Content uploaded by Vladimir Alifanov
Author content
All content in this area was uploaded by Vladimir Alifanov on Mar 14, 2016
Content may be subject to copyright.
621
ISSN 0031-0301, Paleontological Journal, 2008, Vol. 42, No. 6, pp. 621–633. © Pleiades Publishing, Ltd., 2008.
Original Russian Text © V.R. Alifanov, 2008, published in Paleontologicheskii Zhurnal, 2008, No. 6, pp. 49–60.
INTRODUCTION
Bagaceratopidae is a recently established family of
horned dinosaurs of the infraorder Neoceratopsia (Ali-
fanov, 2003, 2004), which flourished in ancient Asia
during the Baruungoyot Time after the period of domi-
nance of Protoceratopidae, which occurred in this con-
tinent mostly in the mid-Cretaceous.
The skull of bagaceratopids differs from that of pro-
toceratopids in the unpaired nasal equipped with a
horny outgrowth; the presence of subnarial fenestrae
behind the nares, which are unusual for dinosaurs; and
in the small number of crests on the enamel side of the
upper teeth. The family includes the following species:
Bagaceratops rozhdestvenskyi, Breviceratops kozlowskii,
Lamaceratops tereschenkoi, Platyceratops tatarinovi
,
and
Magnirostris dodsoni.
The age of the Baruungoyot Formation, which
yielded extensive material of bagaceratopids, remains
questionable. It is evident that it precedes the Nemegt
Formation, which was dated to the boundary between
the Santonian and Campanian (Alifanov, 2000) or
between the Campanian and Maastrichtian (Gradzinski
et al., 1977); however, most of the researchers dated it
to the Maastrichtian (Martinson, 1973; Jerzykiewicz
and Russell, 1991; Shuvalov, 2000; Khand et al., 2000)
or, sometimes, Early Maastrichtian (Gao and Fox,
1996). In some works, the Baruungoyot and Djadochta
formations are considered to be synchronous (Jerzyk-
iewicz and Russell, 1991) or combined into one based
on similarity in lithological and faunal data (Martinson,
1973; Shuvalov, 1982). However, these formations dif-
fer in the composition of horned dinosaurs, indicating
different geological age (Table 1). The Baruungoyot
Formation, which precedes stratigraphically the Nem-
egt Formation and has lost some horned dinosaurs, is
younger. In Central Asia, the Djadochta Formation is
preceded by the Bayanshiree Formation, which has
been dated variously from the Cenomanian (Martinson,
1973) to Campanian (Jerzykiewicz and Russell, 1991),
in Mongolia, and probably its analogue in Central Asia
(Alifanov, 2000), the Bissekty Formation (Turonian–
Coniacian: Nessov, 1995, 1997; Middle–Upper Turo-
nian: Averianov and Sues, 2004), which also contains
neoceratopsians. The last formation is dated more pre-
cisely than the others.
In Asia, bagaceratopids have only been found in the
central part of the Gobi Desert in Mongolia and one
locality, Bayan Mandahu (Dong and Currie, 1993; You
and Dong, 2003) in Inner Mongolia (China), where
Magnirostris dodsoni
and
“Bagaceratops”
sp. (possi-
bly young
Magnirostris
) have been recorded. Judging
from the presence in the Bayan Mandahu beds of an
advanced species of
Protoceratops
(
P. hellenikorhi-
nus
), they should be regarded as intermediate between
the Djadochta and Baruungoyot formations (Table 1).
The North American history of bagaceratopids (Ali-
fanov, 2003) is evidenced by the teeth found in the Milk
River Formation (Lower Campanian) of Canada
(Baszio, 1997).
The post-Baruungoyot deposits lack horned dino-
saurs, which became extinct in Paleoasia long before
the end of the Cretaceous; this coincided with the
beginning of the stage of domination of Hadrosauridae,
the other herbivorous dinosaur group. This conclusion
is supported by the data on the beds younger than the
Baruungoyot Formation, i.e., the Nemegt Formation in
Mongolia and Udurchukan (former Nizhnetsagayan)
formations in the Amur Region. The last was dated with
The Tiny Horned Dinosaur
Gobiceratops minutus
gen. et sp. nov.
(Bagaceratopidae, Neoceratopsia)
from the Upper Cretaceous of Mongolia
V. R. Alifanov
Paleontological Institute, Russian Academy of Sciences, Profsoyuznaya ul. 123, Moscow, 117997 Russia
e-mail: valifan@paleo.ru
Received November 7, 2007
Abstract
—A new horned dinosaur,
Gobiceratops minutus
gen. et sp. nov. (Bagaceratopidae, Neoceratopsia),
from the Upper Cretaceous Baruungoyot Formation of the Khermin Tsav locality (southern Mongolia) is
described based on a 3.5-cm-long skull. The nasal included in the orbital border suggests relationship between
the new taxon and
Bagaceratops rozdestvenskyi.
It is proposed that, unlike other neoceratopsian families, the
family Bagaceratopidae is of Paleoasiatic origin.
DOI:
10.1134/S0031030108060087
Key words
: Dinosauria, Neoceratopsia, Bagaceratopidae, Upper Cretaceous, Mongolia.
622
PALEONTOLOGICAL JOURNAL
Vol. 42
No. 6
2008
ALIFANOV
confidence to the Maastrichtian, namely, the Lower–
“Middle” (Markevich, and Bugdaeva, 2001; Bugdaeva
et al., 2001) or “Middle”-Upper (Godefroit et al., 2004;
Van Itterbeeck et al., 2005) Maastrichtian. The Amur
and Nemegt assemblages differ in the taxonomic com-
position of dinosaurs. In particular, in the first, the had-
rosaurid association is dominated by the subfamily
Lambeosaurinae (
Amurosaurus, Olorotitan, Charono-
saurus
) and, among Ankylosauria, representatives of
the family Nodosauridae, which has not been recorded
in Mongolia, appear (Tumanova et al., 2003, 2004).
The difference is also manifested in the composition of
carnivorous dinosaurs; the theropod assemblage from
the Amur Region looks impoverished (Alifanov and
Bolotsky, 2002). These features suggest different age of
the Nemegt and Udurchukan formations, the second is
younger than the first.
The reason that horned dinosaurs became extinct in
ancient Asia remains uncertain. It is known that, by the
beginning of the Nemegt Time, the inner regions of
Paleoasia had undergone considerable physiographic
changes, including a fall in temperature and humifica-
tion, with the formation of extensive constant lakes
(Shuvalov, 1982). This probably resulted in a decrease
in the food usually consumed by horned dinosaurs and
an increase in competitive pressure of hadrosaurids. If
Asiatic neoceratopsians were closely connected with
pond habitats (this conclusion follows from the high
neural spines of the caudal vertebrae of protoceratopids
and bagaceratopids), they could have been influenced
by predators, such as crocodiles of the family Shamo-
suchidae (Alifanov, 2005).
At the end of Cretaceous, the dominants among the
plant-eating dinosaur association of North America
also changed, but in a different manner than in Asia;
during the Maastrichtian, the diversity of Hadrosau-
ridae decreased, while that of neoceratopsians (mostly
Ceratopidae) increased. This is evidence of regional
faunal differentiation.
To expand the data on Bagaceratopidae,
Gobicer-
atops minutus
gen. et sp. nov. is described below. The
description is based on specimen PIN no. 3142/299, an
almost complete 3.5-cm-long skull. This specimen was
found in the Baruungoyot beds of the Khermin Tsav
locality at the end of the 1970s by employees of the
South Gobi Party of the Joint Soviet–Mongolian Paleon-
tological Expedition (SSMPE) headed by V.Yu. Reshetov.
The first data on this specimen were given in the cata-
logue of exhibits of the Paleontological Museum of the
Russian Academy of Sciences (PIN), where it was dis-
played for several years under the name
Bagaceratops
rozhdestvenskyi
(Kurzanov, 1995).
The holotype of
Gobiceratops minutus
gen. et sp.
nov. makes possible to describe for the first time many
cranial features of a representative of Bagaceratopidae
based on a single relatively complete specimen, to gain
an insight into the determination of individual age
based on the skull of an extinct species (see
Remarks
),
to determine phylogenetic relationships between spe-
cies of the recently established family, and to empha-
size the importance of this family for a better under-
standing of the evolutionary processes within horned
dinosaurs as a whole.
