Life Sciences Contribution 89
Royal Ontario Museum

Redescription of the
skull and mandible of
Parksosaurus from the
Late Cretaceous with

comments on the family
Hypsilophodontidae
(Ornithischia)

P. M. Galton

ROM

NNT ARANETA MRR AU RUHL eH A Nh AAI), it
AE DUM ahah, i ry i; yay ry
| t M H| i? ON f if i, j ; } { i 4 | i | ie | ’

4) ‘ ys ‘is j hy 7 ey ‘\ AL i ; : ‘ ha ¥ ee | i " f by Fy { Pre Abe Oy Mit
TAO MALE Reg OER A PTR RANGE U AYA CE ET es A Dot ed HAE
ny DUEL FEMI RS CIULLAVORED ELMER PLETE PTE NTE PRIA AL AHEAD DYER RR USER Ed SH IEA fed CPA EAD De: UT Ay ea Ait

/ NAM ETSY PoP iy }
dt VER ‘i ; ' i ri

Digitized ihe the internet Archived.) i
in 2011 with fundingfrom a
University of TONG. aii Hi

“http NNW. archive, orgidetails/redescriptionots00galt

i ) |
Oa ot He aiee ts \ 2 : ; nal

4
int i : i :
hee ay i 4 Be ieee {
; , q tai 1 ; us f: 4 ;
set ain pA Pt Pe ee ; ' J ; 051 vg ‘
J 7 an J i) i Ve er Pasa re a
, ti » 4 ih) : : } 17 ‘ iy y MF, Ls

P. M:GALTON

LIFE SCIENCE CONTRIBUTIONS
ROYAL ONTARIO MUSEUM
NUMBER 89

Redescription of the skull and
mandible of Parksosaurus
from the Late Cretaceous with
comments on the family
Hypsilophodontidae
(Ornithischia)

Publication date: 25 April 1973

Suggested citation: Life Sci. Contr., R. Ont. Mus.

ROYAL ONTARIO MUSEUM
PUBLICATIONS IN LIFE SCIENCES

The Royal Ontario Museum publishes three series in the Life Sciences:

LIFE SCIENCES CONTRIBUTIONS, a numbered series of original scientific pub-
lications, including monographic works.

LIFE SCIENCES OCCASIONAL PAPERS, a numbered series of original scientific
publications, primarily short and usually of taxonomic signifi-
cance.

LIFE SCIENCES MISCELLANEOUS PUBLICATIONS, an unnumbered series of
publications of varied subject matter and format.

All manuscripts considered for publication are subject to the scrutiny and
editorial policies of the Life Sciences Editorial Board, and to review by
persons outside the Museum staff who are authorities in the particular field
involved.

LIFE SCIENCES EDITORIAL BOARD

Chairman: R.L. PETERSON

Editors: J.R.TAMSITT, Editor
D. BARR, Associate Editor
E. J. CROSSMAN, Associate Editor

P. M. GALTON is an Assistant Professor of Biology at the University
of Bridgeport, Bridgeport, Connecticut 06602.

PRICE: $1.50
©The Royal Ontario Museum, 1973
100 Queen’s Park, Toronto, Canada

PRINTED AT THE UNIVERSITY OF TORONTO PRESS

Redescription of the skull and mandible
of Parksosaurus from the Late Cretaceous
with comments on the family
Hypsilophodontidae (Ornithischia)

Abstract

The head skeleton of Parksosaurus is more similar to that of
Hypsilophodon from the Lower Cretaceous of England than
to those of Dryosaurus and Dysalotosaurus, two closely re-
lated genera of hypsilophodontids from the Upper Jurassic. The
shallowness of the lower temporal bar is a primitive feature
of the skull of Parksosaurus; specialized features include the
form of the teeth, the small size of the antorbital fenestra, and
the anteroposterior expansion of the posterodorsal end of the
jugal. The presence of foramina for replacement teeth in
several lower ornithopods represents a preadaptation for the
formation of high alveolar walls necessary for the development
of dental batteries in hadrosaurs and ceratopsians. The only
other known hypsilophodont reported from the Late Cre-
taceous of the world is from British Columbia; it may repre-
sent a new species of Hypsilophodon.

Introduction

Gilmore (1913) briefly described Thescelosaurus neglectus, a new genus
and species of ornithopod dinosaur from the Lance Formation of Wyoming,
and referred it to the family Camptosauridae. Later (1915) he gave a more
complete description and reassigned the genus to the family Hypsilopho-
dontidae. Parks (1926) described a new species, Thescelosaurus warreni,
from the Edmonton Formation of Alberta. In his description of the specimen
Parks (1926) only described the left side of the skull (Fig. 1). Sternberg
(1937) subsequently made T. warreni the generotype of Parksosaurus.
Sternberg (1940, p. 492) compared characters of Thescelosaurus and
Parksosaurus and stressed differences in the hind limb. In Thescelosaurus
the femur is longer than the tibia, and the fourth trochanter is below mid-
length. In Parksosaurus the femur is shorter than the tibia, and the fourth
trochanter is above midlength. Sternberg (1940) erected the subfamily
Thescelosaurinae for Thescelosaurus and placed Parksosaurus in the sub-
family Hypsilophodontinae. Parksosaurus was undoubtedly cursorial (see
Galton, 1971a, 1971b), and I retained it in the family Hypsilophodontidiae
with Pisanosaurus Casamiquela, Laosaurus Marsh, Dryosaurus Marsh,
Dysalotosaurus Pompeckj, and Hypsilophodon Huxley. I did not include
Thescelosaurus within the family Hypsilophodontidae but referred it to the
family Iguanodontidae (Galton, 1971a, 1971b, 1972).

