Cretaceous Research 128 (2021) 104994

Contents lists available at ScienceDirect

Cretaceous Research
journal homepage: www.elsevier.com/locate/CretRes

The scaly skin of the abelisaurid Carnotaurus sastrei (Theropoda:

Ceratosauria) from the Upper Cretaceous of Patagonia
Christophe Hendrickx a, *, Phil R. Bell b
a
n 4000, Tucuma

n Miguel Lillo, Miguel Lillo 251, San Miguel de Tucuma
Unidad Ejecutora Lillo, CONICET-Fundacio
n, Argentina
b
School of Environmental and Rural Science, University of New England, Armidale, NSW 2351, Australia

a r t i c l e i n f o a b s t r a c t

Article history: The integument of the theropod dinosaur Carnotaurus sastrei from the Upper Cretaceous of Argentina is
Received 20 January 2021 here described in detail for the first time. The scaly skin of this abelisaurid is the most completely
Received in revised form preserved of any theropod and the only example of this form of integument known outside of Tetanurae
15 July 2021
(excluding footprints). The skin is preserved in the shoulder, thoracic, tail and, possibly, neck regions and
Accepted in revised form 8 August 2021
Available online 13 August 2021
consists of medium to large (20e65 mm in diameter) conical feature scales surrounded by a network of
low and small (<14 mm) non-imbricating basement scales separated by narrow interstitial tissue.
Contrary to previous interpretations, the feature scales are randomly distributed and neither form
Keywords:
Integument
discrete rows nor show progressive variations in their size along parts of the body. They also show little
Scales difference in morphology along the body, although their apices are variously positioned in different body
Abelisauridae parts. Conversely, the basement scales vary from small and elongated, large and polygonal, and circular-
Non-avian Theropoda to-lenticular in the thoracic, scapular, and tail regions, respectively. Given the presumed active lifestyle of
Skin Carnotaurus and the necessity of shedding excess heat, particularly at large body sizes (>1000 kg), we
speculate that the skin may have played a vital role in thermoregulation; a role consistent with integ-
ument function in extant mammals and reptiles.
© 2021 Elsevier Ltd. All rights reserved.

THE 8th PALEONTOLOGICAL Expedition to Patagonia in 1984, led specimen of Carnotaurus, the first body skin to be found in a non-
by famous Argentine palaeontologist Jose
Fernando Bonaparte (14 avian theropod (Czerkas & Czerkas 1989). The holotype of Carno-
June 1928e18 February 2020; Giacchino & Agnolin 2020) and taurus was discovered one year before the 1984 expedition by a
sponsored by the National Geographic Society, marked a turning member of the Sastre family who shared the finding with geologists
point in South American palaeontology. Between the 6th and 29th from the Direccio
n Nacional de Minería who were working in the
of February 1984, Bonaparte and his team (including technician area (Bonaparte 1996a). Because such a discovery was completely
Orlando A. Gutie
rrez and then-undergraduate student Guillermo unexpected at the time, the fossilized integument was neither
W. Rougier) unearthed from two distantly separated localities two recognized by the geologists in 1983 nor by Bonaparte and his team
of the most iconic South American dinosaurs: the spiny dicraeo- during the excavation of the holotype the following year. Skin from
saurid sauropod Amargasaurus cazaui Salgado & Bonaparte 1991 the anterior portion of the tail (Fig. 2F) and near the back of the
from the Barremian‒Aptian La Amarga Formation in Neuque
n lower jaw were not identified until preparation of the specimen at
Province, and the horned abelisaurid theropod Carnotaurus sastrei the Museo Argentino de Ciencias Naturales (MACN) in Buenos Aires
Bonaparte 1985 from the Maastrichtian La Colonia Formation in between 1985 and 1986, although large portions of skin were un-
neighbouring Chubut Province (Bonaparte et al. 1990; Salgado & doubtedly also lost during this period (Czerkas & Czerkas 1997;
Bonaparte 1991; A. Martinelli, pers. comm. 2020; Fig. 1B-D; Guillermo W. Rougier, pers. comm. 2020).
Appendix A1, figure A1A‒C). The expedition in Chubut was pivotal The presence of skin in Carnotaurus was mentioned by Bona-
for its discovery of skin associated with the exquisitely preserved parte during a lecture at the Museum of Natural History of Los
Angeles, California, attended by palaeoartists Stephen A. and Sylvia
J. Czerkas in 1986 (Czerkas & Czerkas 1989). The importance of this
* Corresponding author. discovery led to a second expedition organized by Bonaparte, with
E-mail addresses: [email protected] (C. Hendrickx), pbell23@une. the participation of Rodolfo Coria and the Czerkas’, to the original
edu.au (P.R. Bell).

https://doi.org/10.1016/j.cretres.2021.104994
0195-6671/© 2021 Elsevier Ltd. All rights reserved.
C. Hendrickx and P.R. Bell Cretaceous Research 128 (2021) 104994

excavation site in February 1988 (Czerkas & Czerkas 1989, 1997; Carnotaurus and illustrations of each part of the holotype preser-
Lessem 1989; Fig. 1A; Appendix A1, figure A1C). Additional frag- ving skin is, therefore, provided here for the first time. This detailed
ments of fossil skin associated with bone fragments or impressions treatment of the Carnotaurus integument adds to those recently
of bone that had been removed were collected from the neck, provided for the skull (Cerroni et al. 2020), braincase (Carabajal
shoulder, and rib cage, mainly during the first week of the three 2011; Cerroni & Paulina-Carabajal 2019), cervical vertebrae
week expedition (Lessem 1989; Czerkas & Czerkas 1997; Fig. 2B-E). (Me
ndez, 2012), hand anatomy (Ruiz et al. 2011) and tail myology
The various specimens led to a life-sized restoration of Carnotaurus (Persons & Currie 2011) of this iconic abelisaurid theropod.
commissioned by the Natural History Museum of Los Angeles and Institutional abbreviations. HMNS, Houston Museum of Natural
created by Stephen A. and Sylvia J. Czerkas during the mid-1980s Science, Houston, Texas, USA; JME, Jura Museum Eichsta €tt,
(Fig. 3C; Appendix A1, figures A2‒A3). This sculpture probably Eichst€
att, Germany; MACN, Museo Argentino de Ciencias Naturales
represents the first accurate life restoration of a non-avian ‘Bernardino Rivadavia,’ Buenos Aires, Argentina; MCF-PVPH, Museo
theropod in which the integument was correctly shown (Glut Municipal ‘Carmen Fun ~ es,’ Plaza Huincul, Argentina; ML, Museu da
2000). Lourinha ~, Lourinha ~, Portugal; MPEF-PV, Museo Paleontolo
gico
Carnotaurus sastrei is one of most complete and best preserved ‘Egidio Feruglio,’ Trelew, Argentina; NHMUK PV, Natural History
theropods from Gondwana (Bonaparte 1996b; Ezcurra and Novas, Museum, London, UK; TMP, Royal Tyrrell Museum of Palaeontology,
2016). This large-bodied abelisaurid (~7.8 m in body length, and a Drumheller, Alberta, Canada; UMNH VP, Natural History Museum of
body mass around 1.5e1.7 tonnes; Mazzetta et al. 2004; Benson et Utah, University of Utah, Salt Lake City, USA; ZDM, Zigong Dino-
al. 2014; Campione et al. 2014; Grillo & Delcourt 2017) is charac- saurian Museum, Zigong, Sichuan, China.
terized by a short-snouted and tall cranium adorned with two
prominent frontal horns as well as atrophied forelimbs with four 1. Material and methods
reduced digits (Bonaparte et al. 1990; Ruiz et al. 2011; Cerroni et al.
2020; Fig. 2). The preserved skin of Carnotaurus is unique in being Skin specimens associated with the holotype skeleton of Car-
the only known integument for the whole ceratosaur clade (the notaurus sastrei are deposited in the palaeontological collections of
osteoderms of Ceratosaurus notwithstanding) as well as the only the Museo Argentino de Ciencias Naturales in Buenos Aires (MACN-
scaly skin known from the body other than the crus in a non- CH 894). Although samples of the Carnotaurus skin appear to have
tetanuran theropod (Barrett et al. 2015; Hendrickx et al. in press). left Argentina for analysis during the 1980's (Lessem 1989), all skin
The skin preserved at the level of the tail (Fig. 2F) is also one of the specimens are currently deposited in the collections of the MACN
largest patches of scaly skin available in saurischian dinosaur. and only casts and moulds from the original material are on display
Despite its importance for our understanding of the skin at The Dinosaur Museum in Blanding, Utah (Sylvia Czerkas, pers.
morphology in dinosaurs, the integument of Carnotaurus has only comm. 2015).
been briefly described and illustrated (Bonaparte et al. 1990; The skeleton and associated skin were found in a hematite
Bonaparte 1996a; Czerkas 1997; Czerkas & Czerkas 1997; concretion (Appendix A1, figure A1B) from the Maastrichtian
Hendrickx et al. in press). None of these authors described the skin (Rauhut et al. 2003; Novas et al. 2013; Vera et al. 2019; Clyde et al.
from the different body regions of Carnotaurus comprehensively, 2021) La Colonia Formation at the Estancia “Pocho Sastre” in Bajada
nor provided photos of every specimens preserving skin. In addi- Moreno, Chubut Province, Argentina (43.0 S, 67.5 W; paleo-
tion, a detailed comparison with the skin of other dinosaurs re- coordinates 45.7 S, 53.8 W; The Paleobiology Database;
mains to be given. A comprehensive description of the skin of Bonaparte et al. 1990; Bonaparte 1996b; Fig. 1B-D). The holotype of

Fig. 1. View and geographic location of the Bajado Moreno excavation site, and life-size restoration of Carnotaurus by the Czerkas. (A) Stephen A. Czerkas (far right) looking for skin
impressions at the Carnotaurus excavation site in February 1988. Map of Chubut Province in Argentina (B) showing the La Colonia Formation area (C) and the Bajado Moreno locality
(D) from where the holotype of Carnotaurus was unearthed. (E) Stephen A. Czerkas measuring the head of his restoration of Carnotaurus. Illustrations in BeD are from Cúneo et al.
(2014; modified) and A and E are from Arnold (1993, © Clarions Books, Houghton Mifflin Harcourt Books; used with permission).