SYSTEMATIC PALEONTOLOGY
Suborder Ceratopsia
Infraorder Neoceratopsia
Family Bagaceratopidae Alifanov, 2003
Genus
Gobiceratops
Alifanov, gen. nov.
Etymology. From the Gobi Desert and the
generic name
Ceratops
.
Type species.
Gobiceratops minutus
, sp. nov.
Diagnosis. Tiny horned dinosaur, with large
orbits and large lower temporal fenestrae. Preorbital
region of skull short and low, and occipital region high
and wide. Rostral bone absent. Antorbital fossae
located under orbits; subnarial fenestrae located in line
with their anterior margin, wider and slightly higher
than nares. Supraoccipital large and wide. Dorsally,
prefrontals lacking contact with frontals because of
Table 1.
Horned dinosaurs in terrestrial vertebrate assemblages of Central Asia
Sequence of the major Late Cretaceous
assemblages, including horned dinosaurs Taxa of Ceratopsia
Baruungoyot (Mongolia) Bagaceratopidae:
Bagaceratops
rozhdestvenskyi
,
Breviceratops
kozlowskii
,
Gobiceratops
minutus
gen. et sp. nov.,
Lamaceratops
tereschenkoi
,
Platyceratops
tatarinovi
Protoceratopidae indet.
Bayan Mandahu (Inner Mongolia, China) Bagaceratopidae: “
Bagaceratops
” sp.,
Magnirostris
dodsoni
Protoceratopidae:
Protoceratops
andrewsi
,
Protoceratops
hellenikorhinus
Djadochta (Mongolia) Bagaceratopidae indet.
Protoceratopidae:
Protoceratops
andrewsi
,
Bainoceratops
efremovi
,
Udanoceratops
tchizhovi
Bayanshiree (Mongolia) and
Bissekty (Uzbekistan) Protoceratopidae:
Graciliceratops
mongoliensis
,
Yamaceratops
dorngobiensis
Ceratopidae:
Turanoceratops
tardabilis
PALEONTOLOGICAL JOURNAL
Vol. 42
No. 6
2008
THE TINY HORNED DINOSAUR
GOBICERATOPS MINUTUS
GEN. ET SP. NOV. 623
long posterior processes of nasals, which reaching
orbital border. Nasal horn in shape of slightly project-
ing tubercle with wide base. Frontals with medially
concave orbital borders and massive posterolateral pro-
cesses. Nasofrontal suture V- or U-shaped. Postorbitals
and parietal adjoining each other, excluding frontals
from formation of border of upper temporal fenestrae.
Parietal wide, with short medial crest in occipital part
of bone. Fenestrae in parietal absent. Occipital process
of maxillae long, reaching line of posterior orbital bor-
der. Quadratojugal process of jugals having well-pro-
nounced crest with wide base. Quadratojugals large,
open laterally. Dentaries lacking angular process. Sur-
angulars and angulars approximately equal in area of
lateral surface. Splenials reaching occipitally line of
middle of adductor fossae. Each jaw with six function-
ing teeth.
Species composition. Type species.
Comparison.
Gobiceratops
gen. nov. differs
from all other bagaceratopid genera in the relatively
small body size, the large orbits, the large lower tempo-
ral fenestrae, the short preorbital skull region, the posi-
tions of the antorbital fossae under the orbits and the
subnarial fenestrae in line with their anterior margin,
the presence of close contact between the postorbitals
and parietals behind the frontals, exclusion of the fron-
tals from the formation of the border of the upper tem-
poral fenestrae, very wide parietal, with a poorly pro-
nounced medial crest, wide nasal, relatively small nasal
horn, the long occipital process of maxillae, large
quadratojugals, the presence of a crest on the quadrato-
jugal process of the jugals, the angulars and surangulars
approximately equal in the area of the lateral surface,
and in the fewer upper teeth. In addition, the new genus
differs from the majority of members of the family,
except for
Bagaceratops
, in the absence of contact
between the prefrontals and frontals, the nasal and
maxillae, and in the participation of the nasal in the for-
mation of the orbital border. It differs from
Bagacer-
atops
in the short splenials, which reach occipitally the
midlength of the adductor fossae, and in the shorter
crest of the parietal; from
Magnirostris
and
Brevicer-
atops
, in the presence of contact between the squamos-
als and jugals and in the absence of an angular process
of the dentaries; from
Magnirostris
, in the absence of
tubercles on the postorbitals; from
Lamaceratops
and
Platyceratops
, in the insignificant difference in size
between the nares and subnarial fenestrae, the medially
concave orbital and massive posterolateral margins of
the frontals, and in the U- or V-shaped nasofrontal
suture; and from
Platyceratops
, in the absence of frill
fenestrae.
Remarks. Since no breaks or signs of attachment
of a rostral bone are observed in the specimen, it was
probably lost because of underdevelopment. However,
this is not necessarily a juvenile character, because the
predentary bone, which is opposite to the rostral bone,
is present. It is noteworthy that the rostral bones are
preserved in
Psittacosaurus mongoliensis
with 28-mm-
long and 42-mm-long skulls and in
Protoceratops
andrewsi
with a 54-mm-long skull (Coombs, 1980;
Dong and Currie, 1993). This is the reason why the
absence of the rostral bone is indicated in the diagnosis.
The rostral bone, which is not known in other terres-
trial vertebrates, is usually characterized as a supple-
mentary ossification of uncertain genesis. It may be of
osteodermal origin; however, it could have been formed
due to accelerated development of a rudiment homolo-
gous to the anlage of the rostrale of crossopterygians. It
is also possible that the predentary is a homologue of
the mentomandibular, which ossifies in the anterior part
of the Meckel’s cartilage of crossopterygians.
Gobiceratops minutus
Alifanov, sp. nov.
Plate 4
Etymology. From the Latin
minutus
(small).
Holotype. PIN, no. 3142/299, almost complete
skull (squamosals, right postorbital, and left frontal are
lost, the posterior and partially anterior margin of the
parietal are damaged) with the lower jaw; Mongolia,
Southern Gobi Aimak, Nemegt Basin, Khermin Tsav
locality; Upper Cretaceous, ?Lower Santonian, Bar-
uungoyot Formation.
Description (Fig. 1). The skull is high. The
ratio of the length to the width in the orbital part is
almost 1.5. The preorbital and postorbital parts of the
skull are short. The orbits are very large, the vertical
diameter is almost half as high as the skull, the horizon-
tal diameter is 0.4 of the skull length. The nares and
subnarial fenestrae are in the shape of wide ovals with
symmetrical and vertically extended poles. The nares
are slightly smaller than the subnarial fenestrae. The
antorbital fossae are large and deep. The lower tempo-
ral fenestrae are slightly higher than the orbits; how-
ever, they are approximately half as large as the orbit in
area. The upper temporal fenestrae are small, almost
0.4 as large as the lower temporal fenestrae. The palatal
bone complex lacks suborbital fenestrae.
All bones of the braincase are densely connected to
each other. Only the medial margin of the left exoccip-
ital is turned inside the skull. In occipital view, the
braincase is almost half of the skull height. The
supraoccipital is subtriangular, wide at the base. Ven-
trally, it restricts the entire upper border of the foramen
magnum. It has long contacts with exoccipitals. The
paroccipital processes are moderately long, com-
pressed laterally in the distal part. The lower edge of the
exoccipitals is at the level of the occipital condyle. In
dorsal view, the paroccipital processes diverge at an
angle of 120
°
. The exoccipitals form the dorsal half of
the occipital condyle. The pockets containing the
foramina for XI and XII nerves are narrow and shallow.
The lateral crest of the prootics is wide. The basioccip-
ital is large, forms the lower half of the occipital
condyle. The cross section of the condyle is smaller in
area than the opening of the foramen magnum. The
624
PALEONTOLOGICAL JOURNAL
Vol. 42
No. 6
2008
ALIFANOV
Plate 4
12
7
3
4
8
56
PALEONTOLOGICAL JOURNAL
Vol. 42
No. 6
2008
THE TINY HORNED DINOSAUR
GOBICERATOPS MINUTUS
GEN. ET SP. NOV. 625
sphenoccipital tubercles are small and low, but wide at
the base. A suture between the basioccipital and
basisphenoid is indiscernible. The basipterygoid pro-
cesses are well pronounced, moderately large, rest
against the occipital flanks of the pterygoids. In ventral
view, they are positioned at an almost right angle to
each other. Other structural features of the braincase of
this specimen are uncertain.
