Content uploaded by David R Schwimmer
All content in this area was uploaded by David R Schwimmer on Apr 21, 2015
Content may be subject to copyright.
A NEW GENUS AND SPECIES OF TYRANNOSAUROID FROM THE LATE CRETACEOUS
(MIDDLE CAMPANIAN) DEMOPOLIS FORMATION OF ALABAMA
THOMAS D. CARR
*, THOMAS E. WILLIAMSON
, AND DAVID R. SCHWIMMER
Department of Palaeobiology, Royal Ontario Museum, 100 Queen’s Park, Toronto, Ontario, M5S 2C6, Canada;
New Mexico Museum of Natural History and Science, 1801 Mountain Road, NW, Albuquerque, New Mexico, 87104–1375;
Department of Chemistry and Geology, Columbus State University, 4225 University Avenue, Columbus, Georgia 31907–5645
ABSTRACT—The discovery of a new genus and species of tyrannosauroid from the Demopolis Formation (middle
Campanian) of Alabama increases the known diversity of the clade, although it does not elucidate the place of initial
dispersal. This subadult tyrannosauroid is the most complete non-avian theropod collected and described from the
Cretaceous of eastern North America. In contrast to tyrannosaurids, the new taxon possesses several plesiomorphic
characters, including lacrimals that lack a distinct peaked cornual process, and a dorsoventrally shallow horizontal ramus
of the maxilla. Autapomorphies include a wide jugal process of the ectopterygoid, a caudal pneumatic foramen of the
palatine that pierces the rostral half of the vomeropterygoid process of the bone, an articular surface for the lacrimal on
the palatine that is distally positioned on the dorsolateral process, and pedal unguals that have a distinct proximodorsal
lip over the articular surface. Cladistic analysis indicates the new taxon is a basal tyrannosauroid and its presence in
eastern North America suggests that the recent common ancestor of Tyrannosauridae probably evolved following the
transgression of the Western Interior Seaway. Cladistic analysis indicates that Dryptosaurus aquilunguis is also a basal
tyrannosauroid but is less derived than the new genus.
Eastern North America
is a ‘dark continent’ of Late Creta-
ceous dinosaur paleontology, even though some of the first
named North American dinosaurs were collected from the
northeastern United States (e.g., Cope, 1866; Leidy, 1858). The
ages of Late Cretaceous vertebrate-bearing sedimentary deposits
exposed in eastern North America are similar to those of the
west, extending from the Santonian to the late Maastrichtian
(Baird and Horner, 1979; Baird, 1986). However, eastern expo-
sures are limited and the deposits tend to be marine (Baird and
Horner, 1977; Baird and Galton, 1981; Baird, 1986; Schwimmer,
1986; King et al., 1988; Schwimmer et al., 1993) leading to bias
against articulated remains of terrestrial animals (Schwimmer,
1997). The eastern (or “Appalachian”) theropod fossil record
consists of isolated bones, bone fragments, and teeth (Gilmore,
1920, 1921; Langston, 1960; Baird and Horner, 1979; Baird, 1986;
Schwimmer and Best, 1989; Schwimmer et al., 1993). These are
rarely diagnostic below the family level, and based on this rec-
ord, Tyrannosauridae and Ornithomimidae are the only thero-
pod clades identified with reasonable certainty (Baird, 1986;
Baird and Horner, 1979; Schwimmer et al., 1993; Schwimmer,
1986, 1997). Indeed, the identification of some eastern Ornitho-
mimidae has been questioned (Gallagher, 1995). The most com-
plete large theropod described from Late Cretaceous beds of
eastern North America is Dryptosaurus aquilunguis, which con-
sists of a partial skeleton (Carpenter et al., 1997). In addition,
Campanian “carnosaur” (Baird and Horner, 1977) fossils were
reported from the Woodbury Formation of the New Jersey-
Better understanding of the fossil record of eastern North
American dinosaurs would be invaluable in resolving many bio-
geographic and phylogenetic issues. For example, elucidating the
phylogenetic relationship of eastern North American dinosaurs
with those across the Western Interior Seaway (WIS) can indi-
cate if there was dispersal between land masses following seaway
transgression during Albian-Cenomanian times (Russell, 1995).
This information would further elucidate the faunal shift be-
tween the Early and Late Cretaceous and the relationship be-
tween the faunas of the principal regions of Laurasia. The
Campo-Maastrichtian fossil records of Asia and western North
America dominate the current knowledge and conception of tyr-
annosaurid evolutionary history (e.g., Molnar et al., 1990; Holtz,
1994; Carpenter, 1992, 1997). Recently published time-calibrated
phylogenies indicate Tyrannosauroidea first evolved in the Late
Jurassic (Sereno, 1997) or the Early Cretaceous (Holtz, 2000),
suggesting an unknown but extensive history and high diversity.
The Late Cretaceous fossil record of eastern North America is
woefully incomplete, but recent finds (Chinnery et al., 1998)
indicate a dinosaur fauna comparable to that of the rest of
In 1982, a theropod was found in Montgomery County, south-
eastern Alabama (Fig. 1), at the edge of a road cut (King et al.,
1988) in the marine sediments of the Demopolis Formation
(Fig. 2). The 20.4 square meter quarry was excavated first by
personnel from Auburn University and later by workers of the
Red Mountain Museum (RMM) during September 1984 and in
the next two summers. King et al. (1988) designated the locality
as the Turnipseed Dinosaur Site, in recognition of the landown-
ers. The specimen (RMM 6670) was curated into the collections
of the Red Mountain Museum in 1987, which has since been
relocated to the McWane Center in Birmingham, Alabama. In
1989 the identification of RMM 6670 as Albertosaurus by James
Lamb, then of the Red Mountain Museum, was reported in a
Birmingham newspaper. Carpenter (1992) made mention of the
specimen, then thought to be from the Mooreville Chalk, in his
review of Tyrannosauridae and accepted its informal referral to
Albertosaurus. The specimen was noted by Baird (1989) and was
mentioned in two publications (Schwimmer and Best, 1989;
Schwimmer et al., 1993) in comparison with a metatarsal col-
lected from the Blufftown Formation of Georgia. RMM 6670
consists of portions of the skull, isolated teeth, pelvis, hind limbs,
caudal vertebrae, and ribs (Fig. 3). Thus, RMM 6670 is currently
the most complete, diagnostic Late Cretaceous theropod from
eastern North America and it is also among the few eastern
dinosaurs diagnostic below the suprageneric level.
Herein we describe the skeleton of a new genus of basal tyr-
annosauroid, approximately contemporaneous with the late
Campanian tyrannosaurids of western North America (Fig. 2).
* Present address: Department of Biology, Carthage College, 2001 Al-
ford Park Drive, Kenosha, Wisconsin 53140-1994, firstname.lastname@example.org
Journal of Vertebrate Paleontology 25(1):119–143, March 2005
© 2005 by the Society of Vertebrate Paleontology
Although an assessment of dinosaurian diversity in eastern
North America is still unfeasible, the presence of the new dino-
saur in Alabama presents a more complete view of tyrannosau-
roid evolution and historical biogeography than was possible
We provisionally recognize a node-based Tyrannosauridae
consisting of a dichotomy composed of the lineages Albertosau-
rus and the Daspletosaurus +Tyrannosaurus clade. We provi-
sionally recognize a stem-based Tyrannosauroidea because basal
ingroup relationships of the clade are poorly resolved. These
concepts differ from those of Sereno (1998) and Holtz (2001)
because we consider Aublysodon,Stygivenator, and Nanotyran-
nus to be invalid taxa (Carr, 1999; Carr and Williamson, 2000;
Carr and Williamson, submitted). Thus, as used herein, the term
“tyrannosaurid”refers to members of Tyrannosauridae; “tyran-
nosauroid”refers to members of the more inclusive clade includ-
ing Tyrannosauridae, the new genus, and Dryptosaurus (see Phy-
Osteological terminology is after Baumel (1979), Baumel and
Witmer (1993), Weishampel et al. (1990), Welles and Long
(1974), and Witmer (1997, 2001).
Institutional Abbreviations—AMNH, American Museum of
Natural History, New York; ANSP, The Academy of Natural
Sciences, Philadelphia; BHI, Black Hills Institute of Geological
Research, Inc., Hill City; CMN, Canadian Museum of Nature,
Aylmer; CMNH, Cleveland Museum of Natural History, Cleve-
land; FMNH, Field Museum, Chicago; DMNH, Denver Museum
of Natural History, Denver; MOR, Museum of the Rockies, Boz-
eman; NJSM, New Jersey State Museum, Trenton; NMMNH,
New Mexico Museum of Natural History and Science, Albuquer-
que; PIN, Palaeontological Institute, Moscow; RMM, Red
Mountain Museum (collection curated at the McWane Center),
Birmingham; ROM, Royal Ontario Museum, Toronto; TMP,
Royal Tyrrell Museum of Palaeontology, Drumheller.
THEROPODA Marsh, 1881
TETANURAE Gauthier, 1986
COELUROSAURIA von Huene, 1914
TYRANNOSAUROIDEA (Osborn, 1905) Bonaparte
et al., 1990
APPALACHIOSAURUS, gen. nov.
Type Species—Appalachiosaurus montgomeriensis, sp. nov.
Etymology—Appalachio-, for the occurrence of the specimen
in eastern North America during the Late Cretaceous; -sauros,
Greek, meaning lizard.
Diagnosis—As for the type and only known species.
APPALACHIOSAURUS MONTGOMERIENSIS, sp. nov.
Diagnosis—Tyrannosauroid possessing the following autapo-
morphies: wide jugal process of ectopterygoid; caudal pneumatic
recess of palatine situated rostral to caudal margin of vomerop-
terygoid process; articular surface for lacrimal of palatine situ-
ated distally; and prominent lip extending over dorsal margin of
articular surface of pedal unguals.
Holotype—Incomplete skull and skeleton, RMM 6670
(McWane Center, Birmingham, Alabama).
Locality and Horizon—The Turnipseed Dinosaur site was re-
portedly 9.7 meters above the basal conglomerate of the De-
mopolis Formation, in a marl and calcareous clay stratum (King
et al., 1988). The site is 1.77 km north of US Highway 82 at the
Downing Crossroads; SE1/4, NE1/4, NE1/4, Sec. 35, T14N,
R20E; in Montgomery County, Alabama. The age is determined
as ∼78 million years (Ma) based on stratigraphic position and
nannofossil biostratigraphy (see Chronostratigraphy).