Material and Methods

The crushed and distorted condition of the skull material of Parksosaurus
warreni (ROM 804) warranted a series of stereo photographs with explana-
tory line drawings (Figs. 1-4) in addition to a reconstruction of the skull
(Fig. 5). Parks (1926) described only the left side (Fig. 1) of the skull.
But the right side of the specimen (Figs. 2-4) has been extensively prepared
since 1926, and I recognized several elements not described by Parks when
I compared this side with the disarticulated bones from a nearly complete
skull of Hypsilophodon foxii Huxley (Figs. 6-8). References here to
Hypsilophodon are based on material in the British Museum (Natural
History), a complete account of which will appear elsewhere (Galton, in
press, Br. Mus.). Specimens examined are in the collections of the following
institutions: American Museum of Natural History, New York (AMNH);
British Museum (Natural History), London (BMNH); Carnegie Museum,
Pittsburgh (CM); National Museum of Canada, Ottawa (NMC); Royal
Ontario Museum, Toronto (ROM); University of California Museum of
Paleontology, Berkeley (UCMP); and Yale University Peabody Museum
of Natural History, New Haven (ypM). Radiographs of the skull were
made with a General Electric DXS 350 x-ray machine with an HDN tube
(General Electric, Toronto, Ontario).

Abbreviations used in the figures and descriptive portions of the text
are listed below. Abbreviations for the names of bones are given in upper
case, those for structures in lower case, and foramina for cranial nerves are
indicated by Roman numerals.

angular ANG quadratojugal OJ
basioccipital BO splenial SPL

basisphenoid BSP squamosal SQ
ceratobranchial CB supraoccipital SO

cervical rib 1 or 2 CE RIB supraorbital SOB

cervical vertebra 2 CE2 surangular SA

cervical vertebra 3 CE 3 vomer VO

coronoid C antorbital fenestra ant. f.
dentary D basipterygoid process pt. p.
ectopterygoid ECT cavity c

exoccipital EO foramen lacerum

frontal F posterius for. |. p.
intercentrum of atlas IC 1 foramina f£ (fl, f2, f3)
jugal J fenestra ovalis fen. ov.
lachrymal L fossa subarcuata fs
laterosphenoid LSP internal auditory meatus im
maxilla MX internal jugular vein int. j.v.
nasal N jugular foramen _j. for.
opisthotic OP lagenar recess |
orbitosphenoid OSP maxillary tooth mx
palatine PAL maxillary cavity mx. c.
parasphenoid PSP Meckelian canal mec

4

parietal P
postorbital PO
prearticular PA
predentary PD
prefrontal PF
premaxilla PMX
prootic PRO
pterygoid PT
quadrate Q0

paroccipital process par. p.
post-temporal fossa p. t. fos.
premaxillary cavity pmx.c.
sclerotic plates _ sc. pls.

sella turcica s

supraoccipital, surface for so
supraorbital, surface for sorb
vena capitis dorsalis v. cap. d.

‘SUIpeYs [eotueYysour Aq poj}eorpur st au0g
udsyolg "xX puke O pa][oqe] Souq oy} Jo AjNUAPT oy} JO UOIssnosIp & IOJ
(¢ aded) 1x9} 99g “YOO]q [[NYS JO aps JOT “POS WOU MuaivM SnanvSOsyADg | ‘BIg

Redescription of Parksosaurus warreni
Skull

Bones on the left side (Fig. 1), naturally articulated, were described by
Parks (1926). Those of the right side are displaced (Fig. 2), and most
of the lateral wall of the skull has been removed. The medial surface of the
left wall of the braincase is visible (Fig. 2) near the floor of the braincase
(Fig. 2), the dorsal surface of which is exposed. A fracture in the block
enables certain other structures to be seen (Figs. 3, 4). Certain bones of
the skull of Parksosaurus are described by comparison with well-preserved
material of Hypsilophodon (Figs. 6—8) (Galton, in press, Br. Mus.).

The basioccipital (BO) of Parksosaurus has a subspherical occipital
condyle; laterally there is a broad sutural surface, tapering anteriorly, for
the exoccipital (EO) and opisthotic (OP) bones (Fig. 2). There is no
trace of the suture between the basioccipital and the basisphenoid, the
dorsal surface of which curves dorsally to the posterior rim of the sella
turcica as in Hypsilophodon (Fig. 7c: s). The left basipterygoid process
of Parksosaurus, complete but split into two pieces, is short and resembles
that of Hypsilophodon (Figs. 7B, Cc: bpt. p.). Each process lies close to
the corresponding pterygoid (PT). Only the posterior part of the para-
sphenoid is visible (Fig. 2: PSP) and appears to be suboval in cross-
section, with a gently convex ventral surface and a sharp dorsal edge. The
greatest depth of the paraoccipital process is at midlength, tapering abruptly
distally (Fig. 2: par. p.). Several foramina are visible on the medial surface
of the opisthotic (OP) and prootic (PRO) bones (Fig. 2: cf. Fig. 8B).
The prootic part of the wall of the internal auditory meatus (im) is visible
with the fossa subarcuata (fs) as another prominent groove perpendicular
to the meatus. The sutural surface for the supraoccipital (so) on either
side of the fossa subarcuata and the relationship between the bones of this
region are as shown for Hypsilophodon (Fig. 8). That part of the prootic
of Parksosaurus visible in lateral view is similar in shape to that of Hypsilo-
phodon (Fig. 8A: PRO). The medial surface of the left orbitosphenoid of
Parksosaurus is probably represented by a thin sheet of bone in the vicinity
of the parasphenoid (Fig. 2: OSP, PSP), but it could as well be the lateral
surface of the right orbitosphenoid. The posteroventral ramus of the orbito-
sphenoid is complete, but the anteroventral ramus is not. Gilmore: (1909)
mentioned an ossified orbitosphenoid in Camptosaurus Marsh, but this
element has not been found in Dysalotosaurus (Janensch, 1955) or in
Hypsilophodon. A slender rod of bone (Fig. 2: O) probably lying against
the medial surface of the lachrymal in life as does a similar element in
Hypsilophodon (Fig. 7B: O) may represent part of the prefrontal; in the
prosauropod dinosaur Plateosaurus (AMNH 6810) a slender process of the
prefrontal is medial to the lachrymal.