The prefrontals are small, terminating short of the
midlength of the orbits. In dorsal view, they are wedge-
shaped. In lateral view, the lower edge of the prefrontal
is in line with the orbital center. These bones have a
very small flank. No suture between the prefrontal and
lachrymal is observed on the left side of the skull.
The unpaired nasal is very wide. The horn projec-
tion is in the shape of a low tubercle in line with the
anterior orbital border. Laterally, the bone forms the
upper edge of the nares and partially the subnarial
fenestrae. The area of contact between the nasal and
frontals is heavily damaged; however, the long poste-
rior processes preserved on the nasal suggest that the
nasofrontal suture was U- or V-shaped. The posterior
processes of the nasal reach the orbital borders at their
anterior third. No contact between the nasal and maxil-
lae is seen on the right or left side because of the
absence of a dorsal process of the maxilla. However, if
this connection existed the process of the maxilla was
narrow. In other bagaceratopids (
M. dodsoni, B. rozh-
destvenskyi
), the presence of this process is not always
identifiable because of poor preservation of the bones
in this region.
Judging from the fragment preserved on the left side
and the imprint in the matter, the frontals were narrow
rostrally and wide occipitally. The posterolateral pro-
cesses of the bone are wide. It is impossible to establish
whether or not the frontals were paired.
The width of the unpaired parietal is only slightly
smaller than its length. At the posterior third, the flanks
of this bone expand considerably and are inclined, so
that their upper edges converge, forming a short medial
(dorsal) crest. It is evident that the parietal lacks open-
ings. Rostrolaterally, it comes into distinct contact with
the postorbitals behind the frontals.
The postorbitals are massive, with a very wide fron-
tal process. The squamosal process of these bones is
narrow, crosses occipitally the middle of the upper tem-
poral fenestra. The jugal process of the postorbitals is
wide at the base, pointed ventrally, and closely
approaches the lower orbital border; thus, it frames
almost the entire posterior orbital border. Judging from
the structure of the postorbitals and jugals, the first did
not participate in the formation of the border of the
Explanation of Plate 4
Figs. 1–8.
Gobiceratops minutus
sp. nov., holotype PIN, no. 3142/299: (1–4) skull with mandibles: (1) left, (2) right, (3) rostral,
and (4) occipital views; (5, 6) skull without mandible: (5) dorsal and (6) ventral views; (7) right mandible fragment, labial view;
and (8) left mandible, labial view. Scale bar, 10 mm.
N
Pf
Fr PPo
Sq
Q
Cj
Qj
J
Ept
Mx
Dp
L
Fsn
Pmx
Pd
DA
Sa
C
Fsn
Mx
Dp
N
L
Pf
Fr
J
Po
P
Sq
Cdm
(a) (b)
(c)
Pmx
Fig. 1.
Gobiceratops minutus
sp. nov., reconstruction of the skull: (a) lateral view, (b) dorsal view, and (c) left mandible ramus,
labial view. Designations: bones: (
A
) angular, (
C
) coronoid, (
D
) dentary, (
Ept
) ectopterygoid, (
Fr
) frontal, (
J
) jugal, (
Pmx
) premax-
illa, (
N
) nasal, (
L
) lachrymal, (
Mx
) maxilla, (
P
) parietal, (
Pd
) predentary, (
Pf
) prefrontal, (
Po
) postfrontal, (
Q
) quadrate, (
Qj
) quadra-
tojugal, (
Sq
) squamosal, and (
Sa
) surangular; other structures: (
Cdm
) dorsomedial crest of the parietal, (Cj) crest of the jugal,
(Dp) antorbital fossa, and (Fsn) subnarial fenestra. Scale bar, 10 mm.
626
PALEONTOLOGICAL JOURNAL Vol. 42 No. 6 2008
ALIFANOV
upper temporal fenestra, and the squamosals, which are
not preserved in the specimen, adjoined the jugals.
The lachrymals form the anteroventral orbital bor-
der and the bottom of the antorbital fossae. They lack
contact with the premaxillae. On the left side of the
skull, the lachrymal and prefrontal are fused without
forming a distinct suture.
Rostrally, the jugals terminate short of the anterior
orbital border. The ascending process of the jugal forms
the posterior border of a relatively narrow postorbital
bar, rising to the level of the upper third of the orbits.
The quadratojugal process of these bones is large, with
a low crest on the external surface. The occipital end of
this process is pointed, positioned at an angle of 45° to
the dental edge of the maxillae. It extends ventrally to
the level of the dental edge of the maxillae; posteriorly,
it reaches the quadrate.
The quadratojugals are large, open laterally, form-
ing the border of the lower temporal fenestrae. The ven-
tral edge of these bones is located lower than the jugals
and maxillae.
The quadrates are very high. The mandibular
condyle is transversely extended. The pterygoid pro-
cess is large, formed by the middle part of the bone.
The dorsal process of the premaxillae, which sepa-
rates the subnarial and narial openings, is high, narrow,
slightly widened at the end. On the left side of the skull,
it overlies the nasal. These bones come in contact at the
level of the lower orbital border. On the right side of the
skull, this contact is not seen on the premaxillae. The
maxillary process of the premaxillae is short and wide,
the rostrodorsal process is slightly narrower and lower
than the dorsal process. The suture between the rostro-
dorsal processes is poorly visible. Ventrally, a relatively
small foramen is located at the boundary of the premax-
illae and maxillae, ahead of which the right and left pre-
maxillae are fused.
The shelf of the maxillae is deep, but low; anteriorly,
it is narrowed considerably. In the shelf zone, the bones
have many small foramina. A long occipital process of
the maxillae closely approaches the line of the posterior
orbital border.
The vomers are narrow (it remains uncertain
whether or not this bone is paired), positioned in the
upper part of a high palatal roof, which is supported lat-
erally by the maxillae and palatines. An interpterygoid
depression is absent. The vomer terminates short of the
premaxillae because of connection between the rostro-
medial margins of the maxillae. The choanae are slit-
like, as long as the vomer. The palatines form most of
the lateral walls of the palatal roof. In ventral view, the
sutures between the palatines and pterygoids are indis-
tinct. The ectopterygoids are seen in the region of con-
tact between the occipital ends of the maxillae and
pterygoids. Here they form the occipital part of the
pterygoid flanks, overlapping dorsooccipitally the pos-
terior process of the maxilla. The pterygoid flank is
low; in ventral view, its axis is inclined posteriorly at a
small angle. The pterygoids are large, with a very large
quadrate process, which forms extensive connection
with the quadrate. In line with the posterior orbital bor-
der, the pterygoids adjoin each other medially without
a distinct suture, which disappears completely in the
occipital flank. This flank is massive, U-shaped. The
ventral surface of the base of the quadrate process of the
pterygoids has a crest. The crests of the right and left
pterygoids come into contact at an angle of 120°. The
crests probably contributed to the locking of the phar-
ynx by the tongue base during feeding, for example, in
the water and, hence, participated in the formation of
the pharyngeal valve.
The ventral margin of the mandibular rami is
straight (in B. rozhdestvenskyi, P. tatarinovi, and M.
dodsoni, it is arched). The predentary is incomplete.
However, judging from the preserved upper part and a
trace of attachment of the lost fragment, this bone was
probably narrow and high, inclined rostrally at about
45° to the lower edge of the mandible. The ascending
process of the predentary is preserved; it seems to be
low and blunted. This bone had small but well-pro-
nounced lateral processes at the level of the upper edge
of the dentaries.
Occipitally, the dentaries form two processes, a high
and blunted coronoid process and a short and pointed
surangular process. The dentaries lack an angular pro-
cess. In line with the surangular process of the den-
taries, the postdentary parts of the mandibles are
divided by a wide longitudinally extended crest, which
is formed mostly of the surangular.
In lateral view, the angular bones are positioned
below these crests. They are approximately equal to the
surangulars in the area of the lateral surface (they are
usually larger or smaller, as in the holotype of Brevice-
ratops kozlowskii). The angulars contribute to the for-
mation of the ventral and medial flanks of the posterior
part of the lower jaw rami behind the splenials. Ros-
trally, they extend to the line of the penultimate tooth.
