A ceratopsian dinosaur from the Late Cretaceous of eastern North America, and implications for dinosaur biogeography

Abstract

Tyrannosaurs and hadrosaurs from the Late Cretaceous of eastern North America (Appalachia) are distinct from those found in western North America (Laramidia), suggesting that eastern North America was isolated during the Late Cretaceous. However, the Late Cretaceous fauna of Appalachia remains poorly known. Here, a partial maxilla from the Campanian Tar Heel Formation (Black Creek Group) of North Carolina is shown to represent the first ceratopsian from the Late Cretaceous of eastern North America. The specimen has short alveolar slots, a ventrally projected toothrow, a long dentigerous process overlapped by the ectopterygoid, and a toothrow that curves laterally, a combination of characters unique to the Leptoceratopsidae. The maxilla has a uniquely long, slender and downcurved posterior dentigerous process, suggesting a specialized feeding strategy. The presence of a highly specialized ceratopsian in eastern North America supports the hypothesis that Appalachia underwent an extended period of isolation during the Late Cretaceous, leading the evolution of a distinct dinosaur fauna dominated by basal tyrannosauroids, basal hadrosaurs, ornithimimosaurs, nodosaurs, and leptoceratopsids. Appalachian vertebrate communities are most similar to those of Laramidia. However some taxa-including leptoceratopsids-are also shared with western Europe, raising the possibility of a Late Cretaceous dispersal route connecting Appalachia and Europe.

Full text

A Ceratopsian Dinosaur from the Late Cretaceous of
Eastern North America, and Implications for
Dinosaur Biogeography
Nicholas R. Longrich
Department of Biology and Biochemistry and Milner Centre for Evolution, University of Bath, Claverton Down, Bath, BA2 7AY
United Kingdom
Email address: nrl22@bath.ac.uk
A B S T R A C T
Tyrannosaurs and hadrosaurs from the Late Cretaceous of eastern North America
(Appalachia) are distinct from those found in western North America (Laramidia),
suggesting that eastern North America was isolated during the Late Cretaceous. However,
the Late Cretaceous fauna of Appalachia remains poorly known. Here, a partial maxilla
from the Campanian Tar Heel Formation (Black Creek Group) of North Carolina is
shown to represent the first ceratopsian from the Late Cretaceous of eastern North
America. The specimen has short alveolar slots, a ventrally projected toothrow, a long
dentigerous process overlapped by the ectopterygoid, and a toothrow that curves laterally,
a combination of characters unique to the Leptoceratopsidae. The maxilla has a uniquely
long, slender and downcurved posterior dentigerous process, suggesting a specialized
feeding strategy. The presence of a highly specialized ceratopsian in eastern North
America supports the hypothesis that Appalachia underwent an extended period of
isolation during the Late Cretaceous, leading the evolution of a distinct dinosaur fauna
dominated by basal tyrannosauroids, basal hadrosaurs, ornithimimosaurs, nodosaurs, and
leptoceratopsids. Appalachian vertebrate communities are most similar to those of
Laramidia. However some taxa- including leptoceratopsids- are also shared with western
Europe, raising the possibility of a Late Cretaceous dispersal route connecting
Appalachia and Europe.
Keywords: Dinosauria; Neoceratopsia; Leptoceratopsia; Appalachia; Black Creek Group
1. Introduction
During the Late Cretaceous, a shallow inland sea, the Western Interior Seaway,
extended from the Gulf of Mexico to the Arctic Ocean, splitting North America in two.
The resulting land masses- Laramidia in the west, and Appalachia in the east- each
developed distinct dinosaurian faunas (Fig. 1).

Late Cretaceous dinosaurs from Laramidia (Weishampel et al., 2004) show close
affinities with the dinosaurs of Asia and to a lesser degree, South America. Among
theropods, North America’s tyrannosaurids (Brusatte et al., 2011), alvarezsaurs (Longrich
and Currie, 2009a), caenagnathids (Longrich et al., 2013), microraptorines (Longrich and
Currie, 2009b) and ornithomimids (Xu et al., 2011) all have relatives in Asia. Among
ornithischians, saurolophine (Godefroit et al., 2012) and hadrosaurine (Prieto-Márquez et
al., 2012) duckbills, ceratopsids (Xu et al., 2010b), leptoceratopsids (Ryan et al., 2012),
pachycephalosaurids (Longrich et al., 2010) and ankylosaurids (Sullivan, 1999) show the
same. These patterns show that a high-latitude land corridor joined North America and
Asia in the Late Cretaceous !"#$$%&&'( )**+,, with extensive dispersal between the two
continents. In the Maastrichtian, the appearance of titanosaurs (D' Emic et al., 2010) and
large alethinophidian snakes (Longrich et al., 2012) in Laramidia and saurolophines
(Prieto-Marquez, 2010) and multituberculate mammals (Kielan-Jaworowska et al., 2004)
in South America indicates dispersal between Laramidia and South America, either via a
land bridge or across a narrow ocean channel or archipelago.
The Late Cretaceous dinosaurs of Appalachia are highly distinct from those seen
in Laramidia, however. While Laramidia is dominated by Tyrannosauridae (Weishampel
et al., 2004), Appalachia is dominated by basal tyrannosauroids such as Dryptosaurus and
Appalachisaurus (Brusatte et al., 2011; Carr et al., 2011). Similarly, while Laramidia is
dominated by lambeosaurine and saurolophine hadrosaurs (Weishampel et al., 2004), in
Appalachia hadrosaurine-grade hadrosaurs such as Hadrosaurus and Lophorhoton
(Prieto-Márquez et al., 2012) dominate. These patterns suggest that Appalachia saw an
extended period of isolation beginning in the Late Cretaceous, becoming an island
continent with an endemic fauna, similar to Australia in the Cenozoic.
Unfortunately, our knowledge of Appalachian dinosaurs remains limited(
!-./0122%3'()**45( 6%1$/728%&( %9(7&:'( ;<<=,, with only tyrannosauroids (Brusatte et
al., 2011), hadrosaurids (Prieto-Marquez et al., 2006), ornithomimosaurs (Russell, 1972)
and nodosaurs !>7?@$9A?'()*B<, known from the eastern United States. The discovery of
new dinosaurs from eastern North America is therefore of great interest to understanding
the Appalachian fauna and its origins.
Recently, an unusual dinosaur specimen from the Campanian Black Creek Group
of North Carolina (Fig. 1) was identified in Yale University’s Peabody Museum
collections. The specimen consists of the posterior end of a left maxilla. Although
originally identified and catalogued as a hadrosaur, the specimen shows apomorphies of
the Ceratopsia and specifically the Leptoceratopsidae. This specimen is the first
ceratopsian known from the Late Cretaceous period of eastern North America.
Institutional Abbreviations: AMNH American Museum of Natural History, New York;
NMC, National Museum of Canada, Ottawa; YPM-PU, Yale Peabody Museum,
Princeton University Collections.
2. Systematic Paleontology
Dinosauria Owen 1842
Ornithischia Seeley 1888