2
C. Hendrickx and P.R. Bell Cretaceous Research 128 (2021) 104994

Carnotaurus comes from the second facies association of the La the terminology provided by Bell (2012) and Hendrickx et al. (in
Colonia Formation, whose lithology is defined by laminated mud- press).
stones, pelite-fine sandstones, and banded siltstones and clay-
stones (Pascual et al. 2000). This facies association was interpreted 2. Results
by Pascual et al. (2000, p. 399) as “an estuary, tidal flat or coastal
plain environment influenced both by occasional high freshwater The preserved skin of Carnotaurus sastrei comes from the right
stream flow from the continent and tidal currents from the sea”. side of the animal (Bonaparte et al. 1990; Czerkas & Czerkas 1997).
The La Colonia Formation fauna includes mammals, ophidians, Czerkas and Czerkas (1997, p. 155) wrote that the skeleton of Car-
anurans, turtles, plesiosaurs, and diverse dinosaurs including, notaurus was articulated in a typical death pose with the head
hadrosauroids, ankylosaurs, titanosaur sauropods, and a second pulled back and was “lying upon its right side which is were the
indeterminate taxon of abelisaurid theropod (e.g., Gasparini 2000; skin was preserved”. It is possible that the ‘mummified’ carcass was
Pascual et al. 2000; Rougier et al. 2009; O'Gorman et al. 2013; preserved entirely within the hematite concretion, which limited
Gasparini et al. 2015). weathering of the well-preserved remains (Bonaparte 1996b; Fer-
The original specimens along with two additional casts of the nando Novas, pers. comm. 2020).
skin were examined first hand and imaged with a digital Canon According to Lessem (1989, p. 42), the “skin from essentially the
PowerShot SX60 camera (Fig. 2B‒G; Appendix A1, figures A4‒A12). entire length of the animal, even the skull” was found by Stephen A.
A three-dimensional (3D) model of the large patch of skin from the Czerkas during the first week of the second expedition. The number
anterior portion of the tail (Appendix A1, figures A13‒A14) was of specimens with skin, however, appears to be limited to six frag-
generated using photogrammetric data and the software Agisoft ments, and none are from the skull. Bonaparte et al. (1990, p. 32)
Photoscan 1.3.4 using the protocol detailed by Hendrickx and Bell stated that “one fragment [of skin] is from the anterior cervical re-
(2021). The 3D model was exported, oriented, and scaled in gion, and it is associated with the anterior cervical ribs. Another
Meshlab version 1.3.4BETA (Cignoni et al. 2008) and deposited in fragment is from the scapular area near the glenoid cavity. Two
MorphoMuseuM (https://morphomuseum.com/) where it is freely fragments are from the thoracic region, and they are associated with
downloadable (see Hendrickx & Bell 2021). The morphology of the the mid-proximal area of the ribs. The largest available skin
Carnotaurus scaly skin was compared to that of more than 30 impression corresponds to the lower area of the proximal part of the
dinosaur taxa deposited in scientific collections of France, Germany, tail.”. Specimens from each of these regions were identified in the
Switzerland, Portugal, Spain, Argentina, USA, Canada, and China collection of the MACN (Fig. 2), yet one specimen (Fig. 2D) is assumed
based on personal examination of the specimens, high-resolution to be from the thoracic region based on the statement that two
photographs provided by colleagues, and the primarily literature fragments from this region were found. Likewise, the specimen
(Appendix A2). The integument of live Komodo dragons (Varanus illustrated in Bonaparte et al.’s (1990) figure 37B (Fig. 2G), which did
komodoensis) was observed on captive individuals at Australia Zoo not specify the location on the body in the caption, could not be
(Beerwah, Queensland). The anatomical nomenclature used to positioned with certainty. This specimen is assumed to be from the
describe and annotate the epidermal scales of Carnotaurus follows tail based on the morphology and details of the skin (Fig. 2G). All but

Fig. 2. Location of the specimens with skin (MACN-CH 894) on the body of an accurate reconstruction of Carnotaurus sastrei based on this study. (A) Cranium shortly after being
excavated. (B) Possible skin associated with two cervical ribs from the cervical region (courtesy of Mauricio A. Cerroni; used with permission). (C) Mould with skin from the glenoid
area of the scapular region. (D) Skin possibly from the thoracic region. (E) Skin associated with a dorsal rib from the thoracic region. (F) Large patch of skin associated with three
caudal vertebrae from the anterior tail region. (G) Skin possibly from the tail region. 3D model of Carnotaurus by © Jake Baardse, used with permission.

3
C. Hendrickx and P.R. Bell Cretaceous Research 128 (2021) 104994

one specimen are originals (Appendix A1, figures A4‒A12); the skin (Fig. 3A-B) do not show any obvious pattern of scales. Large por-
from the scapular region being a cast made at the level of the glenoid tions of matrix appear to cover most of the cranium ventral to the
cavity of the scapulocoracoid (Fig. 2C). A cast of the skin from the orbit and the naris, but the skull roof and dorsal surface of the
central part of the large specimen from the anterior tail region was snout are apparently matrix free. Although these photos were
also made. taken prior to preparation (Fig. 3A-B), no unequivocal sca-
lesdincluding feature scales which, if present, should be
2.1. Head obviousdare discernable. Feature scales were, however, said to be
present on the skull by Bonaparte (1996a, b) and Czerkas (1997),
No specimen with skin from the cranium is currently available. and a short row of feature scales were added to the posterior part
A small piece of skin located near the back of the lower jaw and of both upper and lower jaws, at the level of the quadratojugal and
found during the first expedition in 1984 (Czerkas & Czerkas 1997) articular, respectively, in the Czerkas' life restoration of Carno-
appears to be lost. According to Czerkas and Czerkas (1997, p. 155), taurus (Figs. 1E, 3C). As no fossils or supporting photos exist, the
“extensive areas of skin impressions had been preserved over the presence of feature scales on Carnotaurus' head should be seen as
right side of the skull but were largely lost during preparation”. hypothetical. Likewise, if the photos from Bonaparte's (1996a)
These authors also noted that the surface texture of the right side book do indeed show skin covering the right side of the cranium
of the skull is significantly different from the left side and that the of Carnotaurus, the absence of discernable scales on these photos
rugosities observed on the right surface probably reflect the would suggest that particularly small basement scales covered the
presence of tuberculate basement scales. Photographs of the right face.
side of the cranium taken in the field shortly after its extraction in Epidermal structures covering the head of Carnotaurus were
Bonaparte's (1996b, p. 111) book “Dinosaurios de Ame
rica del Sur” recently inferred by Delcourt (2018) and Cerroni et al. (2020) based

Fig. 3. Cranium of Carnotaurus (MACN-CH 894) and life-sized restoration of Carnotaurus' head. (A) Technician Orlando A. Gutie
rrez and (B) then-undergraduate student Guillermo
W. Rougier posing next to the cranium of Carnotaurus shortly after being extracted (from Bonaparte (1996a); modified). The dark material associated with the rostrum may be the
remnants of a cornified pad. (C) Life-size restoration of the head of Carnotaurus by Stephen A. and Sylvia J. Czerkas at the Los Angeles Natural History Museum in 2006 and showing
the feature scales in the posterior part of the upper and lower jaws (© Kolby; downloaded on Flickr; CC BY-NC 2.0).

4
C. Hendrickx and P.R. Bell Cretaceous Research 128 (2021) 104994

on the different zones of bony sculpturing that reflect the miner- Skin from the shoulder region is known from a mould of the
alisation of the underlying dermis and periosteal bone (Sampson & scapulocoracoid around the glenoid cavity (Fig. 7A-B). The skin
Witmer 2007; Hieronymus et al. 2009). The pattern of hummocky from this specimen, the first to be found during the second expe-
rugosities, the density of longitudinal grooves, and intense pitting dition (Czerkas & Czerkas 1989), covers an area of ~135 cm2 over-
on the premaxilla, maxilla, some parts of the dentary, lacrimal, and lying the scapular blade and glenoid cavity of the scapulocoracoid.
postorbital were correlated to a scaly integument such as flat scales Two incomplete yet well-defined feature scales are present, one
on the snoutdsimilar to those inferred to be present in the directly posterior to the glenoid cavity and the second at the level of
tyrannosaurid Daspletosaurus (Carr et al. 2017)d, and large scales the posteriormost portion of the coracoid blade (Fig. 7C). The two
near the orbit (Cerroni et al. 2020). Likewise, the presence of feature scales are ~65 mm in diameter and are separated by
scattered ridges and absence of rugosities, grooves, or pitting on the ~20 mm. Although incomplete, the mould of these feature scales
lateral surface of the jugal also led Cerroni et al. (2020) to infer the reveals that their apices were off-centre and deflected ante-
presence of larger epidermal scales or a small cornual process of the rodorsally (Fig. 7C). Two clusters of a few poorly-defined basement
jugal. Conversely, extensive papillary surfaces over the nasal pro- scales are seen overlying the scapular blade of the scapulocoracoid,
cess of the premaxillae, the dorsal surface and anterior process of directly ventral to the glenoid fossa (Fig. 7D) and anterior to the
the nasals, as well as the dorsolateral processes of the lacrimals and aforementioned feature scales (Fig. 7E). Although they are partic-
postorbitals were correlated with the presence of a cornified pad ularly poorly defined, the basement scales from those two clusters
(Delcourt 2018; Cerroni et al. 2020). Interestingly, dark rocky are polygonal (Fig. 7C). Adjacent to the glenoid fossa, they are
structures covering the nasal and premaxillae are visible in the field somewhat diamond-shaped and approximately 9  7 mm in size
photos from Bonaparte's (1996a) book (Fig. 3A-B). These dark fea- (Fig. 7D), whereas the largest scales from the ventral cluster are
tures appear to differ from the light grey matrix superposing them 13  7 mm and form elongated hexagons (Fig. 7E). Although the
(and also present on the ventral surface of the cranium) and might basement scales from the two clusters are oriented in the same way
in fact correspond to the inferred cornified cover on the nasals and share the same polygonal outline, their morphology is only
(Bonaparte 1996b; Delcourt 2018; Cerroni et al. 2020). If true, this revealed in a certain direction of light and should therefore be seen
would indicate that cornified tissues extended up to the anterior tip as tentative.
of the snout and covered most of the surface of the premaxillae.
2.3. Tail

