Report

An Exceptionally Preserved Three-Dimensional

Armored Dinosaur Reveals Insights into Coloration
and Cretaceous Predator-Prey Dynamics
Graphical Abstract Authors
Caleb M. Brown,
Donald M. Henderson, Jakob Vinther,
Ian Fletcher, Ainara Sistiaga,
Jorsua Herrera, Roger E. Summons

Correspondence
[email protected]

In Brief
Brown et al. report a new, exceptionally
preserved armored dinosaur, showing
bony armor with horn coverings and
organically preserved scales. A reddish-
brown coloration and camouflage in the
form of countershading are indicated.
Crypsis suggests strong visual predation
pressure on this heavily armored
dinosaur, distinct from modern systems.

Highlights
d A new armored dinosaur is described based on an
exceptionally preserved specimen

d Abundant in situ osteoderms with keratinous sheaths and

scales are preserved

d Reddish-brown coloration and crypsis in the form of

countershading are indicated

d Crypsis indicates strong predation pressure on this large,

heavily armored dinosaur

Brown et al., 2017, Current Biology 27, 1–8

August 21, 2017 ª 2017 The Authors. Published by Elsevier Ltd.
http://dx.doi.org/10.1016/j.cub.2017.06.071
 Please cite this article in press as: Brown et al., An Exceptionally Preserved Three-Dimensional Armored Dinosaur Reveals Insights into Coloration and
Cretaceous Predator-Prey Dynamics, Current Biology (2017), http://dx.doi.org/10.1016/j.cub.2017.06.071

Current Biology

Report

An Exceptionally Preserved Three-Dimensional

Armored Dinosaur Reveals Insights into
Coloration and Cretaceous Predator-Prey Dynamics
Caleb M. Brown,1,6,* Donald M. Henderson,1 Jakob Vinther,2,3 Ian Fletcher,4 Ainara Sistiaga,5 Jorsua Herrera,5
and Roger E. Summons5
1Royal Tyrrell Museum of Palaeontology, PO Box 7500, Drumheller, AB T0J 0Y0, Canada
2School of Biological Sciences, University of Bristol, Life Sciences Building, 24 Tyndall Avenue, Bristol BS8 1TQ, UK
3School of Earth Sciences, University of Bristol, Wills Memorial Building, Queens Road, Bristol BS8 1RJ, UK
4School of Mechanical and Systems Engineering, Newcastle University, Newcastle NE1 7RU, UK
5Department of Earth, Atmospheric and Planetary Sciences, Massachusetts Institute of Technology, 45 Carleton St., Cambridge,

MA 02142, USA
6Lead Contact

*Correspondence: [email protected]
http://dx.doi.org/10.1016/j.cub.2017.06.071

SUMMARY possession of massive dorsal and lateral armor,

illustrating a significant dichotomy between Meso-
Predator-prey dynamics are an important evolu- zoic predator-prey dynamics and those of modern
tionary driver of escalating predation mode and terrestrial systems.
efficiency, and commensurate responses of prey
[1–3]. Among these strategies, camouflage is RESULTS AND DISCUSSION
important for visual concealment, with counter-
shading the most universally observed [4–6]. Extant Systematic Paleontology
terrestrial herbivores free of significant predation Dinosauria Owen, 1842 [10].
pressure, due to large size or isolation, do not Ornithischia Seeley, 1888 [11].
exhibit countershading. Modern predator-prey dy- Ankylosauria Osborn, 1923 [12].
namics may not be directly applicable to those Nodosauridae Marsh, 1890 [13].
of the Mesozoic due to the dominance of very Borealopelta markmitchelli gen. et sp. nov. (Figures 1 and 2).
large, visually oriented theropod dinosaurs [7].
Despite thyreophoran dinosaurs’ possessing exten- Etymology
sive dermal armor, some of the most extreme ex- The generic name Borealopelta is derived from ‘‘borealis’’ (Latin,
‘‘northern’’) and ‘‘pelta’’ (Greek, ‘‘shield’’), in reference to the
amples of anti-predator structures [8, 9], little direct
northern locality and the preserved epidermal scales and dermal
evidence of predation on these and other dinosaur osteoderms. The specific epithet markmitchelli honors Mark
megaherbivores has been documented. Here we Mitchell for his more than 7,000 hours of patient and skilled
describe a new, exquisitely three-dimensionally preparation of the holotype.
preserved nodosaurid ankylosaur, Borealopelta
markmitchelli gen. et sp. nov., from the Early Creta- Holotype
ceous of Alberta, which preserves integumentary The holotype is Royal Tyrrell Museum of Palaeontology (TMP)
structures as organic layers, including continuous 2011.033.0001: an articulated specimen preserving the head,
fields of epidermal scales and intact horn sheaths neck, most of the trunk and sacrum, a complete right and a
capping the body armor. We identify melanin in partial left forelimb and manus, partial pes (Figure 1). In situ
the organic residues through mass spectroscopic osteoderms and nearly complete soft tissue integument are
analyses and observe lighter pigmentation of preserved across dorsal and lateral surfaces of the
axial skeleton, posterodorsal surface of forelimbs, and plantar
the large parascapular spines, consistent with
surfaces of a manus and a pes. Specimen is preserved in
display, and a pattern of countershading across
multiple large blocks, including slabs and counter-slabs in
the body. With an estimated body mass exceeding the sacral region.
1,300 kg, B. markmitchelli was much larger than
modern terrestrial mammals that either are counter- Locality and Horizon
shaded or experience significant predation pres- Suncor Millennium Mine, Fort McMurray, Alberta, Canada.
sure as adults. Presence of countershading sug- Wabiskaw Member, Clearwater Formation, Aptian stage.
gests predation pressure strong enough to select Detailed locality data are available at Royal Tyrrell Museum of
for concealment in this megaherbivore despite Palaeontology.