Ceratopsia Marsh 1890
Leptoceratopsidae Nopsca 1923
Leptoceratopsidae sp.
Material: YPM-PU (Yale Peabody Museum, Princeton University collection) 24964,
posterior end of a left maxilla (Fig. 2).
Locality and Horizon: Clifton Farm, Giddensville, Sampson County, North Carolina
(Fig. 2). The same locality has produced a tooth of a tyrannosauroid (YPM PU 23197),
and teeth and scutes of the giant crocodilian Deinosuchus rugosus (YPM-PU 23429).
Although the collections are very limited, vertebrates such as turtles, mosasaurs, fish,
and sharks, which are abundant in the nearby Phoebus Landing locality (Miller, 1967)
were not collected from the assemblage, suggesting a freshwater or estuarine depositional
environment.
Provenance data for the specimen identify it as from the “Black Creek Formation”.
The Black Creek Formation has recently been raised to the level of group, containing
three formations (Fig. 3); from bottom to top, these are the Tar Heel, Bladen, and Donoho
Formations. Maps of outcrop !C0%?$(7?D(-A/&'()*E*,(8#9(9/%(F&1G9A?(H732(&A.7&19I(1?(
9/%(J73(K%%&(HA32791A?:(The Tar Heel was deposited during the Early Campanian
(Harris and Self-Trail, 2006). Previous dates, based on strontium isotopes, suggest an age
of 82.3-73.4 Ma or 74.5-82.6 Ma for the formation, depending on the model used (Harris
and Self-Trail, 2006). (
L(?%73MI(7$$%2M&7@%(G3A2(9/%(J73(K%%&'(Phoebus Landing on the Cape Fear
River, has produced hadrosaurids, a possible ornithomimosaur, and a diverse fauna of
freshwater and marine vertebrates (Miller, 1967). Recent work on Phoebus Landing
suggests that the dinosaurs date to 77.1- 78.5 Ma (Self-Trail et al., 2004), the middle of
the Campanian.
Description: The preserved portion of the maxilla (Fig. 2), is 43 mm long but broken
posteriorly. The anterior end is broken away leaving only the posterior dentigerous
process; the posterior end of this process is also broken off. Comparisons with other
ceratopsians suggest that the complete maxilla may have measured ~120-160 mm.
Although the main body of the maxilla is missing, there is no trace of a jugal
contact (Fig. 2C). The jugal must have articulated well above the toothrow. In primitive
ceratopsians such as Yinlong (Xu et al., 2006) and Psittacosaurus (Osborn, 1923) the
jugal articulates more or less lateral to the toothrow. By contrast, the jugal articulates
more dorsally in primitive neoceratopsians such as Liaoceratops (Makovicky and Norell,
2006) and especially in more advanced neoceratopsians such as Yamaceratops
(Makovicky and Norell, 2006), Protoceratopsidae (Brown and Schlaikjer, 1940;
Maryanska and Osmólska, 1975), Leptoceratopsidae (Chinnery, 2004; Chinnery and
Horner, 2007), and Euceratopsia (Dodson et al., 2004; Wolfe et al., 2010).
The toothrow was also strongly inset relative to the jugal; as can be seen in ventral
view, the lateral surface of the maxilla is sloped inward. This feature is also seen in
Leptoceratopsidae (Fig. 3) and other ceratopsians such as Liaoceratops (Makovicky and
Norell, 2006), Yamaceratops (Makovicky and Norell, 2006) and Protoceratopsidae

(Brown and Schlaikjer, 1940; Maryanska and Osmólska, 1975) but is only very weakly
developed in hadrosaurs.
The posterior dentigerous process is elongate, with room for at least five teeth. By
comparison, primitive neoceratopsians such as Liaoceratops (Xu et al., 2002) and
Auroraceratops (You et al., 2012) have a dentigerous process bearing one or two teeth;
the dentigerous process of protoceratopsids bears up to three teeth (Brown and Schlaikjer,
1940)}(Maryanska and Osmólska, 1975); Leptoceratops has five (Fig. 3), Zuniceratops
has five or six (Wolfe et al., 2010), and Ceratopsidae have even more (Hatcher et al.,
1907). Elongation of the posterior dentigerous process occurs convergently in
hadrosaurids (Horner et al., 2004).
Although the increased number of tooth positions is shared by the Black Creek
ceratopsian and leptoceratopsids, the shape of the dentigerous process is very different. In
other leptoceratopsids, the dentigerous process is very deep, e.g. the height of the process
is 130% of its length in Prenoceratops (Chinnery, 2004) versus 70% or less in the Black
Creek form; in this respect the maxilla is more similar to Euceratopsia (Wolfe et al.,
2010). In lateral view, the dentigerous process has a distinctly downturned end; the very
end of the dentigerous process is downturned by 25º relative to its anterior end. This
distinctive downturn is absent in other leptoceratopsids such as Prenoceratops (Chinnery,
2004) and Cerasinops !F/1??%3I( 7?D( KA3?%3'( ;<<4,, where the ventral margin of the
dentigerous process is straight in lateral view.
The posterior dentigerous process is bowed outward in ventral view, such that the
toothrow is curved in ventral view. This curvature is a derived feature seen in other
Leptoceratopsidae such as Leptoceratops and Prenoceratops, but it is not developed to
the same degree as in the Black Creek ceratopsian (Fig. 3). In Leptoceratops (Figure 3) or
Prenoceratops !F/1??%3I'( ;<<=,, the posterior dentigerous process makes an angle of
approximately 13º with the teeth lying just ahead of the process, whereas this angle is 25º
in the Black Creek ceratopsian.
In addition to being diagnostic of the leptoceratopsids, the outward curvature of
the toothrows indicates a proportionally short, broad skull that would have been
triangular in dorsal view; this skull shape is characteristic of basal Neoceratopsia (You
and Dodson, 2004).
The process is rugose laterally, but there is a smooth dorsal facet where the
pterygoid would have dorsally overlapped the maxilla; pterygoid overlap of the maxilla is
characteristic of neoceratopsians (You and Dodson, 2004).
Teeth would have implanted into alveolar slots, separated by interavleolar ridges
(Fig. 4), as in Neoceratopsia (You and Dodson, 2004) and convergently in Hadrosauridae
(Horner et al., 2004). The alveolar slots are too short to accommodate more than one
replacement tooth, however, ruling out affinities with either Ceratopsidae or
Hadrosauridae, in which the alveolar grooves accommodate a series of replacement teeth
(Dodson et al., 2004) (Horner et al., 2004) below the functional tooth. The shape of the
tooth sockets is very similar to those of Leptoceratops (Fig. 4), and as in Leptoceratops
the interalveolar ridges are poorly developed anteriorly, and become increasingly well-
developed posteriorly such that they tightly embrace the tooth roots. The shape of the
socket suggests that the tooth roots were probably anteroposteriorly compressed, as in
Leptoceratopsidae and Euceratopsia.

