close

Вход

Забыли?

вход по аккаунту

?

Jaw Mechanics in Basal Ceratopsia (Ornithischia Dinosauria).

код для вставкиСкачать
THE ANATOMICAL RECORD 292:1352–1369 (2009)
Jaw Mechanics in Basal Ceratopsia
(Ornithischia, Dinosauria)
KYO TANOUE,1* BARBARA S. GRANDSTAFF,2 HAI-LU YOU,3
4
AND PETER DODSON
1
Canadian Museum of Nature, Ottawa, Ontario, Canada
2
School of Veterinary Medicine, University of Pennsylvania, Philadelphia, Pennsylvania
3
Institute of Geology, Chinese Academy of Geological Sciences, Beijing,
People’s Republic of China
4
School of Veterinary Medicine and Department of Earth and Environmental Science,
University of Pennsylvania, Philadelphia, Pennsylvania
ABSTRACT
Ceratopsian dinosaurs were a dominant group of herbivores in Cretaceous terrestrial ecosystems. We hypothesize that an understanding of
the feeding system will provide important insight into the evolutionary
success of these animals. The mandibular mechanics of eight genera of
basal ceratopsians was examined to understand the variability in shape
of the jaws and the early evolution of the masticatory system in Ceratopsia. Data were collected on lever arms, cranial angles and tooth row
lengths. The results indicate that psittacosaurids had higher leverage at
the beak and in the rostral part of the tooth row than basal neoceratopsians, but lower leverage in the caudal part of the tooth row. Although
the vertebrate mandible is generally considered as a third-class lever,
that of basal neoceratopsians acted as a second-class lever at the caudal
part of the tooth row, as is also true in ceratopsids. When total input force
from the mandibular adductor muscles on both sides of the skull is considered, the largest bite force in basal ceratopsian tooth rows was exerted
in the caudal part of the tooth row at the caudal extremity of the zone
with near-maximum input force. Medially positioned teeth generate
higher leverage than laterally positioned teeth. The largest bite force in
all basal ceratopsians is smaller than the maximum input force, a limit
imposed by the morphology of the basal ceratopsian masticatory system.
In ceratopsids, caudal extension of the tooth row resulted in a much
larger bite force, even exceeding the maximum input force. Anat Rec,
C 2009 Wiley-Liss, Inc.
292:1352–1369, 2009. V
Key words: basal Ceratopsia; masticatory system; mandibular
mechanics; leverage; bite force
Grant sponsors: Summer Research Stipends in Paleontology,
University of Pennsylvania; School of Arts and Sciences
Dissertation Research Fellowship, University of Pennsylvania;
The Jurassic Foundation Research Grant; Government of Canada
Post-Doctoral Research Fellowship; The Basic Outlay of Scientific
Research Work from Ministry of Science and Technology;
Hundred Talents Project of Ministry of Land and Resources of
China; Narayan Avadhani, School of Veterinary Medicine,
University of Pennsylvania.
Grant sponsor: National Natural Science Foundation of China;
Grant number: 40672007.
C 2009 WILEY-LISS, INC.
V
*Correspondence to: Kyo Tanoue, Canadian Museum of Nature, PO Box 3443, Station ‘D’, Ottawa, Canada ON K1P 6P4.
Fax: 1-613-364-4022. E-mail: ktanoue@mus-nature.ca
Received 9 June 2009; Accepted 9 June 2009
DOI 10.1002/ar.20979
Published online in Wiley InterScience (www.interscience.wiley.
com).
JAW MECHANICS IN BASAL CERATOPSIA
1353
Fig. 1. Cladogram of the Ceratopsia compiled from You et al.
(2003, 2005), Chinnery (2004), Makovicky and Norell (2006), and Xu et al.
(2006). Schematic diagrams of the lateral views of mandibles are shown
for Chaoyangsaurus, Psittacosaurus (Psittacosauridae), Auroraceratops
(basal Neoceratopsia), and Triceratops (Ceratopsidae). The outline of Triceratops mandible is modified after Hatcher et al. (1907).
INTRODUCTION
Lindgren et al., 2007). The abundance of some basal
ceratopsian genera (such as Psittacosaurus and Protoceratops) and richness of ceratopsids indicate that ceratopsians, as a group, were ecologically successful
(Dodson, 1990; Dodson and Currie, 1990; Sereno, 1990;
Dodson et al., 2004). Understanding of the effectiveness
of the masticatory apparatus may help to explain how
ceratopsians evolved to become a dominant group of herbivorous dinosaurs. However, only a few biomechanical
analyses of ceratopsian masticatory apparatus have
been carried out. Ostrom’s (1964, 1966) two-dimensional
analyses of the mandibular levers of ceratopsians
showed a progressive improvement of leverage within
Ceratopsia. Norman and Weishampel’s (1991) model of
jaw mechanics in Psittacosaurus, in which they suggested propalinal motion of the mandible, was limited by
the general lack of biological and anatomical knowledge
of the clade due to the incompleteness and poor preparation of the specimens then available.
Since 1997, discoveries of superbly preserved skulls of
basal Ceratopsia, including those of Archaeoceratops
(Dong and Azuma, 1997; You et al., in press), Zuniceratops (Wolfe and Kirkland, 1998), Chaoyangsaurus
(Zhao et al., 1999), Liaoceratops (Xu et al., 2002),
Ceratopsia was one of the dominant herbivorous dinosaur clades in Cretaceous terrestrial ecosystems of Asia
and western North America (Dodson et al., 2004; You
and Dodson, 2004). Among ceratopsians, the Ceratopsidae is the most derived clade (Fig. 1). It is composed of
large bodied (4–8 m in length) genera, known for their
horns and well-developed frills on the skull (Dodson and
Currie, 1990; Dodson et al., 2004). Ceratopsids were
dominant in Late Cretaceous terrestrial ecosystems in
western North America. Basal Ceratopsia include Psittacosauridae, basal Neoceratopsia, and a few basal-most
genera (Zhao et al., 1999; You and Dodson, 2004; Xu
et al., 2006; Zhao et al., 2006). They are generally much
smaller than ceratopsids, typically 1–3 m in length.
Their skulls generally lack horns, and frills are either
undeveloped or poorly developed. Basal ceratopsians
have been found in Oxfordian to Campanian deposits of
Asia, and Turonian to Maastrichtian deposits in North
America (Dodson and Currie, 1990; Wolfe and Kirkland,
1998; You and Dodson, 2004; Xu et al., 2006). Discoveries of European specimens from the early Campanian
were reported recently (Godefroit and Lambert, 2007;
1354
TANOUE ET AL.
TABLE 1. Measurements of various parameters of the mandibular lever
in different ceratopsian specimens
Leverage
Taxon
Ceratopsia
Chaoyangsaurus youngi
Psittacosauridae
Hongshanosaurus houi
P. mongoliensis
P. major
Neoceratopsia
Archaeoceratops oshimai
Archaeoceratops yujingziensis
Auroraceratops rugosus
Leptoceratops gracilis
Liaoceratops yanzigouensis
Protoceratops andrewsi
Ceratopsidae
Chasmosaurus belli
Styracosaurus albertensis
Caudal
end of
tooth row
Tooth
row/jaw
length
0.51
0.94
0.40
0.45
0.40
0.48
0.44
0.39
0.53
0.52
0.53
0.58
0.49
0.85
0.86
0.83
1.03
0.80
0.31
0.30
0.33
0.33
0.31
31
33
41
42
48
40
34
32
42
34
40
32
29
0.28
0.33
<0.35
0.36
0.33
0.25
0.27
0.32
0.37
0.40
0.33
–
0.27
0.39
0.47
<0.53
0.51
0.48
0.41
0.46
0.41
0.43
0.51
0.50
0.48
0.41
0.94
1.07
1.14
1.21
1.23
1.48
1.60
0.97
1.00
1.19
1.11
1.07
0.95
0.43
0.40
–
0.41
0.42
0.44
0.41
0.45
0.49
0.44
0.46
–
0.37
54
57
55
0.32
–
–
0.57
0.53
0.51
4.06
3.57
3.26
0.49
–
–
Specimen
number
Side
Input lever
angle
Beak
tip
IGCAGS V371
r
13
0.45
IVPP V12617
IVPP V12617
AMNH6254
CAGS-IG-VD-004
CAGS-IG-VD-004
lf
r
lfa
lf
r
26
21
31
25
31
IVPP V11114
IVPP V11114
CAGS-IG-VD-003
IG-2004-VD-001
IG-2004-VD-001
CMN8889
CMN8889
IVPP V12738
IVPP V12633
IVPP V12633
AMNH6460
HMNH MPC-D100/530
HMNH MPC-D100/530
lf
r
r
lf
r
lf
r
r
lf
r
r
lf
r
CMN2245
CMN344
CMN344
r
lf
r
Rostral
end of
tooth row
AMNH, American Museum of Natural History, New York, USA; CAGS, IG, IGCAGS, Institute of Geology, Chinese Academy of Geological Sciences, Beijing, China; HMNH, Hayashibara Museum of Natural History, Okayama, Japan; IVPP,
Institute of Vertebrate Paleontology and Paleoanthropology, Academia Sinica, Beijing, China; CMN, Canadian Museum of
Nature, Ottawa, Canada; PKUP, Peking University Paleontological Collections, Beijing, China.
a
Measured from the cast. Lengths are in millimeters and angles in degrees. Jaw length refers to the distance between the
rostral end of the predentary and the center of the glenoid. lf, left mandible; r, right mandible.