Current Biology 27, 1–8, August 21, 2017 ª 2017 The Authors. Published by Elsevier Ltd. 1
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
 Please cite this article in press as: Brown et al., An Exceptionally Preserved Three-Dimensional Armored Dinosaur Reveals Insights into Coloration and
Cretaceous Predator-Prey Dynamics, Current Biology (2017), http://dx.doi.org/10.1016/j.cub.2017.06.071

further differentiated from Sauropelta based on: parietals flat to

slightly convex [51:0]; cervical half ring has 4–6 osteoderms
only [164:1]; medial cervical osteoderms subequal, hexagonal,
and bear prominent median ridge with posterior margin projec-
ting beyond the basal footprint.

Morphological Description
For a morphological description of the head, osteoderms,
epiosteodermal scales (osteoderm/spine horn sheaths), and
epidermal basement scales, see the Supplemental Morpholog-
ical Description.

Phylogenetic Analysis
To determine the phylogenetic position of Borealopelta mark-
mitchelli, we scored it into the morphological character/taxon ma-
trix of Arbour et al. [14] (see also STAR Methods for full phyloge-
netic methods, Data S1 for data, and Supplemental Phylogenetic
Results). The resulting strict consensus tree positions Borealopelta
in a clade with other Albian-aged nodosaurids Pawpawsaurus
campbelli and Europelta carbonensis, with the Santonian-aged
Hungarosaurus tormai as a sister taxon (Figures 3 and S2).