3. Discussion
Affinities. Although fragmentary, the morphology of YPM-PU 24964 is consistent with
referral to Neoceratopsia and specifically to Leptoceratopsidae. A series of characters
support this assignment.
• Transverse expansion of the skull posteriorly (Ceratopsia).
• Strong lateral projection of the jugal relative to the toothrow (Ceratopsia)
• Extensive overlap of the dentigerous process by the ectopterygoid (Neoceratopsia).
• Strong ventral projection of the toothrow below the jugal-maxilla contact
(Yamaceratops, Protoceratopsidae, Leptoceratopsidae, and Euceratopsia).
• Posterior dentigerous process elongate, with 3 or more teeth (Protoceratopsidae,
Leptoceratopsidae, and Euceratopsia).
• Posterior dentigerous process with 5 or more teeth (Leptoceratopsidae, Euceratopsia).
• Laterally deflected posterior dentigerous process (Leptoceratopsidae).
Some of these characters occur convergently in hadrosauroids and hadrosaurs.
Hadrosaurs such as Hadrosaurus (Prieto-Marquez et al., 2006) have both an elongate
dentigerous process and an extensive dorsal overlap of the maxilla by the ectopterygoid,
and alveolar slots. However, the jaw differs from hadrosaurs in many respects. First, in
hadrosaurs (Prieto-Marquez et al., 2006) and hadrosauroids (Prieto‐Márquez, 2011) there
is a prominent contact for the jugal on the lateral surface of the maxilla, the jugal process,
just anterior to the dentigerous process. The Black Creek jaw lacks any evidence for a
jugal attachment, meaning that the jugal must have attached well above the toothrow, as
in leptoceratopsids, protoceratopsids, and euceratopsians; furthermore the toothrow is
strongly inset relative to the body of the maxilla, such that the jugal attachment would
have been well lateral to the toothrow; again this is a ceratopsian feature, not seen in
hadrosaurs.
Second, in hadrosaurs (Prieto-Marquez et al., 2006) and hadrosauroids (Prieto‐
Márquez, 2011) the jugal is supported by a prominent ectopterygoid ridge, a derived
feature of hadrosauroids; no such ridge is present in the Black Creek jaw.
Third, in hadrosaurids the ventral margin of the maxilla is straight in lateral view
and weakly crenellated in ventral view due to reduction of the interdental ridges, a
derived feature of the group. By comparison, in the Black Creek jaw and leptoceratopsids
the maxilla is distinctly crenellated in lateral view and ventral view where prominent
interdental ridges project down and in to wrap around the base of each tooth.
Fourth, in hadrosaurids there are multiple replacement teeth, such that alveolar
ridges form long, narrow slots for teeth (Horner et al., 2004; Prieto-Marquez et al., 2006).
Although small juveniles have proportionately larger teeth and would have
correspondingly wider, shorter alveolar slots, the teeth of comparably sized juvenile
hadrosauroids (Prieto‐Márquez, 2011) are still more tightly packed than in the Black
Creek jaw and would have narrower alveolar slots.
Finally, no hadrosaur is known to show the strong lateral deflection of the
dentigerous process seen in leptoceratopsids. In summary, the jaw exhibits no derived
features of Hadrosauridae that are not also seen in ceratopsians, and exhibits numerous

derived and primitive features that are seen in ceratopsians, but not hadrosaurs; the
available evidence therefore rejects a hadrosaur identification.
The position of the Black Creek ceratopsian within the Leptoceratopsidae is
unclear. It lacks the derived, proportionately short and deep maxilla that characterizes
most leptoceratopsids. Assuming this is a plesiomorphy, then it may represent a relatively
basal divergence. However, it is highly derived with respect to other leptoceratopsids in
terms of the shallow dentigerous process, the strong lateral curvature of the toothrow, and
the strongly downturned dentigerous process; suggesting a high degree of specialization
and a long evolutionary history.
Ecology and Evolution. The maxillae of the Black Creek ceratopsidae are highly derived
relative to other Leptoceratopsidae in terms of the elongation of the posterior dentigerous
process, the strong lateral deflection of the dentigerous process, and the strong downturn
of the process in lateral view. These unusual specializations suggest adaptation for an
unusual diet and/or feeding strategy not seen in other leptoceratopsids or other basal
neoceratopsians.
Leptoceratopsids and other basal neoceratopsians have short, deep jaws that
would be well-suited to shearing tough, fibrous vegetation (Longrich, 2010), and
Leptoceratopsidae in particular are characterized by teeth with a unique combination of
crushing and shearing facets (Ostrom, 1966) and very short, deep, ‘nutcracker’ jaws
(Brown, 1914; Kurzanov, 1992; Chinnery, 2004; Chinnery and Horner, 2007; Ryan et al.,
2012) with the dentigerous process of the maxilla being correspondingly short and deep
(Chinnery, 2004; Chinnery and Horner, 2007). As the strength of a structure in bending
or shearing increases with depth (Gordon, 1978), this jaw structure suggests adaptation to
produce high bite forces and process highly resistant food items.
The Black Creek ceratopsian departs markedly from this trend in having a
relatively long, narrow dentigerous process, more like that of a ceratopsid than a
leptoceratopsid. Presumably, this feature represents an adaptation for processing less
resistant food items. The odd down-and-out bend of the dentigerous process would have
altered the shape of the shearing blade formed by the teeth; it presumably represents a
feeding specialization as well; although its functional significance is less clear, it also
suggests that the animal had evolved a feeding strategy distinct from that of other
leptoceratopsids.
This divergent evolutionary path could result from the distinct biota of the
Appalachian province. Appalachia was part of a distinct palynofloral province, the
Normapolles province (Srivastava, 1981) and so the vegetation found there would have
been distinct from that seen in Laramidia. Perhaps more importantly, many of the
herbivorous dinosaurs found in Laramidia were absent from Appalachia; the absence of
competition may have allowed small ceratopsians to exploit ecological niches and food
items that would have been taken by other lineages of herbivore in Laramidia.
Appalachian biogeography. Together with the basal phylogenetic position of the
tyrannosaurs and hadrosaurs, the highly divergent morphology of the Black Creek
ceratopsian supports the idea that eastern North America was largely isolated from
Laramidia throughout the Campanian and Maastrichtian. This idea is further supported by
the fact that many of the groups that are shared by Laramidia and Asia during the