Hongshanosaurus (You et al., 2003), Magnirostris (You
and Dong, 2003), Prenoceratops (Chinnery, 2004), Auroraceratops (You et al., 2005), Yinlong (Xu et al., 2006),
Xuanhuaceratops (Zhao et al., 2006), Yamaceratops
(Makovicky and Norell, 2006), and Cerasinops (Chinnery
and Horner, 2007) make it possible to study the early
evolution of the ceratopsian masticatory apparatus.
However, articulated skulls and mandibles of type specimens are seldom separated when they are first
described. Further preparation of the specimens, especially involving the detachment and removal of mandibles from the skulls, has revealed new information that
was not accessible in the initial studies.
The effectiveness of a masticatory system, which can
be examined biomechanically, determines the diversity
of food accessible to an animal. Analysis of mandibular
mechanics in basal Ceratopsia is necessary to understand the early evolution of the ceratopsian masticatory
system. Mandibular mechanics should explain not only
the establishment of the ceratopsid masticatory apparatus, but also the variation within basal Ceratopsia,
which is evident from differences in mandibular morphology between psittacosaurids and basal neoceratopsians. In this study, jaw mechanics in basal ceratopsians
is examined in greater detail than has been possible
before.
MATERIALS AND METHODS
Materials
Nineteen mandibular rami of eight basal ceratopsian
genera were observed and measured (Table 1). Representative jaws for basal-most ceratopsians, psittacosaurids,
basal neoceratopsians, and ceratopsids are shown in
Fig. 2. In this study, Chaoyangsaurus (Fig. 2A,B) represents the basal-most Ceratopsia. One Chaoyangsaurus
jaw was measured. Psittacosaurus (Fig. 2C,D) and Hongshanosaurus comprise the Psittacosauridae. Five psittacosaurid
jaws
are
included
in
this
study.
Archaeoceratops (Fig. 2E,F), Auroraceratops, Leptoceratops, Liaoceratops, and Protoceratops represent the basal
Neoceratopsia. Thirteen mandibular rami of basal neoceratopsians were measured. Three rami of mandibles
belonging to two ceratopsid genera, Chasmosaurus (Fig.
2G,H) and Styracosaurus, were included for comparison
with basal ceratopsians. Parameters used in the study
(input lever length, tooth dimension, tooth row length
and lengths from the glenoid to various bite points) were
measured directly from the specimens.
Mandibular Adductor Musculature
M. adductor mandibulae externus (MAME), M. adductor mandibulae internus including M. pseudotemporalis
JAW MECHANICS IN BASAL CERATOPSIA
1355
TABLE 2. Anatomical and mechanical
abbreviations
A
B
B1, B2, B3
Be
G
G0, G1
Gl, Gr
Jl, Jr
Jr1
Jr2
K
lf
MAME
MAMP
MPsT
MPT
P
Pe
PeG
Fig. 2. Mandibles of representative ceratopsians. (A, B) Chaoyangsaurus youngi (IGCAGS V371) in dorsal (A) and right lateral (B) views.
(C, D) Psittacosaurus major (CAGS-IG-VD-004) in dorsal (C) and left
medial (D) views. (E, F) Archaeoceratops oshimai (IVPP V11114) in
dorsal (E) and right medial (F, reflected) views. (G, H) Chasmosaurus
belli (CMN 2245) in dorsal (G) and right lateral (H) views. Scale bars: 5
cm.
(MPsT), and M. pterygoideus (MPT), and M. adductor
mandibulae posterior (MAMP) are the major components
of mandibular adductor musculature in ceratopsians (Table 2; Fig. 3; Haas, 1955; Ostrom, 1964). Based on Haas
(1955), Ostrom (1964), and dissection of Alligator mississippiensis, the origins, insertions, and orientations of the
action of these muscles in basal ceratopsians are postulated as follows: (1) MAME originates in the region of
the supratemporal fenestra and inserts on the apex of
the coronoid process. The line of action is directed caudodorsally. Haas (1955) and Ostrom (1964) assumed the origin of M. adductor mandibulae externus medialis and
profundus to be on the dorsal surface of the frill. In Dodson (1996), the origin was restricted to the base of the
frill. Both suggested origins lie caudodorsal to the mandibular insertion, and both interpretations therefore
support a caudodorsal orientation for the line of action
of MAME. (2) MPsT originates in the postorbital region
on the lateral surface of the laterosphenoid and inserts
on the rostral part of the apex of the coronoid process.
The line of action is nearly vertical. (3) The origin of
MPT is on the mandibular process of the pterygoid, and
r
R0, R2
y
I
II
III
Apex of coronoid process
Bite point at tooth apex
Bite points
Effective bite point; point B projected
onto a line at the level of the tooth
bases
Center of glenoid
Effective fulcrums in the levers PR0G0
and B1R0G1, respectively
Centers of left and right glenoid,
respectively
Positions where the resultant vectors for
left and right mandibular adductor
muscles, respectively, intersect the level
of the tooth row
Original position where the resultant vector for right mandibular adductor
muscles intersect the level of the tooth
row as assumed in this study
Caudally shifted point of application of
the resultant input vector due to
steeper inclination of the vector
Point where line GlR0 and GrR0 intersects
line GrP and GlP, respectively
Left mandible
M. adductor mandibulae externus
M. adductor mandibulae posterior
M. pseudotemporalis
M. pterygoideus
Rostral end of the predentary
Point P projected onto a line at the level
of the tooth bases; the effective rostral
end of the jaw
The line from the glenoid to a point
directly below the tip of the jaw
Right mandible
Positions of resultant vectors for the
muscle forces from both sides of the
mandible
Angle of the input lever arm relative to
the ramus of the mandible in degrees
Region I of the mandible
Region II of the mandible
Region III of the mandible
insertion is on the ventrolateral, ventral, and ventromedial surfaces of the caudalmost part of the mandible.
The orientation of the MPT axis is rostrodorsal. In crocodilians (such as A. mississippiensis) there is a large anterior part of MPT, M. pterygoideus anterior, which was
reduced or entirely absent in ceratopsians because no
anatomical space for the muscle is available in this
group. (4) The origin of MAMP is on the rostral surface
of the quadrate, and its insertion is along and within the
Meckelian fossa. The axis of MAMP is directed caudodorsally. MAME is the largest jaw adductor muscle in
ceratopsians. The dominant direction of pull on the mandible of basal ceratopsians by the combined actions of all
mandibular adductor muscles is caudodorsal. Electromyographic, cinematographic, and cinefluoroscopic studies of mastication pattern in extant reptiles such as
tuatara (Gorniak et al., 1982) and A. mississippiensis
(Busbey, 1989) have recorded that that jaw adductors
act differently during bite cycles. This cannot be determined in fossil taxa and is ignored in this study.
1356
TANOUE ET AL.
Two-Dimensional Analysis
of Mandibular Mechanics
In the following two-dimensional analysis of the mandibular lever system the mandibular ramus is considered in lateral view. Only simple orthal jaw closing is
considered. The vertebrate mandible generally operates
as a third-class lever. The output force acts at the bite
point on the beak or tooth row, and the input force at
the insertion of the jaw closing muscles. Both are rostral
to the glenoid, the fulcrum of the mandibular lever. Development of the lower jaw as a third class lever permits
maximum depression (opening) of the jaw with a mini-
Fig. 3. Diagram of lateral view of basal neoceratopsian Archaeoceratops skull with locations and actions of the mandibular adductor
muscles represented by arrows.
mum length of adductor muscle fibers (Ostrom, 1964). In
ceratopsids, however, the tooth row extends caudal to the
coronoid process, and the caudal portion of the mandible
therefore acts as a second-class lever (Fig. 2G; Ostrom,
1964, 1966). Figure 4 shows a schematic basal ceratopsian
mandible in lateral view. G is the center of the glenoid
and A the apex of the coronoid process. Thus the line GA
is the input lever arm. In this study it is assumed that the
input lever is perpendicular to the resultant vector of all
jaw adductor muscles, indicated by an arrow in Fig. 4.
