsals

(0); fused with metatarsals (1).

171. Proximal coossification of metatarsals

II–IV: absent (0); present (1).

172. Arctometatarsus: absent (0); present

(1). Because the metatarsi of some or-

nithomimosaurs (Garudimimus and

Harpymimus) are not strongly pinched

as in more typical ornithomimids, this

character was coded as (0,1) for Or-

nithomimidae by Holtz (2001) in his

phylogenetic analysis. This character is

coded here as (0, 1), in agreement with

Holtz, instead of being coded as (1) as in

Maryanska et al. (2002).

173. Metatarsals II and IV: not in contact on

the plantar surface (0); contacting distally

(1).

174. Metatarsal I length: more than 50% of

metatarsal II length (0); less than 50% of

metatarsal II length (1); metatarsal I ab-

sent (2).

175. Metatarsal IV length relative to metatarsal

II length: about equal (0); longer (1).

176. Epipterygoid present (0), absent (1)

(new). An epipterygoid is known in al-

losaurids (Madsen, 1976), ornithomimids

(Barsbold, 1981), tyrannosaurids (Molnar,

1973), Dromaeosaurus albertensis (Cur-

rie, 1995), the derived oviraptorid Citipati

osmolskae (Clark et al., 2002) and Nemeg-

tia barsboldi. It is not clear in Inci-

sivosaurus gauthieri, and it is absent in

“Oviraptor” mongoliensis and birds.

177. Metatarsus length to femur length ratio:

0.4–0.6 (0); about 0.3 (1); 0.7–0.8 (2).

178. Thoracic vertebral count: 13–14 (0);

11–12 (1); fewer than 11 (2) (Chiappe,

2002).

179. Ossified uncinate processes: absent (0);

present (1).

180. Coracoid shape: short (0); elongated with

trapezoidal profile (1); strut like (2).

181. The proximal end of metacarpal III:

contacts the distal carpals (0); does not

contact (1) (new).

182. Metacarpals I and II fused proximally:

absent (0); present (1) (new).

183. Ectopterygoid with constricted opening

into fossa (0); or with open ventral fossa

in the main body of the element (1)

126

Lü, Tomida, Azuma, Dong and Lee

Page 33

(Hwang et al., 2004).

184. Flange of pterygoid well developed (0);

or reduced in size or absent (1) (Hwang

et al., 2004).

185. Symphyseal region of dentary broad

and straight, paralleling lateral margin

(0) or medially recurved slightly (1); or

strongly recurved medially (2) (Hwang

et al., 2004).

186. Mandibular articulation surface as long

as distal end of quadrate (0); or twice

or more as long as quadrate surface, al-

lowing anteroposterior movement of

mandible (1) (Hwang et al., 2004).

187. Maxilla toothed (0); or edentulous (1)

(Hwang et al., 2004).

188. The lower margin of the external nasal

opening is below (0), or close or above

(1) the level of the upper corner of the

antorbital fenestra (new). The lower

margin of the external nasal opening is

below the level of the upper corner of the

antorbital fenestra in Coelophysis bauri,

Herrerasaurus ischigualastensis, Veloci-

raptor mongoliensis, troodontids, or-

nithomimids, Tyrannosauridae, Allosaurus

fragilis, Caudipteryx sp., Oviraptor philo-

ceratops, Incisivosaurus gauthieri, Khaan

mckennai, and Ingenia (100/80). It is close

in Citipati osmolskae (Clark et al., 2002),

above in Conchoraptor gracilis (Barsbold

et al., 1990), Nemegtia barsboldi, and

“Oviraptor” mongoliensis.

189. The quadratojugal and quadrate con-

tact area is far from (0), or near (1) the

lateral surface of the quadrate articular

surface (new).

190. Nasals and premaxillae do not form a

crest (0), or form a crest (1) (new).

191. External nasal opening round (0), or

oval (1) (new). In Incisivosaurus

gauthieri, the external nasal opening is

round, in Caudipteryx sp. and Citipati os-

molskae, the external nasal opening is

nearly round (slightly elongated). In other

derived oviraptorids, such as Conchorap-

tor gracilis, “Oviraptor” mongoliensis,

Nemegtia barsboldi and Oviraptor philo-

ceratops, the external nasal openings are

elongate. In majority of theropods, the ex-

ternal nasal openings are oval (elongate),

such as in Herrerasaurus ischigualasten-

sis, Coelophysis bauri, Allosaurus fragilis,

ornithomimids, troodontids and dro-

maeosaurids.

192. The maxillary fenestra is relatively

smaller or absent (0), or larger than the

antorbital fenestra (1) (new). The maxil-

lary fenestra is usually larger, opens later-

ally, and is clearly visible in lateral view

(Rauhut, 2003). It is relatively smaller

in Incisivosaurus gauthieri, Caudipteryx

dongi and “Oviraptor” mongoliensis. It is

relatively larger in Ingenia yanshini

(GIN100/80), Citipati osmolskae, and Ne-

megtia barsboldi. It was thought that in

caenagnathid Chirostenotes pergracilis

(ROM 43250), there was no maxillary

fenestra (Rauhut, 2003), but according to

Clark et al. (2002, Fig. 12), a relatively

larger maxillary fenestra is present in Chi-

rostenotes pergracilis as in Oviraptor

philoceratops.