Etymology—Named for Montgomery County, Alabama,
where the type locality is located; -ensis, Latin, meaning from.
The articular surface for the maxilla is a slot-and-ridge contact
(Fig. 4A, C) that typifies juvenile and subadult specimens of
tyrannosaurids (Carr, 1999). In dorsal view, the surface is rugose
but is smooth over the frontal process (Fig. 4B). The nasals lack
FIGURE 2. Chronostratigraphic correlation chart showing the position
(open circle) of the type locality of Appalachiosaurus montgomeriensis
gen. et sp. nov. (RMM 6670).
FIGURE 1. Type locality of Appalachiosaurus montgomeriensis gen. et
sp. nov. (RMM 6670). Locality indicated by open circle.
JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 25, NO. 1, 2005120
the bony papillae that project from the dorsal surface in tyran-
nosaurids (e.g., Albertosaurus libratus). In Appalachiosaurus
there is a row of six low bumps along the dorsal midline that ends
at the level of the maxillary processes (Figs. 4B, C, 5). Midline
bumps are present in Daspletosaurus (Currie, pers. comm., 2001)
and to a lesser extent in Albertosaurus libratus (ROM 1247) and
Tyrannosaurus bataar (PIN 551–3); six tall bony cusps are pres-
ent in Alioramus (Kurzanov, 1976). Rostrally, the bones taper to
the midline along the premaxillary processes that are separated
from the maxillary processes by the narial fenestrae; the maxil-
lary processes are broken off and missing in RMM 6670. The flat
abutting surfaces of the premaxillary processes are separated by
the open internasal suture, which closes caudal to the level of the
naris (Fig. 4A, B). Distally, the medial surfaces of the processes
deviate dorsolaterally from each other, creating a slot that re-
ceived the nasal processes of the premaxillae (Fig. 4B). The ven-
tral half of the lateral surface of the frontal process is flattened
by the articular surface for the lacrimal that forms a ventrally
open slot as it extends rostrally into the ventral surface of the
bone (Fig. 4A). The ventral surface of the nasal is pierced by a
row of elongate foramina that begins beneath the frontal pro-
cess. The internasal suture is closed but persists as a shallow
groove flanked by a narrow and flat ridge that indicates the
attachment of the cartilaginous internasal septum. The internasal
suture opens at the greatest width of the nasals, separating the
premaxillary processes at the level of the external naris (Fig. 4A).
Taxonomic Variation—The proportions of the nasal of Appa-
lachiosaurus are comparable to those of specimens of Alberto-
saurus libratus (see Table 1, on-line at http://www.vertpaleo.org/
jvp/). In adult Daspletosaurus (CMN 8506), the nasals are wider
through the ‘waist’of the bone. The row of bumps along the
dorsal midline is present in Appalachiosaurus,Albertosaurus li-
bratus,Daspletosaurus, and Tyrannosaurus bataar, and is elabo-
rated into tall cusps in Alioramus.
Many features of the maxilla (Carr, 1999) indicate that RMM
6670 may be a juvenile or subadult: the bone is narrow, the
lateral surface of the interfenestral strut is flat, the antorbital
fenestra is elongate, a ridge encircles the antorbital fossa, the
maxillary fenestra is circular, the rostral margin of the antorbital
fossa is not strut-like, the fossae of the maxillary antrum are
shallow, the sulcus of the ventral jugal process does not breach
the ventral margin of the bone, the articular surface for the
palatine is slot-like, and the dental alveoli are narrow (Fig. 6).
There are 15 dental alveoli. The first alveolus is half the length
of the second, indicating the first maxillary tooth was small and
likely incisiform, as in tyrannosaurids. The rostral alveoli are
elongate and abruptly reduce in length at the eleventh alveolus.
The seventh alveolus is the longest of the tooth row. In Alber-
tosaurus libratus and Daspletosaurus the longest socket is at the
fourth or fifth position. Two rows of foramina pierce the lateral
surface of the bone (Figs. 5, 6A). The maxillary antorbital fossa
extends to the level of the rostral end of the maxillary flange
(Figs. 5, 6A). “Maxillary flange”denotes the convex region of
the dorsal margin situated above the rostral half of the antorbital
fossa (see Figs. 5, 6A). The immutability of this feature in tyr-
annosauroids makes it a suitable point of reference.
A promaxillary fenestra was present in RMM 6670, but crush-
ing has obscured it in lateral view (Fig. 6A). As in Albertosaurus
libratus, the maxillary fenestra is small and does not approach
the rostrodorsal margin of the antorbital fossa as in Daspleto-
saurus and Tyrannosaurus (Figs. 5, 6A, 7). As in tyrannosaurids,
except A.sarcophagus, the interfenestral strut of Appalachiosau-
rus is wide (Figs. 5, 6A, 7). In Appalachiosaurus and Albertosau-
rus, the fenestra is separated from the rostral margin of the
maxillary antorbital fossa by a wide apron of bone (Figs. 5, 6A,
7). The maxillary fenestra of Appalachiosaurus is in a more dor-
sal position than in tyrannosaurids, 42 mm above the ventral
margin of the antorbital fossa. Although the possibility exists
that this state may be an artifact of crushing, the uncrushed right
bone is currently unprepared. This distance is only 19 mm in
similar-sized tyrannosaurids (e.g., ROM 683, ROM 1247).
The base of the ventral jugal process is shallower in Appala-
chiosaurus (20 mm deep) than in similar-sized specimens of Al-
bertosaurus libratus (ROM 683: 34 mm; ROM 1247: 32 mm). In
FIGURE 3. Skeletal reconstruction of Appalachiosaurus montgomeriensis gen. et sp. nov. (RMM 6670) in right lateral view. Gray indicates missing
CARR ET AL.—NEW TYRANNOSAUROID FROM ALABAMA 121
FIGURE 4. Skull bones of Appalachiosaurus montgomeriensis gen. et sp. nov. (RMM 6670). Nasals in ventral (A), dorsal (B), and left lateral
(C) views; left lacrimal in lateral (D), medial (E), and dorsal (F) views; and right jugal (?) in lateral view.
JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 25, NO. 1, 2005122
FIGURE 5. Skull reconstruction of Appalachiosaurus montgomeriensis gen. et sp. nov. (RMM 6670) in right lateral view. Light gray indicates
missing bones; dark gray indicates matrix or damage.
CARR ET AL.—NEW TYRANNOSAUROID FROM ALABAMA 123
medial view, the dorsal margin of the maxillary antrum is hori-
zontal as in A.libratus. This margin is rostroventrally oriented in
Daspletosaurus (Carr, 1999). The postantral strut delimits the
medial margin of the caudal fenestra of the antrum in which the
concave caudal margin of the strut is oriented rostroventrally at
a steep angle (Fig. 6B). In the uncrushed right maxilla, the pala-
tal process is wide and the articular surface is shallower than in
the left bone. The interdental plates contribute to the medial
walls of the dental alveoli beneath the medial alveolar process
Taxonomic Variation—The shallow horizontal ramus (Figs. 5,
6A, 7; Table 2, online at http://www.vertpaleo.org/jvp/) and the
dorsally situated maxillary fenestra sets Appalachiosaurus apart
from tyrannosaurids. In Appalachiosaurus,Albertosaurus libra-
tus, and A.sarcophagus the maxillary fenestra is not close to the
rostral margin of the antorbital fossa (Figs. 5, 6A, 7). In Dasple-
tosaurus and subadult Tyrannosaurus the fenestra and rostral
margin are separated by a narrow rim (Fig. 7) but in adult Tyr-
annosaurus, the fenestra extends rostromedial to the fossa
The absence of a supraorbital ridge, presence of a dished sur-
face around the lacrimal recess, merged antorbital fossa and lat-
eral surface, and concave rostral margin of the rostroventral
lamina indicates that RMM 6670 may be a juvenile or small adult
(Figs. 4D-F, 5, 8; Carr, 1999). In lateral view, the lacrimal is a
T-shaped bone. The lateral surface of the ramus is depressed
around the lacrimal recess and the dorsolateral margin bears a
rugose and sinuous ridge (Figs. 4D, F; 5; 8). The ridge does not
form a cornual process as in Albertosaurus and Daspletosaurus,
but it has a convex profile (Figs. 4D, E, 5, 8). The apex of the
ridge is rostral to the lacrimal recess (Figs. 4D, E, 5; 8), as in
juvenile and subadult Albertosaurus specimens and some speci-
mens of Daspletosaurus (e.g., CMN 11594, TMP 94.143.1). The
lacrimal pneumatic recess is small (length: 23.3 mm). The acces-
sory recess is 32.0 mm ahead of the lacrimal recess (Figs. 4D;
In medial view, a portion of the articular surface for the max-
illa forms a groove in the ventral margin of the rostral ramus
(Fig. 4E). A portion of the flat articular surface for the nasal is
preserved above that of the maxilla (Fig. 4E). The orbitonasal
ridge, a ridge that separates the orbit and paranasal cavity, is
vertically oriented and curves forward distally (Fig. 4E). Proxi-
mally, the strut appears to be aliform with a sharp edge but it is
incompletely prepared. The rostroventral tip of the rostroventral
lamina extends caudoventrally and is rostromedially offset for
contact with the palatine.
FIGURE 6. Left maxilla of Appalachiosaurus montgomeriensis gen. et sp. nov. (RMM 6670) in lateral (A) and medial (B) views.
JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 25, NO. 1, 2005124
Taxonomic Variation—In Appalachiosaurus and Tyranno-
saurus rex, a peaked cornual process is absent; however, Appa-
lachiosaurus retains the fundamental morphology of the dorso-
lateral ridge present in Albertosaurus,Daspletosaurus, and
T.bataar (Figs. 4D, 5, 8). In Appalachiosaurus the apex of the
ridge, like that of the cornual process in Albertosaurus and some
specimens of Daspletosaurus, is rostral to the level of the ventral
ramus (Figs. 4D, 5, 8). In adult A.libratus,T.bataar, and most
FIGURE 7. Taxonomic characters of tyrannosauroid maxillae; parenthetical polarity changes indicate results obtained from the cladistic analysis.