The mandibular condyles are formed by the articular
and prearticular bones; the first is relatively small and
narrow, while the second forms a wider medial part.
The prearticular bones also form a relatively small
medial flank of the mandibles, extending along the
lower margin of the adductor fossae. Rostrally, the sple-
nials terminate short of reaching the symphysis, poste-
rior to a narrow Meckel’s groove (it opens in line with
the fourth replacement tooth). The posterior edge of the
splenials reaches the midlength of the adductor fossae.
These fossae are large and deep. The coronoids are
wide. In lateral view, they are slightly exposed above
the coronoid process of the dentaries; however, in gen-
eral, they are positioned medially and dorsally relative
to this process, terminating short of the edge of the
tooth row.
The maxilla has completely lost its teeth. Young
teeth are only preserved in the anterior part of the left
and right sides; however, their structural features
PALEONTOLOGICAL JOURNAL Vol. 42 No. 6 2008
THE TINY HORNED DINOSAUR GOBICERATOPS MINUTUS GEN. ET SP. NOV. 627
remain uncertain. The tooth row of each jaw has six
alveolar pits, gradually increasing in size posteriorly
(except for the last). The absence of teeth is accompa-
nied by the disruption of the medial edge of the dental
margin of the maxilla, which probably occurred in the
burial. A similar jaw margin is observed in the rostral
part of the left dentary, where one (the first) functioning
tooth is preserved. The most complete set of teeth is
present in the right lower jaw ramus. The horizontal
row consists of six functioning teeth, corresponding to
the number of alveoli. The tooth crowns are worn
heavily labially. The lingual side shows that the enamel
surface of crowns has two depressions unequal in area
(the rostral depression is smaller and the occipital
depression is larger), with a massive central crest
between them. The depressions contain vertical crests,
one on each small tooth and two on each large tooth.
The majority of functioning teeth (except for the first
and last) have a replacement tooth at the base. The
replacement teeth of the second and fourth rows are
large, while those of the third and fifth are small, at the
initial stage of growth. The third replacement tooth is
inside the jaw and is seen through a relatively large
resorption hole. The fourth replacement tooth is the
largest. At high magnification, additional denticles are
visible in its anterior edge.
Measurements, mm. Skull: total length, 33;
occipital depth, 24; width at posterior orbital border,
23; preorbital and postorbital lengths, 10 and 12; hori-
zontal and vertical orbital diameters, 14 and 11; vertical
diameter of nares and antorbital fenestrae, 3 and 5;
depth and width of the braincase in occipital view, 11
and 17; length and width of the nasal, 10 and 7; length
of the frontal, 14; reconstructed width of the frontal in
the rostral and occipital parts, 4 and 12; reconstructed
length of the parietal, 14; reconstructed width of the
parietal in the rostral and occipital parts, 13 and 14;
height of the quadrate, 16; length of the choana, 6;
lower jaw: length of ramus, 27; reconstructed depth at
the attachment of the coronoid bone, 11; dentary length
on the right side, 18; horizontal and vertical diameters
of the adductor fossae, 5 and 4.
Remarks. No distinct criteria for the determina-
tion of individual age based on the skull of reptiles have
been proposed, except for some extreme cases con-
nected with incomplete ossification in embryonic spec-
imens or fusion of certain bones in adult and old indi-
viduals.
Judging from the heavily worn teeth of the holotype
of Gobiceratops minutus gen. et sp. nov., this individual
was capable of feeding by itself. This indicates a later
rather than embryonic stage of its individual development.
The small size and reduced number of teeth are
probably connected with secondary diminution, while
the large orbits may be evidence not only of archaic
nature but also of adaptation to acute vision (this is sup-
ported by the relatively large optic nerve in horned
dinosaurs). These characters alone should not be
regarded as a strong argument for the assignment of the
holotype to a juvenile (immature).
The mature age stage of the holotype of Gobicer-
atops minutus gen. et sp. nov. is supported by the solid
connection of bones of the braincase and skull roof;
incidentally, this provided insignificant deformation of
the specimen in the burial. In addition, the premaxillae
are fused ventrally (this is not observed in some larger
specimens of bagaceratopids), the nasal and prefrontal
are fused on the right side, the prefrontal and lachrymal
are fused on the left side, and fusion is also observed in the
vomers, pterygoids, palatines, and ventral bones of the
braincase. Partial fusion is also observed between the
pterygoids and between the quadrates and quadratojugals.
The young age of the holotype of Gobiceratops
minutus gen. et sp. nov. is supported by the predentary
bone, which is broken in the lower part at the suture
because of incomplete fusion with the dentaries.
The data considered suggest that the holotype is a
young individual.
Material. Holotype.
DISCUSSION
After the description of Protoceratops andrewsi
from the Upper Cretaceous (Djadochta Formation) of
Mongolia, based on the material collected by the Cen-
tral Asian Expedition (1922–1925) of the American
Museum of Natural History in New York (Granger and
Gregory, 1923; Gregory and Mook, 1925; Brown and
Schlaikjer, 1940), an important role in the understand-
ing of the evolution of Asiatic horned dinosaurs was
played by Bagaceratops rozhdestvenskyi (Marya ska
and Osmólska, 1975), the remains of which were col-
lected in the Baruungoyot Formation of the Khermin
Tsav locality by members of the Polish–Mongolian
Paleontological Expedition (1964–1971). As in the
case of P. andrewsi, during the formation of the type
collection of B. rozhdestvenskyi, containing several
dozen specimens, it became evident that all available
specimens differ in certain features from the holotype
(ZPAL, no. MgD-I/126). The generic–species diagno-
sis of this taxon was generalized, bypassing some prob-
lems connected with the polymorphism. In the under-
standing of the results of the study of this taxon, an
important role was played by the reconstruction of the
skull of the holotype, which differed in a number of
details from the specimen examined (see Marya ska
and Osmólska, 1975: text-fig. 6, pls. XLII–XLIX; Ali-
fanov, 2003, text-figs. 2a, 3a).
The polymorphism of Asiatic horned dinosaurs was
previously considered in detail by Brown and
Schlaikjer (1940) based on the example of P. andrewsi.
The researchers remarked that the orbits, frontals, and
lachrymals decrease in relative size with age, the long
axis of the nares becomes more vertical, the nasals are
elongated and narrowed, and the frill increases in size.
n
′
n
′
628
PALEONTOLOGICAL JOURNAL Vol. 42 No. 6 2008
ALIFANOV
Individuals with a high snout, massive prefrontals, and
frontoparietal depression were tentatively determined
as males. Kurzanov (1972) considered more definitely
the manifestation of sexual dimorphism in the cranial
structure of P. andrewsi. In his opinion, the males differ
from females (the skull of which is low in the rostral
and frill regions) in the arched elevated nasals and a
more vertical position and expansion of the frill. How-
ever, in spite of the statements of this author, it is
impossible to indicate characters that divide with cer-
tainty the material described in his work into two
groups; this is evident from the diagrams in this work,
displaying the variation of some cranial elements.
It is also impossible to divide based on sexual char-
acters the specimens from the series of B. rozhdestven-
skyi. At the same time, in the original description, it was
indicated that, in representatives of this species, the
orbits decrease in relative size, the snout becomes
longer, and the frill expands occipitally with age.
In fact, the age changes indicated in the studies cited
are not supported by available material. In particular, in
a tiny form from the Bayan Mandahu locality (Inner
Mongolia, China) determined as “Protoceratops
andrewsi” (specimen IVPP, no. V9606), the snout is
already relatively long (Dong and Currie, 1993, text-
fig. 1) and the orbits are small. The age changes indi-
cated for the parietal are also questionable, since it may
not only increase and expand, but also become nar-
rowed and elongated, as in the holotype of Brevicer-
atops kozlowskii from the Khulsan locality (Marya ska
and Osmólska, 1975). At the same time, judging from
available figures, the simultaneously buried adult and
juvenile Psittacosaurus from the Early Cretaceous of
China (Meng et al., 2004) do not show clear differences
in cranial proportions.