193. Large openings absent (0), or present

(0) on the base of the neural arches of

anterior caudal vertebrae (new). On the

anterior caudal vertebrae of Ingenia yan-

shini and “Oviraptor” mongoliensis, there

are large openings on the lateral surface

near the base of the neural arches, these

openings are larger than the pleurocoels;

In Allosaurus fragilis, Heyuannia huangi,

Conchoraptor gracilis, and Caudipteryx

sp., this kind of opening is absent.

194. Large openings (fossa) present (0), or

absent (1) on the neural arches of cervi-

cal vertebrae (new). Large openings are

present in the neural arches of cervical

vertebrae in Coelophysis bauri, Ar-

chaeopteryx lithographica, Avimimus por-

tentosus, Microvenator celer, Conchorap-

tor gracilis, Heyuannia huangi. They are

New Oviraptorid Dinosaur, Mongolia

127

Page 34

absent in Allosaurus fragilis, Her-

rerasaurus ischigualastensis, Ingenia

(GIN 100/32-02), Caudipteryx sp., Ne-

megtia barsboldi, and “Oviraptor” mon-

goliensis.

195. Constriction between articulated pre-

maxillae and maxillae: absent (0), pre-

sent (1). (Rauhut, 2003). In most of

theropod dinosaurs and a basal ovirap-

torosaur, Incisivosaurus gauthieri, there is

no pronounced constriction, but in derived

forms such as Oviraptor philoceratops,

Conchoraptor gracilis, Citipati osmolskae,

“Oviraptor” mongoliensis, and Nemegtia

barsboldi, this constriction is distinct.

196. Shape of nasals: expanding posteriorly

(0); of subequal width throughout their

length (1). (Rauhut, 2003). The nasals

expand posteriorly in Incisivosaurus gau-

thieri, and they are very broad and widen

posteriorly in Conchoraptor. But the

nasals are of subequal width in Oviraptor

philoceratops, Citipati osmolskae, “Ovi-

raptor” mongoliensis, and Nemegtia bars-

boldi.

197. Postorbital part of the skull roof: as

high as the orbital region (0); deflected

ventrally in adult individuals (1).

(Holtz, 1994). In most dinosaurs, the pari-

etals and the medial parts of the squamos-

als are approximately level with the

frontals above the orbits, and their sur-

faces face dorsally. In Avimimus portento-

sus, Archaeopteryx lithographica, Inci-

sivosaurus gauthieri, derived ovirap-

torosaurs, ornithomimosaurs and troodon-

tids, the postorbital part of the skull roof

deflected ventrally, showing very devel-

oped braincase.

198. Number of pleurocoels in cervicals: ab-

sent (0), two, arranged horizontally (1);

one (2) (modified from Rauhut, 2003).

There are no pleurocoels in Herrerasaurus

ischigualastensis and Caudipteryx sp. Two

pairs of pneumatic openings are present in

Coelophysis bauri, Microvenator celer

(Makovicky and Sues, 1998), Ingenia (GI

100/30), Chirostenotes pergracilis and

Avimimus portentosus. Although in Avim-

imus portentosus, a second pair of open-

ing is present in some vertebrae, it is

coded here as (0), in contrast to Rauhut

(2003). One pair is present in Allosaurus

fragilis, “Oviraptor” mongoliensis and

Heyuannia huangi.

199. Pleurocoels developed as: deep depres-

sions (0); foramina (1) on cervical verte-

brae (Rauhut, 2003). Pleurocoels devel-

op as deep depressions in Coelophysis

bauri, Microvenator celer, “Oviraptor”

mongoliensis, Ingenia yanshini; as forami-

na in Archaeopteryx lithographica, Avim-

imus portentosus, Conchoraptor gracilis,

Heyuannia huangi, Nemegtia barsboldi,

and Allosaurus fragilis.

200. Nasal fusion: absent, nasals separate

(0); present, nasals fused together (1)

(Holtz, 2000).

Appendix 3

Characters and data matrix: Character state 0

is plesiomorphic; character state 1–3 are apomor-

phic. “–” is not comparable character state; “?”

is either not preserved or unknown. [ ] means

two character states included.