The anterior end of the premaxilla of Parksosaurus is absent (Figs. 1, 5:
PMX), so it cannot be determined whether the premaxilla bore teeth. The
posterior process of the premaxilla is broad (Figs. 1, 5). The maxilla is
large, with a long suture with the nasal (Figs. 1, 5: MX, N). The small

‘SUOTIBIADIQGe IO} ¢ pue
Z soded 99§ “YOO]q ][NYs JO opls IYSTY “POX WOU MadiDM snanvsosysvg 7 ‘BIA

URMILA ALLAN ODOLLLEL LOL LLL OOP ITLOL ALL UDELL EOL DIL IAL LE TDLLU RAED OID LLL ALLL LT TT

Us tttae

de eadsidbiicceaiididdccéddsisdsdiiéiia

LM

"SUOTIVIADIQGe IO} ¢ pue 7 sased vag *(Z ‘BI YUM oIedwiod) poaowol zed
JSOWII}INO YIM YOO [[NYS JO opts 1YSTY “pO8 WOU MuasivM snanvsosydvg ¢ ‘BIA

COLLUBIOLOALLAL SOUL SYLLLLLALLLLIDLLLEDSOLOOPOLDLIBOSDSELI SSPE DOOR OPAL A net nepc8 00

till mn

‘ f Ty

_IMETRIC II CM 2i |

EE,

Fig. 4 Parksosaurus warreni ROM 804. Medial view of right dentary showing
teeth. An isolated maxillary tooth in lateral view is indicated by “mx”.

remnant of the antorbital fenestra bounded by the maxilla and lachrymal
(Figs. 1, 5: ant. f., MX, L) was identified as the lachrymal fenestra by
Parks (1926). The suture between the maxilla and the jugal is not visible,
as this region is damaged (Fig. 1: MX, J).

Most of the dorsal part of the nasal is represented only by an impression
(Fig. 1: N), but posteriorly the part covered by the prefrontal (PF) is
preserved. The prefrontal extends about 5 mm along the dorsal margin of
the orbit (Fig. 5: PF). A rugose area of bone occurs on the orbital corner
of the prefrontal (Fig. 1: sorb cf. Fig. 7A: SOB), probably for contact with
the supraorbital (Figs. 5, 6), a bone that is present in all other lower orni-
thopods known from adequate skull material. The right anterolateral part
of the parietal is preserved (Figs. 2, 3: P), so the frontal, about the same
length as the nasal (Fig. 5B: N), occupies most of the missing part of the
orbital rim as in other ornithopods rather than sharing it equally with the
parietal as Parks (1926) thought. The suture shown by Parks (1926, pl. 1)
between the lachrymal and the maxilla is a break in the block (Fig. 1:
photograph), but a horizontal suture is indicated (Figs. 1, 5a: L, MX).
Part of this suture is also visible on the right side, where the lachrymal
extends anterior to the antorbital fenestra (Fig. 2: ant. f.). This portion
of the lachrymal is covered by the descending flange of the prefrontal and
by the maxilla (Fig. 1: L, PF, MX). Dorsal to the right lachrymal is a
thin bone (Fig. 2: X) that is incomplete ventrally but broadens dorsally.
The bone is asymmetrical and, from its position as preserved, should be
part of either the prefrontal or the nasal. But it is more likely a palpebral
osteoderm or part of the medial bony sheet of the right maxilla that has
separated from the rest of the bone.

8

Fig. 5 Parksosaurus warreni ROM 804. Reconstruction of the skull. a. lateral
view. B. dorsal view. See pages 2 and 3 for abbreviations. Evidence from
ROM 804 indicated by——-——-—--— ; hypothetical reconstruction indicated by

Cee ee ee eee ee ee eo

The dorsal part of the postorbital (PO = postfrontal of Parks, 1926)
is round, and in life the upper temporal fenestra probably extended to the
level of the frontoparietal suture (Fig. 5B: F, P) as in Hypsilophodon (Fig.
6B). The dorsal process of the jugal (J) of Parksosaurus is long, and most
of the bone was covered by the postorbital (PO) (Fig. 1) as in Hypsilo-
phodon (Figs. 6A, 7A). Posterodorsally the jugal of Parksosaurus has a
long, broken edge, adjacent to which is an extensive area of smooth matrix
(Fig. 1: J) that represents a natural mould of the internal surface of a
bone. The bone was not the postorbital, for the ventral margin of that bone
is visible dorsal to the area of the natural mould. Radiographs revealed that
the region of smooth matrix was continuous with that part of the jugal
represented by bone. A dorsal expansion of the jugal is unusual (it tapers
in all other ornithopods) and presumably strengthened the postorbital and
dorsal temporal bars. The region between the jugal and the quadratojugal
is damaged, and the suture cannot be found. But I consider the piece of
bone next to the quadrate to be the quadratojugal rather than the jugal
(see Figs. 1, 5A: QJ, J). The dorsal and ventral processes amplify the
contact of the small quadratojugal with the quadrate.

COMMA Va
ayavave ae ees

Fig. 6 Hypsilophodon foxii. Reconstruction of skull based chiefly on BMNH
R197 and R2477. A. lateral view; B. dorsal view. See pages 2 and 3 for ab-
breviations.