In addition, the type collection of B. rozhdestvenskyi
contains specimens of more than one species; this is
supported by the following facts. In specimen ZPAL,
no. MgD-I/129, which is equal in size to the holotype,
the supraoccipital is high, with long lateral processes,
whereas, in the holotype, it is very small and rounded in
shape (Marya ska and Osmólska, 1975, text-figs. 6b, 7).
In the holotype, the dentaries lack an angular pro-
cess, which is well developed in specimen ZPAL,
no. MgD-I/137 (Marya ska and Osmólska, 1975, text-
figs. 6c, 10a). Certain characters, such as the exclusion
(observed in the holotype of B. rozhdestvenskyi) or
involvement (specimen ZPAL, no. MgD-I/123) of the
frontals in the formation of the border of the upper tem-
poral fenestra, are undoubtedly beyond the range of
individual, sexual, or ontogenetic variations. It is note-
worthy that, in P. andrewsi, the first condition is not
expressed. The participation of the frontals in the for-
mation of the border of the upper temporal fenestra is
probably a derived feature, because the frontals form a
wide occipital flank along with obviously secondary
narrowing of the anterior margin of the parietal.
n
′
n
′
n
′
Specimen ZPAL, no. MgD-I/123, considered above,
was characterized as a young B. rozhdestvenskyi. How-
ever, apart from the small size (the skull is 4.5 cm long),
exclusion of the frontals from the formation of the bor-
der of the upper temporal fenestrae, and large orbits, it
differs considerably from the holotype in the different
proportions of the lengths of the frontals and nasal, dif-
ferent topography of the antorbital fossae, the straight
frontonasal suture, and a number of other characters.
The differences listed cast doubt on the identification of
this specimen. In general, the recognition of a wide
range of variation (in addition to allometric changes)
within a single species adds little to the understanding
of morphological differences and, moreover, increases
previously existing contradictions. The history of the
study of Breviceratops kozlowskii is particularly indic-
ative in this respect.
In the original description, this species was repre-
sented by two relatively small skulls from the Baruun-
goyot beds of the Khulsan locality (holotype ZPAL,
no. MgD-I/117 and specimen ZPAL, no. MgD-I/116)
and assigned to the genus Protoceratops. Because of
poor preservation, the determination of the taxonomic
position of ZPAL, no. MgD-I/116 was tentative. More-
over, it differs from the holotype in the general propor-
tions and certain structural details, in particular, those
of the parietal. As was mentioned above, the parietal of
the holotype is narrow, and its dorsal surface as a whole
forms an extended medial crest; on the contrary, in
specimen ZPAL, no. MgD-I/116, this bone is wide and
the medial crest is short. Previously, this character has
not been indicated among the features showing ontoge-
netic changes. In my opinion, the medial crest of the
parietal is a secondary feature, which was formed by
the expansion and rapprochement of its flanks, as in
iguanid lizards; in both large and small bagaceratopid-
like horned dinosaurs, the dorsal surface of the parietal
is usually completely transformed into a crest, while, in
the holotype of B. rozhdestvenskyi, a short medial crest
terminates short of reaching the frontoparietal suture.
Later, Kurzanov (1990) addressed to “Protocer-
atops” kozlowskii, using a series of skulls (specimens
PIN, nos. 3142/1–5) collected by the Joint Soviet–
Mongolian Paleontological Expedition in the Khermin
Tsav locality. Referring to the results of Brown and
Schlaikjer (1940), Kurzanov regarded certain charac-
ters of the initial diagnosis (such as long and narrow
prefrontals, straightened ventral edge of the mandibles,
and the position of the nasofrontal suture occipitally to
the anterior orbital border) as manifestation of ontoge-
netic variation. The diagnosis was emended taking into
account new material, which, however, also showed
some variability.
Subsequently, Sereno (2000) proposed that B. rozh-
destvenskyi is a synonym of Br. kozlowskii. However,
this is hardly probable, since the latter (i.e., the holo-
type) shows a distinct set of characters, which includes
a narrow skull, large orbits, elongated parietal with a
PALEONTOLOGICAL JOURNAL Vol. 42 No. 6 2008
THE TINY HORNED DINOSAUR GOBICERATOPS MINUTUS GEN. ET SP. NOV. 629
dorsal crest which reaches the frontoparietal suture,
unpaired frontals, V-shaped nasofrontal suture, participa-
tion of the frontals in the formation of the border of the
upper temporal fenestrae and the postfrontals in the for-
mation of the border of the lower temporal fenestrae, long
prefrontals, and large and laterally open quadratojugals.
At the same time, some specimens that Kurzanov
(1990) assigned to Br. rozhdestvenskyi, were used for
the description of a new Bagaceratops-like horned
dinosaur, Platyceratops tatarinovi (Alifanov, 2003).
Using material collected later by the Joint Soviet–Mon-
golian Paleontological Expedition, one more species,
Lamaceratops tereschenkoi, was described. As follows
from the original description, the two species are distin-
guished not only by the size but also by a complex set
of diagnostic characters, including cranial proportions,
structural details of dentition, skull bones, frill (in
P. tatarinovi, it is fenestrated), and upper temporal
fenestrae (in L. tereschenkoi, the frontals do not partic-
ipate in the formation of the border of the upper tempo-
ral fenestra). Both species differ distinctly from other
bagaceratopids in the structure of the frontals, with the
laterally curved orbital borders (in P. andrewsi, this fea-
ture has not been reported).
The taxonomic diversity of horned dinosaurs from
the Upper Cretaceous of Mongolia is additionally sup-
ported by the diversity of their egg shells, with two
morphotypes in the Baruungoyot Formation and three
in the Djadochta Formation (Mikhailov, 1994). It is
probable that horned dinosaurs migrated in the regions
of accumulation of the Djadochta and Baruungoyot
deposits during the reproductive season rather than
constantly lived there. The fact that Djadochta and Bar-
uungoyot sands were suitable for hatching the young is
supported by abundant egg shells, shell fragments, and
even dinosaur skeletons (oviraptorids) in a brooding
posture (Norell et al., 1995; Clark et al., 1999).
The Djadochta and Baruungoyot formations (Dja-
dochta Horizon: Barsbold, 1983) have yielded an
extremely rich vertebrate assemblage (including liz-
ards, birds, mammals, and dinosaurs), each taxonomic
group of which is rather diverse. For example, the Bar-
uungoyot Formation has yielded several theropod spe-
cies of the family Oviraptoridae (Citipati osmolskae
Clark et al., 2001; Conchoraptor gracilis Barsbold,
1986; Ingenia yanshini Barsbold, 1981; Khaan mcken-
nae Clark et. al., 2001). This is not surprising because
the Baruungoyot and Djadochta strata are rather thick
and contain fossils at almost all depths of the section.
Note that another horned dinosaur, Udanoceratops
tschizhovi, was described from the Djadochta beds of
the Udan-Sayr locality, situated not far from Bayn Dzak
(Kurzanov, 1992). In addition to P. andrewsi, the Bayn
Dzak locality has yielded Bainoceratops efremovi, with
a unique set of structural characters of the axial skele-
ton and elements of the postcranial skeleton of a bagac-
eratopid-like taxon (Tereshchenko and Alifanov, 2003).
The different magnetization of the host rock with dino-
saur bones (Dashzeveg et al., 2005) in the Bayn Dzak
and Tugrugiin Shireh localities, which yielded
P. andrewsi, supports the long-term formation of
deposits of the Djadochta Formation.
P. andrewsi undoubtedly changed with age; how-
ever, some characters varying within the type collection
of this species that considered to undergo ontogenetic
variation could have been taxonomic differences.
A critical consideration of the data provided by Brown
and Schlaikjer (1940) give an impression that the series
of P. andrewsi contains at least two morphotypes rather
than one. The first is relatively small-sized, with long
frontals and short nasals, large orbits, the absence of
epijugals, and the formation of the occipital condyle
mostly by the basioccipital. The second is large, has
long nasals and short frontals, relatively small orbits,
distinct epijugals, and its exoccipitals contribute signif-
icantly to the formation of the occipital condyle.
It is noteworthy that some of the characters listed are
observed in horned dinosaurs of the psittacosaurid level
of organization. For example, in Psittacosaurus mon-
goliensis, the frontals are large and the nasals are short
(Sereno et al., 1988), as in small P. andrewsi, while, in
small Hongshanosaurus houi, the frontals are short and
the nasals are long (You and Xu, 2005), as in large
P. andrewsi. In addition, psittacosaurid species differ
from each other in the inclusion (or exclusion) of the
frontals in the formation of the border of the upper tem-
poral fenestra.