Allosauroidea

0000[01]000001001[01]10010100[01]11[01]000

01[02]0000001100000000000000000000[01]000

010000000[01]000[01]1000000000000002[01]1

0010000001000001[01]00101002120010102000

101000002000[01]1[01][01]00[02][01]00100000

0010100000010000?000000000001101000110

Ornithomimidae

0000000100000200000[01][01]00011100001110

10020[01]0[01]010[01]0[01]0010000000????01

002011000[01]000[01]10000000002[01][01][01]

00100100[01]011[01]1000001010002001002001

001000[02][01][02]1001120010111?001001000

00111110110[01][01]2101002000000[01]00011

010?10–0

128

Lü, Tomida, Azuma, Dong and Lee

Page 35

Dromaeosauridae

00[01]0000000[01]002100110100101100001100

0001001010000000000000000000010[01]00000

00100011000000000000001011[01]111110[01]1

0000[01]111102[12]1[01]2021021[01]12011112

000011[01]111001002101[01]10[01]011[01]000

001000110001100000000?010010100–0

Troodontidae

002110020000010001001001?11??0?010?10?10

020[01]100001100000?00???????100000001?00

11000??0?0??0?0??1[01]0?0?110001001110110

?12?1?0?112111111011011001110?1000?03?01

10110111???11?01111?10010000[01]?00?0110?

0010–0

Archaeopteryx lithographica

00?1100?00101101100010001??01110?1?00020

?1??000??0?0?0000001?1?????0000112300011–

0000000020000??0101100??1?01110001122210

00?11211111110120110101111?100021–?–11??

?0100?0?110?1011001100100000010000?101?

Avimimus portentosus

?????1???0?????????1?1?1?3?111213??00–1?2

111?1??1101?????????????00???11???1?????10

000???10?1100011011110?102????2??0120011

1112?1???????00000?0110?001?0110??1110010

01101111120?1???01??21???0??00??111?

Caudipteryx zoui

101?100?1??11011111110001??01011200011?1

?1????1?0???0??00??????????01110002?110010

01?101?1011100000000?00[01]0020000100200

00000110–101010100000111011101001111101

110?0100001000010?121010??21101000011??

0–0

Chirostenotes pergracilis

???1???00101????1???0???????????????????2?1

0?12???0??0???????????1121011002111001101

111112?11????0?1?1111111?????1????1???????

??????11????11???1110????0111101??1??1?1??

?0?101000??10???2?1????1?????1??

Nomingia gobiensis

??????????????????????????????????????????

??????1?????????????????????????????????????

??????????0??100011001201?12–???????????

???????????111111011000110011101111111?10

01??????????0????????????????????

Microvenator celer

????????????????????????????????????????????

??2?1??????????????????0????????????????????

????????0??11????01?????0?????1?020011?????

??1??0?0?112?1100???1?0????11?011?000100?

??0?????????2???????00???10?

“Oviraptor” mongoliensis

112?0111110112121111?111121010112111111

12?11122111?011111112111101?111112120120

01111111112111111001111011?010011021111

21120100???????1??011011211100011???????1

1111111?????????1?1????11211111101?111101

GIN 100/42

112001101101101211111111121010112111111

121111221110011111112111101011111212012

001111111112111111011111011111?01?02110

120120111201111111101001211100001111111

01111111110?100001010?11??11211??????????

???

Conchoraptor gracilis

102?011?11011012111101111210101121111

1112111122111?011111112111101?111112120

120011111111121111110111110111?10011011

111201211112001111111010002101000011111

1101111111110?100001010??1??1?2111101?00

101111

Ingenia yanshini

102?011?11011?12111111111210101121111111

??111?2?11???1??11121????1?11111212012001

111111112111?110011110110?1001101011121

121100200110112001001210100001111111011

111111101100001010??1????211010111110110

Heyuannia huangi

10???11????????????1????????10???111111?2?

New Oviraptorid Dinosaur, Mongolia

129

Page 36

1???2?1?????????????????1111?12121?2?01111

11??1??111110111?1?01001?0?1010111?11211

112?112?10??01110211100001111011001?1111

11001000010?011111??211?10??00???11?

Nemegtia barsboldi

1121011111010212111111111210101?2111112

121111221?10011111112111101111111211112

001111111112?111110111?1?01??1?????1?????

??????????????????100021?10000?11??????1?1

11????????????0??????1121111111?1111?11

Citipati osmolskae

112101111101121211111111121010112111100

121111221?1001111111211110111111121?112

0011111111111111???????????1?????????1?12?

111?01?01?1101????????????????11???????????

???00??00?110??1???1121111101??111??1

Incisivosaurus gauthieri

100?00011010021111101011010100213111102

12211?11??1111110010110101102101?20?110?

?100101?1?1000?????????????????????????????

????????????????????????????????????????????

?????????????????????00000??001??1

Oviraptor philoceratops

112?011???011012111111111210101121111???

??????????0???????????????0???????????????1?

11??1??11?????????????????1?????????01??11?

?????00????????????0???1????????????????????

??????????0????10????????????

Khaan mckennai

10210?1?1?0112121111111012001011211?1111

?1????2?????1?????1???????111111212111????

?111??1?1111??01???????????0010???????????

11?0????????0?000?10???????1??01???????????

?????????????1???21101010????1??1

130

Lü, Tomida, Azuma, Dong and Lee

翻译：