Light gray indicates plaster.
CARR ET AL.—NEW TYRANNOSAUROID FROM ALABAMA 125
specimens of Daspletosaurus, the apex is above the ramus
(Fig. 8). In Appalachiosaurus,Daspletosaurus, and Tyrannosau-
rus, the lacrimal recess is small, in contrast to all referred species
of Albertosaurus where the recess is large (Figs. 4D, 5, 8). In
Tyrannosaurus, the accessory recess is located distally on the
rostral ramus (Fig. 8); in Appalachiosaurus and the other tyran-
nosaurids, the recess is proximal (Figs. 4D, 5, 8). Finally, in Ap-
palachiosaurus and Albertosaurus, the rostral ramus is not in-
FIGURE 8. Taxonomic characters of tyrannosauroid lacrimals; parenthetical polarity changes indicate results obtained from the cladistic analysis.
Light gray indicates plaster or damage.
JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 25, NO. 1, 2005126
flated as it is in Daspletosaurus and Tyrannosaurus (Figs. 4D-F,
A 107.8 mm-long fragment that may represent the right jugal
includes the maxillary ramus and the rostral half of the jugal
recess (Fig. 4G). A fragment of the right lacrimal appears to be
in place with the jugal (Fig. 4G). The rostral end of the maxillary
process tapers to a point (Fig. 4G). The external surface is con-
cave and flares outward along its ventral contact with the max-
illa. The dorsolateral surface of the maxillary ramus is flat and
slopes lateroventrally to the coarse lateral surface of the bone.
Distally, the jugal antorbital fossa is convex in cross-section. In
tyrannosaurids, the antorbital fossa excavates the maxillary ra-
mus; this fundamental difference suggests that the fragment con-
sidered here may be of another bone. The bone is convex ventral
to the jugal recess and bears a series of low ridges that trend
caudoventrally to rostrodorsally. The lateral surface extends
ventrolaterally to the ventral margin where it is convex in cross-
section, and becomes bladelike along the contact with the max-
illa. The secondary fossa of the jugal recess is deep such that it is
bounded rostrally by a low strut (Fig. 4G). The rostral margin of
this region of the bone extends caudodorsally along a steep
Taxonomic Variation—Appalachiosaurus differs from tyran-
nosaurids and a new basal genus from New Mexico (NMMNH
P-27469) with regard to the convex surface of its antorbital fossa.
In tyrannosaurids, the lateral surface of the maxillary ramus is
flat ventral to the pneumatic recess and not convex as in Appa-
lachiosaurus. Finally, the maxillary ramus rostral to the pneu-
matic foramen in Appalachiosaurus is dorsoventrally deep as in
Daspletosaurus and Tyrannosaurus.
The incomplete pterygoid (Fig. 9A, B) is comparable with
tyrannosaurids. The articular surface for the quadrate depresses
the lateral surface of the quadrate ramus, which is braced ros-
troventrally by a thick strut (Fig. 9A). The quadrate ramus as
preserved is 77.3 mm long. The articular surface for the epiptery-
goid is indistinct. In medial view, the rostral margin of the quad-
rate ramus is coarsened by a muscle attachment surface. This is
also in tyrannosaurids. In medial view, the dorsoventrally flat
postarticular process extends caudomedially beneath the quad-
rate ramus (Fig. 9B). The process is 25.8 mm wide. In lateral
view, the palatine ramus extends rostroventrally (Fig. 9A) and
has the form of a hemicylinder with the concave surface facing
dorsolaterally. The palatine ramus is 35.2 mm wide and 67.0 mm
deep. The ramus bears a thick lateral strut that extends from the
rostral edge of the quadrate ramus.
In ventral view, a single pneumatic recess pierces the hook-
shaped bone (Fig. 9C). This is also present in Albertosaurus sar-
cophagus and Tyrannosaurus (Fig. 10). The recess extends past
the midlength of the bone toward the articular surface for the
pterygoid (Figs. 9C, 10). The recess in RMM 6670 is 44 mm long
in contrast to the 21-millimeter recess in the larger specimen
ROM 1247 (Albertosaurus libratus; see Table 3, on-line at http://
www.vertpaleo.org/jvp/, for additional measurements). The bone
separating the pneumatic foramen from the medial edge of the
bone is flat as in Albertosaurus (Figs. 9C, 10). This surface is a
convex lip in Daspletosaurus and Tyrannosaurus (Fig. 10).
As in Albertosaurus, the articular surface for the jugal faces
dorsolaterally (Fig. 9D). This surface faces ventrolaterally in
Daspletosaurus (Fig. 10). The bone is hollow and the pneumatic
sinus ends at the base of the jugal process. The base of the caudal
process is deeper (22.8 mm) in RMM 6670 than in similar sized
A.libratus (18 mm). In ventral view, longitudinal-and fine criss-
crossing ridges that are typical of tyrannosaurids reinforce the
flat articular surface for the pterygoid.
Taxonomic Variation—In contrast to that of other tyranno-
sauroids, the pterygoid process of the ectopterygoid of Appala-
chiosaurus is rostrocaudally short (Figs. 9C, D, 10), but this may
be due to damage. As in Albertosaurus, the surface medial to the
pneumatic recess is flat but the recess is situated a greater dis-
tance away from the medial edge of the bone (Figs. 9C, 10).
There are two recesses in A.libratus and Daspletosaurus (Fig.
10). In dorsal or ventral view, Appalachiosaurus and Tyranno-
saurus possess an elongate jugal process such that it is straight-
ened and not strongly backswept (Figs. 9C, D, 10). In Alberto-
saurus and Daspletosaurus, the jugal process is short and curves
strongly caudally (Fig. 10). In Appalachiosaurus the ectoptery-
goid is not inflated. In Albertosaurus, the base of the jugal pro-
cess is inflated; in Daspletosaurus and Tyrannosaurus, the entire
bone is inflated (Figs. 9C, D, 10). In Tyrannosaurus and some
specimens of Daspletosaurus, the caudal surface of the jugal pro-
cess is pierced by a foramen; that of Appalachiosaurus and Al-
bertosaurus is imperforate (Figs. 9C, D, 10).
The palatine of RMM 6670 is not inflated and the choanal
process is shallow, as in subadult Albertosaurus libratus (Carr,
1999). In lateral view, the palatine is 171.6 mm long between the
tips of the maxillary and dorsolateral processes and 182.6 mm
between the tips of the maxillary and pterygoid processes. The
vomeropterygoid process tilts rostrally along a 50-degree angle
and expands distally into the rostrally directed choanal process
(Figs. 5, 9E, H, 11). The bone is 81.9 mm deep through this
region and the incomplete choanal process was over 71.8 mm
long. As in tyrannosaurids, a shallow sulcus crosses the base of
the vomeropterygoid process (Fig. 5). In medial view, a large
foramen pierces the base of the dorsal process at its caudal mar-
gin (Fig. 9H). This is also present in Daspletosaurus (e.g., CMN
In lateral view, the maxillary process is excavated by the dor-
soventrally deep articular surface for the maxilla (Fig. 9E). The
base of the process is 38.3 mm deep. The articular surface is flat
and vertically oriented, eliminating the lip of bone that extends
lateral to the pneumatic foramen (Fig. 11) in tyrannosaurids,
which is absent in one specimen of Albertosaurus sarcophagus
(TMP 81.10.1; Fig. 11) but is developed into a prominent lateral
process in T.rex (e.g., BHI 3033, CMNH 7541, MOR 555;
Fig. 11). Rostrally, the articular surface for the maxilla in Appa-
lachiosaurus reaches the dorsal margin of the bone (Figs. 9E, 11),
which is also present in one Albertosaurus sarcophagus specimen
(TMP 86.64.1). The maxillary process is pierced by a small
(length: 15.7 mm) palatine recess that is rostral in position, ahead
of the caudal margin of the vomeropterygoid process (Figs. 9E,
11). The maxillary process is excavated by a 42.1 mm-long pneu-
matic fossa (Figs. 9E, 11). The dorsolateral process extends cau-
dolaterally from the body of the bone. Distally, its lateral surface
is excavated by the articular surface for the lacrimal. A groove in
the dorsolateral margin secured the contact (Fig. 9E); this deli-
cate region is usually missing in tyrannosaurid fossils; thus, the
polarity of the character is equivocal. Unlike tyrannosaurids, the
articular surface for the lacrimal is positioned distally on the
dorsolateral process (Figs. 9E, 11).
Taxonomic Variation—In subadult and adult Albertosaurus
and juvenile Daspletosaurus (e.g., TMP 94.143.1), the choanal
process is dorsoventrally deep and has a sinuous rostroventral
margin for articulation with the vomer; this margin is not sinuous
CARR ET AL.—NEW TYRANNOSAUROID FROM ALABAMA 127
FIGURE 9. Palate of Appalachiosaurus montgomeriensis gen. et sp. nov. (RMM 6670). Right pterygoid in lateral (A) and medial (B) views; right
ectopterygoid in ventral (C) and dorsal (D) views; right palatine in lateral (E), dorsal (F), ventral (G), and medial (H) views.
JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 25, NO. 1, 2005128
in Appalachiosaurus (Fig. 11). However, the tip of the process is
missing. As in T.rex and Daspletosaurus, the vomeropterygoid
process of Appalachiosaurus is tall (Figs. 9E, H, 11); in Alberto-
saurus, the process is short in lateral view (Fig. 11). In Appala-
chiosaurus and Albertosaurus, the palatine recess is situated ros-
tral to the caudal margin of the vomeropterygoid process (Fig.
11); in Daspletosaurus and Tyrannosaurus it approaches or is
beneath the caudal margin of the process (Fig. 11).
FIGURE 10. Taxonomic characters of tyrannosauroid ectopterygoids; parenthetical polarity changes indicate results obtained from the cladistic
CARR ET AL.—NEW TYRANNOSAUROID FROM ALABAMA 129
In lateral view, the caudal process of the angular is deep, the
ventral margin is convex, and the external mandibular fenestra is
expressed as a concave margin in the rostrodorsal corner of the
bone (Figs. 5, 12A, B). The lateral surface beneath the external
mandibular fenestra is depressed by the articular surface for the
dentary (Figs. 5, 12A). Unlike T.bataar, the caudal margin of the
surface is not braced by a ridge (Hurum and Currie, 2000). In
FIGURE 11. Taxonomic characters of tyrannosauroid palatines; parenthetical polarity changes indicate results obtained from the cladistic analysis.