2.2. Neck, shoulder and thorax The largest patch of skin comes from the anterior portion of the
tail (Fig. 8A). This specimen and an additional small fragment with
The supposed skin from the anterior cervical region and asso- skin assumed to be from the same tail region (see rationale below),
ciated with two anterior cervical ribs does not show any particular provide the best information on the integumentary architecture of
pattern (Fig. 4A-B). The skin patch is found mainly dorsal to the Carnotaurus. Definition of the basement and feature scales is best
second (as preserved) of the two cervical ribs and covers an area of preserved in the largest fragment where the total integumentary
~80 cm2 (16  7 cm for the longest length and width on the pre- covering (a natural cast; negative epirelief) measures >1200 cm2
served portion of the rock fragment). No basement or feature scales (Figs. 8B-C, 9). Basement scales form poorly-defined ellipses
could be identified despite the fact that feature scales were said to (4.4  3.1 mm in size but up to 5.7  5.1 mm; Table 1) that gradually
be present over the neck by Czerkas (1997; Fig. 4A-B). It is unknown merge with the surrounding interstitial tissue, giving the skin an
whether this specimen preserves scaleless skin or some other soft overall ‘hummocky’ appearance (Fig. 8D). The basement scales
tissue. strongly vary in size and shape but do not form cluster areas (sensu
Conversely, two fragments from the thoracic region, one asso- Bell 2012) or rosettes as was previously suggested (Czerkas &
ciated with a large mid-proximal portion of a thoracic rib, clearly Czerkas 1997; Fig. 8F). Instead, groups of basement scales are par-
show basement and feature scales (Figs 5-6). The first fragment titioned by sub-parallel vertical ridges (forming furrows in life) of
directly associated with the rib (Fig. 5A) shows two incomplete interstitial tissue ~15e17 mm apart (Figs. 8D-9; Table 1). Between
feature scales ~10 cm apart anterior and posterior to the rib one and four basement scales may sit between the vertical ridges,
(Fig. 5B-C). Their dorsoventral diameter ranges from 52 to 58 mm. the long axes of which are roughly parallel to the ridges. A latex
The anteroposterior radius of the anterior and best-preserved peel taken from the portion with the best-preserved skin (Fig. 10A-
feature scale is 28 mm and its apex appears to be centrally posi- B) shows that the basement scales form convex non-imbricating
tioned (Fig 5A-B). None of the two feature scales show any partic- tubercles, which are elliptical to lenticular in outline, and sepa-
ular surface ornamentation, such as corrugations. The skin surface rated by relatively wide interstitial tissue (Fig. 10C). The largest
surrounding the feature scales, which covers an area of ~220 cm2 basement scales are typically dorsoventrally taller than wide
(16  14 cm), does not show any pattern and no basement scales whereas the smallest basement scales tend to be more subcircular.
can be identified (Fig. 5B-C). This differs from the second fragment Feature scales form blunt cones ~20e50 mm in diameter and are
in which several clusters of basement scales can be seen (Fig. 6A-C). spaced ~20e50 mm apart (Figs. 8E, 9, 10C). The largest and best-
The patch of skin covers an area of ~192 cm2 (16  12 cm) and two preserved feature scales are ~10 mm in height, although Czerkas
incomplete feature scales ~45 mm apart appear to be present at the & Czerkas (1997) speculated they may have been taller
edge of the sample (Fig. 6B). One seems to be larger (38 mm in (30e50 mm in height) in life; the reasoning behind such specula-
diameter) than the other one (22 mm in diameter), the latter of tion was not stated. Their surfaces are rather irregular-to-smooth
which is, however, particularly incomplete. It is unknown whether but otherwise unornamented, as clearly seen on the latex mould
their apices were centred or not. The basement scales on this (Fig. 10C). Interestingly, the apex of each feature scale from the tail
specimen are poorly defined and difficult to identify. Some of them, is off-centre and posterodorsally deflected (Figs. 8E, 10C). The latex
however, seem to be oblong and significantly longer than wide peel and 3D model of the skin additionaly show that the feature
(5.3  2.6 mm), therefore showing a clear polarity (Fig. 6B-C). These scales vary morphologically, some of them having a prominent and
oblong scales appear relatively uniform and all oriented in the same pointed apex whereas other are simple flared domes (Appendix A1,
direction (Fig. 6C). The orientation of the block, however, is un- figures A13‒A14). The external limit of the feature scales is often
known and therefore the direction of polarity undeterminable. difficult to identify given that the surrounding basement scales are
5
C. Hendrickx and P.R. Bell Cretaceous Research 128 (2021) 104994

Fig. 4. (AeB) Possible integument in positive epirelief from the cervical region of Carnotaurus sastrei (MACN-CH 894), with (B) close up of the presumed integument dorsal to an
anterior cervical rib. Abbreviation: acr, anterior cervical rib. Scale bar in B represents: 3 cm.

poorly defined and often appear to merge with the margin of the in diameter) on the tail is flanked by two feature scales that are
feature scales. These peripheral basement scales are numerous, twice as large (Fig. 8C). Likewise, there is no evidence that the
randomly distributed, variable in shape and size, and do not form a feature scales progressively changed in size either ante-
wall bounding the feature scale. The peripheral surface of the roposteriorly or dorsoventrally along the tail (Table 1).
feature scales can be flat and elevated from the rest of the basement The morphology of the basement and feature scales from the
scales or slightly concave and lower than the peripheral basement smaller specimen (positive epirelief; Fig. 11A) is almost the same as
scales. Not enough of the skin is preserved to identify whether that of the large patch of skin from the tail. This specimen is,
these feature scales were distributed randomly or in distinct pat- however, not a natural cast of the large block. Although its precise
terns as has been previously suggested (Czerkas & Czerkas 1997); position on the body is unknown, it is regarded here as likely
however, it is clear that the distribution of feature scales of different derived from the tail given its similarity to the aforementioned
sizes is random (Fig. 8B). Indeed, the smallest feature scale (20 mm large tail block. The skin on the small block (17.5  14.5 mm;
6
C. Hendrickx and P.R. Bell Cretaceous Research 128 (2021) 104994

Fig. 5. Skin from the thoracic region. (A) Natural mould (negative epirelief) of skin from the right thoracic region of Carnotaurus sastrei (MACN-CH 894) showing the feature scales in
dotted lines with the apex marked by a white cross. (BeC) Close up of the anterior (B) and posterior (C) feature scales. Abbreviations: fes, feature scale; thr, thoracic rib.

~220 cm2) preserves three feature scales in a rough line spaced 52 numerous and poorly defined peripheral basement scales. None of
and 20 mm apart (Fig. 11A). The feature scales are somewhat the feature scales, however, show a concave peripheral surface
elliptical; the diameters of the largest and smallest feature scales lower than the basement scales in this specimen. The feature scales
are 52  44 mm and 44  35 mm, respectively. As in the large patch also have a strongly deflected apex oriented in the same (unknown)
of skin, they are not well-delimited and are surrounded by direction, which would be posterodorsally if this small block comes

7
C. Hendrickx and P.R. Bell Cretaceous Research 128 (2021) 104994

Fig. 6. Skin from the? thoracic region. (A) Natural mould (negative epirelief) of skin from the right? thoracic region of Carnotaurus sastrei (MACN-CH 894). (BeC) Close up of the
feature scales (dotted lines) (B) and oblong basement scales (C). Abbreviations: bas, basement scales; fes, feature scale.

from the anterior tail region. The apical region of the largest feature tissue and many of them are poorly-defined. They differ from the
scale shows three short and poorly delimited parallel ridges, which basement scales of the large patch of skin from the tail in being
may be ornamentation or taphonomic in origin (Fig. 11B). Basement much larger (11 mm in their longest dimension), in having a weakly
scales, well-visible in this specimen, are once again non-uniform in granular surface texture for the largest ones, and in being more
shape and size (Fig. 11C). They are separated by narrow interstitial irregular in size over a small area, with the largest basement scales

8
C. Hendrickx and P.R. Bell Cretaceous Research 128 (2021) 104994

Fig. 7. Skin from the scapular region. (A) Latex mould of skin in negative relief from the right scapular region of Carnotaurus sastrei (MACN-CH 894), with outline of the scap-
ulocoracoid in grey. (B) Region of the scapulocoracoid from which the mould was made (photo courtesy of Rafael Delcourt, used with permission; modified). (CeE) Close up of the
feature scales (dotted lines) with their apices marked by a white cross (C) and diamond-shaped and hexagonal basement scales ventral to the glenoid fossa (D) and anterior to the
ventral feature scale (E). Abbreviations: bas, basement scale; fes, feature scales; gl, glenoid fossa; sad, subacromial depression. Scale bars represent: 10 cm (B), 5 cm (C), 2 cm (DeE).