The left quadrate (Fig. 1: Q) was rotated counterclockwise about its
long axis before preservation. The external ridge mentioned by Parks
(1926) is the anterior edge of this bone, and the faint depression men-
tioned by Parks sutured against the quadratojugal. The medial surface of
the pterygoid flange of the right quadrate is visible (Fig. 2: Q; the quadrate
has been rotated 180° about its vertical axis before preservation; see also
Figs. 7, 8D). Most of the central part of the pterygoid flange is slightly con-
cave and was probably covered by the pterygoid as in Hypsilophodon (Fig.
7A). Proximally, the shaft of the quadrate of Parksosaurus is triangular in
cross-section, with a sharp posterior edge that becomes less acute distally
and near the mandible becomes slightly convex. The quadrate, lacking its
ventral end, is 65 mm long. In my opinion only a few millimeters of its total
length are missing, and the length of 77 mm estimated by Parks (1926)
may have been too long. The part of the quadrate projecting ventral to
the quadratojugal was probably shorter and, judging from its position in

10

Hypsilophodon (Fig. 64) and Thescelosaurus (Sternberg, 1940, fig. 16),
the adjacent part of the lower jaw was probably deeper (Fig. 54) than that
shown in Parks’ restoration (1926, pl. 1).

Part of the right pterygoid of Parksosaurus is visible (Fig. 2: PT). The
distal edge of the alar process for the quadrate is broken and originally was
probably larger, as in Hypsilophodon (Figs. 7A, 7B, 8D). The alar process
for the palatine is anterior to a constricted region and perpendicular to the
quadrate process, also as in Hypsilophodon. The exact outline of the alar
process for the palatine cannot be determined, and the process may include
part of the palatine. Part of the ectopterygoid is preserved on the posterior
angle of the maxilla (Fig. 2: ECT, MX), but its sutural relationship with
other cranial elements is indeterminate.

Mandible

The preserved part of the left ramus undoubtedly includes the surangular
and angular in addition to the dentary, but sutures cannot be determined
(Figs. 1, 5A: SA, ANG, D). The rod-like bone next to the posteroventral
part of the jaw that Parks (1926) regarded as the angular is probably part
of the left ceratobranchial; part of the right ceratobranchial is also pre-
served (Fig. 1: CB). The well-developed convexity on the posterodorsal
margin of the jaw (Fig. 1) is also present on the jaw of Hypsilophodon.
In medial view the splenial of Parksosaurus (Fig. 3: SPL) is similar to that
of Hypsilophodon. Ventrally the right splenial (Fig. 1: SPL) becomes thick
and has a longitudinal depression that fits against the dentary. The small
piece of bone posterior to the splenial may represent part of the prearticular
(Bigs ls PA):

Teeth

The anterior part of the premaxilla of Parksosaurus is missing, and whether
the premaxilla bore teeth is unknown. Counts for maxillary and dentary
teeth for each side were about 18 and 20, respectively, for Parksosaurus
(Fig. 5A), as opposed to 10-11 and 13 in Hypsilophodon (Fig. 6A). The
lateral surface of the crown of a maxillary tooth of Parksosaurus (Fig. 4:
mx) has low, vertical ridges (Parks, 1926, fig. 1B), and the wear surfaces
of three dentary teeth are visible on the left side (Fig. 1). Each surface is
flat, and a layer of enamel occurs on the lateral and medial surfaces of the
crown. The wear surfaces of the maxillary teeth are not visible, but the
arrangement of the teeth is similar to that in Hypsilophodon, with maxillary
teeth curving medially and the dentary teeth curving laterally. The lateral
surface of a dentary tooth of Parksosaurus has about 10 vertical ridges
that become indistinct toward the middle of the crown (Fig. 1, second
tooth from posterior end of row; Parks, 1926, fig. 1A). The apical ridge of
the medial surface of the crown is not clearly differentiated from other
parallel ridges because the more lateral ridges are almost equally large. The
medial view (Fig. 4) of the dentary teeth of the right side reveals that tooth
replacement was alternate as in Hypsilophodon (Galton, in press, Br. Mus.)
and most lower tetrapods (Edmund, 1960).

|

VV RALOERR

ea are —
Tehkaotode bales

MmxX.C.

5cm

Fig. 7 Hypsilophodon foxii. Partly disarticulated skull of BMNH R2477, modified
from Galton (in press, Br. Mus.). A. lateral view; B. as in A but with outer
bones of left side removed; c. medial view of most of bones of right side
(compare with B). See pages 2 and 3 for abbreviations. Scale = 5 cm.

12

Fig. 8 Hypsilophodon foxii. Braincase, modified from Galton (in press, Br.
Mus.).A-c. lateral wall of braincase, composite reconstruction: A. latero-
ventral view; B. mediodorsal view with laterophenoid and supraoc-
cipital removed; c. mediodorsal view; D. occipital view of partly
disarticulated skull BMNH R2477 (compare with Fig. 7).

Comparisons

With large orbits, narrow elongate frontals, and a nasal-maxillary suture,
the skull of Parksosaurus (Fig. 5) resembles that of Hypsilophodon (Fig.
6). Frontals and nasals are about the same length (Fig. 5B) as in skulls
of Hypsilophodon (Fig. 68), Dryosaurus (CM 3392), and Dysalotosaurus
(Janensch, 1955), whereas in Camptosaurus (Gilmore, 1909) the frontals
are less than half the length of the nasals. The maximum width of the
frontals is about two-thirds of their length (Fig. 5B: F), as in Hypsilophodon
(Fig. 6B), but in Thescelosaurus (Sternberg, 1940), Dryosaurus, and
Camptosaurus these distances are approximately the same.