Deposition and Preservation History

The specimen was discovered in the Suncor Millennium Mine
(open pit, oil sands) in northeastern Alberta, Canada during over-
burden removal and was subsequently collected by staff of the
Royal Tyrrell Museum and Suncor. The hosting rock is the Al-
bian-aged Wabiskaw Member of the Clearwater Formation over-
lying the bitumen-rich McMurray Formation. Several plesiosaurs
Figure 1. Photographs of the Holotype of Borealopelta markmitch- and ichthyosaurs have been recovered from the Wabiskaw pre-
elli, TMP 2011.033.0001 viously [15–17], but never a dinosaur. This member records a
Top: anterodorsolateral view; bottom: anterodorsal view. Scale bar, 10 cm. lower shoreface or proximal offshore marine environment [15].
See also Figure S1. The carcass arrived at the seabed on its back and with sufficient
force to impact and deform the immediately underlying sedimen-
tary layers. Despite the trace fossils left by burrowing animals in
Diagnosis the hosting sediments, implying at least a partially oxygenated
A nodosaurid ankylosaur characterized by the following autapo- environment, the specimen lacks any evidence of scavenging.
morphies (*) and suite of characters [character/state]: cranial: When found, the fossil was completely encased in a very dense
dorsal skull ornamentation expressed as a large hexagonal and strong but brittle siderite concretion that ranged in thickness
dermal plate in frontoparietal region [52:1] and multiple (>20) around the carcass from 20 cm on the upper side to 40 cm on the
small dermal plates in frontonasal region [21:2]*; external nares lower, seabed side. Broken surfaces through the concretion
excluded from view dorsally (shared with Pawpawsaurus) reveal sedimentary features above the fossil that allow for infer-
[16:1]; supraorbital ornamentation forming sharp lateral rim dor- ring the natural collapse of the body after burial and before
sal to orbits (shared with Gargoyleosaurus and Kunbarrasaurus) consolidation (Figure S4i). About 15 cm of sediment was laid
[38:2]; jugal (suborbital) horn triangular with pointed apex (shared down prior to release of internal body fluids and collapse (Fig-
with Gastonia, Gargoyleosaurus, and Polocanthus) [47:2]; jugal ure S4i), evidenced from a collapse/fluid escape structure in
(suborbital) horn base longer than orbit length [49:2]*; osteo- the sacral region (Figure S4i). The body cavity was injected
derms: cervical and thoracic osteoderms form continuous with almost homogeneous sand, with no apparent sedimentary
(abutting) transverse rows completely separated by continuous features. Formation of the concretion must have commenced
transverse rows of polygonal basement scales; parascapular shortly after the carcass arrived at the seabed, preventing any
spine is the largest osteoderm, recurved, and projects postero- scavenging and allowing all of the scales and osteoderms to
laterally and horizontally (potentially shared with Sauropelta); retain their original configurations and morphology, with minimal
osteoderm count for transverse rows: cervicals: C1-3, C2-3, dorsoventral compression.
C3-3, transition: TR-2, thoracic: T1-6*; third and sixth transverse The dorsal integument is well preserved as an organic film
thoracic osteoderm rows expressed medially but pinch out derived from the keratin sheaths over the osteoderms, integu-
laterally*. mentary scales, and the epicuticle of hinge regions between
The new taxon can be further differentiated from Pawpawsau- scales. The distribution of the film correlates well to the expected
rus based on: dermal plate in frontonasal region (central dermal distribution of melanin, a pigment that has been found to
plates) flat [22:1]; absence of ciliary osteoderm [41:0]. Can be preserve in a number of vertebrate integumentary structures

2 Current Biology 27, 1–8, August 21, 2017

 Please cite this article in press as: Brown et al., An Exceptionally Preserved Three-Dimensional Armored Dinosaur Reveals Insights into Coloration and
Cretaceous Predator-Prey Dynamics, Current Biology (2017), http://dx.doi.org/10.1016/j.cub.2017.06.071

Figure 2. Schematic Line Drawing of TMP 2011.033.0001, the Holotype of Borealopelta markmitchelli, Illustrating Preservation of the
Different Tissue Types
(A) Schematic of complete specimen in dorsal view.
(B and C) Skull in dorsal (B) and left lateral (C) views.
(D) Close-up view of the neck, illustrating alternating cervical osteoderm bands (and preserved keratinous sheaths) and polygonal scales.
(E) Close-up view of flank illustrating lateral thoracic osteoderms (with keratinous coverings) and polygonal scales.
(F) Close-up view of sacral shield counterpart illustrating osteoderms and scales.
(G) Close-up view of antebrachium including osteoderms and keratinous coverings.
(D’–G’) Interpretive line drawings of the corresponding panels (D)–(G).
Scale bars in (B)–(G), 10 cm. See also Figure S1.

Current Biology 27, 1–8, August 21, 2017 3

 Please cite this article in press as: Brown et al., An Exceptionally Preserved Three-Dimensional Armored Dinosaur Reveals Insights into Coloration and
Cretaceous Predator-Prey Dynamics, Current Biology (2017), http://dx.doi.org/10.1016/j.cub.2017.06.071

A B C D E

Key:
Rock

Denversaurus schlessmani
Cranial bone
Postcranial endochondral bone
ANKYLOSAURINAE