9
C. Hendrickx and P.R. Bell Cretaceous Research 128 (2021) 104994

Fig. 8. Natural mould (negative epirelief) of skin from the right side of the anterior tail region of Carnotaurus sastrei (MACN-CH 894). (A) Position of the specimen on the skeleton of
Carnotaurus (© Scott Hartman; used with permission). (B, C) Natural mould of the large patch of skin with feature scales in dotted lines. (DeF) Close ups of the skin showing the area
of vertical ridges (D), the largest feature scale (central feature scale from the dorsal row) (E), and basement scales (F). Abbreviations: fes, feature scale; hae, haemal arch; isc,
irregular basement scales; tvp, transverse process of the anterior caudal vertebra; ver, vertical ridges. Scale bars represent: 2 cm (DeE), 1 cm (F).

being often adjacent to particularly small ones. The basement scales representing the only example of theropod skin outside of Tetanurae
are subcircular, elliptical or lenticular in outline, with the largest (excluding footprints; Hendrickx et al. in press). As described by
basement scales being mainly lenticular (Fig. 11C). The large elon- Bonaparte et al. (1990), the basement scales of Carnotaurus are (on
gated basement scales are parallel to each other and the direction of average) small (<10 mm), low, and typically rounded tubercles
their long axis is perpendicular to the orientation of the feature separated by narrow bands of interstitial tissue. However, contrary to
scales (as defined by the deflected apex in the latter). Bonaparte et al.’s (1990) and Czerkas and Czerkas' (1997) observa-
tions on the uniformity of these “tubercular scales”, we note that the
3. Discussion basement scales significantly differ in size and morphology in
various parts of the body. The basement scales are indeed strongly
3.1. General skin morphology elongated (oblong), relatively uniform, and small (<6 mm in length)
in the thoracic region, but are diamond-shaped, uniform, and large
At ~1200 cm2, the biggest patch of skin from the tail of Carno- (up to 13 mm in length) in the shoulder region, whereas they are
taurus (Fig. 8) is also one of the largest single patches of squamous subcircular to elliptical in outline and strongly variable in shape and
integument available for any saurischian, competing with the largest size in the anterior tail region (Fig. 12A-C). In addition, if the small
patch of skin recently described by Gallagher et al. (2021) from block with well-defined scales is indeed from the anterior tail region,
Diplodocus sp. from the Morrison Formation of Montana. Together then variations in the size and morphology of the basement scales
with patches from the neck, shoulder girdle, and thorax, the holo- also occur along the length of the tail i.e. having a smooth surface and
type of Carnotaurus also provides the most comprehensive overview being uniformly small (<8 mm) in one area of the tail, and poorly
of the squamous skin from any non-avian theropod as well as defined, larger (up to 12 mm) and with an uneven surface in another.

10
C. Hendrickx and P.R. Bell Cretaceous Research 128 (2021) 104994

Fig. 9. Locations of measurements taken from several integumentary structures on the large patch of skin from the anteroventral portion of the tail. Numbers correspond to the
features listed in Table 1. Scale bar on close up image represents: 1 cm.

Unlike the basement scales, the feature scales are relatively surfaces of the feature scales are smooth and lack sculpturing. The
uniform in morphology along the body of Carnotaurus, being dome- ridges seen on the apical surface of a single feature scale from the
like and not exceeding 10 mm in height and 65 mm in diameter in (?)tail region have no parallel among other known dinosaur feature
the thoracic, scapular, and tail regions. Smaller feature scales scales (pers. obs.) but these features may be due to taphonomy.
(~20 mm diameter) are interspersed, apparently at random, on the Contrary to Czerkas & Czerkas’ (1997) observations, we find no
thorax and proximal tail. The feature scales consist of a dome-like evidence that the feature scales were arranged into irregular rows
structure that flares adjacent to the surrounding basement scales or that they exhibited distinct size trends in different parts of the
so that the peripheral zone of each feature scale is weakly convex, body (e.g., they do not decrease in diameter towards the ventral
flat, or even slightly concave. As pointed out by Czerkas & Czerkas part of the body). Instead, the presence of both large and small
(1997), there is no indication of a bony core or osteoderm within the feature scales on the thorax and tail suggests that they were
feature scales (contra Lindoso et al. (2013), who implied that the randomly distributed (Fig. 2). Likewise, despite statements by
rugose keeled protuberances in the skin of Carnotaurus might be Bonaparte (1996a) and Czerkas (1997), there is currently no evi-
similar to Ceratosaurus' osteoderms). Interestingly, feature scales dence for the presence of feature scales on the head and neck of
from the scapular and tail regions are characterized by off-centre Carnotaurus (Fig. 2).
apices (contra Bonaparte 1996b), which are oriented differently in
these two body parts: anterodorsally in the shoulder and poster- 3.2. Comparison with other dinosaur scaly integuments
odorsally in the anterior tail regions. Not all feature scales of Car-
notaurus have off-centre apices, however; the apex of at least one The small basement scales (5e8 mm in average) covering most
feature scale from the thoracic region is centrally positioned. The of the skin of the large-bodied Carnotaurus (~7.8 m in length) are
11
C. Hendrickx and P.R. Bell Cretaceous Research 128 (2021) 104994

Table 1 the smallest basement scales show a similarly random pattern. The
Measurements of different integumentary structures visible on the large patch of basement scales of Tehuelchesaurus are, however, more polygonal
skin from the anterior tail region (Fig. 9).
and the size variation does not seem to be as important in this
Position Measurements in mm sauropod as that in Carnotaurus (Fig. 12O).
1 37  32a The granular surface texture seen on the basement scales on
2 56  43a the? tail of Carnotaurus is also distinctive among known theropods.
3 43  39a Indeed, nearly all other theropods with skin apparently had smooth
4 45  39a
scales. Longitudinal ridges and concentrically-ringed features (the
5 22  22a
6 41  40a latter interpreted as integumentary sense organs by Bell &
7 52 Hendrickx 2020), which are found together on some scales on the
8 53 tail of Juravenator, are the only other reported forms of ornamen-
9 44
tation in non-avian theropod scales. In comparison, surface orna-
10 44
11 24
mentation is more commonly present in iguanodontian
12 101 ornithopods (particularly hadrosaurids), which may have corru-
13 40 gations or ridges (Bell 2014), or eusauropods, which have a
14 80 distinctive papillose texture (Czerkas 1994; Gime
nez 2007; Foster
15 17
& Hunt-Foster 2011; Gallagher et al. 2021) that may be exclusive
16 17
17 3.1  4.4a to the clade. It is notable that papillose basement scales are seen in
a
diplodocoids and titanosaurs (Czerkas 1994; Gime
nez 2007;
Anteroposterior length vs. dorsoventral height of feature scales (positions 1 to 6)
and basement scale (position 17). Refer to Fig. 9 for positions.
Gallagher et al. 2021), which, together with the evidence from
Carnotaurus, suggests that some form of irregular surface texture
may have been plesiomorphic for saurischian body scales.
present in other large theropods such as tyrannosaurids as well as Feature scales are relatively rare among saurischians, having
some stegosaurids and hadrosaurids (Christiansen and Tschopp, been reported only in Carnotaurus, the tyrannosaurid Albertosaurus
2010; Bell 2014; Bell et al. 2017; Hendrickx et al., in press). These (Bell et al. 2017; Fig. 12H), embryos of titanosaur sauropods
small basement scales differ from significantly larger basement (Hendrickx et al. in press; Coria & Chiappe 2007; Fig. 12P) and
scales seen in mature individuals of sauropods, ankylosaurs, and possibly Diplodocus (the dome scales of Gallagher et al. 2021),
ceratopsians (Brown 1917; Sternberg 1925; Czerkas 1994; Gime
nez although they are more common in stegosaurian, ceratopsian, and
2007; Arbour et al. 2014; Brown 2017; Hendrickx et al., in press; Fig. iguanodontian ornithischians (Brown 1917; Sternberg 1925;
13). The largest basement scales (13 mm) from the scapular region Lingham-Soliar & Plodowski 2010; Bell 2012, 2014; Currie et al.
of Carnotaurus are unlike those of other theropods but are similar in 2016). In both Carnotaurus and Albertosaurus, the feature scales
size to those present in the belly area of some sauropods are sub-conical but in most (but not all) other taxa, including
(Mamenchisaurus, Diplodocus), ceratopsians (Nasutoceratops), thy- Albertosaurus (Fig. 12H), the feature scales also bear radial corru-
reophorans (Gigantspinosaurus, Dyoplosaurus), and hadrosaurs gations, unlike the relatively smooth condition in Carnotaurus
(Ouyang & Ye 2002; Xing et al. 2008; Lund 2010; Bell 2014; (Fig. 12D) and the embryonic titanosaurs (Fig. 12P). The feature
Gallagher et al. 2021; Fig. 13). Among theropods, the largest base- scales of Carnotaurus and Albertosaurus also differ in their external
ment scales of tyrannosaurids and juvenile tetanurans (i.e., Jura- boundary. In Albertosaurus, the basement scales surrounding the
venator, Allosaurus, and Lourinhanosaurus) are much smaller and do feature scale are large, well-delimited, and form a tall wall around
not exceed 5 mm in diameter (Hendrickx et al., in press; Fig. 13). the feature scale (Fig. 12L). Conversely, the basement scales
Only the basement scales of a specimen said to be from the anterior bordering the feature scales of Carnotaurus are small, poorly-
tail region of the adult specimen of Allosaurus “Big Al Two” shares defined, numerous, randomly distributed, and typically placed on
similarly large basement scales (Hendrickx et al., in press; Fig. 13). the same plane as or only slightly higher than the peripheral sur-
Given its size and the close association with diplodocid material in face of the feature scale (Figs. 11B, 12D).
the same quarry as “Big Al Two”, this specimen was considered by The only other non-tetanuran theropod with evidence of body
Hendrickx et al. (in press) as probably belonging to a sauropod. The integument is Ceratosaurus, which preserves osteoderms in a
presence of large basement scales (>10 mm) on some parts of sagittal row on parts of the neck and tail (Gilmore 1920; Hendrickx
Carnotaurus is therefore peculiar to that taxon among known et al. in press). This form of tegument is peculiar among theropods
theropods. and clearly differs from the unossified integument of Carnotaurus
Morphologically, the oblong basement scales from the thoracic (Czerkas & Czerkas 1997; contra; Lindoso et al. 2013). No cerato-
region of Carnotaurus best match the large, elongated tubercles saurs preserve skin at the level of the pes but the presence of
forming parallel rows on the skin of the titanosaur embryos from reticulate scales in non-averostran theropod trackmakers (Grallator,
Auca Mahuevo, Patagonia (Coria & Chiappe 2007; C. H. pers. obs.) as Eubrontes; Hendrickx et al. in press; Gatesy 2001; Milner et al.
well as the subrectangular dome-scales of Diplodocus (Gallagher 2006), the allosauroid Concavenator (Cuesta et al. 2015), the
et al. (2021), figure 8B). Likewise, the diamond-shaped basement tyrannosauroids Santanaraptor and Tarbosaurus (Currie et al. 2003;
scales covering the shoulder region of Carnotaurus (Fig. 12B, C) are Hendrickx et al. in press), and various paravians such as Sinorni-
somewhat reminiscent to the diamond-shaped basement scales of thosaurus (Ji et al. 2001) and Anchiornis (Wang et al. 2017) clearly
the tyrannosaurid Albertosaurus (Fig. 12K) as well as the hexagonal indicates that the plantar surface of the foot of Carnotaurus and
basement scales seen on the scapular area of the titanosauriform other ceratosaurs was covered in small reticulate scales (Bell &
Tehuelchesaurus. However, they differ strongly from Tehuelchesau- Hendrickx 2021; Hendrickx et al. in press). The presence of
rus’ basement scales in their small size (<14 mm) and the absence scutellate scales in Kulindadromeus (Godefroit et al. 2020) and the
of a distinctive papillose surface texture (see below). The combi- tyrannosauroid Santanaraptor (Hendrickx et al. in press) also sug-
nation of subcircular, lenticular and elliptical basement scales seen gests that the pes of Carnotaurus and other stem-averostrans likely
in the tail regiondwhich strongly vary in size and shape over a had scutellate scales at the level of the metatarsals. It is, however,
small areadof Carnotaurus appears to be unique among dinosaurs. unknown whether a bird-like podotheca was present in Carno-
To our knowledge, only portions of the skin of Tehuelchesaurus with taurus and other non-averostran theropods given that a podotheca
12
C. Hendrickx and P.R. Bell Cretaceous Research 128 (2021) 104994