The length of the output lever arm is the distance from
the glenoid to the bite point along the line GPe, which is
parallel to the long axis of the mandibular ramus. It is
assumed that the resistant force at the bite point (bite
force), provided by contact with the food, acts perpendicular to the long axis of the mandibular ramus. For example,
if B is the bite point and Be is the effective bite point projected onto the mechanical axis of the jaw, the line GBe is
the output lever arm. P is the beak tip, the rostral end of
the predentary. The line GPe is the functional jaw length
as well as the output lever arm when biting food at the tip
of the beak. Friction is ignored in this analysis to simplify
calculations. Any possible rostrocaudal movement of the
mandible can also be ignored since only resistant force
acting perpendicular to the long axis of the mandibular
ramus is considered here.
Measurements were made of the length of the input
lever arm, and the angle between the input lever arm
and the long axis of the mandibular ramus. The distances from the glenoid to the tip of the predentary and to
the rostral and caudal ends of the tooth rows were measured. These represent the lengths of the output lever
arms for bite points at three different points along the
tooth row. The length of tooth row was also measured.
In a mandibular lever system,
ðTotal input forceÞ ðInput lever lengthÞ
¼ ðBite forceÞ ðOutput lever lengthÞ
Fig. 4. Diagram of the mandibular lever of ceratopsians. The lateral view of the right mandible of
Archaeoceratops is shown. The arrow pointing caudodorsally from the apex of the coronoid process is
the resultant vector of the right jaw adductor musculature.
JAW MECHANICS IN BASAL CERATOPSIA
1357
If the total input force exerted by the mandibular
adductor muscles is set at 1 unit, bite force can be
described by the ratio of input lever length to output lever length:
Bite force ¼
Input lever length
Output lever length
for any given bite point along the jaw. To compare the
mandibular mechanics of the different ceratopsian taxa,
bite forces at the tip of the beak and at the rostral and
caudal ends of the tooth row were calculated for each
taxon included in the study.
Three-Dimensional Analysis
of Mandibular Mechanics
Two-dimensional analysis is limited to consideration of
forces acting on the working side of the head, on which
an object is bitten, and ignores the effects of muscles on
the balancing side of the head, the opposite side of the
working side. During mastication, mandibular adductor
muscles on both sides are active. Since both rami of the
ceratopsian mandible are embraced firmly by the predentary, the input muscle force from the balancing side
can be transmitted to the working side and will contribute to the total input forces acting on the working side.
Greaves (1978) and Druzinsky and Greaves (1979) developed a method for three-dimensional analysis of mandibular mechanics that makes it possible to understand the
total bite force produced by the combined activity of
mandibular adductor muscles on both sides of the jaw.
This three-dimensional model has been used to analyze
bite forces in a wide variety of vertebrates (Druzinsky
and Greaves, 1979; Spencer and Demes, 1993; Bryant
and Russell, 1995; Greaves, 1978, 1988, 1995; Spencer,
1998, 1999). The models developed by Greaves (1978)
and Druzinsky and Greaves (1979) are employed here.
Figure 5A shows a dorsal view of the ceratopsian mandible. Gl and Gr are the centers of the left and right glenoid, respectively. P is the rostral end of the predentary.
Vectors of the mandibular adductor muscle forces acting
on the left and right mandibles intersect the level of the
tooth rows at Jl and Jr, respectively. Only vertical components of the vectors are considered in this model. If
the muscles on both sides produce equal forces, the resultant muscle force can be assumed to be acting at R0,
the midpoint of a line connecting Jl and Jr. This model
treats each tooth as a separate bite point, and assumes
only one bite point is active at any one time. When the
animal bites with the tip of the predentary (P) with the
left and right jaw muscles supplying equal forces to the
bite, the total resistance force at the glenoids can be represented by a force acting at an effective fulcrum G0
halfway between the glenoids, and the muscle forces can
be assumed to act at R0. The lever PR0G0 can be
regarded as a third-class lever with the fulcrum at G0.
Moving the bite point caudally along the working side
tooth row changes the location of the effective fulcrum,
causing it to migrate toward the opposite (balancing
side) glenoid. When the bite point is at B1 and both muscle forces are equal, G1 is the effective fulcrum needed to
keep R0 on the output lever. As the bite point shifts caudally along the right tooth row, as shown in Fig. 5A, the
Fig. 5. Diagram of the dorsal view of ceratopsian mandible.
Archaeoceratops mandible is shown.
effective fulcrum moves toward Gl. Since the effective
fulcrum cannot be positioned lateral to the glenoid, any
bite point caudal to the line GlK (K is the point where
the line GlR0 intersects the line GrP). For instance, at
1358
TANOUE ET AL.
B2, would be on a lever which does not incorporate R0.
Instead, the resultant muscle force will shift laterally toward Jr, and will be located on the line GlB2 at R2, the
point where GlB2 intersects the line JlJr. As the bite
point shifts caudally and the resultant force input point
shifts toward the working side, the force exerted by the
balancing side decreases. Thus, when the bite point is at
B2 the mandibular adductor muscles on the working
side still exert maximum force, but the force produced
by the muscles on the balancing side is reduced. The lateral shift in the effective force input point (from R0 to
R2) is proportional to the reduction of input from the
balancing side jaw adductor muscles. When the resultant input vector reaches Jr, effective muscle force input
from the left side drops to zero and no longer contributes
to mastication. At all points caudal to the line JlJr, for
instance at B3 in Fig. 5A, only the muscle force from the
right jaw adductors is applied, and leverage can be analyzed two-dimensionally. The zones of bite points lying
rostral to K, between K and Jr, and caudal to Jr are
termed Region I, II, and III, respectively, by Spencer
and Demes (1993) and Spencer (1998). In Region I and
the rostral part of Region II the jaw functions as a thirdclass lever; in the caudal part of Region II and in Region
III it functions as a second-class lever.
For this analysis the better-preserved side of the
mandible was reflected about the midline to form a
symmetrical reconstructed mandible, which minimizes
taphonomic distortion. Complete reconstructed mandibles, including articulated left and right rami, were then
analyzed three-dimensionally. The apices of teeth were
selected as the bite points used to calculate the maximum bite force at each tooth position. The center of the
tooth in occlusal view substituted for the apex if the apical portion of the crown is not preserved. The tip of the
predentary was used as the predentary bite point. When
the bite point is in Region I, mandibular adductor
muscles on both sides can exert maximum input force.
The total input force is set as 1 unit in this analysis,
with half of the total input force being contributed by
left adductors and half by right adductor muscles. Based
on Druzinsky and Greaves (1979), at point B1:
Bite force ¼
G1 R0
G1 B1
where G1 is the fulcrum for the lever arm passing
through B1, R0 is the effective force application point,
G1R0 is input lever arm length, and G1B1 is output lever
arm length.
This equation can be expanded to apply to Region II,
where the working side can exert 0.5 units of muscle
force, but the balancing side cannot exert the maximum
input force, to give the following equation when the bite
point is at B2:
JrR2
GlR2
Bite force ¼ 0:5 1 þ
JrR0
GlB2
where Jr is the working side muscle insertion point, Gl
is the balancing side jaw joint, R0 is the effective force
application point when input force from both sides is
maximal, and R2 is the effective force application point
when the bite point is at B2. The line JrR2 describes
total muscle input at point R2 and the line JrR0
describes total muscle input at point R0. The ratio
between the two gives effective muscle input from both
sides when the input lever passes through R2. The line
GlR2 is input lever arm length, and the line GlB2 is output lever arm length, when the bite point is at B2. The
bite force in Region III was calculated as in two-dimensional analysis. However, since all of the muscle force
being applied comes from the working side, the maximum input force available is 0.5 units.
A striking contrast between psittacosaurids and some
basal neoceratopsians lies in the geometry of the tooth
row. The tooth row of psittacosaurids is nearly straight,
but that of small basal neoceratopsians is concave laterally (Fig. 2C,E). To examine the possible effect on bite
force produced by the curved tooth rows of Archaeoceratops oshimai (IVPP V11114), Auroraceratops (IG-2004VD-001), and Liaoceratops (IVPP V12738), the maximum bite force of the teeth of a hypothetical straight
tooth row like that of psittacosaurids, lying on a line
between the first and the last teeth of the actual tooth
row, was calculated. Bite forces produced along the hypothetical straight tooth row were then compared to bite
forces generated on the actual curved tooth row
(Fig. 5B). Each tooth in the hypothetical tooth row is
placed the same distance from the glenoid as the corresponding tooth of the actual curved tooth row.