Small/undifferentiated scales/ossicle fields

Struthiosaurus austriacus

Edmontonia longiceps
Edmontonia rugosidens
Large polygonal scales

Panoplosaurus mirus
Osteoderms
Hungarosaurus tormai

Keratinous sheaths of osteoderms

Dongyangopelta yangyanensis

Niobrarasaurus coleii
Gobisaurus domoculus
Period
Epoch

Peloroplites cedrimontanus
Cedarpelta bilbeyhallorum

Ma
Tatankacephalus cooneyorum

Animantarx ramaljonesi

Borealopelta markmitchelli

Stegopelta landerensis
Age
Pawpawsaurus campbelli
Chuanqilong chaoyangensis

Nodosaurus textilis
Shamosaurus scutatus

66.0
Kunbarrasaura ieversi

Maastrichtian
Silvisaurus condrayi
Liaoningosaurus paradoxus

Europelta carbonensis

72.1
Sauropelta edwardsi

Campanian
Sauroplites scutiger

83.6
Upper

Santonian
86.3
Coniacian
89.8
Polacanthus foxii
Gastonia burgei
Hylaeosaurus armatus

Turonian
93.9
Cretaceous

Cenomanian
Gargoyleosaurus parkpinorum

100
Albian
Mymoorapelta maysi

113
Aptian
126
Lower

Barremian
Huayangosaurus taibaii

131
Hauterivian
134
Valanginian
139
Berriasian
145
Tithonian
Lesothosaurus diagnosticus
Scelidosaurus harrisonii

152
Upper

Kimmeridgian
157
Oxfordian ANKYLOSAURIDAE
164
Callovian 166
NODOSAURIDAE
Middle

Bathonian 168
Jurassic

Bajocian 170
Aalenian
174
ANKYLOSAURIA
Toarcian
183
Lower

Pliensbachian
191
Sinemurian
199
Hettangian 201

(legend on next page)

4 Current Biology 27, 1–8, August 21, 2017

 Please cite this article in press as: Brown et al., An Exceptionally Preserved Three-Dimensional Armored Dinosaur Reveals Insights into Coloration and
Cretaceous Predator-Prey Dynamics, Current Biology (2017), http://dx.doi.org/10.1016/j.cub.2017.06.071