Fig. 10. Latex mould of the scaly skin from the anterior portion of the tail of the abelisaurid Carnotaurus sastrei (MACN-CH 894). (A) Position of the mould, and (B) mould of the skin
and haemal arch. (C) Close up of the basement scales and two feature scales (dotted lines; apices marked by a black cross) posterior to the haemal arch. Abbreviations: fes, feature
scale; hae, haemal arch. Scale bars represent: 10 cm (AeB), 5 cm (C).

13
C. Hendrickx and P.R. Bell Cretaceous Research 128 (2021) 104994

Fig. 11. Skin from the? caudal region. (A) Natural mould (positive epirelief) of skin from the right side of the? tail of Carnotaurus sastrei (MACN-CH 894) showing the feature scales in
dotted lines. (BeC) Close ups of the largest feature scale (B) and basement scales (C). Abbreviations: bas, basement scales; fes, feature scale; rid, ridges on the feature scale. Scale
bars represent: 2 cm (B), 1 cm (C).

made of scutate, scutellate, and reticulate scales is currently authors compared the feature scales of Carnotaurus to those present
restricted to the averostrans Concavenator and Avialae (e.g., on the neck of the green iguana Iguana iguana (known as “nuchal
Yanornis; Hendrickx et al. in press; Cuesta et al. 2015) whereas non- tubercles”; Breuil 2013), noting that male iguanas confront and battle
theropod dinosaurs such as the neornithischian Kulindadromeus over mates. However, these structures would have afforded very
and the ceratopsian Psittacosaurus lack scutate scales (Godefroit little protection from any bites sustained during agonistic encounters
et al. 2020; Hendrickx et al. in press). in Carnotaurus (Mazzetta et al. 2009). Alternatively, in Carnotaurus
With scales showing multiple morphologies (i.e., oblong, and more broadly among dinosaurs, feature scales may simply have
diamond-shaped, hexagonal, subcircular, elliptical, and lenticular served a display/colouration function. It is worth mentioning, how-
basement scales), size, textures (smooth, granular) and arrange- ever, that different scale types are not intrinsically linked to different
ments (oriented, irregular) in different body areas, the scaly skin of colours in modern reptiles (e.g., Charette 1995; Burton 2004) and
Carnotaurus is revealed to be particularly diverse. Similarly com- scale patterns do not imply like-for-like colour patterns (Osborn
plex scale architecture has been described recently in Kulinda- 1912), as revealed in Psittacosaurus (Vinther et al. 2016).
dromeus (Godefroit et al. 2020), Diplodocus (Gallagher et al. 2021), The arrangement of parallel ridges (forming vertical grooves or
and Juravenator (Bell & Hendrickx 2021), indicating that the scaly wrinkles in life) of interstitial tissue and the uneven surface of the
integument of non-maniraptoriform dinosaurs were as complex as basement scales on the tail of Carnotaurus recall the intricate wrin-
at least some extant lizards (e.g., Moloch horridus, Phrynosoma spp; kling pattern of the African elephant (Loxodonta africana), which at
Sherbrooke et al. 2007). up to 6000 kg (Langman et al. 1995), provides a potential modern
analogue for the condition in Carnotaurus. As the largest organ in the
3.3. Speculations on skin function in Carnotaurus body, skin plays an important role in thermoregulation, a well-
known phenomenon in reptiles, but also, more uniquely among
The functional anatomy of squamous dinosaur skin has, until mammals, in elephants. Although Loxodonta (a mammal) and Car-
recently, been largely ignored (Bell & Hendrickx 2020, 2021). Despite notaurus (a dinosaur) would have had potentially very different
inherent difficulties in extrapolating and testing the functional sig- thermal physiologies, elephants are virtually hairless, lack sweat
nificance of dinosaur scales, its pursuit, through the lens of the extant glands, and have a thick, keratinous stratum corneum (Spearman
phylogenetic bracket (EPB; Witmer 1995) and modern analogues, 1970), all features that are more similar to a reptile's integument
lends a novel and complementary direction for investigating the than to a ‘typical’ mammal. African elephants combine a macro-scale
palaeobiology of these animals. Czerkas and Czerkas (1997) specu- wrinkling pattern with an intricate network of micrometre-wide
lated that the conical feature scales of Carnotaurus would have cracks in the heavily-cornified skin (Martins et al. 2018) that helps
provided “some degree of protection during confrontation, especially to retain water, which is adsorbed while wallowing, and effects
against other theropods” (Czerkas & Czerkas 1997, p. 156). These thermoregulation through transepidermal water loss (Wright & Luck
14
C. Hendrickx and P.R. Bell Cretaceous Research 128 (2021) 104994

Fig. 12. Epidermal scales in saurischian dinosaurs. (A) Oblong basement scales from the thoracic region of the abelisaurid Carnotaurus sastrei (MACN-CH 894). (B) Hexagonal
basement scales from the shoulder region of Carnotaurus sastrei (MACN-CH 894). (C) Lenticular basement scales possibly from the tail region of Carnotaurus sastrei (MACN-CH 894).
(D) Feature scales from the anterior tail region of the Carnotaurus sastrei (MACN-CH 894). (E) Pebbly basement scales from the thoracic region of a juvenile allosaurid Allosaurus
jimmadseni (plastic cast; UMNH VP C481). (F) Pebbly basement scales from the scapular region of an embryo referred to the putative metriacanthosaurid Lourinhanosaurus antunesi
(ML 565; courtesy of Oct
avio Mateus). (G) Arrowhead-shaped basement scales on the dorsal portion of the tail of the compsognathid Juravenator starki (JME Sch 200). (H)
Ornamented basement scales on the ventral portion of the tail of Juravenator starki (JME Sch 200). (I) Irregular basement scales on an unidentified area of the body of the
tyrannosaurid Daspletosaurus torosus (TMP 2001.036.0001). (J) Polygonal basement scales and epidermal midribs from the region of the neck of the tyrannosaurid Tyrannosaurus rex
(HMNS 2006.1743.01). (K) Diamond-shaped basement scales from an unidentified area of the body of the tyrannosaurid Albertosaurus sarcophagus (TMP 1994.186.0001). (L) Feature
scale and surrounding basements scales from the thoracic region of Albertosaurus sarcophagus (TMP 1994.186.0001). (M) Polygonal basement scales from the pelvis region of the
mamenchisaurid sauropod Mamenchisaurus youngi (ZDM 00083; from Xu and Guo (2009); modified). (N) Polygonal basement scales from the fore limb region of the basally
branching macronarian Haestasaurus becklesii (NHMUK R1868; courtesy of Mike Taylor, used with permission). (O) Rosette patterns of polygonal basement scales from the thoracic
region of the titanosauriform Tehuelchesaurus benitezii (MEF-PV 1125/3; courtesy of Paul Barrett, used with permission). (P) Feature scale with smaller basement scales in a rosette-
like arrangement in an embryonic titanosaur from Auca Mahuevo, Patagonia (MCF-PVPH-130; from Coria and Chiappe (2007), fig. 1.6; modified). Scale bars represent: 2 cm (M),
1 cm (A‒D, I, K‒L, NeO), 5 mm (E, J), 1 mm (FeH), 0.2 mm (P).