Campanian and Maastrichtian- including ceratopsids, pachycephalosaurs, thescelosaurs,
lambeosaurs, saurolophines and ankylosaurids among the Ornithischia, and
saurornitholestines, dromaeosaurines, caenagnathids, alvarezsaurids and titanosaurs
among the Saurischia (Weishampel et al., 2004)- are currently unknown from the fauna
(Weishampel et al., 2004). Furthermore, even taxa known from the mid-Turonian of
Laramidia, such as euceratopsians (Wolfe et al., 2010) and therizinosaurs (Kirkland and
Wolfe, 2001) are unknown from Appalachia. Their absence would suggest that the
physical isolation of Appalachia and Laramidia had already occurred at this time.
Sampling clearly remains an issue. Compared to the rich fauna found in
Laramidia, Appalachia’s dinosaur fauna is known from a far more limited number of
specimens, mostly from marine depositional settings (Schwimmer, 1997). Yet while
individual assemblages are poorly sampled compared to Western North America, Late
Cretaceous dinosaurs have been reported from localities in many Eastern states, including
New Jersey, Delaware, Maryland, North and South Carolina, Tennessee, Georgia,
Alabama, Mississippi, and Missouri (Fig.6). Furthermore these strata range from the early
Santonian to the Late Maastrichtian in age, a period of approximately 20 million years
(Schwimmer, 1997; Weishampel et al., 2004).
And while the associated remains are admittedly limited, isolated teeth and bones
are typically diagnostic to family or subfamily level; the existence of ceratopsids,
ankylosaurs, or titanosaurs could be confirmed by even a single tooth, bone, or scute.
Further collection and study could easily reveal previously unknown dinosaur lineages in
Appalachia, but even so, over a century of sampling from numerous localities has
consistently painted a picture of a fauna dominated by hadrosaurines, tyrannosauroids,
ornithomimosaurs, and nodosaurs— one that is low in diversity, even depauperate,
relative to Laramidia.
Despite the absence of many characteristic Laramidian taxa, the known vertebrate
fauna of Appalachia is most similar to that of the Late Cretaceous of Laramidia. Taxa of
fish (Grandstaff et al., 1992), amphibians (Denton Jr and O'Neill, 1998), reptiles
(Grandstaff et al., 1992; Denton and O'Neill, 1995), and mammals (Grandstaff et al.,
1992) are all shared with Late Cretaceous faunas known from Western North America.
Semiaquatic reptiles such as the crocodilians Deinosuchus and Borealosuchus and the
turtle Adocus (Grandstaff et al., 1992) may have been able to swim across the Western
Interior Seaway, and small mammals, lizards and even frogs could conceivably have
rafted. However, the fact that Laramidia and Appalachia share salt-intolerant aquatic
forms such as amiid fish and salamanders is strong evidence for an ancient land
connection between the two. Given this, most of the Appalachian fauna can be interpreted
as resulting from (i) dispersal across the Western Interior seaway following isolation
from Laramidia; (ii) a mid-Cretaceous land connection between Eastern and Western
North America, or (iii) a combination of these processes.
How ceratopsians arrived in Appalachia (Fig. 1) remains unclear. Ceratopsian
teeth are known from the Lower Cretaceous Arundel Formation of Maryland (Chinnery
et al., 1998). However, the teeth are primitive compared to leptoceratopsids; the teeth
lack a strong offset of the primary ridge, whereas they are strongly offset in more derived
forms such as leptoceratopsids, protoceratopsids, and euceratopsians; likewise the central
ridges extend only around halfway down the face of the crown or less, whereas the
secondary ridges extend nearly to the cingulum in more derived forms. Given this, the

Arundel ceratopsian does not appear to be closely related to the Black Creek form or any
other known Late Cretaceous ceratopsian.
Instead, given the high diversity of neoceratopsians in Asia, the Black Creek
ceratopsian is likely to represent a lineage that dispersed to Eastern North America.
Conceivably, leptoceratopsids could have traversed a land bridge between western and
eastern North America during the mid-Cretaceous, before Appalachia was fully isolated
by the Western Interior Seaway. An alternative is that ceratopsians dispersed from
Laramidia to Appalachia following the separation of the two via the Western Interior
Seaway. Although it seems improbable that animals as large as ceratopsians could have
colonized Appalachia via oceanic rafting, animals as large as iguanas are known to
successfully cross oceanic barriers on floating vegetation (Censky et al., 1998) and
juvenile ceratopsians would have been relatively small animals and could conceivably
have rafted between the two land masses. Furthermore, the dispersal of mammals from
Africa to South America during the Cenozoic (Poux et al., 2006) shows that trans-
oceanic dispersal can and does occur in large terrestrial animals.
Yet although the Appalachian fauna shows a strong affinity with Laramidia,
Appalachia and Europe also share a number of taxa. These include cimolomyid
multituberculates (Grandstaff et al., 1992), nortedelphid marsupials (Martin et al., 2005),
and leptoceratopsids (Fig. 1), which are known from the Late Cretaceous of Sweden
(Lindgren et al., 2007). Furthermore, the neoceratopsian Craspedodon lonzeensis
(Godefroit and Lambert, 2007) appears to represent another European leptoceratopsid, as
it shares an inset primary ridge with the leptoceratopsids, as well as an anteroposteriorly
compressed, figure-8 shaped tooth root (seen also in the euceratopsian Turanoceratops
(Sues and Averianov, 2009) but not in protoceratopsids or more primitive
neoceratopsians).
Thus, dispersal between Appalachia and Europe- with dinosaurs colonizing
Europe from North America, or vice versa- is also possible. If so, then a high-latitude
land corridor connecting North America and Europe via Greenland, the Thulian route,
may have been established towards the end of the Cretaceous. Leptoceratopsids could
conceivably have traveled via this route from Europe into Appalachia, rather than through
Laramidia. New discoveries from both eastern North America and western Europe will be
needed to test these hypotheses and to better understand the Appalachian fauna and its
origins.
Conclusions. The Black Creek ceratopsian represents a highly derived member of the
Leptoceratopsidae. Along with basal tyrannosauroids, hadrosaurines, and nodosaurs, it
was part of a distinctive fauna that emerged during the Late Cretaceous in eastern North
America. The unusual composition and low diversity of this fauna are likely the result of
prolonged isolation of eastern North America by the Western Interior Seaway to form the
island continent of Appalachia. The origins of this fauna remain poorly understood.
Overall the fauna is most similar to that of Laramidia, but similarities with the fauna of
western Europe raise the possibility of dispersal events between Europe and eastern
North America.

Acknowledgments. Thanks to the curators and staff of the Yale Peabody Museum,
National Museum of Canada, and American Museum of Natural History for specimen
access, and to the Yale Institute for Biospheric Studies for funding. Thanks to Jordan
Mallon (National Museum of Canada) and Albert Prieto-Marquez (University of Bristol)
for specimen photos, to Ron Blakey for the map in Fig. 1, and to Marilyn Fox and Don
Brinkman (Yale) for information on provenance.