Makovicky and Norell (2006) cast doubt on the
expediency of the establishment of new species of
bagaceratopid horned dinosaurs, because for them it
seemed impossible to compare the specimens used for
the establishment of these species with the type speci-
mens of B. rozhdestvenskyi. They proposed that the
type specimens represented a juvenile stage, poorly
preserved and showing individual variation similar to
that of P. andrewsi.
However, the statement that all specimens of the
type collection of B. rozhdestvenskyi are poorly pre-
served is not true, since a large proportion of them are
morphologically complete skulls (or complete enough
to establish polymorphism). The assignment of all
specimens of the type collection of B. rozhdestvenskyi
to juveniles was not supported by any justification;
however, this statement was probably based on the
large size of some recently described Bagaceratops-
like horned dinosaurs.
The presence in members of the Bagaceratopidae of
morphological variation that have not been marked in
P. andrewsi (exclusion of the frontals from the forma-
tion of the border of the upper temporal fenestra, the
absence of an angular process of the dentary, the exter-
nally curved lateral margins of the frontals) was dis-
cussed above. The same can be illustrated based on the
example of Magnirostris dodsoni (You and Dong,
2003) from the Bayan Mandahu locality mentioned
above. The good preservation facilitates the recognition
in this taxon of the development of massive postorbit-
als, the upper margin of which expands above the level
630
PALEONTOLOGICAL JOURNAL Vol. 42 No. 6 2008
ALIFANOV
of the frontal and parietal in the shape of very high pro-
jections, and a large nasal horn, the apex of which is at
the level of the anterior orbital border. It is noteworthy
that the skull, excluding the frill, of this animal is 35 cm
long, i.e., this is the largest representative of the family
Bagaceratopidae.
It is evident that doubts of Makovicky and Norell
about the validity of new taxa are in fact an attempt at
passing over discussions concerning the morphological
diversity of late Asiatic horned dinosaurs, because other
interpretations of available material of Bagaceratops-
like taxa contradict the results of their cladistic analy-
sis, namely, of the hypothesis of relationships between
P. andrewsi and B. rozhdestvenskyi. The same is true of
the holotype of Protoceratops hellenikorhinus from
Bayan Mandahu, the skull of which is more than 70 cm
long (Lambert et al., 2001), which was ignored by
Mackovicky and Norell. Taking into account this giant
animal and following the concept of these researchers,
all specimens of the type collection of P. andrewsi
should be regarded as juveniles and the close affinity of
this species to B. rozhdestvenskyi should be questioned
because it was established using young (i.e., morpho-
logically unstable) specimens. On the other hand, a log-
ical construction assuming unpredictable deviation of
the bone arrangement in the skull roof of Bagacer-
atops-like horned dinosaurs from theoretically conceiv-
able (based on, for example, P. andrewsi) definitive
morphogenetic stage provided a basis for unrestricted
treatment of any morphological features.
REMARKS ON THE PHYLOGENY
AND PALEOZOOGEOGRAPHY
OF NEOCERATOPSIANS
AND BAGACERATOPIDS
Using cladistic analysis of relationships, the major-
ity of neoceratopsians compose parallel clades,
approaching in similarity the family Ceratopidae (Ser-
eno, 2000; Xu et al., 2002; You and Dodson, 2004;
Makovicky and Norell, 2006). Ceratopidae form a sis-
ter group of either Protoceratopidae sensu stricto, usu-
ally including Protoceratops and Bagaceratops (Ma-
kovicky, 2001; Xu et al., 2002; Makovicky and Norell,
2006), sometimes, Graciliceratops (Sereno, 2000;
Chinnery, 2004), or the group including Leptoceratops
and Montanoceratops (You and Dodson, 2004), which
is sometimes assigned to the family Leptoceratopidae
Makovicky, 2001. The two points of view are concep-
tually identical. Because of the poor understanding of
B. rozhdestvenskyi and the evolutionary trend of horned
dinosaurs, which is represented by this taxon, it is only
possible to gain tentative conclusions concerning the
phylogeny of Neoceratopsia.
The cladistic version of the evolution of neoceratop-
sians as a whole is questionable. Like Ceratopidae, the
families Bagaceratopidae and Protoceratopidae sensu
lato are characterized by a stable set of characters. For
example, the last displays distinct antorbital fossae, the
formation of connection between the nasals and maxil-
lae, large lower temporal fenestrae, the loss of contact
between the supraorbitals and postorbital and between
the quadratojugals and squamosals, and a short
postquadrate part of the squamosals. This set of mor-
phological features existed for a long time, since it
occurred not only in the Late Cretaceous, but also,
judging from the data from China, in the Early Creta-
ceous (Archaeoceratops, Liaceratops, Auroraceratops)
and even in the Late Jurassic, which is supported by the
data on Yinlong downsi. The last species displays cer-
tain archaic characters (for example, its frill is undevel-
oped); however, its cranial structure fits well in the gen-
eral morphotype of Protoceratopidae sensu lato rather
than representing a basal member of Ceratopsia, with
transitional features, as Xu et al. (2006) proposed in the
original description. Based on the set of the protocer-
atopian characters listed, it was possible to develop cra-
nial features of Bagaceratopidae. If Protoceratopidae
sensu lato and Bagaceratopidae are sister groups, the
isolation of Ceratopidae occurred prior to the beginning
of their divergence (Fig. 2a).
The distribution of Protoceratopidae sensu lato in
the Early Cretaceous and Late Jurassic of Paleoasia,
along with the data on the presence of this group in the
Late Cretaceous of Europe (Lindgren et al., 2007), sug-
gest that it initially inhabited Laurasia. After the expan-
sion of the Turgai Sea in the later half of the Jurassic
(Kalandadze and Rautian, 1992, 1997; Russell, 1993;
Alifanov, 2000), the initially single center of origin of
horned dinosaurs was divided. As is known, Ceratopi-
dae and Bagaceratopidae radiated in North America
and Paleoasia, respectively. This concept is not contrary
to the occurrence of neoceratopid teeth in the Upper
Aptian of Maryland, North America (Chinnery et al.,
1998), which were probably buried before the connec-
tion of North America and Paleoasia through the Bering
land bridge in the Albian–Cenomanian. The establish-
Ceratopidae
Protoceratopidae
Bagaceratopidae
Bagaceratops rozhdestvenskyi
Gobiceratops minutus
Breviceratops kozlowskii
Magnirostris dodsoni
Platyceratops tatarinovi
Lamaceratops tereschenkoi
(a) (b)
Fig. 2. Phylogenetic relationships of (a) the family Bagac-
eratopidae within Neoceratopsia and (b) Gobiceratops
minutus gen. et sp. nov. within Bagaceratopidae. For expla-
nation, see the text.
PALEONTOLOGICAL JOURNAL Vol. 42 No. 6 2008
THE TINY HORNED DINOSAUR GOBICERATOPS MINUTUS GEN. ET SP. NOV. 631
ment of the first Asian–American link resulted in a fau-
nal exchange between the two continents; as a result,
Ceratopidae (Turanoceratops) penetrated into Paleoa-
sia and Bagaceratopidae penetrated into North America
for the first time (Alifanov, 2003). This scenario means
that the major Neoceratopsia stems diverged in the later
half of the Jurassic rather than in the Cretaceous, as was
proposed in a number of works. This conclusion pro-
vides the basis for the understanding of reports about
the presence of Neoceratopsia in the Lower Cretaceous
of Australia (Rich and Vickers-Rich, 1993, 2003),
which suggest that this group was established in Pangea
in the Triassic.
The data on Gobiceratops minutus gen. et sp. nov.
actualize the problem of relationships within Bagacer-
atopidae. At present, it is possible to perform a similar
analysis using comparable data on the bone structure of
the skull roof. Based on this, it is possible to recognize
three groups within the family.
The first includes M. dodsoni, Br. kozlowskii, and
specimen ZPAL, no. MgD-I/123, which display long
prefrontals, the loss of contact between the jugals and
squamosals, massive postorbitals, which participate in
the formation of the border of the lower temporal fenes-
tra, and a distinct angular process of the dentaries.