1984; Lillywhite & Stein 1987). Smaller Asian elephants (Elephas (acarodomatia; Arnold 1986), the functional role of these folds re-
maximus) too exhibit this wrinkling, although to a lesser degree than mains largely unexplored. Instead, large varanids and crocodylians
their African cousins, suggesting that thermoregulation (by evapo- employ a variety of mouth gaping, gular pumping (high-amplitude
rative cooling) is an important but less critical requirement for large- movements) and gular fluttering (high-frequency movements) be-
bodied heterotherms (see Dunkin et al. 2013) in more mesic envi- haviours in addition to ‘sun shuttling’ (Spellerberg 1972) to aid in
ronments (Wright & Luck 1984; Lillywhite & Stein 1987). Reptilian evaporative cooling (Harlow et al. 2010 and references therein).
analogues for the skin of Carnotaurus may be less appropriate since Since, like both reptiles and elephants (Wright & Luck 1984), Car-
the largest members of the clade (crocodylians; >1000 kg body notaurus would have lacked sweat glands (based on the extant
mass) are semi-aquatic and employ alternative thermoregulatory phylogenetic bracket), it would have been faced with similar chal-
behaviours (Seebacher et al. 1999). Skin wrinkling is also not as lenges in shedding excess heat, especially given its predicted highly-
obvious in reptiles, such as the Komodo dragon (~85 kg; Jessop et al. cursorial lifestyle (Mazzetta et al. 1998; Persons & Currie 2011). The
2006), and is typically restricted to the length of the neck as well as La Colonia Formation where Carnotaurus was found, was deposited
the limb joints (Frost 1992; P.R.B. pers. obs.). Beyond mite pockets at ~43 S palaeolatitude (Pascual et al. 2000; Matthews et al. 2016) on

15
C. Hendrickx and P.R. Bell Cretaceous Research 128 (2021) 104994

Fig. 13. Log-plot of basement scale length (taken from the largest scale) versus body mass in Kulindadromeus (violet dot) as well as non-avian theropod (red squares), sauropod
(green squares), thyreophoran (orange dots), ceratopsian (pink dots ), and ornithopod (blue dots) dinosaurs. Only theropod taxa are labelled. Data on body masses and largest
basement scales are provided in Appendix A3. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

a low-energy, marine-influenced (estuarine) coastal floodplain 4. Conclusions

(Gasparini et al. 2015) under Greenhouse conditions (Vera et al.
2019) and a seasonal climate with alternating periods of humidity The scaly integument of the abelisaurid Carnotaurus is the most
and aridity (Ardolino & Delpino 1987; Ardolino et al. 1995). African completely preserved of any theropod and the only example of this
elephants live in a similar seasonal climate (in at least part of their type of integument outside of Tetanurae (excluding footprints). The
range), albeit at lower latitudes, which dictates their thermoregula- holotype of Carnotaurus therefore contributes important informa-
tory behaviour (Dunkin et al. 2013). While we do not claim that tion on the scale architecture among relatively basally-branching
Carnotaurus and elephants necessarily thermoregulated in the same non-avian theropods, albeit late in the evolution of the clade. Con-
ways (i.e. using evaporative cooling), we note here that they share trary to previous interpretations, the basement scales are
distinct gross morphological similarities in their integument, despite morphologically diverse and we find no evidence for discrete rows
one having scales and the other having a highly modified mamma- of feature scales or progressive variations in their size along the
lian epidermis. As a ~1600 kg (Benson et al. 2014) predator with body (e.g., larger scales dorsally, smaller scales ventrally). Instead,
enhanced cursorial abilities (Mazzetta et al. 1998; Persons & Currie both feature and basement scales appear to be randomly distributed
2011), Carnotaurus would have had significant requirements for on the available patches of skin. Despite the large size of Carno-
shedding excess heat (Seebacher 2003). The skin is therefore a likely taurus, its average basement scales are small (<10 mm), a feature
site to search for potential adaptations, in addition to other physio- consistent among squamous-skinned theropods. The largest base-
logical strategiesdsuch as vascularisation of the paranasal air sinus ment scales, found in the shoulder region, however, indicate that
(Porter & Witmer 2020)dthey might have employed. The skin plays some portions of the body had relatively large scales, a characteristic
an important role in thermoregulation in lepidosaurs and archo- more prevalent among sauropods and ornithischians. The basement
saurs, which, like large-bodied mammals (e.g., elephants, rhinoc- scales on the tail, where they are best preserved, are strongly vari-
eros; Lillywhite & Stein 1987; Endo et al. 2009; Dunkin et al. 2013) able in shape and size over a small area. They also exhibit a unique
combine behavioural and physiological adaptations to moderate irregular surface texture in some parts of the tail, which is unknown
their body temperature despite fundamentally different thermal from other non-avian theropod skin occurrences, but which bears
physiologies (Seebacher & Franklin 2005). Folds in the skin would some similarity to the papillose texture on the scales of many
have increased the available surface area for heat transfer (Endo et al. eusauropods. Given the important role of the integument in ther-
2009), which would have been particularly critical in a large-bodied moregulation, particularly among extant reptiles and large-bodied
dinosaur such as Carnotaurus (Seebacher et al. 1999; Seebacher extant mammals, a similar function can be postulated for Carno-
2003). Dinosaurs are likely to have had a uniquedbut not neces- taurus and probably other dinosaurs. In particular, wrinkling in the
sarily ubiquitousdthermal physiology (Benson 2018), therefore skin of Carnotaurus is reminiscent of the characteristic wrinkling
estimating the thermoregulatory requirements of any dinosaur is not pattern in modern elephants, which contributes to their unique
likely to be straightforward (Seebacher et al. 1999). Regardless of the physiological and behavioural thermoregulation strategy (by
specifics, the skin is likely to have played some role in thermoreg- increasing surface area and helping to trap water, adsorbed during
ulation complementary to cephalic heat transfer, which has been wallowing behaviour). Although speculative in the case of Carno-
suggested as an important thermoregulatory mechanism in large taurus and the potential role of skin in thermoregulation, we
theropods (Porter & Witmer 2020). advocate for more detailed assessments of scaly dinosaur skin that

16
C. Hendrickx and P.R. Bell Cretaceous Research 128 (2021) 104994

go beyond external gross morphological descriptions. Incorporation Natural History Museum of Los Angeles County Contributions in Science 416,
1e42.
of comparative methods, together with theoretical modelling (e.g.,
risation morphologique de l’iguane commun Iguana iguana
Breuil, M., 2013. Caracte
thermal or biomechanical modelling) are likely to provide novel (Linnaeus, 1758), de l’iguane des Petites Antilles Iguana delicatissima Laurenti, 1768
insights into the palaeobiology of squamous skinned theropods and et de leurs hybrides. Bulletin de la Socie
te