References
N3A0?'(N:'()*)=:(Leptoceratops'(7(?%0(@%?#$(AG(F%379A8$17(G3A2(9/%(OD2A?9A?(
F3%97.%A#$(AG(L&M%397:(N#&&%91?(AG(9/%(L2%31.7?(P#$%#2(AG(Q79#37&(K1$9A3I(
++'(RB4SRE<:(
N3A0?'(N:'()*=;:(J/%($T%&%9A?(AG(Leptoceratops(019/(9/%(D%$.31891A?(AG(7(?%0($8%.1%$:(
L2%31.7?(P#$%#2(QAU1979%$())B*'()S)R:(
N3A0?'(N:'(-./&71TV%3'(O:P:'()*=<:(J/%($93#.9#3%(7?D(3%&791A?$/18$(AG(Protoceratops:(
L??7&$(AG(9/%(Q%0(WA3T(L.7D%2I(AG(-.1%?.%$(=<'()++S;BB:(
N3#$799%'(-:>:'(N%?$A?'(":N:X:'(QA3%&&'(P:L:'(;<)):(J/%(7?79A2I(AG(Dryptosaurus.
aquilunguis(!Y1?A$7#317Z(J/%3A8AD7,(7?D(7(3%U1%0(AG(19$(9I37??A$7#3A1D(
7GG1?191%$:(L2%31.7?(P#$%#2(QAU1979%$(+4)4'()SR+:(
F733'(J:Y:'(61&&172$A?'(J:O:'(N3199'(N:N:'(-97D927?'([:'(;<)):(OU1D%?.%(GA3(/1@/(
97\A?A21.(7?D(2A38/A&A@1.(9I37??A$7#3A1D(D1U%3$19I(1?(9/%(>79%(F3%97.%A#$(
!>79%(F7287?17?,(AG(9/%(L2%31.7?(-A#9/0%$9(7?D(7(?%0($/A39S$T#&&%D(
9I37??A$7#31D(G3A2(9/%([71873A019$(HA32791A?(AG(]97/:(Q79#301$$%?$./7G9%?(
*E'(;=)S;=B:(
F%?$TI'(O:X:'(KAD@%'([:'(Y#D&%I'(X:'()**E:(CU%3S079%3(D1$8%3$7&(AG(&1^73D$(D#%(9A(
/#331.7?%$:(Q79#3%(+*R'(RRBSRRB:(
F/1??%3I'(N:'(>18T7'(J:":'([13T&7?D'(X:'(_7331$/'(X:'(N3%99S-#327?'(P:'()**E:(
Q%A.%379A8$17?(9%%9/(G3A2(9/%(>A0%3(9A(21DD&%(F3%97.%A#$(AG(QA39/(L2%31.7:(
>A0%3(7?D(P1DD&%(F3%97.%A#$(J%33%$9317&(O.A$I$9%2$:(Q%0(P%\1.A(P#$%#2(
AG(Q79#37&(K1$9A3I(7?D(-.1%?.%(N#&&%91?()='(;*4S+<;:(
F/1??%3I'(N:X:'(;<<=:(Y%$.31891A?(AG.Prenoceratops.pieganensis.@%?:(%9($8:(?AU:(
!Y1?A$7#317Z(Q%A.%379A8$17,(G3A2(9/%(J0A(P%D1.1?%(HA32791A?(AG(PA?97?7:(
XA#3?7&(AG(`%39%M379%(_7&%A?9A&A@I(;='(R4;SR*<:(
F/1??%3I'(N:X:'(KA3?%3'(X:":'(;<<4:(L(?%0(?%A.%379A8$17?(D1?A$7#3(&1?T1?@(QA39/(
L2%31.7?(7?D(L$17?(97\7:(XA#3?7&(AG(`%39%M379%(_7&%A?9A&A@I(;4'(B;RSB=):(
Ya(O21.'(P:Y:'(61&$A?'(X:L:'(J/A28$A?'(Y:'(;<)<:(J/%(%?D(AG(9/%($7#3A8AD(D1?A$7#3(
/179#$(1?(QA39/(L2%31.7:(_7&7%A@%A@378/I'(_7&7%A.&1279A&A@I'(_7&7%A%.A&A@I(
;*4'(=EBS=*<:(
Y%?9A?(X3'(":[:'(CaQ%1&&'(":F:'()**E:(Parrisia.neocesariensis'(7(?%0(M7937./A$7#3A1D1D(
$7&727?D%3(7?D(A9/%3(728/1M17?$(G3A2(9/%(F7287?17?(AG(%7$9%3?(QA39/(
L2%31.7:(XA#3?7&(AG(`%39%M379%(_7&%A?9A&A@I()E'(=E=S=*=:(
Y%?9A?'(":'(CaQ%1&&'(":F:'()**R:(Prototeius.stageri'(@%?:(%9($8:(?AU'(7(?%0(9%11D(&1^73D(
G3A2(9/%(]88%3(F3%97.%A#$(P73$/7&&9A0?(HA32791A?(AG(Q%0(X%3$%I'(019/(7(
83%&121?73I(8/I&A@%?%91.(3%U1$1A?(AG(9/%(J%11D7%:(XA#3?7&(AG(`%39%M379%(
_7&%A?9A&A@I()R'(;+RS;R+:(
YAD$A?'(_:'(HA3$9%3'(F:L:'(-728$A?'(-:Y:(;<<=:(F%379A8$1D7%:(b?(6%1$/728%&'(Y:N:'(
YAD$A?'(_:(7?D(C$2A&$T7'(K:(!OD$,:(J/%(D1?A$7#317:(]?1U%3$19I(AG(F7&1GA3?17(
_3%$$'(N%3T%&%I'(88(=*=SR)+:(
cAD%G3A19'(_:'(NA&A9$TI'(W:>:'(>7#9%3$'(_:'(;<);:(L(?%0($7#3A&A8/1?%(D1?A$7#3(G3A2(9/%(
&79%$9(F3%97.%A#$(AG(H73(O7$9%3?("#$$17:(_>A-(CQO(4'(%+BE=*:(
cAD%G3A19'(_:'(>72M%39'(C:'(;<<4:(L(3%S788371$7&(AG(Craspedodon.lonzeensis.YA&&A'()EE+(
G3A2(9/%(]88%3(F3%97.%A#$(AG(N%&@1#2Z(9/%(G13$9(3%.A3D(AG(7(?%A.%379A8$17?(
D1?A$7#3(1?(O#3A8%:(N#&&:(b?$919:("AI:(-.1:(Q79:(N%&@:(-.1:(J%33%(44'(E+S*+:(