The second group, including L. tereshchenkoi and
P. tatarinovi, is distinguished by the narrow nasal, wide
occipital end of the prefrontals, and by the frontals with
rounded nasal and orbital margins, a set of characters
that is atypical for horned dinosaurs.
The third group comprises B. rozhdestvenskyi and
Gobiceratops minutus gen. et sp. nov., the holotypes of
which are distinguished by the unusual position of the
posterior ends of the nasal, which reach (probably sec-
ondarily) the orbital borders, separating small prefron-
tals from the frontals. This character correlates with the
underdevelopment of the medial crest of the parietal.
Based on the formation of the medial crest of the
parietal, which reaches rostrally the line of the fronto-
parietal suture, it is possible to establish links between
the pairs Lamaceratops tereshchenkoi–Platyceratops
tatarinovi and Magnirostris dodsoni–Breviceratops
kozlowskii. Figure 2b shows a phylogram showing the
results of the present study.
ACKNOWLEDGMENTS
This study was supported by the Russian Founda-
tion for Basic Research, project no. 07-04-00306, and
the Program of the President of the Russian Federation
(NSh-6228.2006.4).
REFERENCES
1. V. R. Alifanov, “Macrocephalosaurs and Early Evolu-
tionary Stages of Lizards of Central Asia,” Tr. Paleontol.
Inst. Ross. Akad. Nauk 272, 1–126 (2000).
2. V. R. Alifanov, “Two New Dinosaurs of the Infraorder
Neoceratopsia (Ornithischia) from the Upper Cretaceous
of the Nemegt Depression, Mongolian People’s Repub-
lic,” Paleontol. Zh., No. 5, 77–88 (2003) [Paleontol. J. 37
(5), 524–534 (2003)].
3. V. R. Alifanov, “Evolution of Horned Dinosaurs in Asia,”
in Problems of Paleontology of Central Asia: To 35th Anni-
versary of the Joint Russian–Mongolian Paleontological
Expedition (Paleontol. Inst. Ross. Akad. Nauk, Moscow,
2004), pp. 9–11 [in Russian].
4. V. R. Alifanov, “Surprising Horned Dinosaurs,” Priroda,
No. 10, 58–69 (2005).
5. R. Barsbold, “Carnivorous Dinosaurs of Mongolia,”
Tr. Sovm. Sovet–Mongol. Paleontol. Eksped., No. 19,
1−120 (1983).
6. V. R. Alifanov and Y. L. Bolotsky, “New Data about the
Assemblages of the Upper Cretaceous Carnivourous
Dinosaurs (Theropoda) from the Amur Region,” in
IV Internal Sympsium IGCP 434 on the Cretaceous Con-
tinental Margin of East Asia: Stratigraphy, Sedimenta-
tion, and Tectonics: Program and Abstracts, Ed. by
G. L. Kirillova (Khabarovsk, 2002), pp. 25–26 [in Rus-
sian].
7. A. O. Averianov and C. D. Sues, “Late Cretaceous Car-
nivorous Dinosaurs of Central Asia,” in Problems of
Paleontology of Central Asia: To 35th Anniversary of the
Joint Russian–Mongolian Paleontological Expedition
(Paleontol. Inst. Ross. Akad. Nauk, Moscow, 2004),
pp. 7–8 [in Russian].
8. S. Baszio, “Systematic Palaeontology of Isolated Dino-
saur Teeth from the Latest Cretaceous of South Alberta,
Canada,” Cour. Forsch. Senckenb. 196, 33–77 (1997).
9. B. Brown and E. M. Schlaikjer, “The Structure and Rela-
tionships of Protoceratopidae,” Ann. NY Acad. Sci. 40
(3), 133–266 (1940).
10. E. V. Bugdaeva, V. S. Markevich, A. P. Sorokin, et al.,
“Stratigraphy,” in Flora and Dinosaurs from the Creta-
ceous–Paleogene Boundary of the Zeya–Bureya Basin,
Ed. by E. V. Bugdaeva (Dal’nauka, Vladivostok, 2001),
pp. 25–43 [in Russian].
11. B. J. Chinnery, “Description of Prenoceratops pieganen-
sis gen. et sp. nov. (Dinosauria: Neoceratopsia) from the
Two Medicine Formation of Montana,” J. Vertebr. Pale-
ontol. 24 (3), 572–590 (2004).
12. B. J. Chinnery, Th. R. Lipka, J. I. Kirkland, et al., “Neo-
ceratopsian Teeth from the Lower to Middle Cretaceous
of North America,” Bull. New Mexico Mus. Natur. Hist.
Sci., No. 14, 297–302 (1998).
13. J. M. Clark, M. A. Norell, and L. M. Chiappe, “An Ovi-
raptorid Skeleton from the Late Cretaceous of Ukhaa
Tolgod, Mongolia, Preserved in an Avianlike Brooding
Position over an Oviraptorid Nest,” Am. Mus. Novit.,
No. 3265, 1–36 (1999).
14. W. P. Coombs, “Juvenile Ceratopsians from Mongolia—
the Smallest Known Dinosaur Specimens,” Nature 283,
380–381 (1980).
15. D. Dashzeveg, L. Dingus, D. B. Loope, et al., “Environ-
ment, and Age Estimate for the Upper Cretaceous Dja-
dokhta Formation, Southern Ulan Nur Basin, Mongo-
lia,” Am. Mus. Novit., No. 3498, 1–31 (2005).
632
PALEONTOLOGICAL JOURNAL Vol. 42 No. 6 2008
ALIFANOV
16. Z. Dong and P. J. Currie, “Protoceratopsian Embryos
from Inner Mongolia, People’s Republic of China,” Can.
J. Earth Sci. 30 (10–11), 2248–2254 (1993).
17. K. Gao and R. Fox, “Taxonomy and Evolution of the
Late Cretaceous Lizards (Reptilia: Squamata) from
Western Canada,” Bull. Carnegie Mus. Natur. Hist.,
No. 33, 1–107 (1996).
18. P. Godefroit, Y. L. Bolotsky, and J. Van Itterbeeck, “The
Lambeosaurine Dinosaur Amurosaurus riabinini, from
the Maastrichtian of Far Eastern Russia,” Acta Palaeon-
tol. Polon. 49 (4), 585–618 (2004).
19. R. Gradzinski, Z. Kielan-Jaworowska, and T. Marianska,
“Upper Cretaceous Djadokhta, Baruungoyot and Nem-
egt Formations of Mongolia, Including Remarks on Pre-
vious Subdivisions,” Acta Geol. Polon. 27 (3), 281–318
(1977).
20. W. Granger and W. R. Gregory, “Protoceratops andrewsi, a
Preceratopsian Dinosaur from Mongolia,” Am. Mus.
Novit., No. 72, 1–9 (1923).
21. W. K. Gregory and C. C. Mook, “On Protoceratops, a
Primitive Ceratopsian Dinosaur from the Lower Creta-
ceous of Mongolia,” Am. Mus. Novit., No. 156, 1–9
(1925).
22. T. Jerzykiewicz and D. A. Russell, “Late Mesozoic
Stratigraphy and Vertebrates of the Gobi Basin,” Creta-
ceous Res. 12, 345–377 (1991).
23. N. N. Kalandadze and A. S. Rautian, “Mammalian Sys-
tem and Historical Zoogeography,” Phylogeny of Mam-
mals, Ed. by O. L. Rossolimo (Mosk. Gos. Univ., Mos-
cow, 1992), pp. 44–152 (Tr. Zool. Mus. MGU, Vol. 29)
[in Russian].
24. N. N. Kalandadze and A. S. Rautian, “Historical Zooge-
ography of Terrestrial Tetrapods and New Method of
Global Palaeogeographical Reconstructions,” Rec. Queen
Victoria Mus., No. 104, 95–98 (1997).
25. Yo. Khand, D. Badamgarav, Ya. Ariunchimeg, et al.,
“Cretaceous System in Mongolia and Depositional Envi-
ronments,” in Cretaceous Environments of Asia, Ed. by
H. Okada and N. J. Mateer (Elsever, 2000), pp. 49 –79.
26. S. M. Kurzanov, “On the Sexual Dimorphism of Proto-
ceratopses,” Paleontol. Zh., No. 1, 104–112 (1972).