Herpe

tologique de France 147, 309e346.
other dinosaurs. Brown, B., 1917. A complete skeleton of the horned dinosaur Monoclonius, and
description of a second skeleton showing skin impressions. Bulletin of the
American Museum of Natural History 37, 298e306.
Acknowledgements Brown, C.M., 2017. An exceptionally preserved armored dinosaur reveals the
morphology and allometry of osteoderms and their horny epidermal coverings.
PeerJ 5, e4066.
We thank Martín Ezcurra (MACN) for access to the specimens of Burton, F., 2004. Color and pattern in Grand Cayman blue iguanas. Iguana 11, 149e151.
Carnotaurus with skin under his care. We also warmly thank Campione, N.E., Evans, D.C., Brown, C.M., Carrano, M.T., 2014. Body mass estimation
in non-avian bipeds using a theoretical conversion to quadruped stylopodial
Waylon Rowley for sharing Lessem (1989)'s article from Discover; proportions. Methods in Ecology and Evolution 5, 913e923.
Michael Pittman (Hong Kong University) for sharing photos of the Carabajal, A.P., 2011. The braincase anatomy of Carnotaurus sastrei (Theropoda:
skin and scans of a book with illustrations on the Carnotaurus' Abelisauridae) from the Upper Cretaceous of Patagonia. Journal of Vertebrate
Paleontology 31, 378e386.
discovery; as well as Ryan DeLuca, Mauricio A. Cerroni (MACN), Carr, T.D., Varricchio, D.J., Sedlmayr, J.C., Roberts, E.M., Moore, J.R., 2017. A new
Paul Barrett (NHMUK), Scott Hartman (Uni. of Wisconsin-Madison), tyrannosaur with evidence for anagenesis and crocodile-like facial sensory
and Rafael Delcourt (Uni. of Campinas) for allowing us to use their system. Scientific Reports 7, 1e11.
Cerroni, M.A., Paulina-Carabajal, A., 2019. Novel information on the endocranial
photos/artworks in the figures. We are particularly indebted to Jake morphology of the abelisaurid theropod Carnotaurus sastrei. Comptes Rendus
Baardse for creating a beautiful and accurate 3D model of Carno- Palevol 18, 985e995.
taurus as well as Sylvia J. Czerkas, Agustín G. Martinelli (MACN), Cerroni, M.A., Canale, J.I., Novas, F.E., 2020. The skull of Carnotaurus sastrei Bona-
parte 1985 revisited: insights from craniofacial bones, palate and lower jaw.
Fernando Novas (MACN), and Guillermo W. Rougier (University of
Historical Biology 1e42, 0.
Louisville) for providing information on the discovery of Carno- Charette, R., 1995. CITES identification guide e Crocodilians: Guide to the identifi-
taurus and Houghton Mifflin Harcourt Books for accepting to use cation of crocodilian species controlled under the convention of international
some illustrations in the figures. We finally thank reviewer Paul trade in endangered species of wild fauna and flora. Environment Canada,
Ottawa, Canada.
Barrett and editor Eduardo Koutsoukos for their valuable com- Christiansen, N.A., Tschopp, E., 2010. Exceptional stegosaur integument impressions
ments and suggestions that improved the final version of this from the Upper Jurassic Morrison Formation of Wyoming. Swiss Journal of
manuscript. This research was supported by the Consejo Nacional Geosciences 103 (2), 163e171. https://doi.org/10.1007/s00015-010-0026-0.
Cignoni, P., Callieri, M., Corsini, M., Dellepiane, M., Ganovelli, F., Ranzuglia, G., 2008.
de Investigaciones Científicas y Te
cnicas (CONICET) and Agencia MeshLab: an open-source mesh processing tool. In: Eurographics Italian
Nacional de Promocio
n Científica y Tecnolo
gica, Argentina to C.H. Chapter Conference, 2008, pp. 129e136.
(Beca Pos-doctoral CONICET Legajo 181417). C.H. dedicates this Clyde, W.C., Krause, J.M., De Benedetti, F., Ramezani, J., Cúneo, N.R., Gandolfo, M.A.,
Haber, P., Whelan, C., Smith, T., 2021. New South American record of the Cre-
paper to his newborn son Julian Hendrickx and to the memory of taceousePaleogene boundary interval (La Colonia Formation, Patagonia,
Dr. J. F. Bonaparte and Dr. A. Milner. Argentina). Cretaceous Research 126, 104889. https://doi.org/10.1016/
j.cretres.2021.104889.
Coria, R.A., Chiappe, L.M., 2007. Embryonic skin from Late Cretaceous sauropods
References (Dinosauria) of Auca Mahuevo, Patagonia, Argentina. Journal of Paleontology 81,
1528e1532.
Arbour, V.M., Burns, M.E., Bell, P.R., Currie, P.J., 2014. Epidermal and dermal integu- Cuesta, E., Díaz-Martínez, I., Ortega, F., Sanz, J.L., 2015. Did all theropods have
mentary structures of ankylosaurian dinosaurs. Journal of Morphology 275, 39e50. chicken-like feet? First evidence of a non-avian dinosaur podotheca. Cretaceous
Ardolino, A., Delpino, D., 1987. Senoniano (continental-marino). Comarca nordpa- Research 56, 53e59.
tago
nica Provincía del Chubut, Argentina. Congreso Geolo
gico Argentino 10, Cúneo, N.R., Gandolfo, M.A., Zamaloa, M.C., Hermsen, E., 2014. Late Cretaceous
193e196. Aquatic Plant World in Patagonia, Argentina. PloS One 9, e104749.
Ardolino, A., Busteros, A., Cucchi, R., Franchi, M., Lema, H., Remesal, M., 1995. Currie, P.J., Badamgarav, D., Koppelhus, E., 2003. The first Late Cretaceous footprints
Cuerpos alcalinos b
asicos paleo
genos del sur de Somún Cur
a (Argentina) y su from the Nemegt Locality in the Gobi of Mongolia. Ichnos 10, 1e13.
marco estratigr
afico. Publicacio
n Electro
nica de la Asociacio
n Paleontolo
gica Currie, P.J., Holmes, R.B., Ryan, M.J., Coy, C., 2016. A juvenile chasmosaurine cera-
Argentina 3. topsid (Dinosauria, Ornithischia) from the Dinosaur Park Formation, Alberta,
Arnold, E.N., 1986. Mite pockets of lizards, a possible means of reducing damage by Canada. Journal of Vertebrate Paleontology 36, e1048348.
ectoparasites. Biological Journal of the Linnean Society 29, 1e21. Czerkas, S.A., 1994. The history and interpretation of sauropod skin impressions.
Arnold, C., 1993. Dinosaurs All Around: An Artist’s View of the Prehistoric World. Gaia 10, 173e182.
Clarion Books, New York. Czerkas, S.A., 1997. Skin. In: Currie, P.J., Padian, K. (Eds.), Encyclopedia of Dinosaurs.
Barrett, P.M., Evans, D.C., Campione, N.E., 2015. Evolution of dinosaur epidermal Academic Press, San Diego, California, pp. 669e675. San Diego, California.
structures. Biology Letters 11, 20150229. Czerkas, S., Czerkas, S., 1989. My Life with the Dinosaurs. Aladdin, New York.
Bell, P.R., 2012. Standardized terminology and potential taxonomic utility for Czerkas, S.A., Czerkas, S.J., 1997. The integument and life restoration of Carnotaurus.
hadrosaurid skin impressions: a case study for Saurolophus from Canada and Dinofest International. In: Proceedings of the Symposium at Arizona State Uni-
Mongolia. PloS One 7, e31295. versity, pp. 155e158.
Bell, P.R., 2014. A review of hadrosaurid skin impressions. In: Hadrosaurs. Indiana Delcourt, R., 2018. Ceratosaur palaeobiology: new insights on evolution and ecology
University Press, Bloomington, Indiana, pp. 572e590. of the southern rulers. Scientific Reports 8, 1e12.
Bell, P.R., Hendrickx, C., 2020. Crocodile-like sensory scales in a Late Jurassic Dunkin, R.C., Wilson, D., Way, N., Johnson, K., Williams, T.M., 2013. Climate in-
theropod dinosaur. Current Biology 30, R1068eR1070. fluences thermal balance and water use in African and Asian elephants:
Bell, P.R., Hendrickx, C., 2021. Epidermal complexity in the theropod dinosaur physiology can predict drivers of elephant distribution. Journal of Experimental
Juravenator from the Late Jurassic of Germany. Palaeontology 64, 203e223. Biology 216, 2939e2952.
Bell, P.R., Campione, N.E., Persons, W.S., Currie, P.J., Larson, P.L., Tanke, D.H., Endo, H., Kobayashi, H., Koyabu, D., Hayashida, A., Jogahara, T., Taru, H., Oishi, M.,
Bakker, R.T., 2017. Tyrannosauroid integument reveals conflicting patterns of Itou, T., Koie, H., Sakai, T., 2009. The morphological basis of the armor-like
gigantism and feather evolution. Biology Letters 13, 20170092. folded skin of the greater Indian rhinoceros as a thermoregulator. Mammal
Benson, R.B.J., 2018. Dinosaur macroevolution and macroecology. Annual Review of Study 34, 195e200.
Ecology Evolution and Systematics 49, 379e408. Ezcurra, M.D., Novas, F.E., 2016. Theropod dinosaurs from Argentina. Contribuciones
Benson, R.B.J., Campione, N.E., Carrano, M.T., Mannion, P.D., Sullivan, C., del Museo Argentino de Ciencias Naturales 6, 139e156.
Upchurch, P., Evans, D.C., 2014. Rates of dinosaur body mass evolution indicate Foster, J.R., Hunt-Foster, R.K., 2011. New occurrences of dinosaur skin of two types
170 Million years of sustained ecological innovation on the avian stem lineage. (Sauropoda? and Dinosauria indet.) from the Late Jurassic of North America
PLoS Biology 12, e1001853. (Mygatt-Moore Quarry, Morrison Formation). Journal of Vertebrate Paleon-
Bonaparte, J.F., 1985. A horned Cretaceous carnosaur from Patagonia. National tology 31, 717e721.
Geographic Research 1, 149e151. Frost, D.R., 1992. Phylogenetic analysis and taxonomy of the Tropidurus group of
Bonaparte, J.F., 1996a. Dinosaurios de America del Sur. Buenos Aires. lizards (Iguania: Tropiduridae). American Museum of Natural History 3033,
Bonaparte, J.F., 1996b. Cretaceous tetrapods of Argentina. Münchener Geo- 1e68.
wissenschaftliche Abhandlungen, A 30, 73e130. Gallagher, T., Poole, J., Schein, J.P., 2021. Evidence of integumentary scale diversity in
Bonaparte, J.F., Novas, F.E., Coria, R.A., 1990. Carnotaurus sastrei Bonaparte, the the late Jurassic sauropod Diplodocus sp. from the Mother's Day Quarry, Mon-
horned, lightly built carnosaur from the Middle Cretaceous of Patagonia. tana. PeerJ 9, e11202.