cA3DA?'(X:O:()*4E:(-93#.9#3%$Z(A3'(6/I(J/1?@$(YA?a9(H7&&(YA0?:(_&%?#2'(Q%0(WA3T(
+*R(88:(
c37?D$97GG'(N:-:'(_7331$'(Y:F:'(Y%?9A?(X3'(":[:'(c7&&7@/%3'(6:N:'()**;:(Alphadon(
!P73$#817&17,(7?D(P#&919#M%3.#&797(!L&&A9/%317,(1?(9/%(F3%97.%A#$(AG(%7$9%3?(
QA39/(L2%31.7:(XA#3?7&(AG(`%39%M379%(_7&%A?9A&A@I();'(;)4S;;;:(
K7331$'(6:N:'(-%&GSJ371&'(X:P:'(;<<B:(>79%(F3%97.%A#$(M7$%(&%U%&(&A0%31?@(1?(F7287?17?(
7?D(P77$931./917?(D%8A$191A?7&($%d#%?.%$'([#3%(N%7./'(QA39/(F73A&1?7:(
-93791@378/I(+'()*RS;)B:(
K79./%3'(X:N:'(P73$/'(C:F:'(>#&&'(":-:'()*<4:(J/%(F%379A8$17:(]?19%D(-979%$(c%A&A@1.7&(
PA?A@378/(=*'()S+<<:(
KA3?%3'(X:":'(6%1$/728%&'(Y:N:'(HA3$9%3'(F:L:(;<<=:(K7D3A$7#31D7%:(b?(6%1$/728%&'(
Y:N:'(YAD$A?'(_:(7?D(C$2A&$T7'(K:(!OD$,:(J/%(D1?A$7#317:(]?1U%3$19I(AG(
F7&1GA3?17(_3%$$'(N%3T%&%I'(88(=+ES=B+:(
[1%&7?SX70A3A0$T7'(e:'(F1G%&&1'(":>:'(>#A'(e:Sf:(;<<=:(P7227&$(G3A2(9/%(L@%(AG(
Y1?A$7#3$Z(C31@1?$'(OUA&#91A?(7?D(-93#.9#3%:(FA&#2M17(]?1U%3$19I(_3%$$'(Q%0(
WA3T(B=E(88:(
[13T&7?D'(X:b:'(6A&G%'(Y:c:'(;<<):(H13$9(D%G1?191U%(9/%31^1?A$7#31D(!Y1?A$7#3175(
J/%3A8AD7,(G3A2(QA39/(L2%31.7:(XA#3?7&(AG(`%39%M379%(_7&%A?9A&A@I(;)'(=)<S
=)=:(
[#3^7?AU'(-:'()**;:(L(@17?9(83A9A.%379A8$1D(G3A2(9/%(]88%3(F3%97.%A#$(AG(PA?@A&17:(
_7&%A?9A&A@1./%$T11(e/#3?7&(E)S*+:(
>7?@$9A?'(6:'()*B<:(J/%(U%39%M379%(G7#?7(AG(9/%(-%&27(HA32791A?(AG(L&7M727:(_739(
`b:(J/%(D1?A$7#3$:(H1%&D17?7Z(c%A&A@I(P%2A13$(+'(+)RS+B):(
>1?D@3%?'(X:'(F#331%'(_:X:'(-1U%3$A?'(P:'("%%$'(X:'(F%D%3$93g2'(_:'(>1?D@3%?'(H:'(;<<4:(J/%(
G13$9(?%A.%379A8$17?(D1?A$7#3(3%271?$(G3A2(O#3A8%:(_7&7%A?9A&A@I(R<'(*;*S
*+4:(
>A?@31./'(Q:":(;<)<:(J/%(G#?.91A?(AG(&73@%(%I%$(1?(_3A9A.%379A8$Z(7(?A.9#3?7&(
.%379A8$17?h(:(b?(OM%39/'(Y:L:(7?D("I7?'(P:X:(!OD$,:(KA3?%D(Y1?A$7#3$:(b?D17?7(
]?1U%3$19I(_3%$$'(N&AA21?@9A?'(88:(
>A?@31./'(Q:":'(N73?%$'([:'(F&73T'(-:'(P1&&73'(>:'(;<)+:(F7%?7@?79/1D7%(G3A2(9/%(#88%3(
F7287?17?(L@#V7(HA32791A?(AG(0%$9(J%\7$'(7?D(7(3%U1$1A?(AG(9/%(
F7%?7@?79/1?7%:(_%7MADI(P#$%#2(N#&&%91?(R='(;+S=*:(
>A?@31./'(Q:":'(N/#&&73'(L:SN:-:'(c7#9/1%3'(X:'(;<);:(P7$$(%\91?.91A?(AG(&1^73D$(7?D(
$?7T%$(79(9/%(F3%97.%A#$S_7&%A@%?%(MA#?D73I:(_3A.%%D1?@$(AG(9/%(Q791A?7&(
L.7D%2I(AG(-.1%?.%$()<*'(;)+*BSS;)=<):(
>A?@31./'(Q:":'(F#331%'(_:X:'(;<<*7:(Albertonykus.borealis'(7(?%0(7&U73%^$7#3(
!Y1?A$7#317Z(J/%3A8AD7,(G3A2(9/%(O73&I(P77$931./917?(AG(L&M%397'(F7?7D7Z(
b28&1.791A?$(GA3(9/%($I$9%2791.$(7?D(%.A&A@I(AG(9/%(L&U73%^$7#31D7%:(
F3%97.%A#$("%$%73./(+<'(;+*S;R;:(
>A?@31./'(Q:":'(F#331%'(_:X:'(;<<*M:(L(21.3A3789A31?%(!Y1?A$7#317SY3A27%A$7#31D7%,(
G3A2(9/%(>79%(F3%97.%A#$(AG(QA39/(L2%31.7:(_3A.%%D1?@$(AG(9/%(Q791A?7&(
L.7D%2I(AG(-.1%?.%$()<B'(R<<;SR<<E:(
>A?@31./'(Q:":'(-7?T%I'(X:J:'(J7?T%'(Y:K:'(;<)<:(Texacephale.langstoni'(7(?%0(@%?#$(AG(
87./I.%8/7&A$7#31D(!Y1?A$7#317Z(C3?19/1$./17,(G3A2(9/%(#88%3(F7287?17?(
L@#V7(HA32791A?'($A#9/%3?(J%\7$'(]-L:(F3%97.%A#$("%$%73./(+)'(;4=S;E=:(

P7TAU1.TI'(_:X:(;<)<:(L(3%D%$.31891A?(AG(9/%(Montanoceratops.cerorhynchus(/A&A9I8%'(
019/(7(3%U1%0(AG(3%G%33%D(279%317&:(b?("I7?'(P:X:'(F/1??%3I'(N:X:(7?D(OM%39/'(
Y:L:(!OD$,:(Q%0(8%3$8%.91U%$(A?(/A3?%D(D1?A$7#3$Z(J/%("AI7&(JI33%&&(
P#$%#2(F%379A8$17?(-I28A$1#2:(88(BESE;:(
P7TAU1.TI'(_:X:'(QA3%&&'(P:L:'(;<<B:(Yamaceratops.dorngobiensis'(7(?%0(8312191U%(
.%379A8$17?(!Y1?A$7#317Z(C3?19/1$./17,(G3A2(9/%(F3%97.%A#$(AG(PA?@A&17:(
L2%31.7?(P#$%#2(QAU1979%$(+R+<'(=;:(
P7391?'(X:O:'(F7$%'(X:L:'(X7@9'(X:6:P:'(-./#&8'(L:-:'(P#&D%3'(O:6:L:'(;<<R:(L(Q%0(
O#3A8%7?(P73$#817&(b?D1.79%$(7(>79%(F3%97.%A#$(K1@/S>7919#D%(J37?$79&7?91.(
Y1$8%3$7&("A#9%:(XA#3?7&(AG(P7227&17?(OUA&#91A?();'(=*RSR)):(
P73I7?$T7'(J:'(C$2i&$T7'(K:'()*4R:(_3A9A.%379A8$1D7%(!Y1?A$7#317,(AG(L$17:(
_7&7%A?9A&A@17(_A&A?1.7(++'()++S)E):(
P1&&%3'(K:6:'()*B4:(F3%97.%A#$(U%39%M379%$(G3A2(_/A%M#$(>7?D1?@'(QA39/(F73A&1?7:(
_3A.%%D1?@$(AG(9/%(L.7D%2I(AG(Q79#37&(-.1%?.%$(AG(_/1&7D%&8/17(;)*S;+*:(
P1I7$/197'(J:'(F#331%'(_:X:'(F/1??%3ISL&&@%1%3'(N:X:(;<)<:(H13$9(M7$7&(?%A.%379A8$17?(
G3A2(9/%(C&D27?(HA32791A?(!N%&&I("1U%3(c3A#8,'($A#9/%3?(L&M%397:(b?,:(Q%0(
_%3$8%.91U%$(A?(KA3?%D(Y1?A$7#3$Z(J/%("AI7&(JI33%&&(P#$%#2(F%379A8$17?(
-I28A$1#2:(b?D17?7(]?1U%3$19I(_3%$$'(E+(88:(
C$MA3?'(K:H:()*;+:(J0A(&A0%3(F3%97.%A#$(D1?A$7#3$(AG(PA?@A&17:(L2%31.7?(P#$%#2(
AG(Q79#37&(K1$9A3I:(
C$93A2'(X:K:'()*BB:(H#?.91A?7&(2A38/A&A@I(7?D(%UA&#91A?(AG(9/%(.%379A8$17?(
D1?A$7#3$:(OUA&#91A?(;<'(;*<S+<E:(
C99'(F:X:(;<<4:(F37?17&(7?79A2I(7?D(M1A@%A@378/I(AG(9/%(G13$9(Leptoceratops.gracilis(
!Y1?A$7#317Z(C3?19/1$./17,($8%.12%?$(G3A2(9/%(K%&&(F3%%T(HA32791A?'(
$A#9/%7$9(PA?97?7:(b?(F738%?9%3'([:(!%D,:(KA3?$(7?D(N%7T$Z(F%379A8$17?(7?D(
C3?19/A8AD(Y1?A$7#3$:(b?D17?7(]?1U%3$19I(_3%$$'(N&AA21?@9A?'(88(;)+S;++:(
C0%?$'(X:_:'(-A/&'(Q:H:()*E*:(F7287?17?(7?D(P77$931./917?(D%8A$191A?7&($I$9%2$(AG(
9/%(N&7.T(F3%%T(c3A#8(AG(9/%(F73A&1?7$:(QF(c%A&A@1.7&(-#3U%I:(
_A#\'(F:'(F/%U3%9'(_:'(K#./A?'(Y:'(Y%(XA?@'(6:6:'(YA#^%3I'(O:X:'(;<<B:(L331U7&(7?D(
D1U%3$1G1.791A?(AG(.7U1A2A38/(3AD%?9$(7?D(8&79I33/1?%(831279%$(1?(-A#9/(
L2%31.7:(-I$9%2791.(N1A&A@I(RR'(;;ES;==:(
_31%9ASP73d#%^'(L:'(;<)<:(L(3%S%U7&#791A?(AG(Secernosaurus.koerneri.7?D(Kritosaurus.
australis(!Y1?A$7#317'(K7D3A$7#31D7%,(G3A2(9/%(>79%(F3%97.%A#$(AG(L3@%?91?7:(
XA#3?7&(AG(`%39%M379%(_7&%A?9A&A@I(+<'(E)+SE+4:(
_31%9ASPj3d#%^'(L:'(F/1788%'(>:P:'(XA$/1'(-:K:'(;<);:(J/%(>72M%A$7#31?%(Y1?A$7#3(
Magnapaulia.laticaudus(G3A2(9/%(>79%(F3%97.%A#$(AG(N7V7(F7&1GA3?17'(
QA39/0%$9%3?(P%\1.A:(_>A-(CQO(4'(%+E;<4:(
_31%9ASP73d#%^'(L:'(6%1$/728%&'(Y:N:'(KA3?%3'(X:":'(;<<B:(J/%(D1?A$7#3(Hadrosaurus.
foulkii'(G3A2(9/%(F7287?17?(AG(9/%(O7$9(FA7$9(AG(QA39/(L2%31.7'(019/(7(
3%%U7&#791A?(AG(9/%(@%?#$:(L.97(_7&7%A?9A&A@1.7(_A&A?1.7(R)'(44S*E:(
_31%9A‐Pj3d#%^'(L:'(;<)):(F37?17&(7?D(788%?D1.#&73(A?9A@%?I(AG(N7.93A$7#3#$(
VA/?$A?1'(7(/7D3A$7#3A1D(D1?A$7#3(G3A2(9/%(>79%(F3%97.%A#$(AG(?A39/%3?(
F/1?7:(_7&7%A?9A&A@I(R='(44+S4*;:(
"#$$%&&'(Y:L:'()*4;:(C$931./(D1?A$7#3$(G3A2(9/%(>79%(F3%97.%A#$(AG(6%$9%3?(F7?7D7:(
F7?7D17?(XA#3?7&(AG(O739/(-.1%?.%$(*'(+4RS=<;:(