27. S. M. Kurzanov, “New Data on ?Protoceratops kozlowskii
from Khermin Tsav, Mongolia,” Paleontol. Zh., No. 4,
91–97 (1990).
28. S. M. Kurzanov, “A Giant Protoceratopid from the
Upper Cretaceous of Mongolia,” Paleontol. Zh., No. 3,
81–93 (1992).
29. S. M. Kurzanov, “The Horned Dinosaurs,” in Palaeonto-
logical Institute, Palaeontological Museum Named after
Academician Yu.A. Orlov (Acad. Sci. Russia, Moscow,
1995), pp. 86–87.
30. O. Lambert, P. Godefroit, H. Li, et al., “A New Species
of Protoceratops (Dinosauria, Neoceratopsia) from the
Late Cretaceous of Inner Mongolia (P. R. China),” Bull.
Inst. Roy. Sci. Natur. Belg. Sci. Terre. 71 (Suppl.), 5–28
(2001).
31. J. Lindgren, Ph. J. Currie, M. Siverson, et al., “The First
Neoceratopsian Dinosaur Remains from Europe,” Pale-
ontology 50 (4), 929–937 (2007).
32. P. J. Makovicky, “A Montanoceratops cerorhynchus
(Dinosauria: Ceratopsia) Braincase from the Horseshoe
Canyon Formation of Alberta,” in Mesozoic Vertebrate
Life, New Research Inspired by the Paleontology of
Philip J. Currie, Ed. by D. Tanke and K. Carpenter (Indi-
ana Univ. Press, Bloomington, 2001), pp. 243–262.
33. P. J. Makovicky and M. A. Norell, “Yamaceratops dorn-
gobiensis, a New Primitive Ceratopsian (Dinosauria:
Ornithischia) from the Cretaceous of Mongolia,” Am.
Mus. Novit., No. 3530, 1–42 (2006).
34. D. S. Markevich and E. V. Bugdaeva, “Correlation of
Plant-bearing Upper Cretaceous and Paleogene Beds of
the Russian Far East,” in Flora and Dinosaurs from the
Cretaceous–Paleogene Boundary of the Zeya–Bureya
Basin, Ed. by E. V. Bugdaeva (Dal’nauka, Vladivostok,
2001), pp. 79–96 [in Russian].
35. G. G. Martinson, “On Stratigraphy and Cretaceous
Deposits of Mongolia,” Izv. Akad. Nauk SSSR, Ser.
Geol., No. 12, 89–95 (1973).
36. T. Marya ska and H. Osmólska, “Protoceratopsidae
(Dinosauria) of Asia,” Palaeontol. Polon. 33, 133–181
(1975).
37. Q. Meng, J. Liu, D. J. Varriccio, et al., “Parental Care in
an Ornithischian Dinosaur: A Dramatic Fossil May Shed
Light on How Modern Archosaurs Became Devoted Par-
ents,” Nature 431 (7005), 145–146 (2004).
38. K. E. Mikhailov, “Eggs of Theropod and Protoceratop-
sian Dinosaurs from the Cretaceous of Mongolia and
Kazakhstan,” Paleontol. Zh., No. 2, 81–96 (1994).
39. L. A. Nessov, Dinosaurs of Northern Eurasia: New Data
on the Composition of Assemblages, Ecology, and Pale-
obiogeography (St. Peterb. Gos. Univ., St. Petersburg,
1995) [in Russian].
40. L. A. Nessov, Nonmarine Vertebrates of the Cretaceous
Period of Northern Asia (Botan. Inst. Ross. Akad. Nauk,
St. Petersburg, 1997) [in Russian].
41. M. A. Norell, J. M. Clark, L. M. Chiappe, et al., “A Nest-
ing Dinosaur,” Nature 378 (6559), 774–776 (1995).
42. Th. H. Rich and P. Vickers-Rich, “Neoceratopsians and
Omithomimosaurs: Dinosaurs of Gondwana Origin?,”
Nat. Geograph. Res. Explor. 10 (1), 129–31 (1994).
43. Th. H. Rich and P. Vickers-Rich, “Protoceratopsian?
Ulnae from Australia,” Rec. Queen Victoria Mus.,
No. 113, 1–12 (2003).
44. D. A. Russell, “The Role of Central Asia in Dinosaurian
Biogeography,” Can. J. Earth. Sci. 30, 2002–2013 (1993).
45. P. C. Sereno, “The Fossil Record, Systematics and Evo-
lution of Pachycephalosaurs and Ceratopsians from
Asia,” in The Age of Dinosaurs in Russia and Mongolia,
Ed. by M. J. Benton, M. A. Shishkin, D. M. Unwin, and
E. N. Kurochkin (Cambridge Univ. Press, Cambridge,
2000), pp. 480–516.
46. P. C. Sereno, S. C. Chao, Z. Chang, et al., “Psittacosau-
rus meileyingensis (Ornithischia: Ceratopsia), a New
Psittacosaur from the Lower Cretaceous of Northeastern
China,” J. Vertebr. Paleontol. 8 (2), 366–377 (1988).
47. V. F. Shuvalov, “Paleogeography and History of Devel-
opment of Lake Systems of Mongolia in the Jurassic and
Cretaceous Time,” in Mesozoic Lacustrine Basins of
Mongolia (Nauka, Leningrad, 1982), pp. 18–68 [in Rus-
sian].
48. V. F. Shuvalov, “The Cretaceous Stratigraphy and Paleo-
biogeography of Mongolia,” in The Age of Dinosaurs in
Russia and Mongolia, Ed. by M. J. Benton, M. A. Shish-
n
′
PALEONTOLOGICAL JOURNAL Vol. 42 No. 6 2008
THE TINY HORNED DINOSAUR GOBICERATOPS MINUTUS GEN. ET SP. NOV. 633
kin, D. M. Unwin, and E. N. Kurochkin (Cambridge
Univ. Press, Cambridge, 2000), pp. 256–278.
49. V. S. Tereshchenko and V. R. Alifanov, “Bainoceratops
efremovi, a New Protoceratopid Dinosaur (Protoceratop-
idae, Neoceratopsia) from the Bayn Dzak Locality
(South Mongolia),” Paleontol. Zh., No. 3, 71–80 (2003)
[Paleontol. J. 37 (3), 293–302 (2003)].
50. T. A. Tumanova, Yu. L. Bolotsky, and V. R. Alifanov,
“Armored Dinosaurs are Discovered in Russia for the
First Time,” Priroda, No. 3, 69–70 (2003).
51. T. A. Tumanova, Yu. L. Bolotsky, and V. R. Alifanov,
“The First Finds of Armored Dinosaurs in the Upper
Cretaceous of Russia (Amur Region),” Paleontol. Zh.,
No. 1, 68–72 (2004) [Paleontol. J. 38 (1), 73–77 (2004)].
52. J. Van Itterbeeck, Yu. Bolotsky, P. Bultynck, et al.,
“Stratigraphy, Sedimentology and Palaeoecology of the
Dinosaur-bearing Kundur Section (Zeya–Bureya Basin,
Amur Region, Far Eastern Russia),” Geol. Mag. 142 (6),
735–750 (2005).
53. X. Xu, C. A. Forster, J. M. Clark, et al., “A Basal Cer-
atopsian with Transitional Features from the Late Juras-
sic of Northwestern China,” Proc. R. Soc., Ser. B. 273,
2135–2140 (2006).
54. X. Xu, P. J. Makovicky, X. Wang, et al., “Ceratopsian
Dinosaur from China and the Early Evolution of Cer-
atopsia,” Nature 416 (6879), 314–317 (2002).
55. H. You and P. Dodson, “Basal Ceratopsia,” in The Dino-
sauria 2nd ed., Ed. by D. Weishampel, P. Dodson, and
H. Osmólska (Univ. California Press, Berkley, 2004),
pp. 478–493.
56. H. You and Z. Dong, “A New Protoceratopsid (Dinosau-
ria, Neoceratopsia) from the Late Cretaceous of Inner
Mongolia, China,” Acta Geol. Sin. 77 (3), 299–303
(2003).
57. H. You and X. Xu, “An Adult Specimen of Hongshano-
saurus houi (Dinosauria: Psittacosauridae) from the
Lower Cretaceous of Western Liaoning Province,
China,” Acta Geol. Sin. 79 (2), 168–173 (2005).