Additionally, the bite force in mandibles with different
angles between the two rami was examined because
observed intermandibular angles vary among different
species of psittacosaurids. Difference in intermandibular
angle might affect the bite forces generated by the teeth
by changing the distance from the tooth row to the midline of the mandibles, and it was desired to discover how
much of an effect differences in the intermandibular
angle might have on bite force. The left mandible of Psittacosaurus major (CAGS-IG-VD-004) was used for this
analysis (Fig. 2C). If it is reflected onto the right side to
form a symmetrical mandible, the angle between the
mandibular rami is only 18 degrees. The holotype of
P. major (LHPV1; Sereno et al., 2007) has an intermandibular angle of 30 degrees. In other Psittacosaurus
species including P. sinensis (IVPP V738) and P. lujiatunensis (PKUP 1054), this angle reaches about 60
degrees. Differences in intermandibular angle could be
an artifact of preservation, or could reflect real differences in skull morphology. Any real differences in this
angle will affect the bite forces generated by the jaws.
The bite force of hypothetical mandibles of P. major with
intermandibular angles of 30 and 60 degrees, representing, respectively, the holotype mandible of P. major and
the mandibles of other psittacosaurids with widelydiverging mandibles, were compared with that of the
jaw of P. major (CAGS-IG-VD-004; Fig. 6A).
RESULTS
Two-Dimensional Analysis
of Mandibular Mechanics
Table 1 shows the calculated values of bite force relative to the input force at three points along the ceratopsian mandible. Bite force is represented in Table 1 by
the ratio between input and output lever arm lengths.
Average bite force at the rostral end of predentary and
JAW MECHANICS IN BASAL CERATOPSIA
1359
Fig. 6. (A) Diagram of the dorsal views of the psittacosaurid mandibles. Reconstructed left mandible of Psittacosaurus major (CAGS-IGVD-004) in heavy line. Right mandible (dashed) is the reflected left
mandible. The actual mandible has an intermandibular angle of 18
degrees. Hypothetical mandibles with intermandibular angles of 30
and 60 degrees, reflecting jaw angles of the type specimen of P. major
(LHPV1) and other Psittacosaurus species with widely-diverging man-
dibles, respectively, are shown with dotted lines. (B) Regions in the
dentary tooth row of P. major (CAGS-IG-VD-004). (C) Regions in the
dentary tooth row of hypothetical mandible reflecting the holotype of
P. major (LHPV1). (D) Regions in the dentary tooth row of hypothetical
mandible reflecting Psittacosaurus species with widely divergent
mandibles.
rostral and caudal ends of the tooth rows in Choayangsaurus (basal-most Ceratopsia), Psittacosauridae, basal
Neoceratopsia, and Chasmosaurus (Ceratopsidae) are
compared in Fig. 7. Leverage at the rostral ends of the
predentary (the beak tip) and tooth row is greater in the
five psittacosaurid jaws measured than at the same two
points in the jaws of basal neoceratopsians. At the caudal end of the tooth row, in contrast, the leverage is
greater in basal neoceratopsians than in psittacosaurids.
With the exception of the left mandible of P. major, the
output lever arm from the glenoid to the caudal end of
the tooth row in psittacosaurids is longer than the input
1360
TANOUE ET AL.
Fig. 7. Average bite force at the rostral end of predentary and rostral and caudal ends of tooth rows in Chaoyangsaurus (basal-most
Ceratopsia), Psittacosauridae, basal Neoceratopsia, and Chasmosaurus (Ceratopsidae). Error bars indicate 95% confidence intervals. See
Table 1 for the bite force in each species.
lever, and leverage at the caudal end of the tooth row is
less than 1.0. In most basal neoceratopsians the length
of this output lever is shorter than the input lever, and
leverage at the caudal end of the tooth row is greater
than 1.0. The leverage in Chaoyangsaurus, the basalmost ceratopsian studied, is similar to that of psittacosaurids. Measurements of ceratopsids, while few in number for this study, indicate higher leverage throughout
the tooth row than in basal neoceratopsians. At the beak
tip, the leverage of Chasmosaurus is lower than that of
Chaoyangsaurus and psittacosaurids.
The proportional lengths of tooth rows relative to the
distances between the rostral end of predentary and the
center of glenoid, referred to as jaw lengths in this
study, differ among basal ceratopsians (Table 1). The
tooth row in Chaoyangsaurus is relatively longer than
that of psittacosaurids, but shorter than that of most basal neoceratopsians. The tooth row in psittacosaurids is
very short. The tooth row in basal neoceratopsians is relatively longer than that of psittacosaurids. The tooth
row is shorter relative to the jaw length in Protoceratops
than in Chaoyangsaurus. Correspondingly, the postdentary length of the mandible is greater in psittacosaurids
than in basal neoceratopsians, and the input lever arm
is therefore longer in psittacosaurids.
Three-Dimensional Analysis
of Mandibular Mechanics
Bite forces at each tooth position were calculated to
provide a detailed picture of how bite force varies along
the jaw in ceratopsians, and to test how reduced muscle
input force effects the bite force in the caudal part of the
jaw, where the balancing side muscle contribution to
input force decreases. Table 3 shows bite force as a proportion of the combined input force from muscles on
both sides of the head as calculated for each tooth position. The number of teeth falling into each of the three
functional regions of the jaw varies among taxa. All
teeth of Chaoyangsaurus and psittacosaurids lie in
Regions I and II, with a majority of the teeth located in
Region I. The teeth of basal neoceratopsians are distributed over all three regions, with more teeth located in
Region I than in the other two regions. However, in
Chasmosaurus (CMN 2245) the majority of the teeth (11
teeth, plus three additional tooth positions) are located
in Region III because the toothless rostral part of the
jaw is very elongated. Only three teeth are located in
Region I, and seven in Region II. In basal ceratopsians,
except for Protoceratops (AMNH 6460), the highest bite
force is exerted in the caudal half of the tooth row, but
not at the last tooth. In all basal ceratopsian specimens
studied the teeth with the largest bite force are positioned at the boundary between Region I and II. In
Chasmosaurus, the bite force increases caudally to a
maximum at the fourth and fifth teeth, which lie within
Region II, then decreases caudally in the remainder of
Region II. Bite force increases again in Region III, and
is highest at the caudal end of the tooth row. Although
the last three teeth are not preserved, the last two teeth
analyzed exerted a bite force larger than the maximum
input force. The bite force in the caudalmost region of
the tooth row exceeds the highest bite force in Region II.
Tooth size varies among ceratopsian species, and also
varies along the tooth row of an individual animal. Differences in tooth size within a single tooth row might be
related to the differences in bite force along the tooth
row found using three-dimensional analysis. Calculated
bite force at each tooth was compared with the size of
that tooth to test whether there is any correlation
between bite force and tooth size. The bite force and the
product of labiolingual and mesiodistal diameters in occlusal view, as a proxy for tooth area, of the dentary
teeth in Chaoyangsaurus (IGCAGS V371), Hongshanosaurus (IVPP V12617), P. major (CAGS-IG-VD-004), A.
oshimai (IVPP V11114), and Liaoceratops (IVPP V12738)
are plotted in Fig. 8. Bite force, tooth area in occlusal
view, and the correlation coefficient of the bite force and
tooth area for teeth in Regions I and II in these five taxa
are given in Table 4. The tooth area of Chaoyangsaurus
(IGCAGS V371), Hongshanosaurus (IVPP V12617), A.
oshimai (IVPP V11114), and Liaoceratops (IVPP V12738)
is significantly correlated with bite force (a ¼ 0.01 or
0.05). The tooth area of P. major (CAGS-IG-VD-004) is
not as well correlated with bite force as in the other four
genera (r ¼ 0.53; a ¼ 0.10). In Region III, bite force
increases caudally even though the tooth diameters
decrease in A. oshimai (IVPP V11114) and Liaoceratops,
the only two of the five taxa tested which have teeth in
Region III. Since the teeth of these five specimens are in
the same size range, tooth area and bite force of the
teeth in Regions I and II are plotted together in Fig. 9.
The comparison of bite forces in the curved tooth row
found in A. oshimai (IVPP V11114), Auroraceratops (IG2004-VD-001), and Liaoceratops (IVPP V12738) with bite
force along a hypothetical straight tooth row positioned
lateral to the actual tooth row, and anchored to the real
tooth row at its rostral and caudal ends, is shown in Fig.