"#$$%&&'(Y:L:'()**+:(J/%(3A&%(AG(F%?937&(L$17(1?(D1?A$7#317?(M1A@%A@378/I:(F7?7D17?(
XA#3?7&(AG(O739/(-.1%?.%$(+<'(;<<;S;<);:(
"I7?'(P:X:'(OU7?$'(Y:F:'(F#331%'(_:X:'(N3A0?'(F:P:'(N31?T27?'(Y:'(;<);:(Q%0(
&%89A.%379A8$1D$(G3A2(9/%(]88%3(F3%97.%A#$(AG(L&M%397'(F7?7D7:(F3%97.%A#$(
"%$%73./(+R'(B*SE<:(
-./0122%3'(Y:":()**4:(>79%(F3%97.%A#$(D1?A$7#3$(1?(%7$9%3?(]-LkL(978/A?A21.(
7?D(M1A@%A@378/1.(2AD%&(AG(A..#33%?.%$:(b?,:(Y1?AG%$9(b?9%3?791A?7&(
-I28A$1#2:(
-%&GSJ371&'(X:P:'(F/31$9A8/%3'(":L:'(_3A0%&&'(Y:F:'(6%%2$'(":O:(;<<=:(J/%(7@%(AG(
D1?A$7#3SM%731?@($93797(79(_/A%M#$(>7?D1?@'(F78%(H%73("1U%3'(QA39/(F73A&1?7:(
b?,:(c%A&A@1.7&(-A.1%9I(AG(L2%31.7(LM$937.9$(019/(_3A@372$:(`A&#2%(+B'())4(
88:(
-31U7$97U7'(-:[:'()*E):(OUA&#91A?(AG(]88%3(F3%97.%A#$(8/I9A@%A83AU1?.%$(7?D(9/%13(
8A&&%?(G&A37:("%U1%0(AG(_7&7%AMA97?I(7?D(_7&I?A&A@I(+R'()RRS)4+:(
-#%$'(K:Y:'(LU%317?AU'(L:'(;<<*:(Turanoceratops.tardabilisl(9/%(G13$9(.%379A8$1D(
D1?A$7#3(G3A2(L$17:(Q79#301$$%?$./7G9%?(*B'(B=RSBR;:(
-#&&1U7?'(":P:'()***:(Nodocephalosaurus.kirtlandensis'(@%?:(%9($8(?AU:'(7(?%0(
7?TI&A$7#31D(D1?A$7#3(!C3?19/1$./17Z(L?TI&A$7#317,(G3A2(9/%(]88%3(
F3%97.%A#$([139&7?D(HA32791A?(!]88%3(F7287?17?,'(-7?(X#7?(N7$1?'(Q%0(
P%\1.A:(XA#3?7&(AG(`%39%M379%(_7&%A?9A&A@I()*'();BS)+*:(
6%1$/728%&'(Y:N:'(N733%99'(_:P:'(FA317'(":L:'(>%(>A%#GG'(X:'(f#'(f:'(e/7A'(f:'(-7/?1'(L:'(
cA27?1'(O:'(QA9A'(F:":(;<<=:(Y1?A$7#3(Y1$931M#91A?:(b?(6%1$/728%&'(Y:N:'(
YAD$A?'(_:(7?D(C$2A&$T7'(K:(!OD$,:(J/%(D1?A$7#317:(]?1U%3$19I(AG(F7&1GA3?17(
_3%$$'(N%3T%&%I'(88(R)4SB<B:(
6A&G%'(Y:c:'([13T&7?D'(X:b:'(-219/'(Y:'(_AA&%'([:'(F/1??%3ISL&&@%13'(N:'(P.YA?7&D'(L:(;<)<:(
Zuniceratops.christopheriZ(9/%(QA39/(L2%31.7?(.%379A8$1D($1$9%3(97\A?(
3%.A?$93#.9%D(A?(9/%(M7$1$(AG(?%0(D797:(b?("I7?'(P:X:'(F/1??%3I'(N:X:(7?D(
OM%39/'(Y:L:(!OD$,:(Q%0(_%3$8%.91U%$(A?(KA3?%D(Y1?A$7#3$Z(J/%("AI7&(JI33%&&(
P#$%#2(F%379A8$17?(-I28A$1#2:(b?D17?7(]?1U%3$19I(_3%$$'(N&AA21?@9A?'(88(
*)S*E:(
Xu,$L.,$Kobayashi,$Y.,$Lü,$J.,$Lee,$W:SQ:'(>1#'(W:'(J7?7T7'([:'(e/7?@'(f:'(X17'(-:'(e/7?@'(X:'(
;<)):(L(?%0(A3?19/A2121D(D1?A$7#3(019/(QA39/(L2%31.7?(7GG1?191%$(G3A2(9/%(
>79%(F3%97.%A#$(m1#87(HA32791A?(1?(K%?7?(_3AU1?.%(AG(F/1?7:(F3%97.%A#$(
"%$%73./(+;'(;)+S;;;:(
f#'(f:'(HA3$9%3'(F:L:'(F&73T'(X:P:'(PA'(X:'(;<<B:(L(M7$7&(.%379A8$17?(019/(937?$191A?7&(
G%79#3%$(G3A2(9/%(>79%(X#37$$1.(AG(?A39/0%$9%3?(F/1?7:(_3A.%%D1?@$(AG(9/%(
"AI7&(-A.1%9I(NZ(N1A&A@1.7&(-.1%?.%$(;4+'(;)+RS;)=<:(
f#'(f:'(P7TAU1.TI'(_:X:'(67?@'(f:S&:'(QA3%&&'(P:L:'(WA#'(K:S&:'(;<<;:(L(.%379A8$17?(
D1?A$7#3(G3A2(F/1?7(7?D(9/%(%73&I(%UA&#91A?(AG(F%379A8$17:(Q79#3%(=)B'(+)=S
+)4:(
f#'(f:'(67?@'([:'(e/7A'(f:'(-#&&1U7?'(F:'(F/%?'(-:'(;<)<7:(L(?%0(&%89A.%379A8$1D(
!C3?19/1$./17Z(F%379A8$17,(G3A2(9/%(]88%3(F3%97.%A#$(AG(-/7?DA?@'(F/1?7(7?D(
19$(128&1.791A?$(GA3(?%A.%379A8$17?(%UA&#91A?:(_>A-(CQO(R'(%)+E+R:(
f#'(f:'(67?@'([:SN:'(e/7A'(f:SX:'(>1'(Y:SX:'(;<)<M:(H13$9(.%379A8$1D(D1?A$7#3(G3A2(F/1?7(
7?D(19$(M1A@%A@378/1.7&(128&1.791A?$:(F/1?%$%(-.1%?.%(N#&&%91?(RR'()B+)S)B+R:(