Gray indicates plaster.
JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 25, NO. 1, 2005130
medial view, the ventral margin of the bone bears a narrow shelf
that supported the prearticular (Fig. 12B).
The splenial of Appalachiosaurus is an elongate bone pierced
by a large rostral mylohyoid foramen and bearing a prominent
dorsal process (Fig. 12C, D; see Table 4, on-line at http://www.
vertpaleo.org/jvp/). Unlike tyrannosaurids, the rostral margin is
not notched for reception of the lingual bar of the dentary, but
this may be due to damage (Fig. 12C, D). Unlike that of tyran-
nosaurids, the rostrodorsal margin is blade-like and not flattened
by the rostral articular surface for the dentary. The lateroventral
articular surface for the dentary diminishes before reaching the
caudal tip of the bone (Fig. 12C). As in juvenile tyrannosaurids,
the articular surface is smooth, unlike the ridged contact of
adults. In lateral view, the floor of the rostral mylohyoid foramen
is flat and extends medially along the caudal half of the foramen.
This flat surface extends rostrally in Albertosaurus and Tyran-
nosaurus. The biconcave “waist”of the dorsal and ventral mar-
gins of the caudal process is rostral in position relative to tyran-
nosaurids (Fig. 12C, D). The splenial of Appalachiosaurus has a
shallower rostral mylohyoid foramen than tyrannosaurids (Fig.
12C, D; Table 4).
Only the lateral surface of the bone has been prepared and is
described. The shallow dentary of RMM 6670 (Figs. 5,12E) is
typical of juvenile and subadult tyrannosaurids (see Table 5,
on-line at http://www.vertpaleo.org/jvp/). The preserved portion
of the bone is 538 mm long. The dorsal margin of the dentary
follows the typical tyrannosaurid pattern: convex rostrally adja-
cent to the first several alveoli, then concave to the caudal mar-
gin of the bone (Figs. 5, 12E). The bone deepens caudally at
midlength (see Table 5). The rostral margin is convex and ex-
tends caudoventrally, meeting the sinuous ventral margin at the
level of the fourth alveolus, forming a ‘chin’at the rostroventral
margin of the bone (Figs. 5, 12E). The external surface is pierced
by two main rows of foramina (Figs. 5, 12E) and numerous
minute foramina pierce the alveolar border of the bone.
At least ten isolated teeth are preserved with RMM 6670.
These include one premaxillary tooth (Fig. 13) and nine lateral
teeth. The carinae of all teeth in Appalachiosaurus are denticu-
late, and as in tyrannosaurids (Carr and Williamson, 2000), the
number of denticles per five mm intervals (denticle density) in-
creases toward the base of the crown (see Tables 6, 7, on-line at
FIGURE 12. Mandibular bones of Appalachiosaurus montgomeriensis gen. et sp. nov. (RMM 6670). Right angular in lateral (A) and medial
(B) views; right splenial in lateral (C) and medial (D) views; and right dentary in lateral (E) view.
CARR ET AL.—NEW TYRANNOSAUROID FROM ALABAMA 131
http://www.vertpaleo.org/jvp/). The denticle densities of RMM
6670 are comparable to those of similar sized teeth from western
North America (Carr and Williamson, 2000). Very similar teeth
have been collected in the late-early Campanian Blufftown
Formation in nearby Georgia (DRS pers. obs.). The basal crown
lengths of some teeth are elongate relative to similar-sized
A.libratus (e.g., ROM 1247). As in Albertosaurus and juvenile of
a new genus (NMMNH P-25049), the mesial carina of the lateral
teeth shifts lingually toward the crown base. A lingual ridge is
present on the premaxillary tooth (Fig. 13).
Eight caudal vertebrae of RMM 6670 (Fig. 14) are preserved.
A proximal caudal vertebra, still in its field jacket, has a trace-
able neurocentral suture. A study of neurocentral suture closure,
as done for crocodylians (sensu Brochu, 1996), has not been
conducted for tyrannosaurids. Although the ontogenetic se-
quence of suture closure is unknown for the clade, open sutures
suggest RMM 6670 was not an adult at death.
In lateral view, two proximal caudal vertebrae are coössified,
possibly the result of injury (Fig. 14A-D). The neurocentral su-
ture is traceable on the left side along the proximal half of the
proximal bone. The cranial zygapophysis is dorsoventrally deep
and is situated above the level of the articular surface of the
caudal zygapophysis. The proximal centrum is 95 mm long on the
left and 93 mm long on the right. The distal centrum is 105 long
mm on the left and 88 mm long on the right. The transverse
processes are craniocaudally shorter than transversely wide and
they expand abruptly at their tips (Fig. 14A, B). The proximal
end of the intervening hemal arch is in articulation between the
vertebrae (Fig. 14B-D).
A largely complete proximal caudal vertebra has a maximum
height of 205.6 mm (Fig. 10E-H). The spinous process is 103.2
mm high caudally. The neurocentral suture is traceable on both
sides. Traceable sutures in proximal caudal vertebrae are present
in a juvenile T.bataar (e.g., PIN 552–2).
In lateral view, an incomplete distal caudal vertebra has a
deep, elongate, and horizontally oriented cranial zygapophysis
(Fig. 14K-O). The articular surface is proximal in location and
the zygapophysis extends ahead of the articular surface. The
caudal zygapophysis is short and extends caudodorsally (Fig.
14M, N). In lateral view, the ventral margin of the centrum is
concave and transverse processes are absent (Fig. 14M, N). The
caudal surface of the centrum is 49.9 mm wide and 34.4 mm deep.
The centrum is 106.0 mm long on the complete right side. Finally,
there are fragments of two other caudal vertebrae: half of a distal
caudal centrum (Fig. 14I, J), and an isolated neural arch of a
proximal caudal vertebra (Fig. 14P-R).
As in tyrannosaurids, the shaft of the pubis is bowed caudally
such that it is convex caudally and concave rostrally in lateral
view (Fig. 15A, B). The width at the midshaft is 38.3 mm and the
length is 54.2 mm (width/length ratio: 0.71). The circumference
above and below the pubic apron is 157.3 mm and 58.5 mm,
respectively. A transversely compressed ala forms the dorsolat-
eral margin of the rostral half of the pubic boot.
As in tyrannosaurids, the ischium of Appalachiosaurus is elon-
gate with a rod-like and curved distal ramus, possesses a convex
oval scar that interrupts the dorsal edge of the bone, and has a
triangular obturator process that diminishes at the midlength of
the rod-like distal process (Fig. 15C, D; Table 8, on-line at http://
www.vertpaleo.org/jvp/). The iliac and pubic peduncles are miss-
ing in RMM 6670. An oval muscle scar, presumably for the third
division of the internal tibial flexor (Carrano and Hutchinson,
2002), interrupts the dorsal margin of the bone, is coarse and is
bounded laterally by a prominent lip (Fig. 15D). This is present
in other tyrannosauroids, including Dryptosaurus. The obturator
process lacks the rostrally-directed distal projection (Fig. 15C,
D) present in some tyrannosaurid specimens (e.g., PIN 552–2).
As in Albertosaurus (e.g., AMNH 5664, ROM 1247) and juvenile
Tyrannosaurus (e.g., PIN 552–2), the ventral margin of the bone
is concave (Fig. 15C, D). The margin is straight in adult Alber-
tosaurus (e.g., CMN 2120; Lambe, 1917) and Tyrannosaurus
(e.g., FMNH PR2081).
The maximum length of the femur is 786 mm and the circum-
ference is 260 mm (Table 9, on-line at http://www.vertpaleo.org/
jvp/). Using the equation of Anderson et al. (1985), a mass of 623
kilograms (1,374 pounds) is estimated for RMM 6670. The ven-
tral margin of the neck of the femoral head extends dorsomedi-
ally, but the femoral head is not elevated (Fig. 16A). A femoral
head that extends perpendicular to the long axis of the bone is
also found in Shanshanosaurus, a juvenile tyrannosaurid (Currie
and Dong, 2001). The femoral head and the greater trochanter of
Appalachiosaurus are confluent as in tyrannosaurids (Fig. 16A;
Lambe, 1917; Maleev, 1974). In lateral view, the lesser trochanter
is prominent with a convex rostral margin (Fig. 16C), with its
lateral surface deeply excavated by a depression that is textured
by rostrodorsally extending ridges. A wide low ridge delimits the
rostral margin of the fossa. The fourth trochanter extends from
the caudomedial margin of the bone. The trochanter is triangular
(Fig. 16B, C) and has a concave and rugose medial surface.
The tibia of RMM 6670 (Fig. 16E-L) is more circular in cross
section at midlength (ratio: 0.85) than in Albertosaurus libratus
(0.70 in ROM 1247). The tibia is 97% of the length of the femur
and the ratio of combined length of the tibioastragalus and femur
is 1.02, proportions consistent with other tyrannosaurids (Currie,
2000; see Table 10, on-line at http://www.vertpaleo.org/jvp/). As
in tyrannosaurids (cf. Lambe, 1917), the distal end of the tibia is
wider than the astragalus. The articular surface for the astragalus
occupies 23% of shaft length on the right and 21% on the left. In
cranial view, a vertical ridge is present on the articular surface for
the astragalus. When the bones are joined, the ridge fits into a
groove in the astragalus. This is also in tyrannosaurids and Alec-
trosaurus. Molnar et al. (1990) noted that tyrannosaurids and
Allosaurus possess a proximal “cranial projection”(1990:197),
similar to the lateral cranial process of birds, which extends ros-
trally adjacent to the articular surface for the fibula; the process
FIGURE 13. Right premaxillary tooth of Appalachiosaurus montgom-
eriensis gen. et sp. nov. (RMM 6670) in lingual (A), mesial (B), and distal
JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 25, NO. 1, 2005132
is present in Appalachiosaurus (Fig. 16I, K) but is absent in
Dryptosaurus. In lateral view, the lateral proximal condyle for
the femur is dorsoventrally deep; the condyle is shallow in ty-
In RMM 6670, the tubercle for the insertion of the iliofibularis
muscle (Carrano and Hutchinson, 2002) displays the bipartite
morphology also present in tyrannosaurids (Mader and Bradley,
1989; Fig. 17A-C). As in other theropods, the medial surface of
the proximal end of the bone is excavated by a deep concavity of
which the rostral margin is widely separated from the rostral
margin of the bone proximally (Fig. 17C). The cranial surface
distal to the bipartite scar is wide and flat (Fig. 17B). In lateral
view, the cranial and caudal margins of the bone dilate proxi-
mally toward the articular surface for the femur (Fig. 17A, C).