10. In Region I the curved tooth row has slightly greater
leverage than the hypothetical straight tooth row, and in
Region II the leverage is also better for the curved tooth
0.73
0.66
0.53
Region II
Ceratopsidae
0.80
0.78
0.73
0.68
0.62
0.47
0.62
0.65
0.69
0.73
0.76
r
0.84
0.79
0.74
0.42
0.57
0.60
0.63
0.66
0.71
0.75
–
lf
0.47
0.50
0.81
0.79
0.77
0.73
0.63
0.38
0.55
0.58
0.61
0.65
0.69
0.72
0.76
r
0.74
0.70
0.65
0.61
0.31
0.47
0.49
0.51
0.54
0.59
0.62
0.68
0.74
lf
0.40
0.44
0.47
0.82
0.79
0.75
0.67
0.42
0.56
0.58
0.60
0.62
0.65
0.69
0.73
0.78
r
0.41
0.45
–
0.52
0.78
0.77
0.71
0.63
0.43
0.68
0.71
0.74
–
0.81
r
0.47
0.50
0.53
0.57
0.61
0.66
0.72
0.83
0.92
1.03
1.14
–
–
–
0.82
0.82
0.77
0.77
–
0.67
0.62
0.43
0.75
0.77
0.80
r
Psittacosaurus
Archaeoceratops
Hongshanosaurus
major
oshimai
Auroraceratops
Liaoceratops
Protoceratops Chasmosaurus
(IVPP V12617)
(CAGS-IG-VD-004) (IVPP V11114) (IG-2004-VD-001) (IVPP V12738) (AMNH 6460)
(CMN2245)
Basal Neoeratopsia
The bite force at the rostral end of the predentary is indicated by the top row in the table. The number of bite force values in each column represents the number
of teeth in each region in that species.
Region III
0.45
0.53
0.54
0.54
0.58
0.62
0.65
0.70
0.75
r
Region I
Side
Chaoyangsaurus
(IGCAGS V371)
Psittacosauridae
TABLE 3. Bite force relative to maximum input force at each tooth in the three regions of the jaw
Basalmost Ceratopsia
JAW MECHANICS IN BASAL CERATOPSIA
1361
1362
TANOUE ET AL.
Fig. 8. Bite force and tooth area in occlusal view of the teeth of five basal ceratopsian specimens.
Tooth area in mm2.
rows. The difference is largest where medial deflection of
the tooth row is greatest. In Region III the bite force is
equal in both types of the tooth row, since balancing side
muscles do not contribute to the bite and only the distance of the teeth from the glenoid along the mandibular
ramus determines the leverage. Differences in bite force
between medially and laterally positioned teeth at the
same tooth position along the jaw can only be revealed
by three-dimensional analysis.
The analysis of bite force in P. major (CAGS-IG-VD004) mandibles with narrow and wide divergence produced results similar to those in the comparison of
curved and straight tooth rows in basal neoceratopsians
(Fig. 11). As the intermandibular angle increases, boundaries among the three regions shift rostrally along the
tooth row, incorporating more teeth in Region II.
Because the position where the vector of input muscle
force (Jl in Fig. 6B) intersects the jaw is fixed, increases
in intermandibular angle shift the line JlJr and the
point R0 rostrally relative to the tooth row, so that the
boundaries between regions (marked by the lines GrR0K
and GrJl) pass through more rostral teeth as the intermandibular angle increases (Fig. 6B–D). In Region I,
bite force is only slightly affected by changes in intermandibular angle. The bite force in Region II shows a
clear association with jaw separation, and the difference
is much larger and more consistent than that seen in
Region I. In Region II, bite force decreases as the intermandibular angle increases.
DISCUSSION
Leverage
To date, the oldest known basal ceratopsians are
known only from Asia, as are the oldest and most basal
neoceratopsians. In most basal neoceratopsians, the
6.2
4.9
7.4
15.2
19.0
24.2
26.2
23.8
18.9
8.3
0.70
0.75
0.73
0.66
0.53
Tooth area
0.53
0.54
0.54
0.58
0.62
Bite force
Tooth areas are in mm2.
Chaoyangsaurus
r
a
0.96
0.01
16.0
18.0
16.3
17.5
15.2
14.9
11.9
0.73
0.76
0.80
0.78
0.73
0.68
0.42
Hongshanosaurus
r
a
0.87
0.01
14.6
Tooth area
0.65
Bite force
Right mandible
Right mandible
Correlation test results
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Tooth
position
r
0.53
0.84
0.79
0.74
0.57
0.60
0.63
0.66
0.71
0.75
a
0.1
20.7
21.6
13.7
14.1
16.6
17.5
19.8
23.6
20.3
Tooth area
P. major
Bite force
Left mandible
Psittacosaurus major
(CAGS-IG-VD-004)
Psittacosauridae
Hongshanosaurus
(IVPP V12617)
Chaoyangsaurus
(IGCAGS V371)
Basalmost Ceratopsia
r
0.65
0.58
0.61
0.65
0.69
0.72
0.76
0.81
0.79
0.77
0.73
0.43
0.47
a
0.05
4.9
11.4
15.2
18.4
13.9
25.9
18.0
21.6
29.0
8.9
17.9
15.1
Tooth area
A. oshimai
Bite force
Right mandible
Archaoeceratops oshimai
(IVPP V11114)
6.8
4.3
6.5
8.4
16.3
18.7
24.1
18.3
22.9
29.4
23.0
17.6
10.4
6.9
Tooth area
Liaoceratops
r
a
0.79
0.01
0.58
0.60
0.62
0.65
0.69
0.73
0.78
0.82
0.79
0.34
0.36
0.40
0.44
0.47
Bite force
Right mandible
Liaoceratops
(IVPP V12738)
Basal Neoceratopsia
TABLE 4. Correlation of bite force with tooth area in occlusal view of the teeth in Region I and II of the five specimens shown in Figure 8
JAW MECHANICS IN BASAL CERATOPSIA
1363
1364
TANOUE ET AL.
Fig. 9. Plot of bite force versus tooth area of the teeth in Regions I and II. Data for teeth of all five basal ceratopsian specimens shown in Fig. 8 are plotted together. Tooth area in mm2. Solid line, regression
for the data (y ¼ 48.916.9, r2 ¼ 0.44); dashed lines, 95% confidence band.
mandible acted as a second-class lever in the caudal
portion of the tooth row due to the geometry of the jaw
(Table 1). If the vector of the input muscle force were
steeper than is assumed in this study, the point of application of the resultant input vector would be shifted caudally (Fig. 12). As a result, the input lever arm would
have been shorter, which reduces the leverage. Nevertheless, two-dimensional analysis shows that basal neoceratopsians could still produce a bite force nearly equal
to the input muscle force at the caudal end of the tooth
row even if the input vector was steeper than assumed
in this study. Most basal neoceratopsian mandibles
would be restricted to third-class lever action by this
variation.
Caudal displacement of the coronoid process in neoceratopsians shortens the input lever arm significantly. At
the same time, it increases the slope of the input lever
arm. The angle of the input lever arm relative to the
mandibular ramus in neoceratopsians, with the exceptions of Archaeoceratops and Protoceratops, is higher
than the angles in Chaoyangsaurus and the psittacosaurids (Table 1). In ceratopsids, depression of the glenoid also affects jaw mechanics. The glenoid of
Chaoyangsaurus is almost at the same level as the dentary tooth row in lateral or medial view (Fig. 2B). That
of psittacosaurids and basal neoceratopsians lies slightly
ventral to the level of the tooth row (Fig. 2D,F). The glenoid of the more derived ceratopsids, in contrast, is well
below the level of the tooth row. Although the coronoid
process is even more displaced in a caudal direction
than is that of basal ceratopsians, the increased depression of the glenoid and increased height of the coronoid
process lengthen the input lever arm to increase leverage in ceratopsids (Ostrom, 1964).
Three-dimensional analysis of ceratopsian mandibles
produced a pattern of bite forces along the jaw that differed from the steady increase in bite force caudally
seen in the two-dimensional analysis. At the rostral end
of the predentary the bite forces calculated using the
three-dimensional method are generally higher than
those calculated using the two-dimensional method. Bite
forces calculated for the rostral ends of the tooth row are
also generally higher using the three-dimensional
method. However, in bite forces calculated for the caudal
end of the tooth row using the three-dimensional
method, calculations are much lower than expected from
the two-dimensional analysis. The low caudal bite force
results because input force in the caudal part of the jaw
comes only from the working side adductor muscles, and
is therefore only half of the total input force from both
sides applied in Region I. Bite forces calculated using
the three-dimensional method vary along the jaw
because input forces decrease caudally in Region II,
reaching a minimum value of 50% of the maximum
input force at the caudal end of Region II. In contrast,
bite forces calculated using two-dimensional analysis
increase caudally because they depend only on the
decreasing distance from the jaw hinge, while input
force remains constant over the entire jaw length.