The posterior process of the premaxilla of Parksosaurus (Figs. 1, 5A)
is short, so that the maxilla contacts the nasal as in skulls of Hypsilophodon
(Fig. 6A) and Fabrosaurus Ginsburg (Thulborn, 1970); but the posterior
process of the premaxilla of Parksosaurus is uniquely deep. The maxilla
and nasal bones are separated by the elongate, posterior process of the

[3

premaxilla in the skulls of Dryosaurus (CM 3392), Dysalotosaurus, and
Camptosaurus. Variation in sutural contact may reflect the independent
walling in of the large antorbital fenestra of thecodonts in different evolu-
tionary lines of ornithopods. The antorbital fenestra is large in the Triassic
ornithopods Fabrosaurus (Thulborn, 1970) and Heterodontosaurus Cromp-
ton and Charig (1962; Galton, 1970, fig. 4c). The size of this fenestra is
variously reduced in skulls of Hypsilophodon (Fig. 6A), Dysalotosaurus,
Dryosaurus, Parksosaurus (Fig. 54), and Camptosaurus.

The dorsal end of the jugal of Parksosaurus was apparently expanded
anteroposteriorly (Fig. 5A: J) in a manner unique in any ornithopod. The
deep, lower temporal fenestra and low jugal-quadratojugal bar (Fig. 5a:
J, QJ) may be considered primitive, for this region is similar to that found
in thecodonts and in the Triassic ornithischians Fabrosaurus (Thulborn,
1970) and Heterodontosaurus (Galton, 1970). The lower temporal bar
is deeper in the above mentioned genera than in Camptosaurus, Dysaloto-
saurus, Dryosaurus, Laosaurus (Gilmore, 1925), and Hypsilophodon. The
large size of the quadratojugal of Hypsilophodon (Fig. 6A: QJ) is a
specialization in a genus that is primitive in most other characters.

Sternberg (1940) first noted that the crowns of the teeth of Parksosaurus
and Thescelosaurus were enamelled on both sides, as they also are in
Hypsilophodon. But in Parksosaurus patterns of ridging of the thickly
enamelled surface of the teeth (lateral for maxillary, medial for dentary
teeth), with numerous, low, rounded ridges, differ from those of any other
known ornithopod. These ridges are parallel, as in Hypsilophodon and
Dysalotosaurus, rather than forming two series of concentral curves, one
on each side of the apical ridge, as in Thescelosaurus (Sternberg, 1940,
figs. 4-8). In Thescelosaurus (NMC 8537) the thinly enamelled surfaces
of the crowns of teeth have an even more prominent series of ridges than
do those of the teeth of Parksosaurus. In Hypsilophodon they are barely
discernible.

Discussion

Edmund (1957) discussed the function of the special foramina (one per
tooth position) on the medial surface of the maxilla and dentary of hadro-
saurs and ceratopsians. He concluded that the foramina were for the admis-
sion of parts of the dental lamina or for the admission of young replacement
teeth produced by lamina. Edmund (1957) found no trace of these fora-
mina on the jaws of Parksosaurus, and as foramina are absent in all other
hypsilophodonts and iguanodonts, he suggested that the special foramina
are associated with the development of a high alveolar wall on the lingual
side. But special foramina are present in the skulls of Echinodon Owen
(Thulborn, 1970, p. 430), Hypsilophodon (Fig. 7B), Dryosaurus (CM
3392), Dysalotosaurus (Janensch, 1955, pl. x1, figs. 2b, 3b), Camptosaurus
(YPM 1886), and Iguanodon Mantell (Hooley, 1925, figs. 2, m; 4, 1).
Ventral margins of foramina are visible on the lower jaw of Thescelosaurus
(NMC 8537). In skulls of Hypsilophodon, Dysalotosaurus, Dryosaurus, and
Iguanodon these foramina occur on the medial surface, are arranged in a

14

regular line (one per tooth position), and are connected by a definite groove
that was suggested by Edmund (1957) to accommodate the dental lamina.
Foramina in these genera (except [guanodon) are oval rather than circular
in outline as in hadrosaurs and ceratopsians. The absence of these foramina
in lower ornithopods other than Fabrosaurus probably reflects the state of
preservation of the material. The region where these foramina should occur
in the dentaries of Parksosaurus is badly preserved (Figs. 3, 4). The appear-
ance of these jaws closely resembles that of one of Hypsilophodon (BMNH
R196), although a series of foramina is present in other specimens of
Hypsilophodon (Fig. 7c). As Edmund (1957) emphasized, the special
foramina undoubtedly provided access for replacement teeth to the base
of the series. The presence of these foramina in lower ornithopods represents
a preadaptation for the development of a battery of teeth consisting of verti-
cal series, for high alveolar walls may then be developed. The potential
for development of dental batteries was realized independently in two lines
of ornithischians, the hadrosaurs and ceratopsians, in which special foramina
are the best developed.

In my opinion Sternberg (1937) was justified in establishing the genus
Parksosaurus for Thescelosaurus warreni Parks. But the wear on the crown
of a maxillary or dentary tooth of Thescelosaurus as two surfaces set per-
pendicular to each other (Sternberg, 1940, fig. 6) is peculiar to specimen
NMC 8537. Wear on the two surfaces probably resulted from malocclusion,
with dentary teeth occluding between rather than directly against maxillary
teeth. Isolated teeth (UCMP 31815, 73086, 83000, 83001) from the Lance
Formation of Wyoming, undoubtedly from Thescelosaurus, have one wear
surface as in most ornithopods including Parksosaurus, and double wear
facets occur on the teeth of other lower ornithopods (Galton, 1973). In
addition to the differences between Thescelosaurus and Parksosaurus given
by Sternberg (1940, p. 492), the frontals of Thescelosaurus (Sternberg,
1940, fig. 1) are proportionally much broader than are those of Parkso-
saurus (Fig. 5B: F).