The cranial dilation is absent or not pronounced in tyrannosau-
rids (comparative measurements are in Table 11, on-line at
FIGURE 14. Caudal vertebrae of Appalachiosaurus montgomeriensis gen. et sp. nov. (RMM 6670) in lateral (C,D,G,H,I,J,M,N), dorsal (A,
K,P), ventral (B,L,Q), proximal (E,O,R), and distal (F) views. See text for explanation.
CARR ET AL.—NEW TYRANNOSAUROID FROM ALABAMA 133
The astragalus of Appalachiosaurus is typical of that of other
tyrannosauroids (Fig. 18A-F). The distal condyles are approxi-
mately the same depth (Fig.18A; Table 12, on-line at http://www.
vertpaleo.org/jvp/). The ratio of the width of the lateral condyle
to that of the medial condyle is higher (0.69) in RMM 6670 than
in Albertosaurus libratus (0.63 in ROM 1247). A deep ligament
pit excavates the base of the ascending process next to the ven-
trolateral buttress (sensu Welles and Long, 1974:197) (Fig. 18A).
In contrast, the buttress is weakly developed in Dryptosaurus.
The distal end of the buttress is prominent in Alectrosaurus and
adult T.rex (e.g., DMNH 2827, MOR 555). Medial to the liga-
ment pit is a shallow rugose concavity as in other tyrannosau-
roids (Fig. 18A). The dorsal margin of the condylar surface is
delimited by the horizontal groove (Fig.18A). The lateral margin
of the ascending process ascends at a steep dorsomedial angle
(Fig. 18A, B). The medial margin of the ascending process is
convex and extends dorsolaterally to the peaked dorsal margin
of the process (Fig. 18A, B). The rostrolateral surface of the
ascending process is flattened by the articular surface for the
fibula, which extends caudal to the ventrolateral buttress (Fig.
In caudal view, a vertical slot incises the caudal surface of the
ascending process that receives the ridge from the tibia (Fig.
18B). In lateral view, the articular surface for the calcaneum
includes a dorsal socket that receives a peg from the calcaneum
(Fig.18D). This is also found in T.rex (DMNH 2827). The socket
is absent in Dryptosaurus, in which the astragalus overlaps the
dorsomedial edge of the calcaneum with a lip of bone. Along the
rostrodorsal margin of the articular surface for the calcaneum in
Appalachiosaurus, a series of stout pegs fit into shallow pits in
the calcaneum (Fig. 18A, D). In rostral view, a tongue of bone
from the calcaneum overlaps the ventral half of the external
surface of the astragalus (Fig. 18A). Also, a lenticular concavity
scours the caudal margin of the articular surface for the cal-
caneum that is opposed by a rugose patch on the calcaneum
In lateral view, the calcaneum has convex rostral and ventral
margins that contribute to the crurotarsal joint, a concave cau-
dodorsal margin to receive the fibula, and a concave caudal mar-
gin that overlaps the caudolateral surface of the tibia (Fig. 18G).
The lateral surface of the bone is excavated by the depression of
the lateral epicondyle below mid height (Fig. 18G). The maxi-
mum width of the bone is 24% that of the astragalus (Table 13,
on-line at http://www.vertpaleo.org/jvp/).
The crurotarsal joint surface widens over the rostrolateral sur-
face of the bone and is widest rostral to the depression of the
lateral epicondyle (Fig. 18G, H). The joint surface reaches the
caudoventral margin of the bone and continues onto the caudo-
ventral corner of the astragalus (Fig. 18G, J). The caudoventral
corner of the calcaneum extends caudoventrally from the joint
surface as a low ‘heel’(Fig. 18G, J). The rostromedial surface of
the articular surface for the fibula is visible in lateral view but the
rest is blocked from view by the dorsolateral margin of the bone
(Fig. 18G). In medial view, the calcaneum bears the articular
surfaces for the astragalus and tibia that cover the rostral and
caudal halves of the bone, respectively (Fig. 18I). The articular
surface for the astragalus parallels the long axis of the bone and
that for the tibia extends caudolaterally behind it (Fig. 18I).
The articular surface for the astragalus is complex, with a stout
peg that extends medially from the caudodorsal margin of the
articular surface (Fig. 18I); the peg fits into a concavity in the
astragalus (Fig. 18D). The ventral surface of the peg is smooth,
possibly indicating the former presence of a cartilaginous cap.
The rostrodorsal surface of the peg is scoured by a rugose con-
cavity that opposes a pit in the astragalus (Fig. 18I, D). This gap,
formed between the two bones, possibly indicates the attach-
ment of a ligament that secured the distal end of the fibula.
The rostral half of the articular surface for the astragalus is
excavated by deep pits that receive low pegs from the astragalus
(Fig. 18I). Ventrally, the contact extends medially such that the
calcaneum overlaps the rostrolateral margin of the astragalus
(Fig. 18A, I, K). Caudal to this contact, the articular surface for
the astragalus is convex and coarse (Fig. 18I). The articular sur-
face for the tibia is dorsoventrally concave and coarse (Fig. 18I. J).
In dorsal view, the articular surface for the fibula is rostrocau-
dally elongate with its long axis oriented rostromedially to cau-
dolaterally (Fig. 18L). It is divided into a wide, convex rostral
surface and a narrow, concave caudal surface, separated by an
oblique groove (Fig. 18L). In lateral view, the rostral half is
convex and extends rostrodorsally at a low angle (∼40°) whereas
the caudal half is horizontal (Fig. 18G). In Albertosaurus libratus
(ROM 1247) the suture with the fibula is horizontal in lateral
view. In Appalachiosaurus, the lateral margin of the caudal half
is elevated, creating a ridge that prevented lateral displacement
of the fibula (Fig. 18G). The angulation in the articular surface
would have prevented rostrocaudal motion of the fibula on the
otherwise smooth surface. In dorsal view, the caudal half of the
articular surface for the fibula is saddle shaped (Fig. 18L). The
opposing surface of the fibula reflects this morphology in that the
FIGURE 15. Pelvic bones of Appalachiosaurus montgomeriensis gen.
et sp. nov. (RMM 6670). Right pubis in lateral (A) and medial (B) views,
left ischium in medial (C) and lateral (D) views.
JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 25, NO. 1, 2005134
FIGURE 16. Hind limb bones of Appalachiosaurus montgomeriensis gen. et sp. nov. (RMM 6670). Right femur in caudal (A), medial (B), lateral
(C), and distal (D) views; left tibia in cranial (E), caudal (F), proximal (I), and distal (J) views; right tibia in cranial (G), caudal (H), proximal (K),
and distal (L) views).
CARR ET AL.—NEW TYRANNOSAUROID FROM ALABAMA 135
wide rostral half is concave and the medial edge is elaborated
into a low ridge. The contact is loose, suggesting that a ligament
secured the contact.
The crurotarsal condylar surface intrudes onto the caudoven-
tral surface of the articular surface for the tibia, extending the
joint onto the caudomedial surface of the bone (Fig. 18J). When
joined with the astragalus, the calcaneum forms the lateral third
of the articular surface of metatarsal II. In rostral view, the as-
tragalocalcaneal contact is sinuous. In dorsal view, the calca-
neum overlaps the rostrolateral surface of the tibia and com-
pletes the shelf formed by the astragalus that underlaps the tibia.
In RMM 6670, metatarsals I and V are missing (Fig. 3). The
metatarsus of Appalachiosaurus displays the arctometatarsalian
condition and the metatarsals conform to the description given
by Lambe (1917) for Albertosaurus libratus (Fig. 19A-R). The
third metatarsal is 64% the length of the femur (Tables 9, 15,
on-line at http://www.vertpaleo.org/jvp/), falling within the range
given by Russell (1970) and Currie (2000) for tyrannosaurids.
The length proportions between the metatarsals are consistent
with those in tyrannosaurids (Lambe, 1917; Parks, 1928; Tables
14–16, on-line at http://www.vertpaleo.org/jvp/).
The pedal phalanges are comparable with those of other tyr-
annosaurids (Tables 14–16, on-line at http://www.vertpaleo.org/
jvp/). In contrast, the caudodorsal margins of the pedal unguals
are elongate, and are concave in lateral view cranial to the proxi-
mal articular surface, creating a ‘lipped’appearance (Fig. 20A-J)
similar to the manual unguals of oviraptorosaurians (Currie,
1990). The concavity is situated cranial to the articular surface
such that the dorsal margin extends cranially along a convex arc
from the articular surface into the concavity. This condition is
absent in western North American tyrannosaurids and this re-
gion in the ungual of digit III of Alectrosaurus is elongate but flat
in lateral view.
Taxonomic Variation—The flexor tubercles of the unguals of
Appalachiosaurus are low and not developed as in Alectrosaurus
(Mader and Bradley, 1989); the ‘lipped’articular surface dis-
tinguishes Appalachiosaurus from tyrannosaurids and Alectro-
The possibility that RMM 6670 might be referable to Drypto-
saurus aquilunguis or Alectrosaurus olseni (thought to be a basal
form; Currie and Eberth, 1993; Holtz, 2001) motivated first-hand
study of their type specimens by TDC.
The phylogenetic position of the type specimen (ANSP 9995)
is tentative (Carpenter et al., 1997). This issue and others require
comment. Carpenter et al. (1997) suggested the glenoid region of
the right surangular might the prearticular or angular. As in
tyrannosaurids, the caudal surangular foramen is large and a
bipartite muscle scar on the rostral margin of the fibula is present
FIGURE 17. Left fibula of Appalachiosaurus montgomeriensis gen. et sp. nov. (RMM 6670) in lateral (A), cranial (B), medial (C), caudal (D),
proximal (E), and distal (F) views.
JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 25, NO. 1, 2005136
FIGURE 18. Proximal tarsus of Appalachiosaurus montgomeriensis gen. et sp. nov. (RMM 6670). Left astragalus in cranial (A), caudal (B), medial
(C), lateral (D), ventral (E) and dorsal (F) views. Right calcaneum in lateral (G), cranial (H), and ventral (K) views; left calcaneum in medial (I),
caudal (J), and dorsal (L) views.
CARR ET AL.—NEW TYRANNOSAUROID FROM ALABAMA 137
FIGURE 19. Right metatarsus of Appalachiosaurus montgomeriensis gen. et sp. nov. (RMM 6670). Metatarsal IV in dorsal (A), proximal (D), distal
(G), plantar (J), lateral (M), and medial (P) views; metatarsal III in dorsal (B), proximal (E), distal (H), plantar (L), lateral (N), and medial (Q)
views; metatarsal II in dorsal (C), proximal (F), distal (I), plantar (L), lateral (O), and medial (R) views.
JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 25, NO. 1, 2005138
in D.aquilunguis, a feature thought to be diagnostic for tyran-
nosaurids (Mader and Bradley, 1989). Also, the associated
fourth metatarsal is comparable to that of tyrannosaurids: the
long axis of the proximal end is transverse, is notched medially
for metatarsal III, the caudolateral margin travels rostrolaterally,
and, the distal end is not transversely compressed as in Allosau-
rus. We therefore consider D.aquilunguis a member of Tyran-
nosauroidea. Unlike Appalachiosaurus,D.aquilunguis lacks a
socket in the dorsal margin of the astragalus for the calcaneum
and the cranial process of the tibia is absent. RMM 6670 is there-
fore not referable to D.aquilunguis.
The lectotype specimen (AMNH 6554) of Alectrosaurus olseni
is different from RMM 6670. In medial view, the medial distal
condyle of the femur in A.olseni has a spike-like dorsal process;
in distal view, the intercondylar groove is shallow cranially. In
proximal view, the rostral margin of the lateral condyle (⳱cra-
nial process) of the tibia does not extend cranially and in lateral
view the condyle slopes at a low angle cranioventrally. In cranial
view, the calcaneum is narrow. The distal articular surfaces of
metatarsals II and IV and most phalanges are elevated on ped-
icles and in cranial view the proximal margin of the distal articu-
lar surface of metatarsal III is horizontal. In plantar view, the
distal articular surface of metatarsal III is hyperextended proxi-
mally (Mader and Bradley, 1989). These characters are absent in
RMM 6670, precluding its referral to Alectrosaurus.
We coded 31 morphological characters (Appendix 1, on-line at
http://www.vertpaleo.org/jvp/) for six ingroup taxa and 12 out-
group taxa to construct a data matrix (Appendix 2, on-line at
http://www.vertpaleo.org/jvp/) in MacClade (Maddison and
Maddison, 1992) to specify the phylogenetic position of Appala-
chiosaurus. Twenty-four characters were limited to the bones
preserved in RMM 6670 except for one (po. 12), which was
included to support the monophyly of Albertosaurus. The out-
group topology was constrained following Holtz (2000) and the
ingroup was collapsed into a polytomy. The ingroup topology
was first obtained excluding Appalachiosaurus; when included,
Appalachiosaurus did not alter the ingroup relationships. Analy-
sis of the data matrix in the program Phylogenetic Analysis Us-
ing Parsimony (PAUP) version 3.1.1 (Swofford, 1993) resulted in
a single minimum length tree with a length of 102 steps, a con-
sistency index (C.I.) of 0.54, a Homoplasy Index (H.I.) of 0.62,
and a Retention Index (R.I.) of 0.59. The analysis was run under
accelerated (ACCTRAN) and delayed (DELTRAN) optimiza-
tions (see Appendix 3, on-line at http://www.vertpaleo.org/jvp/).
The basal position of Appalachiosaurus is supported by three
unambiguous character states: the apex of the dorsolateral ridge
of the lacrimal is rostral to the level of the ventral ramus, the
lacrimal pneumatic recess is small, and the accessory lacrimal
recess is proximal in position.
Three unambiguous character states distinguish Appalachio-
saurus (see Diagnosis; Appendix 3). All other North American
forms evaluated (aside from Dryptosaurus), and the Asian Ty-
rannosaurus bataar, comprise Tyrannosauridae. The clade is
composed of two lineages, Albertosaurus and Daspletosaurus +
Tyrannosaurus. Three unambiguous character states unite Ty-
rannosauridae: a maxillary fenestra that is close to the ventral
margin of the antorbital fossa, possession of a dorsoventrally
deep horizontal ramus of the maxilla, and the presence of an
apex on the cornual process of the lacrimal (Figs. 7, 8).
Dryptosaurus-Dryptosaurus aquilunguis was included in the
data matrix (Appendix 2); in the constraint tree, D.aquilunguis
was included in the polytomy comprised of tyrannosauroids.
Three additional characters were required to place D.aquilun-
guis phylogenetically (Appendix 1). A single most parsimonious
tree (Fig. 21) was obtained of 106 steps in length with a C.I. of
0.55 in which D.aquilunguis is the basalmost tyrannosauroid
(Fig. 21). The large caudal surangular foramen unites D.aqui-
lunguis with all other tyrannosauroids and a cranial process on
the proximal end of the tibia unites Appalachiosaurus and Tyr-
annosauridae (see Appendix 4, on-line at http://www.vertpaleo.
The sister relationship between Tyrannosaurus rex and T.
bataar has been demonstrated recently by numerical cladistic
analysis (Holtz, 2000); our analyses obtain the same result (see
Appendix 3 for synapomorphies). We therefore are not com-
pelled, at this time, to argue that T.bataar is generically distinct.
Complete Data Set—Finally, a larger data set (156 charac-
ters), for analyzing the ingroup relationships of Tyrannosauroi-
dea (Carr and Williamson, pers. obs.), was run to determine if
the phylogenetic position of Appalachiosaurus would change in
view of the incompleteness of the holotype. The analysis resulted
in the same topology as the smaller data matrix. This larger study
is in progress and will be published elsewhere (Carr and Willi-
amson, pers. obs.).
Because the age of RMM 6670 may be informative regarding
its basal position, we corroborated the stratigraphic field obser-
vations of its occurrence reported in King et al. (1988). A sample
of matrix from the specimen was sent to C. C. Smith of the
Alabama Geologic Survey for identification of calcareous nan-
nofossils. The results show that RMM 6670 is from the upper half
of the Calculites ovalis Zone (upper Zone 19 in Sissingh, 1977),
locally termed Zone 19b (Smith, 1998). Assignment of the
sample to this subzone is based on the presence of a suite of
species in the sample including C.ovalis, and by the absence of
the index species Bukryaster hayi (which is ubiquitous in the
lower zone, Zone 19a), and Ceratolithoides aculeus (which first
occurs in Zone 20). The nannofossil stratigraphic age assignment
places the occurrence between the uppermost 1.0-meter of the
FIGURE 20. Pedal unguals of Appalachiosaurus montgomeriensis gen.
et sp. nov. (RMM 6670). Right digit III, phalanx IV in proximal (A),
lateral (B), medial (C), dorsal (D), and ventral (E) views; and right digit
IV, phalanx V in proximal (F), lateral (G), medial (H), dorsal (I), and
plantar (J) views.
CARR ET AL.—NEW TYRANNOSAUROID FROM ALABAMA 139
Arcola Limestone to the lowermost 5.0 m of the Demopolis
Formation (Mancini, et al., 1996; C. C. Smith, 2000 pers. comm.
to DRS). This is the lowermost upper Campanian interval, and
the absolute age of occurrence is ∼77.5 +/- 1.0 Ma, contempora-
neous with the Oldman Formation of Alberta and the lower Two
Medicine and Judith River Formations of Montana (Fig. 2)
Optimization of palaeobiogeographic data (Appendix 5, on-
line at http://www.vertpaleo.org/jvp) was run in PAUP on a con-
strained topology of tyrannosauroid relationships within coelu-
rosaurian phylogeny, reconstructed after a recently published
theropod cladogram (Holtz, 2000). Parts of the topology are
modified to incorporate the most recent cladistic work on several
clades (e.g., Xu et al., 1999; Xu et al., 2000; Norell et al., 2000).
With or without Dryptosaurus included, a North American ori-
gin of dispersal for tyrannosauroids is indicated. However,
whether or not the recent common ancestor of the clade initially
evolved in Asia or North America is unknown. There was no
difference in results using ACCTRAN and DELTRAN optimi-
To redress the problem of a lack of Asian taxa in our analysis
and its potential effect on our reconstruction of historical bio-
geography, we included Shanshanosaurus huoyanshanensis,a
taxon recently identified as a juvenile tyrannosaurid and docu-
mented in enough detail to include in our data matrix (Currie
and Dong, 2001). Three additional characters are required to
specify the phylogenetic position of Shanshanosaurus (Appendix
1). Two 118-step (C.I.: 0.53) trees are obtained in which Shan-
shanosaurus is either the sister taxon to Tyrannosaurus or the
sister species to T.bataar (see Appendix 6, on-line at http://www.
vertpaleo.org/jvp/). This result differs from that of Holtz, in
which Shanshanosaurus is in a less derived position, as a basal
tyrannosaurine (sensu Holtz, 2001). The center of origin for a
Shanshanosaurus +Tyrannosaurus clade is equivocal as well as
for the genus Tyrannosaurus. With Shanshanosaurus as the sister
species of T.bataar, their recent common ancestor dispersed
from North America to Asia; when Shanshanosaurus is excluded
from the analysis, the dispersal of T.bataar into Asia from North
America is indicated. However, these results are tentative be-
cause we have not examined the holotype of Shanshanosaurus
first hand and the Asian taxa Alectrosaurus and Alioramus are
not included in our analysis.