17
C. Hendrickx and P.R. Bell Cretaceous Research 128 (2021) 104994

Gasparini, Z., 2000. Tortugas y plesiosauridos de la Formacio
n La Colonia (Cret
acico Novas, F.E., Agnolín, F.L., Ezcurra, M.D., Porfiri, J., Canale, J.I., 2013. Evolution of the
Superior) de Patagonia, Argentina. Revista Espanola de Paleontologia 15, 23e36. carnivorous dinosaurs during the Cretaceous: The evidence from Patagonia.
Gasparini, Z., Sterli, J., Parras, A., O’gorman, J.P., Salgado, L., Varela, J., Pol, D., 2015. Cretaceous Research 45, 174e215.
Late Cretaceous reptilian biota of the La Colonia Formation, central Patagonia, O’gorman, J.P., Salgado, L., Varela, J., Parras, A., 2013. Elasmosaurs (Sauropterygia,
Argentina: Occurrences, preservation and paleoenvironments. Cretaceous Plesiosauria) from the La Colonia Formation (CampanianeMaastrichtian),
Research 54, 154e168. Argentina. Alcheringa: An Australasian Journal of Palaeontology 37, 259e267.
Gatesy, S.M., 2001. Skin impressions of Triassic theropods as records of foot Osborn, H.F., 1912. Part II. Integument of the iguanodont dinosaur Trachodon.
movement. Bulletin of the Museum of Comparative Zoology 156, 137e149. American Museum of Natural History Memoir, pp. 33e54.
Giacchino, A., Agnolin, F., 2020. Obituario: Jose
Fernando Bonaparte (14/06/1928- Ouyang, H., Ye, Y., 2002. The first mamenchisaurian skeleton with complete skull,
18/02/2020). Historia Natural 10, 175e180. Mamenchisaurus youngi. Sichuan Publishing House of Science and Technology,
Gilmore, C.W., 1920. Osteology of the carnivorous Dinosauria in the United State Chengdu, China.
National museum: with special reference to the genera Antrodemus (Allosaurus) Pascual, R., Goin, F.J., Gonz
alez, P., Ardolino, A., Puerta, P.F., 2000. A highly derived
and Ceratosaurus. Bulletin of the United States National Museum 110, 1e159. docodont from the Patagonian Late Cretaceous: evolutionary implications for
Gime
nez, O. del V., 2007. Skin impressions of Tehuelchesaurus (Sauropoda) from the Gondwanan mammals. Geodiversitas 22, 395e414.
Upper Jurassic of Patagonia. Revista del Museo Argentino de Ciencias Naturales Persons, W.S.I., Currie, P.J., 2011. Dinosaur speed demon: the caudal musculature of
9, 119e124. Carnotaurus sastrei and implications for the evolution of South American abe-
Glut, D.F., 2000. Dinosaurs: The Encyclopedia. Supplement 1. Jefferson, N.C. lisaurids. PloS One 6, e25763.
Godefroit, P., Sinitsa, S.M., Cincotta, A., Mcnamara, M.E., Reshetova, S.A., Porter, W.R., Witmer, L.M., 2020. Vascular patterns in the heads of dinosaurs: evi-
Dhouailly, D., 2020. Integumentary structures in Kulindadromeus zabaikalicus, a dence for blood vessels, sites of thermal exchange, and their role in physio-
basal neornithischian dinosaur from the Jurassic of Siberia. In: Foth, C., logical thermoregulatory strategies. The Anatomical Record 303, 1075e1103.
Rauhut, O.W.M. (Eds.), The Evolution of Feathers: From Their Origin to the Rauhut, O.W.M., Cladera, G., Vickers-Rich, P., Rich, T.H., 2003. Dinosaur remains
Present. Springer International Publishing, Cham, pp. 47e65. from the Lower Cretaceous of the Chubut Group, Argentina. Cretaceous
Grillo, O.N., Delcourt, R., 2017. Allometry and body length of abelisauroid theropods: Research 24, 487e497.
Pycnonemosaurus nevesi is the new king. Cretaceous Research 69, 71e89. Rougier, G.W., Forasiepi, A.M., Hill, R.V., Novacek, M., 2009. New mammalian re-
Harlow, H.J., Purwandana, D., Jessop, T.S., Phillips, J.A., 2010. Size-related differences mains from the Late Cretaceous La Colonia Formation, Patagonia, Argentina.
in the thermoregulatory habits of free-ranging Komodo dragons. International Acta Palaeontologica Polonica 54, 195e212.
Journal of Zoology 2010, e921371. Ruiz, J., Torices, A., Serrano, H., Lo
pez, V., 2011. The hand structure of Carnotaurus
Hendrickx, C., Bell, P.R., 2021. 3D model related to the publication: The scaly skin of sastrei (Theropoda, Abelisauridae): implications for hand diversity and evolu-
the abelisaurid Carnotaurus sastrei (Theropoda: Ceratosauria) from the Upper tion in abelisaurids. Palaeontology 54, 1271e1277.
Cretaceous of Patagonia. MorphoMuseuM 7, 1e3. Salgado, L., Bonaparte, J.F., 1991. Un nuevo sauropodo Dicraeosauridae, Amargasaurus
Hendrickx, C., Bell, P.R., Pittman, M., Milner, A.R.C., Cuesta, E., O’Connor, J., Loewen, cazaui gen. et sp. nov., de la Provincia del Neuque
n, Argentina, 28, 333e346.
M.A., Currie, P.J., Mateus, O., Kaye, T.G., Delcourt, R., in press. Morphology and Sampson, S.D., Witmer, L.M., 2007. Craniofacial anatomy of Majungasaurus cren-
distribution of scales, dermal ossifications, and other non-feather integumen- atissimus (Theropoda: Abelisauridae) from the Late Cretaceous of Madagascar.
tary structures in non-avialan theropod dinosaurs. Biological Reviews. Journal of Vertebrate Paleontology 27, 32e104.
Hieronymus, T.L., Witmer, L.M., Tanke, D.H., Currie, P.J., 2009. The facial integument Seebacher, F., 2003. Dinosaur body temperatures: the occurrence of endothermy
of centrosaurine ceratopsids: morphological and histological correlates of novel and ectothermy. Paleobiology 29, 105e122.
skin structures. The Anatomical Record: Advances in Integrative Anatomy and Seebacher, F., Franklin, C.E., 2005. Physiological mechanisms of thermoregulation in
Evolutionary Biology 292, 1370e1396. reptiles: a review. Journal of Comparative Physiology B 175, 533e541.
Jessop, T.S., Madsen, T., Sumner, J., Rudiharto, H., Phillips, J.A., Ciofi, C., 2006. Seebacher, F., Grigg, G.C., Beard, L.A., 1999. Crocodiles as dinosaurs: behavioural
Maximum body size among insular Komodo dragon populations covaries with thermoregulation in very large ectotherms leads to high and stable body
large prey density. Oikos 112, 422e429. temperatures. Journal of Experimental Biology 202, 77e86.
Ji, Q., Norell, M.A., Gao, K.-Q., Ji, S.-A., Ren, D., 2001. The distribution of integu- Sherbrooke, W.C., Scardino, A.J., De Nys, R., Schwarzkopf, L., 2007. Functional
mentary structures in a feathered dinosaur. Nature 410, 1084e1088. morphology of scale hinges used to transport water: convergent drinking ad-
Langman, V.A., Roberts, T.J., Black, J., Maloiy, G.M., Heglund, N.C., Weber, J.M., aptations in desert lizards (Moloch horridus and Phrynosoma cornutum). Zoo-
Kram, R., Taylor, C.R., 1995. Moving cheaply: energetics of walking in the African morphology 126, 89e102.
elephant. Journal of Experimental Biology 198, 629e632. Spearman, R.I.C., 1970. The epidermis and its keratinisation in the African elephant
Lessem, D., 1989. Skinning the Dinosaur. Discover 10, 38e44. (Loxodonta africana). Zoologica Africana 5, 327e338.
Lillywhite, H.B., Stein, B.R., 1987. Surface sculpturing and water retention of Spellerberg, I.F., 1972. Temperature tolerances of Southeast Australian reptiles
elephant skin. Journal of Zoology 211, 727e734. examined in relation to reptile thermoregulatory behaviour and distribution.
Lindoso, R.M., Marinho, T. da S., Santucci, R.M., Medeiros, M.A., Carvalho, I. de S., Oecologia 9, 23e46.
2013. A titanosaur (Dinosauria: Sauropoda) osteoderm from the Alca ^ntara Sternberg, C.M., 1925. Integument of Chasmosaurus belli. Canadian Field-Naturalist
Formation (Cenomanian), S~ ao Luís Basin, Northeastern Brazil. Cretaceous 39, 108e110.
Research 45, 43e48. Vera, E.I., Perez Loinaze, V.S., Llorens, M., Paez, M., Passalia, M.G., 2019. Fossil woods
Lingham-Soliar, T., Plodowski, G., 2010. The integument of Psittacosaurus from with coniferalean affinities from the Upper Cretaceous (Campa-
Liaoning Province, China: taphonomy, epidermal patterns and color of a cera- nianeMaastrichtian) Puntudo Chico Formation, Chubut Province, Argentina.
topsian dinosaur. Naturwissenschaften 97, 479e486. Cretaceous Research 99, 321e333.
Lund, E.K., 2010. Nasutuceratops titusi, a new basal centrosaurine dinosaur (Orni- Vinther, J., Nicholls, R., Lautenschlager, S., Pittman, M., Kaye, T.G., Rayfield, E.,
thischia: Ceratopsidae) from the Upper Cretaceous Kaiparowits Formation. MSc. Mayr, G., Cuthill, I.C., 2016. 3D camouflage in an ornithischian dinosaur. Current
Dissertation. southern Utah, p. 158. Biology 26, 2456e2462.
Martins, A.F., Bennett, N.C., Clavel, S., Groenewald, H., Hensman, S., Hoby, S., Joris, A., Wang, X., Pittman, M., Zheng, X., Kaye, T.G., Falk, A.R., Hartman, S.A., Xu, X., 2017.
Manger, P.R., Milinkovitch, M.C., 2018. Locally-curved geometry generates Basal paravian functional anatomy illuminated by high-detail body outline.
bending cracks in the African elephant skin. Nature Communications 9, 3865. Nature Communications 8, 14576.
Matthews, K.J., Maloney, K.T., Zahirovic, S., Williams, S.E., Seton, M., Müller, R.D., Witmer, L.M., 1995. The extant phylogenetic bracket and the importance of
2016. Global plate boundary evolution and kinematics since the late Paleozoic. reconstructing soft tissues in fossils. Functional morphology in vertebrate
Global and Planetary Change 146, 226e250. paleontology 1, 19e33.
Mazzetta, G.V., Farin ~ a, R.A., Vizcaíno, S.F., 1998. On the palaeobiology of the South Wright, P.G., Luck, C.P., 1984. Do elephants need to sweat? South African Journal of
American horned theropod Carnotaurus sastrei Bonaparte. Gaia 15, 192. Zoology 19, 270e274.
Mazzetta, G.V., Christiansen, P., Farin~ a, R.A., 2004. Giants and bizarres: body size of some Xing, L.D., Peng, G.Z., Shu, C.K., 2008. Stegosaurian skin impressions from the Upper
southern South American Cretaceous dinosaurs. Historical Biology 16, 71e83. Jurassic Shangshaximiao Formation, Zigong, Sichuan, China: A new observation.
Mazzetta, G.V., Cisilino, A.P., Blanco, R.E., Calvo, N., 2009. Cranial mechanics and Geological Bulletin of China 27, 1049e1053.
functional interpretation of the horned carnivorous dinosaur Carnotaurus sas- Xu, X., Guo, Y., 2009. The origin and early evolution of feathers: insights from recent
trei. Journal of Vertebrate Paleontology 29, 822e830. paleontological and neontological data. Vertebrata Palasiatica 47, 311e329.
Me
ndez, Ariel H., 2012. The cervical vertebrae of the Late Cretaceous abelisaurid
dinosaur Carnotaurus sastrei. Acta Palaeontologica Polonica 59 (3), 569e579.
https://doi.org/10.4202/app.2011.0129.
Milner, A.R.C., Lockley, M.G., Johnson, S., 2006. The story of the St. George Dinosaur
Discovery Site at Johnson Farm: an important new Lower Jurassic dinosaur Appendix A. Supplementary data
tracksite from the Moenave Formation of southwestern Utah. The Triassic-
Jurassic Terrestrial Transition. New Mexico Museum of Natural History and Sci- Supplementary data to this article can be found online at https://doi.org/10.
ence Bulletin 37, 329e345. 1016/j.cretres.2021.104994.

18