Three-dimensional analysis shows that the largest bite
force is exerted at or close to the Region I/Region II
boundary, near the caudal end of the zone receiving
maximum total input force from the muscles on both
sides of the skull in basal ceratopsians (Table 3). Bite
force never exceeds the maximum total input force in
any of the basal ceratopsians analyzed, with the largest
ratio of 0.84 relative to the maximum total input force
being found in P. major. The bite force in Region III
increases caudally. In basal neoceratopsians it remains
smaller than the bite force in Region II throughout the
entire length of Region III, which contains only two to
four teeth. Even in basal neoceratopsians, however, the
caudalmost Region III teeth exert a larger bite force
than teeth in the rostral part of Region III. In Chasmosaurus, more than half of the teeth (14 of 24 teeth) lie
within Region III. From the 18th tooth (the eighth tooth
JAW MECHANICS IN BASAL CERATOPSIA
1365
Fig. 10. Comparison of bite forces at the teeth of the actual curved tooth row and a hypothetical
straight tooth row for three basal neoceratopsians.
in Region III) and the bite force in the caudal part of
Region III exceeds that of Region II. The two most caudal teeth analyzed exert a bite force higher than the
maximum total input force.
Position and Shape of the Tooth Row
The mandibles of ceratopsians are much wider than
the labiolingual diameters of the teeth. The tooth rows
of basal neoceratopsians lie medially on the jaw, while
those of Chaoyangsaurus and psittacosaurids lie more
laterally. Moreover, the tooth rows of basal neoceratopsians such as A. oshimai and Liaoceratops are often
strongly curved; that is, they are labially concave
(Fig. 2E). Comparison of bite strength between the
curved tooth row and a hypothetical straight tooth row
with its ends anchored at the ends of the real tooth row
reveals the advantage associated with a medial shift of
the tooth row (Fig. 10). The effect of mediolateral tooth
position on bite force can be found only using the threedimensional method of analysis; it cannot be discovered
using two-dimensional analysis, which considers only
the distance from the glenoid along the mandibular
ramus in calculations of bite force. Changes in the
mediolateral position of the bite point in Region I do not
change the leverage (Greaves, 2002). However, in Region
II the leverage increases as a tooth shifts medially due
to shortening of the output lever as pointed out in
Greaves (2002). In Region III, the input force is fixed at
the 0.5 units contributed by the working-side mandibular adductors regardless of the mediolateral position of a
tooth, and bite force is linked only to distance from the
glenoid. Taken as a whole, a more medial position of the
tooth row results in greater leverage than is produced
by the laterally positioned tooth row, and curvature of
the tooth row increases the leverage of the tooth row as
a whole by increasing leverage of the teeth in Region II
even without any increase in leverage for Regions I and
III. The ceratopsian tooth row becomes greatly elongated
relative to mandibular length in derived taxa. Medial
displacement had to take place in order for the tooth
row to extend caudal to the laterally-located coronoid
process in ceratopsids.
1366
TANOUE ET AL.
Fig. 11. Comparison of bite forces at the teeth of actual and hypothetical mandibles of P. major. See
Fig. 6 for the caudal divergence of the rami.
Fig. 12. Caudal shift of the point of application of the resultant input vector due to the change in slope
of vector. The lateral view of the right mandible of Archaeoceratops is shown. The input vector used in
this study is shown by the solid arrow. A steeper input vector is shown by a dashed arrow.
Differences in the intermandibular angle, like those
seen in psittacosaurids, may also have affected bite
force. The three-dimensional comparison of mandibles
with narrow and wide divergence angles also indicates
the advantage of positioning the teeth closer to the midline (Fig. 11). In Region I, the teeth of widely diverging
mandibles can produce bite forces slightly higher than
those produced by the narrow jaws of the actual specimen due to the shorter distance of the teeth from the
effective fulcrum G0 in the wider forms. However, the
bite force is much higher in Region II of the narrowlydiverging jaws than in the widely divergent jaws due to
JAW MECHANICS IN BASAL CERATOPSIA
1367
the shorter output lever arm lengths when the teeth lie
closer to the midline. As a whole, P. major had a more
efficient masticatory apparatus than P. lujiatunensis and
P. sinensis, which have widely-diverging mandibles.
Another interesting characteristic of psittacosaurids is
their very short tooth rows in comparison to jaw length,
when compared to the basal-most ceratopsian Chaoyangsaurus and to basal neoceratopsians (Table 1). The tooth
row in psittacosaurids occupies no more than one-third
of the mandible. In basal neoceratopsians, such as Liaoceratops, it may occupy nearly one half of the jaw length.
Psittacosaurus have been found with gastroliths
(Osborn, 1923; Xu, 1997; Wings, 2007), and may have
needed only a short tooth row for oral processing of food.
The angle y between the input lever arm and the mandibular ramus is higher in neoceratopsians, with the
exceptions of Archaeoceratops and Protoceratops, than in
Chaoyangsaurus and the psittacosaurids. Slope of the
input lever arm is linked to the vertical distance
between the adductor muscle insertion and the teeth;
increase in this angle means that the tooth row lies farther below the point where MAME and MPsT insert on
the apex of the coronoid process in most basal neoceratopsians and ceratopsids than in Chaoyangsaurus,
Archaeoceratops, Protoceratops, and the psittacosaurids.
One possible result of an increase in y, and the consequent increase in vertical distance between the adductor
muscle insertions and the tooth row, could be an
increase in gape.
Gape
Input Force of Mandibular Adductor Muscles
Gape refers to the distance between the upper and
lower jaws when they are open, and is associated with
the size of the largest food particles which can enter the
mouth. In terms of bite force generated, it is mechanically more efficient to have the jaw adductor muscle
attachment far rostral to the jaw joint to lengthen the
input lever as much as possible (Ostrom, 1964). This,
however, results in reduction of the gape since the angle
to which the jaw can open is limited by the length of the
jaw adductor muscles. In ceratopsians the jaw adductors
insert onto the coronoid process. The position of the coronoid process along the length of the jaw is correlated
with beak morphology in basal ceratopsians. The dorsal
margin of the predentary is nearly horizontal and the
broadly squared off rostral tip is no higher than the dentary tooth row in psittacosaurids, whose coronoid process
is positioned just caudal to the midlength of the mandible (Fig. 2D). This morphology limits intrusion of the
beak tip into the mouth opening, and therefore allows
maximizing the gape even with the relatively more limited degree of jaw depression permitted by the rostral
extent of the jaw adductors in psittacosaurs. In contrast,
the narrow, wedge-shaped predentary extends rostrodorsally in basal neoceratopsians, whose coronoid process is
in the caudal third of the mandible (Fig. 2F). The predentary is especially long in relatively large forms such
as Leptoceratops and Protoceratops, and extends dorsal
to the dentary tooth row as it does in ceratopsids
(Fig. 2H). An exception is Auroraceratops, whose predentary has a horizontal dorsal margin (You et al., 2005).
The hooked beaks of neoceratopsians extend into the
mouth opening, and these animals may have had a
decreased gape compared to those forms with a horizontal dorsal margin of the predentary. At the same time,
the caudal shift of the coronoid process permitted more
madibular depression than in psittacosaurids.
The difference in beak shape may reflect differences in
diet between psittacosaurids and basal neoceratopsians.
The narrow, pointed beaks of basal neoceratopsians
might have been able to scrape off and penetrate hard
plant materials including stems and large seeds, which
were not available to the taxa with a broad, flat beak
morphology. On the other hand, the wide semicircular
beaks of psittacosaurids might have been suitable for
plucking more foliage, fruits, and possibly small seeds
with each bite than the narrow neoceratopsian beaks,
which suggest selective feeding.
One way to increase the bite force is to increase the
mass or the cross sectional area of the mandibular
adductor muscles. The size of the mandibular adductor
muscles and the magnitude of input muscle force were
not considered in this study because it is difficult to estimate the mass of muscles, or their effective cross sectional area, especially in fossil taxa. However, similar
greatest bite forces relative to total input force in most
basal ceratopsian genera examined here suggest that
the magnitude of input forces differentiated the actual
bite force among them. Studies on extant vertebrates
involving actual measurements of bite force and resistance of their food prove that larger skull size is associated with a larger mass of mandibular adductor
muscles, which produce higher bite forces (Kiltie, 1982;
Herrel et al., 2002; van der Meij and Bout, 2008). The
absolute skull size of large basal neoceratopsians such
as Protoceratops, Leptoceratops, and Udanoceratops
exceeds the size of the largest psittacosaurids, indicating
larger mandibular adductor muscle mass in these basal
neoceratopsians compared to psittacosaurids and smaller
basal neoceratopsians (Kurzanov, 1992; You and Dodson,
2004; Sereno et al., 2007; You et al., 2008). Furthermore,
the expanded dorsal margin of the coronoid process in
Leptoceratops and ceratopsids implies the attachment of
mandibular adductor muscles that were much larger
than those of most basal ceratopsians. The larger absolute skull size of ceratopsids compared to basal ceratopsians means that ceratopsids would have had a larger
mass of mandibular adductor muscles than did the
smaller basal ceratopsians. This study shows that
derived ceratopsians also had greater leverage than basal ceratopsians. These factors clearly indicate the evolution of a highly effective and powerful masticatory
system in the Ceratopsidae.