The basal position of Appalachiosaurus and Dryptosaurus is
congruent with their isolation in the east by transgression of the
WIS during the late Albian (100 Ma) (Russell, 1995) and sug-
gests dispersal did not occur between the east and west of North
America before the middle Campanian (cf. Schwimmer, 1997) or
up to the late Maastrichtian. We agree that the eastern dinosaur
fauna represents relict forms isolated by the initial transgression
of the WIS (Schwimmer, 1997). A full phylogenetic treatment of
eastern and western dinosaurs is required to test this hypothesis,
an undertaking limited by the current state of the eastern fossil
record. The presence of Appalachiosaurus and Dryptosaurus in
the east indicates there were no fewer than two main tyranno-
sauroid lineages in North America before the transgression. Our
results indicate that the recent common ancestor of Tyrannosau-
ridae was present in western North America, an occurrence that
may have been precipitated by the WIS transgression. Later, by
the Campanian, there were no fewer than two lineages in west-
ern North America. Although Appalachiosaurus increases our
knowledge of tyrannosaurid diversity and establishes the taxo-
nomic distinctness of the eastern fauna, it does not clarify the
center of origin for tyrannosauroids.
FIGURE 21. Cladogram of tyrannosauroid ingroup relationships obtained from a numerical cladistic analysis. Clade names indicate terminology
used in text; undefined supraspecific names are of provisional usage only. Gray indicates bones that are present.
JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 25, NO. 1, 2005140
Purported premaxillary teeth of tyrannosauroids have been
reported from Valanginian-Hauterivian strata in Japan (Barrett
and Manabe, 2000) and Kimmeridgean sediments of Portugal
(Rauhut, 2000). These fossils, if they are tyrannosauroids, sug-
gest a Eurasian center of origin for the clade.
CONCLUSIONS AND SIGNIFICANCE
Tyrannosauroids from eastern North America are shown to be
less derived than their western counterparts and thus provide a
window onto the morphology of the forms that predate the WIS
transgression. The presence of basal tyrannosauroids in the
Campanian and Maastrichtian of the east suggests that dispersal
did not occur between the west and the east immediately prior to
or during these ages. However, given the absence of complete or
full comparative descriptions of pre-transgression tyrannosau-
roid fossils from western North America and Asia, new infor-
mation about Appalachiosaurus and Dryptosaurus still does not
delimit the center of origin for Tyrannosauroidea.
Comparative measurements of A.montgomeriensis (Tables
1–16), character descriptions (Appendix 1), character data ma-
trix (Appendix 2), apomorphy lists (Appendices 3–4), taxa and
geographic locations for reconstruction of historical biogeogra-
phy (Appendix 5), and list of synapomorphies that support the
sister relationship between Shanshanosaurus and Tyrannosaurus
are available on the Society of Vertebrate Paleontology web site
as supplemental information at http://www.vertpaleo.org/jvp/. A
Jurassic Foundation Grant to TDC and TEW supported this
project. We thank Susan Henson for access to RMM 6670 and
for assistance during visits to Red Mountain Museum Collections
housed at McWane Center in Birmingham, Alabama. DRS ex-
pressly thanks Charles C. Smith of the Geological Survey of
Alabama for identifications of calcareous microfossils from the
matrix of RMM 6670. He also thanks Caitlan Kiernan for initial
measurements and study of the specimen. Thanks also go to Ken
Carpenter (DMNH), Philip Currie (TMP), Ted Daeschler
(ANSP), Edward Gilmore (ANSP), John Horner (MOR), Dave
Parris (NJSM), Mark Norell (AMNH), Kevin Seymour (ROM),
Kieran Shepherd (CMN), Hans-Dieter Sues (ROM), and Mi-
chael Williams (CMNH) for access to specimens under their
care. We are grateful to Michael Ryan for assistance during our
visit to the TMP. Aspects of this manuscript were improved by
discussions with Ryosuke Motani (ROM) and Sean Modesto
(ROM). Ben Creisler assisted with etymology and name con-
struction. Dino Puleràrendered the carbon-dust plate of Appa-
lachiosaurus. TDC thanks Cinzia Puleràand Daniel Luca Pulerà
for sharing their home and Dino for the duration of the illustra-
tion project. Pen-and-ink illustrations are by TDC. Eleanor Le-
Blanc helpled construct the bibliographic database and proof
read earlier drafts of the manuscript. Jean-Bernard Caron
(ROM) digitized pen and ink illustrations, and Jerry Harris and
Tracey O’Kelly provided room and board during a visit to Phila-
delphia. Gary R. Williams (editor and publisher of Dinosaur
World™magazine) provided access to his archives. TDC’s costs
were defrayed in part by an AMNH Collections Study Grant and
by NSERC funding awarded to Chris McGowan. Chris Brochu,
Philip Currie, and Thomas Holtz, Jr., provided insightful and
constructive reviews of the original manuscript.
Anderson, J. F., Hall-Martin, A., and D. A. Russell. 1985. Long-bone
circumference and weight in mammals, birds and dinosaurs. Journal
of Zoology, Series A 207:53–61.
Baird, D. 1986. Upper Cretaceous reptiles from the Severn Formation of
Maryland. The Mosasaur 3:63–85.
Baird, D. 1989. Medial Cretaceous carnivorous dinosaur and footprints
from New Jersey. The Mosasaur 4:53–63.
Baird, D., and J. R. Horner. 1977. A fresh look at the dinosaurs of New
Jersey and Delaware. New Jersey Academy of Sciences Bulletin
Baird, D., and J. R. Horner .1979. Cretaceous dinosaurs of North Caro-
lina. Brimleyana 2:1–28.
Baird, D., and P. M. Galton. 1981. Pterosaur bones from the Upper
Cretaceous of Delaware. Journal of Vertebrate Paleontology 1:
Barrett, P. M., and M. Manabe. 2000. The dinosaur fauna from the
earliest Cretaceous Tetori Group of Central Honshu, Japan. Journal
of Vertebrate Paleontology 20(3, supplement):28A.
Barsbold, R. 1974. Saurornithoididae, a new family of small theropod
dinosaurs from central Asia and North America. Palaeontologia
Baumel, J. J. 1979. Osteologica; pp. 53–122 in J. J. Baumel, A. S. King, A.
M. Lucas, J. E. Breazile, and H. E. Evans (eds.), Nomina Anatomica
Avium. An Annotated Anatomical Dictionary of Birds. Academic
Baumel, J. J., and L. M. Witmer. 1993. Osteologica; pp. 45–132 in J. J.
Baumel, A. S. King, J. E. Breazile, H. E. Evans, and J. C. Vanden
Berge (eds.), Handbook of Avian Anatomy: Nomina Anatomica
Avium, Publications of the Nuttall Ornithological Club, No. 23,
Bonaparte, J. F., F. E. Novas, and R. Coria. 1990. Carnotaurus sastrei
Bonaparte, the horned, lightly built carnosaur from the middle Cre-
taceous of Patagonia. Contributions in Science of the Natural His-
tory Museum of Los Angeles County 416:1–42.
Brochu, C. A. 1996. Closure of neurocentral sutures during crocodilian
ontogeny: implications for maturity assessment in fossil archosaurs.
Journal of Vertebrate Paleontology 16:49–62.
Carpenter, K. D. 1992. Tyrannosaurids (Dinosauria) of Asia and North
America; pp. 250–268 in N. Mateer and P. J. Chen (eds.), Aspects of
Nonmarine Cretaceous Geology. China Ocean Press, Beijing.
Carpenter, K. D. 1997. Tyrannosauridae; pp. 766–768 in P. J. Currie and
K. Padian (eds.), Encyclopedia of Dinosaurs. Academic Press,
Carpenter, K. D., Russell, D., Baird, D., and R. Denton. 1997. Redescrip-
tion of the holotype of Dryptosaurus aquilunguis (Dinosauria:
Theropoda) from the Upper Cretaceous of New Jersey. Journal of
Vertebrate Paleontology 17:561–573.
Carr, T. D. 1999. Craniofacial ontogeny in Tyrannosauridae (Dinosauria,
Theropoda). Journal of Vertebrate Paleontology 19:497–520.
Carr, T. D, and T. E. Williamson. 2000. A review of Tyrannosauridae
(Dinosauria: Coelurosauria) from New Mexico. New Mexico Mu-
seum of Natural History Bulletin 17:113–146.
Carrano, M. T. and J. R. Hutchinson. 2002. Pelvic and hindlimb muscu-
lature of Tyrannosaurus rex (Dinosauria: Theropoda). Journal of
Chinnery, B. J., Lipka, T. R., Kirkland, J. I., Parrish, J. M., and M. K.
Brett-Surman. 1998. Neoceratopsian teeth from the lower to Middle
Cretaceous of North America. New Mexico Museum of Natural
History Bulletin 14: 297–302.
Clark, J. M., Altangerel, P., and M. A. Norell. 1994. The skull of Erli-
cosaurus andrewsi, a Late Cretaceous “Segnosaur”(Theropoda:
Therizinosauridae) from Mongolia. American Museum Novitates
Colbert, E. H. 1989. The Triassic dinosaur Coelophysis. Bulletin of the
Museum of Northern Arizona 57:1–174.
Cope, E. D. 1866. Discovery of a gigantic dinosaur in the Cretaceous of
New Jersey. Proceedings of the Academy of Natural Sciences of
Currie, P. J. 1990. Elmisauridae; pp. 245–248 in D. B. Weishampel, P.
Dodson, and H. Osmólska (eds.), The Dinosauria. University of
California Press, Berkeley and Los Angeles, California.
Currie, P. J. 2000. Possible evidence of gregarious behavior in tyranno-
saurids. Gaia 15:271–277.
Currie, P. J., and D. A. Eberth. 1993. Palaeontology, sedimentology and
palaeoecology of the Iren Dabasu Formation (Upper Cretaceous),
Inner Mongolia, People’s Republic of China. Cretaceous Research
CARR ET AL.—NEW TYRANNOSAUROID FROM ALABAMA 141
Currie, P. J., and X.-J. Zhao. 1993. A new carnosaur (Dinosauria,
Theropoda) from the Jurassic of Xinjiang, People’s Republic of
China. Canadian Journal of Earth Sciences 30:2037–2081.
Currie, P. J., and K. Carpenter. 2000. A new specimen of Acrocantho-
saurus atokensis (Theropoda, Dinosauria) from the Lower Creta-
ceous Antlers Formation (Lower Cretaceous, Aptian) of Oklahoma,
USA. Geodiversitas 22:207–246.