Thescelosaurus parallels the earlier Camptosaurus in its graviportal adap-
tations. I consider that the family Iguanodontidae is polyphyletic, represent-
ing a grade that includes all graviportal ornithopods except hadrosaurs and
pachycephalosaurs. Actual relationships cannot be established now, but I
suggest that at least four separate evolutionary lines of hypsilophodonts
became graviportal (Fig. 9). One line probably led to Camptosaurus that,
as Ostrom (1970) noted, has a unique skull that is low and broad with a
short, forward-inclined quadrate. A second line (lines?) led to the more
typical iguanodonts such as Tenontosaurus Ostrom (1970) and Iguanodon,
in which the skull is deep and narrow with an elongate, vertically-inclined
quadrate. Another line that retained many hypsilophodont features led to
Thescelosaurus; this locomotory transition may have occurred in the Lower
Cretaceous or else Thescelosaurus was a persistently conservative iguano-
dont. The pachycephalosaurids were probably also derived from the hypsilo-
phodontids (Galton, 1971c).

I characterize the family Hypsilophodontidae as follows: head small with

15

CRE TACEOUS

nN
> an as op ames

reese Lg

JURASSIC

;
‘
v
i
U

>

Ea FEI EAM Re) CAN oe CRAP RRR

Fig.9 Phylogeny of the Ornithopoda, from Galton (1972). Diagram to show
phylogenetic relationships and the nature of the fossil record of lower
ornithopods. Ages of genera are based on data in Charig (In Appleby
et al. 1967), and the stratigraphic distribution is by stages, the initials
of which are given in the third column.

Abbreviations: Classificationary units: ANKYL, Ankylosauria; CERAT, Cera-
topsia; FABR, Fabrosauridae; HADR, Hadrosauridae; HETER, He-
terodontosauridae; HYPSIL, Hypsilophodontidae; IGUAN, Iguanodon-
tidae; PACHY, Pachycephalosauridae,; PSITT, Psittacosauridae; STEG,
Stegosauria. Genera: C, Camptosaurus; CA, “Camptosaurus” leedsi,
D, Dryosaurus and Dysalotosaurus,; E, Echinodon; F, Fabrosaurus,; H,
Heterodontosaurus; HY, Hypsilophodon; 1, Iguanodon; L, Laosaurus;
LM, “Laosaurus” minimus; P, Pisanosaurus; PA, Pachycephalosaurus;
PK, Parksosaurus; PS, Psittacosaurus; S, Stegoceras; ST, Stenopelix;
T, Tenontosaurus; TH, Thescelosaurus; W, Wealden hypsilophodont
Galton (in press, Palaeont.); Y, Yaverlandia Galton (1971c). Actual
fossil record for lower ornithopods indicated by - no fossil record
indicated by —---—-—-—- . Genera in the same vertical line are closely
related, and postulated relationships are indicated by ............. :

16

short snout, large orbits, no caniniform teeth or rostral beak, maxillary and
dentary teeth inset (longitudinal recess to maxilla, massive dentary) with
randomly formed wear surfaces not all in the same plane, cursorial with
distal part of hind limb elongate. To this family I refer the following genera
that are represented by adequate material: Pisanosaurus (Late Triassic);
Laosaurus, Dryosaurus, Dysalotosaurus (Late Jurassic); Hypsilophodon
(Early Cretaceous); and Parksosaurus (Late Cretaceous) (Galton, 1972).
A comparison of the skull of cM 3392 from Utah with the holotype (YPM
1876) of Dryosaurus altus Marsh (1894) confirms the provisional identi-
fication by Gilmore (1925). Dryosaurus (CM 3392), however, does not
have a supraorbital bar as was described by Gilmore (1925, fig. 3). The
supraorbital gently tapers to a point lateral to the postorbital, and there is
no sutural contact with this bone. The skull (CM 3392) of Dryosaurus altus
Marsh (1894) is similar to that of Dysalotosaurus lettow-vorbecki Pompeckj
(1920, p. 120) from Tanzania as described by Janensch (1955), and these
two taxa may eventually prove to be congeneric, as also may Laosaurus celer
Marsh (1878) and Hypsilophodon foxii Huxley (1869).

The skull of Fabrosaurus from the Upper Triassic of Lesotho is so primi-
tive that it is probably similar to the archetypal ornithischian from which
all known ornithischians can be derived (Thulborn, 1970; Galton, 1972,
1973). The skull has a flat maxilla, a slender dentary, and marginally
situated maxillary and dentary teeth. Because of these uniquely primitive
features (Galton, 1973), I have removed Fabrosaurus from the Hypsilopho-
dontidae and established a separate family, the Fabrosauridae, for this
genus and Echinodon (Galton, 1972). Affinities of Tatisaurus Simmons
(1965) from the Upper Triassic of China and Nanosaurus Marsh (Huene
and Lull, 1908) from the Upper Jurassic of North America are uncertain;
it is not even certain that these genera are hypsilophodonts. Romer (1966,
p. 370) listed Macelognathus Marsh from the Upper Jurassic of North
America as a hypsilophodont, but Ostrom (1971) showed that it is prob-
ably a crocodile.