Potential Food for Basal Ceratopsians
Basal ceratopsians date from the Late Jurassic to the
end of the Cretaceous. Angiosperms, which have often
been regarded as the most nutritious plants, did not constitute a significant part of the Early Cretaceous flora.
They were the most taxonomically diverse plants in the
Late Cretaceous, but they may not have been abundant
even in the Late Cretaceous, with conifers and ferns providing the dominant vegetational cover (Wing et al.,
1993; Tiffney, 1997; Barrett and Willis, 2001). Although
1368
TANOUE ET AL.
pteridophytes and gymnosperms are often dismissed as
having low nutrition quality, the foliage of Equisetum and
some conifers and ferns has an energy content comparable
to that of angiosperms when fermented (Hummel et al.,
2006; Hummel et al., 2008). In fact, Equisetum yields
more energy than angiosperms. Ginkgo foliage contains a
high amount of protein (Hummel et al., 2008). The foliage
of cycads is less nutritious than other plants mentioned
above, but their stems store starch in the pith (Jones,
1993; Whitelock 2002). Palms, which emerged in the Late
Cretaceous, may have contained starch in their stems as
well (Daghlian, 1981; Crabtree, 1987; Wing et al., 1993;
Pei-Lang et al., 2006). Palm stems, like cycads, would
have been another source of starch for ceratopsians. The
pre-angiosperm flora would have provided enough nutrition for contemporary herbivores, even for sauropods
(Hummel et al., 2008).
Basal ceratopsians were obligatory browsers on low
vegetation due to their small body size. They probably
could reach no higher than 1.5 m above ground, even if
a large bipedal basal ceratopsian stretched its neck
upward or a large quadrupedal basal neoceratopsian
could stand on its hindlimbs. All basal ceratopsians
must have fed on foliage and fruits within this range.
Availability of seeds depends on the bite force necessary
to overcome their hardness. Large basal neoceratopsians
could have consumed some seeds not available to smaller
basal ceratopsians. Cycad phytoliths have been extracted
from ceratopsid teeth, indicating that ceratopsids
actually fed on cycads (Krauss, 2001). Ceratopsids probably could scrape off the outer layers of cycad stems to
reach the starchy pith inside using their high bite force
and hooked beaks. Large basal neoceratopsians may
have been able to feed on the stem as well.
Implications for Ceratopsian Evolution
The mandibular mechanics of basal ceratopsians
mainly from Asia was examined to understand the early
evolution of the ceratopsian masticatory system. Complete mandiblular ramus is necessary for two-dimensional analysis and articulated left and right mandibular
rami for three-dimensional analysis, but the preservation of North American specimens does not allow these
analyses. Hence well-preserved Leptoceratops (CMN
8889) is the only North American basal ceratopsian specimen analyzed in this study. It is worth noting that in
two-dimensional analysis, bite force is always greatest at
the caudal end of the tooth row and lowest at the predentary in all taxa. In basal neoceratopsians bite force
at the caudal end of the tooth row equals or slightly
exceeds input force, and in ceratopsids bite force at the
caudal end of the tooth row greatly exceeds the input
muscle force. In psittacosaurs, however, the bite force at
the caudal end of the tooth row may be less than the
input force. Three-dimensional analysis shows that in
basal ceratopsians the largest bite force is exerted near
the caudal end of the zone receiving maximum total
input force from the muscles on both sides of the skull,
not at the caudal end of the tooth row. Leverage of the
labially concave tooth rows of some basal neoceratopsians and of medially-positioned teeth in psittacosaurids
with narrow intermandibular angles is higher than the
leverage of teeth with more lateral positions due to the
shorter output lever arm length in forms with medially-
positioned teeth. Elongation and caudal extension of the
tooth row took place in basal neoceratopsians and was
associated with a caudal shift of the coronoid process,
resulting in greater leverage at the caudal end of the
tooth row (Tanoue et al., in press). Nevertheless, bite
force in all basal ceratopsians examined does not ever
greatly exceed total input force, which shows the limit of
the basal ceratopsian masticatory apparatus. Within the
Ceratopsia, it was only in ceratopsids that bite force
greatly exceeded the total input force. The exceptionally
high bite forces of ceratopsids were made possible by the
caudal extension of the tooth row, reaching even caudal
to the coronoid process. Ceratopsids also had larger
mandibular adductor muscles than basal ceratopsians.
These factors combined with rapidly-replaced dental batteries mean that ceratopsids had a much more effective
masticatory system than basal ceratopsians. However,
the early stages of medial displacement of the teeth and
of caudal elongation of the tooth rows, both of which are
characteristic of the derived ceratopsid masticatory apparatus, can be seen in basal neoceratopsians. This
study also demonstrates that the early evolution of the
ceratopsian masticatory system was not a monolithic
march leading only to the ceratopsid form, but rather
reflected diversity of feeding adaptations within the basal ceratopsians.
ACKNOWLEDGMENTS
The authors thank Carl Mehling (American Museum of
Natural History), Kieran Shepherd and Margaret Feuerstack (Canadian Museum of Nature), Xing Xu (Institute
of Vertebrate Paleontology and Paleoanthropology),
Keqin Gao (Peking University), and Ken-ichi Ishii, Shigeru Suzuki, Mahito Watabe, and Yukihide Matsumoto
(Hayashibara Museum of Natural History) for providing
access to their collections. They thank George Duda (Hill
International), David Grandstaff (Temple University),
Edward Doheny, Hermann Pfefferkorn, and David Vann
(University of Pennsylvania) for valuable discussions.
They also thank Brenda Chinnery-Allgeier (University of
Texas in Austin) and Xiao-chun Wu (Canadian Museum
of Nature) for reviewing the manuscript.
LITERATURE CITED
Barrett PM, Willis KJ. 2001. Did dinosaurs invent flowers? Dinosaur-angiosperm coevolution revisited. Biol Rev 76:411–447.
Bryant HN, Russell AP. 1995. Carnassial functioning in nimravid
and felid sabertooths: theoretical basis and robustness of inferences. In: Thomason JJ, editor. Functional morphology in vertebrate paleontology. Cambridge, UK: Cambridge University Press.
p 116–135.
Busbey AB, III. 1989. Form and function of the feeding apparatus
of Alligator mississippiensis. J Morphol 202:99–127.
Chinnery B. 2004. Description of Prenoceratops pieganensis gen. et
sp. nov. (Dinosauria: Neoceratopsia) from the Two Medicine Formation of Montana. J Vert Paleontol 24:572–590.
Chinnery BJ, Horner JR. 2007. A new neoceratopsian dinosaur linking North American and Asian taxa. J Vert Paleontol 27:625–641.
Crabtree DR. 1987. Angiosperms of the northern Rocky Mountains:
Albian to Campanian (Cretaceous) macrofossil floras. Ann Mo Bot
Gard 74:707–747.
Daghlian CP. 1981. A review of the fossil record of monocotyledons.
Bot Rev 47:517–555.
Dodson P. 1990. Counting dinosaurs—how many kinds were there?
Proc Natl Acad Sci USA 87:7608–7612.
JAW MECHANICS IN BASAL CERATOPSIA
Dodson P. 1996. The horned dinosaurs. Princeton: Princeton University Press.
Dodson P, Currie PJ. 1990. Neoceratopsia. In: Weishampel DB, Dodson P, Osmolska H, editors. The Dinosauria. Berkeley: University
of California. p 593–618.
Dodson P, Forster CA, Sampson SD. 2004. Ceratopsidae. In:
Weishampel DB, Dodson P, Osmolska H, editors. The Dinosauria. 2nd ed. Berkeley: University of California Press. p 494–513.
Dong Z-M, Azuma Y. 1997. On a primitive Neoceratopsian from the
Early Cretaceous of China. In: Dong Z-M, editor. Sino-Japanese silk
road dinosaur expedition. Beijing: China Ocean Press. p 68–89.
Druzinsky RE, Greaves WS. 1979. A model to explain the posterior
limit of the bite point in reptiles. J Morphol 160:165–168.