WA#'(K:'(YAD$A?'(_:(;<<=:(N7$7&(F%379A8$17:(b?(6%1$/728%&'(Y:N:'(YAD$A?'(_:(7?D(
C$2A&$T7'(K:(!OD$,:(J/%(D1?A$7#317:(]?1U%3$19I(AG(F7&1GA3?17(_3%$$'(N%3T%&%I'(
88(=4ES=*+:(
WA#'(K:S>:'(PA3$.//7#$%3'(O:'(YAD$A?'(_:'(>1'(Y:Sm:'(;<);:(Auroraceratops($8:(
!Y1?A$7#317ZQ%A.%379A8$17,(G3A2(9/%(O73&I(F3%97.%A#$(AG(9/%(P7^A?@$/7?(
73%7(1?(QA39/0%$9%3?(F/1?7:(`%39%M3797(_7&L$1791.7(R<'()4<S)E<:(

Figure Captions
Fig. 1. Distribution of Leptoceratopsidae and possible dispersal routes: (1),
Udanoceratops and Zhuchengceratops (Eastern Asia) (Kurzanov, 1992; Xu et al., 2010a),
(2), Leptoceratops, Montanoceratops, Gryphoceratops, Unescoceratops and cf.
Prenoceratops (Alberta) (Brown, 1914; Makovicky, 2010; Miyashita et al., 2010; Ryan et
al., 2012); (3) Montanoceratops, Prenoceratops, Cerasinops, and Leptoceratops (Brown,
1942; Chinnery and Horner, 2007; Ott, 2007) !P1I7$/197( %9( 7&:'( ;<)<, (Montana); (4)
Black Creek ceratopsian (this paper); (5) Kristianstaad ceratopsian (Sweden) (Lindgren et
al., 2007); (6) Craspedodon lonzeensis (Belgium) (Godefroit and Lambert, 2007).

Fig. 2. Map showing outcrops of Upper Cretaceous Black Creek Group and Peedee
Formation strata, and the locality of the Black Creek ceratopsian. The specimen comes
from the Clifton Farm locality, south of Giddensville, Sampson County, N 35.13, W
78.22. The Lower Campanian Tar Heel Formation outcrops in this area. Map after Owens
and Sohl !C0%?$( 7?D( -A/&'( )*E*,5( $93791@378/1.( .A&#2?( 7G9%3( K7331$( 7?D( -%&GSJ371&(
!K7331$(7?D(-%&GSJ371&'(;<<B,:(QA9%(9/79(9/%(8A$191A?(AG(9/%( $8%.12%?(1?(9/%(J73(K%%&(
1$(.#33%?9&I(#?.A?$9371?%D:

Fig. 3, Black Creek ceratopsian, YPM-PU 24964, left maxilla. A, medial, B, ventral, C,
lateral, D, dorsal view. Abbreviations: ag, alveolar groove; dp, dentigerous process.

Fig. 4, A, NMC 8889, Leptoceratops gracilis; B1, divergent posterior dentigerous
process of Leptoceratops; B2 posterior dentigerous process of Black Creek Ceratopsian
YPM-PU 24964, showing the more highly divergent process versus Leptoceratops
(dashed).

Fig. 5, A, AMNH 5205 Leptoceratops gracilis posterior dentigerous process (reversed
for comparison) showing alveolar ridges and anteroposteriorly compressed tooth sockets.
B, alveolar ridges of YPM-PU 24964.

Fig. 6. Summary figure showing distribution of dinosaurs in Appalachia. 1, Missouri,
Hadrosauridae; 2, Tennessee, Hadrosauridae; 3, Mississippi, Hadrosauridae,
Tyrannosauroidea, and Ornithimimosauria, 4, Alabama, Hadrosauridae,
Ornithomimosauria, and Nodosauridae; 5, Georgia, Hadrosauridae, Tyrannosauroidea,
and Ornithomimosauria; 6, South Carolina, Hadrosauridae, 7, North Carolina,
Hadrosauridae, Tyrannosauroidea, and Lepticeratopsidae 8, Maryland, Hadrosauridae and
Ornithomimosauria; 9, Delaware, Hadrosauridae and Ornithomimosauria; Hadrosauridae
and Ornithomimosauria, 10, New Jersey, Hadrosauridae, Tyrannosauroidea,
Ornithomimosauria, Nodosauridae. Map after Schwimmer !-./0122%3'()**4,(019/(
D797(G3A2(-./0122%3(!-./0122%3'()**4,(7?D(6%1$/728%&(%9(7&:(!6%1$/728%&(%9(7&:'(
;<<=,:(J/%(A..#33%?.%$(%\9%?D(G3A2(9/%(%73&I(-7?9A?17?(9A(9/%(&79%(P77$931./917?'(7(
8%31AD(AG(73A#?D(;<(21&&1A?(I%73$:(