All hypsilophodonts were probably derived from an ancestor similar to
Pisanosaurus Casamiquela (1967) from the Upper Triassic of Argentina.
Dryosaurus—Dysalotosaurus, Laosaurus—Hypsilophodon, and Parksosaurus
probably represent three separate lineages (Fig. 9). Femora of the
Dryosaurus—Dysalotosaurus type are also found in England in: the Upper
Jurassic (“Camptosaurus” leedsi; Lydekker, 1889) and in the Lower Cre-
taceous (Wealden hypsilophodont) (Galton, in press, Palaeont.). Parkso-
saurus resembles the Laosaurus—Hypsilophodon line more closely than it
does the Dryosaurus—Dysalotosaurus line in the elongate nature of the
frontals and in the form of the femur. But the shallow lower temporal arch
and extensive nasal-maxillary contact in the skull of Parksosaurus represent
the retention of conservative features that are lost in Hypsilophodon.

Apart from Parksosaurus, remains of only one other taxon of the family
Hypsilophodontidae have been reported from rocks of Late Cretaceous age.
A partial hind limb (originally at the University of British Columbia, now
NMC 9483) was described by Gilmore (1924b) as Laosaurus minimus. The

17

specimen is from Southern Alberta near Burmis and was from the Blairmore
Formation, the age of which was thought to be Lower Cretaceous. Allan and
Rutherford (1932) concluded that the rocks concerned represented the
so-called Belly River Formation, more properly the lower part of the Allison
Formation (Upper Cretaceous). Russell (1949) compared the structure of
Laosaurus minimus with that of other lower ornithopods and concluded that
its closest affinities were with Hypsilophodon from the Lower Cretaceous
of England. A fifth metatarsal is apparently absent in L. minimus (NMC
9483) but present in Hypsilophodon and Parksosaurus (Parks, 1926). The
proximal end of the first metatarsal of L. minimus is slender and greatly
reduced (Gilmore, 1924b, pl. 1, fig. 1). In Hypsilophodon, as in Parkso-
saurus (Parks, 1926, figs. 15, 16), the proximal part of the first metatarsal
is robust. The cleft separating the lesser and greater trochanters of the femur
of L. minimus is shallow as in Hypsilophodon and Parksosaurus. Facets on
the articular surfaces of the ungual phalanges of Laosaurus minimus are
symmetrical as in Hypsilophodon, rather than markedly asymmetrical as in
Parksosaurus (Parks, 1926, fig. 18). Bones of the hind limb of specimen
NMC 9483 are slender and obviously from a cursorial animal; as Russell
(1949) noted, the bones may represent a new genus that is at the moment
undefinable. Nonetheless, the specimen appears to be closely related to
Parksosaurus and Hypsilophodon, and, although it could represent a new
species of Hypsilophodon, it is probably best referred to as the Burmis
hypsitophodontid (?Laosaurus minimus).

Acknowledgments

Most work was done at Yale University, and I thank Mr. A. Coleman for
taking the photographs, Rosanne Rowen for labelling figures, Nancy Wallace
(University of Bridgeport) for typing the manuscript, and the following
for their comments on the paper: Drs. A. G. Edmund, C. McGowan, and
L. S. Russell of the Royal Ontario Museum; D. A. Russell of the National
Museum of Canada; and J. H. Ostrom of Yale University. I thank the fol-
lowing for the loan of material: Drs. C. Black of the Carnegie Museum;
FE. H. Colbert of the American Museum of Natural History; A. G. Edmund
of the Royal Ontario Museum; J. H. Ostrom of Yale University; D. A.
Russell of the National Museum of Canada; and S. P. Welles of the Univer-
sity of California at Berkeley. Radiographs of the skulls of Parksosaurus were
taken by Dr. A. G. Edmund of the Royal Ontario Museum, and use of the
radiographic facilities of the Laboratory of Analytical Systematics in the
Royal Ontario Museum, established through a grant from the National Re-
search Council of Canada to the Department of Zoology of the University of
Toronto, is gratefully acknowledged. The manuscript was completed while
I was in receipt of a Faculty Research Grant from the University of Bridge-
port.

18

Literature Cited

ALLAN, J. A. AND R. L. RUTHERFORD

1932 The stratigraphical horizon of Laosaurus minimus Gilmore. Am. J.
Sci., ser. 5, vol. 24, no. 141, pp. 225-227.

APPLEBY, R.M., A.J. CHARIG, C.B. COX, K. A. KERMACK AND L. B. H. TARLO

1967 Reptilia. In Harland, W. B. et al., eds. The fossil record. A symposium
with documentation jointly sponsored by the Geological Society of
London and the Palaeontological Association. London, Geological
Society of London, pp. 695-731.

CASAMIQUELA, R. M.

1967 Un nuevo dinosaurio ornitisquio triasico (Pisanosaurus mertii;
Ornithopoda) de la formacién Ischigualasto, Argentina. Ameghi-
niana, vol. 5, no. 2, pp. 47—64. [In Spanish, English summary].

CROMPTON, A. W. AND A. J. CHARIG

1962 A new ornithischian from the Upper Triassic of South Africa.

Nature, vol. 196, no. 4859, pp. 1074-1077.
EDMUND, A. G.

1a On the special foramina in the jaws of many ornithischian dino-
saurs. Life Sci. Contr., R. Ont. Mus., no. 48, pp. 1-14.
1960 Tooth replacement phenomena in the lower vertebrates. Life Sci.

Contr., R. Ont. Mus., no. 52, pp. 1-90.
GALTON, P. M.

1970 Ornithischian dinosaurs and the origin of birds. Evolution, vol. 24,
no. 2, pp. 448-462.

1971a Hypsilophodon, the cursorial non-arboreal dinosaur. Nature, vol.
231, no. 5299, pp. 159-161.

1971b The mode of life of Hypsilophodon, the supposedly arboreal or-
nithopod dinosaur. Lethaia, vol. 4, no. 4, pp. 453-465.