Godefroit P, Lambert O. 2007. A re-appraisal of Craspedodon lonzeensis Dollo, 1883 from the Upper Cretaceous of Belgium: the
first record of a neoceratopsian dinosaur in Europe? Bull Inst R
Sci Naturelles Belgique Sci de la Terre 77:83–93.
Gorniak GC, Rosenberg HI, Gans C. 1982. Mastication in the
tuatara, Sphenodon punctatus (Reptilia: Rhynchocephalia):
structure and activity of the motor system. J Morphol 171:321–353.
Greaves WS. 1978. The jaw lever system in ungulates: a new model.
J Zool 184:271–285.
Greaves WS. 1988. The maximum average bite force for a given jaw
length. J Zool 214:295–306.
Greaves WS. 1995. Functional predictions from theoretical models
of the skull and jaws in reptiles and mammals. In: Thomason JJ,
editor. Functional morphology in vertebrate paleontology. Cambridge: Cambridge University Press. p 99–115.
Greaves WS. 2002. Modeling the distance between the molar tooth
rows in mammals. Can J Zool 80:388–393.
Haas GK. 1955. The jaw musculature in Protoceratops and in other
ceratopsians. Am Mus Novit 1729:1–24.
Hatcher JB, Marsh OC, Lull RS. 1907. The Ceratopsia. US Geol
Surv Monogr 49:1–300.
Herrel A, O’Reilly JC, Richmond AM. 2002. Evolution of bite performance in turtles. J Evol Biol 15:1083–1094.
Hummel J, Gee CT, Südekum K-H, Sander PM, Nogge G, Clauss
M. 2008. In vitro digestibility of fern and gymnosperm foliage:
implications for sauropod feeding ecology and diet selection. Proc
R Soc B 275:1015–1021.
Hummel J, Südekum K-H, Streich WJ, Clauss M. 2006. Forage fermentation patterns and their implications for herbivore ingesta
retention times. Funct Ecol 20:989–1002.
Jones DL. 1993. Cycads of the world. Washington, DC: Smithsonian
Institution Press.
Kiltie RA. 1982. Bite force as a basis for niche differentiation
between rain forest peccaries (Tayassu tajaacu and T. pecari). Biotropica 14:188–195.
Krauss DA. 2001. An analysis of the feeding habits of herbivorous
dinosaurs through the examination of phytoliths trapped on tooth
grinding surfaces. J Vert Paleontol 21:69A.
Kurzanov SM. 1992. A gigantic protoceratopsid from the Upper Cretaceous of Mongolia. Paleontol J 26:103–116.
Lindgren J, Currie PJ, Siverson M, Rees J, Cedestöm P, Lindgren
F. 2007. The first neoceratopsian dinosaur remains from Europe.
Palaeontol 50:929–937.
Makovicky PJ, Norell MA. 2006. Yamaceratops dorngobiensis, a
new primitive ceratopsian (Dinosauria: Ornithischia) from the
Cretaceous of Mongolia. Am Mus Novit 3530:1–42.
Norman DB, Weishampel DB. 1991. Feeding mechanisms in some
small herbivorous dinosaurs: processes and patterns. In: Rayner
JMV, Wootton RJ, editors. Biomechanics in evolution. Cambridge:
Cambridge University Press. p 161–181.
Osborn HF. 1923. Two Lower Cretaceous dinosaurs of Mongolia. Am
Mus Novit 95:1–10.
Ostrom JH. 1964. A functional analysis of jaw mechanics in the dinosaur Triceratops. Postilla 88:1–35.
Ostrom JH. 1966. Functional morphology and evolution of the ceratopsian dinosaurs. Evolution 20:290–308.
Pei-Lang AT, Mohamed AMD, Karim AA. 2006. Sago starch and
composition of associated components in palms of different growth
stages. Carbohydr Polym 63:283–286.
1369
Sereno PC. 1990. Psittacosauridae. In: Weishampel DB, Dodson P,
Osmólska H, editors. The Dinosauria. Berkeley: University of California Press. p 579–592.
Sereno PC, Zhao X-J, Brown L, Lin T. 2007. New psittacosaurid
highlights skull enlargement in horned dinosaurs. Acta Palaeontol Pol 52:275–284.
Spencer MA. 1998. Force production in the primate masticatory system: electromyographic tests of biomechanical hypotheses. J Hum
Evol 34:25–54.
Spencer MA. 1999. Constraints on masticatory system evolution in
anthropoid primates. Am J Phys Anthropol 108:483–506.
Spencer MA, Demes AB. 1993. Biomechanical analysis of masticatory system configuration in Neandertals and Inuits. Am J Phys
Anthropol 91:1–20.
Tanoue K, You H, Dodson P. In press. Mandibular anatomy in basal
Ceratopsia. In: Ryan MJ, Chinnery-Allgeier BJ, Eberth DA, editors. New perspectives on horned dinosaurs: The Royal Tyrrell
Museum Ceratopsian Symposium. Bloomington: Indiana University Press.
Tiffney BH. 1997. Land plants as food and habitat in the age of
dinosaurs. In: Farlow JO, Brett-Surman MK, editors. The complete dinosaur. Bloomington, Indiana: Indiana University Press.
p 352–370.
van der Meij MAA, Bout RG. 2008. The relationship between shape of
the skull and bite force in finches. J Exp Biol 211:1668–1680.
Whitelock LM. 2002. The Cycads. Portland, Oregon: Timber Press.
Wing SL, Hickey LJ, Swisher CC. 1993. Implications of an exceptional
fossil flora for Late Cretaceous vegetation. Nature 363:342–344.
Wings O. 2007. A review of gastrolith function with implications for
fossil vertebrates and a revised classification. Acta Palaeontol Pol
52:1–16.
Wolfe DG, Kirkland JI. 1998. Zuniceratops christopheri n. gen. and
n. sp., a ceratopsian dinosaur from the Moreno Hill Formation
(Cretaceous, Turonian) of west-central New Mexico. N M Mus
Nat Hist Sci Bull 14:303–317.
Xu X. 1997. A new psittacosaur (Psittacosaurus mazongshanensis
sp. nov.) from Mazongshan Area, Gansu Province, China. In:
Dong Z-M, editor. Sino-Japanese silk road dinosaur expedition.
Beijing: China Ocean Press. p 48–67.
Xu X, Forster CA, Clark JM, Mo J. 2006. A basal ceratopsian with
transitional features from the Late Jurassic of Northwestern
China. Proc R Soc B 273:2135–2140.
Xu X, Makovicky PJ, Wang X-l, Norell MA, You H. 2002. A Ceratopsian dinosaur from China and the early evolution of Ceratopsia.
Nature 416:314–317.
You H, Dodson P. 2004. Basal Ceratopsia. In: Weishampel DB, Dodson P, Osmólska H, editors. The dinosauria. 2nd ed. Berkeley:
University of California Press. p 478–493.
You H, Li D, Ji Q, Lamanna MC, Dodson P. 2005. On a new genus
of basal neoceratopsian dinosaur from the Early Cretaceous of
Gansu Province, China. Acta Geol Sin 79:593–597.
You H, Dong Z. 2003. A new protoceratopsid (Dinosauria: Neoceratopsia) from the Late Cretaceous of Inner Mongolia. Acta Geol
Sin 77:299–304.
You H, Tanoue K, Dodson P. 2008. New data on cranial anatomy of
ceratopsian dinosaur Psittacosaurus major. Acta Palaeontol Pol
53:183–196.
You H, Tanoue K, Dodson P. In press. A new species of Archaeoceratops (Dinosauria: Neoceratopsia) from the Early Cretaceous of the
Mazongshan area in Northwestern China. In: Ryan MJ, Chinnery-Allgeier BJ, Eberth DA, editors. New perspectives on horned
dinosaurs: The Royal Tyrrell Museum Ceratopsian Symposium.
Bloomington: Indiana University Press.
You H, Xu X, Wang X. 2003. A new genus of Psittacosauridae (Dinosauria: Ornithopoda) and the origin and early evolution of marginocephalian dinosaurs. Acta Geol Sin 77:15–20.
Zhao X, Cheng Z, Xu X. 1999. The earliest ceratopsian from the
Tuchengzi Formation of Liaoning, China. J Vert Paleontol
19:681–691.
Zhao X, Cheng Z, Xu X, Makovicky PJ. 2006. A new ceratopsian
from the Upper Jurassic Houcheng Formation of Hebei, China.
Acta Geol Sin 80:467–473.
Документ
Категория
Без категории
Просмотров
3
Размер файла
593 Кб
Теги
dinosaurs, jaw, mechanics, ceratopsia, ornithischia, basal
1/--страниц
Пожаловаться на содержимое документа