Currie, P. J., and Dong, Z. 2001. New information on Shanshanosaurus
huoyanshanensis, a juvenile tyrannosaurid (Theropoda, Dinosauria)
from the Late Cretaceous of China. Canadian Journal of Earth Sci-
Dong, Z., Chang Y., Li X., and S. Zhou. 1978. [Note on a new carnosaur
(Yangchuanosaurus shangyouensis gen. et. sp. nov.) from the Juras-
sic of Yangchuan District, Szechuan Province]. Kexue Tongbao 23:
Gallagher, W. G. 1995. Evidence for juvenile dinosaurs and dinosaurian
growth stages in the Late Cretaceous deposits of the Atlantic
Coastal Plain. Bulletin of the New Jersey Academy of Science 40:
Gauthier, J. 1986. Saurischian monophyly and the origin of birds; pp.
1–55 in K. Padian (ed.), The Origin of Birds and the Evolution of
Flight. Memoirs of the California Academy of Science 8.
Gilmore, C. W. 1920. Osteology of the carnivorous Dinosauria in the
United States National Museum, with special reference to the gen-
era Antrodemus (Allosaurus) and Ceratosaurus. Bulletin of the
United States National Museum 110:1–154.
Gilmore, C. W. 1921. The fauna of the Arundel Formation of Maryland.
Proceedings of the United States National Museum 59:581–594.
Harris, J. D. 1998. A reanalysis of Acrocanthosaurus atokensis, its phy-
logenetic status, and paleobiogeographic implications, based on a
new specimen from Texas. New Mexico Museum of Natural History
and Science Bulletin 13:1–75.
Holtz, T. R., Jr. 1994. The phylogenetic position of the Tyrannosauridae:
implications for theropod systematics. Journal of Palaeontology 68:
Holtz, T. R., Jr. 2000. A new phylogeny of the carnivorous dinosaurs.
Holtz, T. R., Jr. 2001. The phylogeny and taxonomy of the Tyrannosau-
ridae; pp. 64–83 in D. H. Tanke and K. Carpenter (eds.), Mesozoic
Vertebrate Life. Indiana University Press, Bloomington and India-
Huene, F. von. 1914. Das natürliche System der Saurischia. Zentralblatt
Mineralogie, Geologie, und Palaeontologie B 1914:154–158.
Hurum, J. H., and P. J. Currie. 2000. The crushing bite of tyrannosaurids.
Journal of Vertebrate Paleontology 20:619–621.
King, D. T., Abbott-King, J. P., Bell, Jr., G. L., Lamb, Jr., J. P., Dobie,
J. L., and D. R. Womochel. 1988. Stratigraphy and depositional en-
vironments of the Turnipseed dinosaur site in the Upper Cretaceous
Demopolis Chalk of Montgomery County, Alabama. Journal of the
Alabama Academy of Science 59:34–48.
Kurzanov, S. M. 1976. [A new Late Cretaceous carnosaur from Nogon-
Tsav, Mongolia.] Joint Soviet-Mongolian Paleontological Expedi-
tion Transactions 3:93–104. [Russian with English summary]
Lambe, L. M. 1917. The Cretaceous theropodous dinosaur Gorgosaurus.
Memoirs of the Geological Survey of Canada 100:1–84.
Langston, W., Jr. 1960. The vertebrate fauna of the Selma Formation of
Alabama; Part VI, The Dinosaurs. Fieldiana Geology Memoirs 3:
Leidy, J. 1858. Hadrosaurus foulkii, a new saurian from the Cretaceous
of New Jersey. Proceedings of the Academy of Natural Sciences
Mader, B. J. and R. L. Bradley. 1989. A redescription and revised diag-
nosis of the syntypes of the Mongolian tyrannosaur Alectrosaurus
olseni. Journal of Vertebrate Paleontology 9:41–55.
Maddison, W. P., and D. R. Maddison. 1992. MacClade, version 3.0.
Computer program distributed by Sinauer Associates, Inc., Sunder-
Madsen, J. H., Jr. 1976. Allosaurus fragilis: a revised osteology. Bulletin
of the Utah Geological and Mineral Survey 109:1–163.
Maleev, E. A. 1974. [Gigantic carnosaurs of the family Tyrannosauridae.]
Joint Soviet-Mongolian Palaeontological Expedition, Transactions
1:132–191. [Russian with English summary]
Mancini, E. A., Puckett, T. M., and B. H. Tew. 1996. Integrated Biostrati-
graphic and Sequence Stratigraphic Framework for Upper Creta-
ceous and Paleogene Biostratigraphy and Lithostratigraphy of the
Eastern Gulf Coastal Plain. 28th International Geological Congress
Field Trip Guidebook, Y372, 122 pp.
Marsh, O. C. 1881. Classification of the Dinosauria. American Journal of
Science (ser. 3) 23:81–86.
Marsh, O. C. 1878. Notice of new dinosaurian reptiles. American Journal
of Science (ser. 3) 15:241–244.
Molnar, R. E., Kurzanov, S. M., and Z. Dong. 1990. Carnosauria; pp.
169–209 in D. B. Weishampel, P. Dodson, and H. Osmólska (eds.),
The Dinosauria. University of California Press, Berkeley and Los
Norell, M. A., Makovicky, P. J., and J. M. Clark. 2000. A new troodontid
theropod from Ukhaa Tolgod, Mongolia. Journal of Vertebrate Pa-
Novas, F. 1993. New information on the systematics and Postcranial
skeleton of Herrerasaurus ischigualastensis (Theropoda: Herre-
rasauridae) from the Ischigualasto Formation (Upper Triassic) of
Argentina. Journal of Vertebrate Paleontology 13:400–423.
Osborn, H. F. 1905. Tyrannosaurus and other Cretaceous carnivorous
dinosaurs. Bulletin of the American Museum of Natural History
Osborn, H. F. 1916. Skeletal adaptations of Ornitholestes,Struthiomimus,
Tyrannosaurus. Bulletin of the American Museum of Natural His-
Osborn, H. F. 1924. Three new Theropoda, Protoceratops Zone, central
Mongolia. American Museum Novitates 144:1–12.
Ostrom, J. H. 1969. Osteology of Deinonychus antirrhopus, an unusual
theropod from the Lower Cretaceous of Montana. Bulletin of the
Peabody Museum of Natural History, Yale University 30:1–165.
Parks, W. A. 1928. Albertosaurus arctunguis, a new species of theropo-
dous dinosaur from the Edmonton Formation of Alberta. University
of Toronto Studies (Geological Series) 25:3–42.
Russell, D. A. 1970. Tyrannosaurs from the Late Cretaceous of western
Canada. National Museum of Natural Science Publications in Pal-
Russell, D. A. 1995. China and the lost worlds of the dinosaurian era.
Historical Biology 10:3–12.
Rauhut, O. W. M. 2000. The dinosaur fauna from the Guimarota mine;
pp. 75–82 in T. Martin and B. Krebs (eds.), Guimarota: A Jurassic
Ecosystem. Verlag Dr. Friedrich Pfeil, Munich.
Schwimmer, D. R. 1986. Late Cretaceous fossils from the Blufftown
Formation (Campanian) in Western Georgia. The Mosasaur 3:
Schwimmer, D. R. 1997. Late Cretaceous dinosaurs in Eastern USA:
A taphonomic and biogeographic model of occurrences. Dino-
fest International, Academy of Natural Sciences, Philadelphia: 203–
Schwimmer, D. R., and R. H. Best. 1989. First dinosaur fossils from
Georgia, with notes on additional Cretaceous fossils from the State.
Georgia Journal of Sciences 47:147–157.
Schwimmer, D. R.,Williams, G. D., Dobie, J. L., and W. G. Siesser. 1993.
Upper Cretaceous dinosaurs from the Blufftown Formation, west-
ern Georgia and eastern Alabama. Journal of Paleontology 67:288–
Sereno, P. C. 1997. The origin and evolution of dinosaurs. Annual Re-
view of Earth and Planetary Sciences 25:435–489.
Sereno, P. C. 1998. A rationale for phylogenetic definitions, with appli-
cation to the higher-level taxonomy of Dinosauria. Neues Jahrbuch
für Geologie und Paläontologie Abhandlungen 210:41–83.
Sereno, P. C., and F. Novas. 1993. The skull and neck of the basal
theropod Herrerasaurus ischigualastensis. Journal of Vertebrate Pa-
Sissingh, W. 1977. Biostratigraphy of Cretaceous calcareous nannoplank-
ton. Geologie en Mijnbouw 56:37–65.
Smith, C. C. 1998. Regional Upper Cretaceous stratigraphy; pp. 6–10 in
E. A. Mancini, T. M. Puckett, W. C. Parcell, S. J. Crow, and C. C.
Smith (eds.), Sequence Stratigraphy and Biostratigraphy of Upper
Cretaceous Strata of the Alabama Coastal Plain. Alabama Geologi-
cal Society Guidebook to 35th Field Trip.
Swofford, D. L. 1993. PAUP: Phylogenetic Analysis Using Parsimony.
Version 3.1. Illinois Natural History Survey.
JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 25, NO. 1, 2005142
Weishampel, D., P. Dodson, and H. Osmólska (eds.), The Dinosauria.
University of California Press, Berkeley.
Welles, S. P., and R. A. Long. 1974. The tarsus of theropod dinosaurs.
Annals of the South African Museum 64:191–218.
Witmer, L. M. 1997. The evolution of the antorbital cavity of archosaurs:
a study in soft-tissue reconstruction in the fossil record with an
analysis of the function of pneumaticity. Journal of Vertebrate Pa-
leontology 17(1, supplement):1–73.
Witmer, L. M. 20001. Nostril position in dinosaurs and other vertebrates
and its significance for nasal function. Science 293:850–853.
Xu, X., Wang, X.-L., and X.-C. Wu. 1999. A dromaeosaurid dinosaur
with a filamentous integument from the Yixian Formation of China.
Xu, X., Zhou, Z. and X. Wang. 2000. The smallest known non-avian
theropod dinosaur. Nature 408:705–708.
Submitted 1 February 2001; accepted 18 May 2004.
CARR ET AL.—NEW TYRANNOSAUROID FROM ALABAMA 143