1971c A primitive dome-headed dinosaur (Ornithischia: Pachycephalo-
sauridae) from the Lower Cretaceous of England and the function
of the dome of pachycephalosaurids. J. Paleont., vol. 45, no. 1,

pp. 40-47.

1972 Classification and evolution of ornithopod dinosaurs. Nature, vol.
239, no. 5373, pp. 464-466.

1973 On the cheeks of ornithischian dinosaurs. Lethaia, vol. 6, no. 1,
pp. 67-89.

In press The ornithischian dinosaur Hypsilophodon from the Wealden of the
Isle of Wight. Bull. Br. Mus. Nat. Hist., Geol.

In press A new hypsilophodontid from the Lower Cretaceous of England
with comments on ornithopod dinosaurs. Palaeontology.

GILMORE, C. W.

1909 Osteology of the Jurassic reptile Camptosaurus with a revision of
the species of the genus, and description of two new species. Proc.
U.S. Natn. Mus., vol. 36, no. 1666, pp. 197-332.

1913 A new dinosaur from the Lance Formation of Wyoming. Smithson.
Misc. Collns., vol. 61, no. 5, pp. 1-5.
1915 Osteology of Thescelosaurus, an orthopodous dinosaur from the

Lance Formation of Wyoming. Proc. U.S. Nat. Mus., vol. 49, no.
2127, pp. 591-616.

19

1924a On Tréodon validus, an orthopodous dinosaur from the Belly River
Cretaceous of Alberta, Canada. Bull. Dep. Geol. Univ. Alberta,
now il pp: 143.

1924b A new species of Laosaurus, an ornithischian dinosaur from the
Cretaceous of Alberta. Trans. R. Soc. Can., ser. 3, vol. 18, sec. 4,
pp. 1-6.

1925 Osteology of ornithopodous dinosaurs from the Dinosaur National
Monument, Utah. Mem. Carneg. Mus., vol. 10, no. 4, pp. 385-410.

HOOLEY, R. W.
11925 On the skeleton of Iguanodon atherfieldensis sp. nov., from the
Wealden Shales of Atherfield (Isle of Wight). Q. J. Geol. Soc. Lond.,
vol, él, pt. i ne, 321, pp. tor

HUENE, F. V. AND R. S. LULL
1908 Neubescheibung des Originals von Nanosaurus agilis Marsh. Neues
Jb. Miner. Geol. Palaont., band 1, pp. 134-144.

HUXLEY, T. H.

1869 On Hypsilophodon, a new genus of Dinosauria. Proc. Geol. Soc.,
no. 204, pp. 3—4 [Abstract].

JANENSCH, W.
1955 Der Ornithopod Dysalotosaurus der Tendaguruschichten. Palaeonto-
graphica Suppl., band 7, tiel 3, lief 3, pp. 105-176.

LYDEKKER, R.
1889 On the remains and affinities of five genera of Mesozoic reptiles.
Q. J. Geol. Soc. Lond., vol. 45, pt. 1, no. 177, pp. 41-59.

MARSH, O. C.
1878 Notice of new dinosaurian reptiles. Am. J. Sci., ser. 3, vol. 15, no. 87,
pp. 241-244.
1894 The typical Ornithopoda of the American Jurassic. Am. J. Sci., ser. 3,
vol. 48, no. 283, pp. 85-90.

OSTROM, J. H.
1970 Stratigraphy and paleontology of the Cloverly Formation (Lower
Cretaceous) of the Bighorn Basin area, Wyoming and Montana.
Bull. Peabody Mus. Nat. Hist., no. 35, pp. 1-234.

SFA On the systematic position of Macelognathus vagans. Postilla, no.
153; pp. t—10:
PARKS, W.A.
1926 Thescelosaurus warreni, a new species of orthopodous dinosaur from

the Edmonton Formation of Alberta. Univ. Toronto Stud. Geol. Ser.,
no: 21, pp, 1-42.

POMPECKJ, J. F.

1920 Das angebliche Vorkommen und Wandern des Parietal foramens
bei dinosauriern. Sber. Ges. Naturf. Freunde Berl. 1920, pp.
109-129"

20

ROMER, A. S.
1966 Vertebrate paleontology (3d ed.). Chicago, University of Chicago
Press. 468 pp.

RUSSELL, L. S.
1949 The relationships of the Alberta Cretaceous dinosaur “Laosaurus”
minimus Gilmore. J. Paleont., vol. 23, no. 5, pp. 518-520.

SIMMONS, D.J.
1965 The non-therapsid reptiles of the Lufeng Basin, Yunnan, China.
Fieldiana, Geol., vol. 15, no. 1, pp. 1-93.

STERNBERG, C.M.

1937 A classification of Thescelosaurus, with a description of a new
species. Proc. Geol. Soc. Am. 1936, pp. 375 [Abstract].
1940 Thescelosaurus edmontonensis, n. sp., and the classification of the

Hypsilophodontidae. J. Paleont., vol. 4, no. 5, pp. 481-494.
THULBORN, R. A.

1970 The skull of Fabrosaurus australis, a Triassic ornithischian dinosaur.
Palaeontology, vol. 13, pt. 3, pp. 414-432.

rs |

a he nig Je. ¢ ae ~t Vv lps alien 4G fl j <1 a
.! elietid th eke Yt, pe renkerh
| < pak hee aoe ;

-
cg ix? 5 seasies Ws es = 1) eachigt au nn gt ‘Se = A

. ot Piller fay Craak, BR: Bice .. tod Tee Aiet wm
nk} Geen. siozopint.) -srTnilA. edt $6 epidseiteee ST

. Pu } ; ‘ 7 ne :
Lhe Mes EA ote? OR RA eS: nities . peppy! hs? wayFaGuy >

~