[6, 18, 19]. Scanning electron microscopy (SEM) and energy- lie in the same plane as the inferred direction of growth of
dispersive X-ray spectroscopy (EDAX) analyses reveal that the the sheath (i.e., parallel to the stratum germinativum) and may
organic material is present as solid to finely granular material represent some preserved artifact of appositional growth of these
with desiccation cracks scattered in a matrix largely composed epiosteodermal scales, as seen in living crocodilians [24, 25].
of siderite cement (Figure S3ii; see also STAR Methods and Sup- The largest of the horn sheaths, the parascapular spines, are
plemental Observations Under the Electron Microscope). No distinct from the remaining sheaths and epiosteodermal scales
apparent melanosomes are preserved. Time-of-flight secondary in being both lighter colored in visible light and slightly fluo-
ion mass spectrometry (TOF-SIMS) reveals spectra of negative rescing under UV light (Figures S4iiiE–H). This is most simply in-
secondary ions that have an overall relative secondary ion inten- terpreted as having lower concentrations of melanin incorpo-
sity similar to previously studies of fossil melanins (Figure S3iiiA; rated into the horn sheath and likely reflects a distinct lighter
see also STAR Methods and Supplemental Results from the TOF color of these spines in life.
SIMS). Principal-component analyses demonstrate that there is The contrast in fossilization between the ventral and dorsal
a significant contribution of sulfur-bearing secondary ions to the surface provides a further case for melanin preservation, as
organic material in TMP 2011.033.0001, which spread the this transition is best interpreted as countershading. Other
nodosaur samples distinctly from other fossil melanin samples epidermal structural molecules such as keratin [26, 27] or
(Figure S3iiiB). These secondary ions have been identified for collagen [28] would have had a very similar distribution in the
sulfur-bearing pheomelanin (benzothiazole) previously [19]. Py- epidermis on both top and bottom surfaces, and no unique pro-
rolysis-gas chromatography-mass spectroscopy (py-GC-MS) tein markers (amides, succinimides, diketopiperazines) were
afforded pyrolysates with assemblages of small nitrogen-, oxy- recovered in the py-GC-MS data to the exclusion of markers
gen- and sulfur-containing heterocyclic and aromatic molecules overlapping with melanins [29] (Figure S3iv).
characteristic of eumelanin (e.g., pyrrole, indole, N-methylpyr- The countershading transition can be traced from cross-
role, and methylphenol). Of special note is the presence of sectional views of the sacrum (Figure S4ii) and neck. The organic
significant amounts of benzothiazole (Figure S3iv), which is diag- film terminates a little beyond the ventralmost lateral osteo-
nostic for pheomelanin. Although sulfur may be incorporated into derms. Projecting the melanin distribution to a retrodeformed
melanin secondarily [20] to yield thiophenes, which are also body outline suggests a transition from highly pigmented to
observed and could similarly be derived from pheomelanin, less pigmented integument on the lateral flank (Figure S4ii; see
this process is not known to give rise to benzothiazoles [20] also Supplemental Discussion on the Chemical Preservation of
(see also STAR Methods and Supplemental Results of Pyrolysis Melanin).
GC-MS).
Hence, we argue that the integument was pigmented reddish- Implications for Paleobiology
brown by pheomelanin-rich melanin. This may also explain the The discovery of a three-dimensionally preserved ankylosaurian
lack of melanosome preservation, as pheomelanin-rich melano- provides new evidence for understanding the anatomy, soft tis-
somes have been shown to be less stable in heat/pressure sue outline, and arrangement of dermal armor in thyreophoran
autoclave experiments [18], as well as in enzymatic extraction dinosaurs.
procedures [21]. The preservation of the nearly complete integument, along
with a suite of in situ pre-caudal osteoderms and their horn
Pigment Distribution sheaths, allows multiple novel inferences regarding the
Hardened keratinous tissues such as claws, scales, and feathers epidermis of the ancient animal. Across all preserved regions,
have reinforcing calcium phosphate deposits [22], which often the epidermal covering (‘‘epiosteodermal scales’’ sensu [30])
preserve well in fossil tissues and can be identified using fluores- associated with the osteoderms is highly congruous in having
cence imaging [6, 23]. However, the keratinized tissues in TMP a 1:1 correlation in count and basal shape between epidermal
2011.033.0001 are heavily pigmented, which masks calcium scale and underlying osteoderm. This epidermal scale/osteo-
phosphate fluorescence [6]. These keratinous sheaths are inert derm association is most analogous, and potentially deeply
(non-fluorescing and non-reflecting) under UV light, with two homologous, to that observed in extant crocodilians [31, 32].
major exceptions. The epidermal coverings for the thoracic and sacral osteo-
Several cervical osteoderm sheaths show underlying longitudi- derms are best interpreted as single, sub-centimeter-thick, ker-
nal ridges that appear to have been unpigmented, as they are atinized scales (scutes) that slightly exaggerate the keels and
dramatically lighter in visible light and exhibit strong fluorescence spines. In contrast, the epidermal components of the spine-
in UV (Figures S4iiiA–D). These unpigmented longitudinal striae like cervical, transitional, and parascapular osteoderms are

Figure 3. Time-Calibrated Strict Consensus Tree Showing Position of Borealopelta markmitchelli within Ankylosauria, with Representative
Well-Preserved Ankylosaurs Shown Above
Bottom: time-calibrated strict consensus tree illustrating position of Borealopelta markmitchelli within Ankylosauria scaled to Jurassic and Cretaceous stages.
Top: line drawings of representative well-preserved ankylosaur specimens with in situ armor and/or skin. Scale bars, 1 m. See also Figure S2 and Data S1.
(A) Kunbarrasaurus, QM F18101.
(B) Euoplocephalus, NHMUK 5161.
(C) Sauropelta, AMNH 3035 and 3036 composite.
(D) Borealopelta, TMP 2011.033.0001 (this study).
(E) Edmontonia, AMNH 5665.

Current Biology 27, 1–8, August 21, 2017 5

 Please cite this article in press as: Brown et al., An Exceptionally Preserved Three-Dimensional Armored Dinosaur Reveals Insights into Coloration and
Cretaceous Predator-Prey Dynamics, Current Biology (2017), http://dx.doi.org/10.1016/j.cub.2017.06.071

Figure 4. Chart Illustrating the Loss of

Panthera pardis
Puma concolor

Panthera onca

Panthera tigris
Countershading as Body Mass Increases in

Panthera leo
Canis lupus
Uncia uncia
Terrestrial Mammal Herbivores
Chart includes pooled data for artiodactyls,
perissodactyls, and proboscideans divided into
body-mass bins, showing relative proportion of
species that exhibit countershading. The diago-
Borealopelta nally hatched area represents the mass above
which significant predation of adults does not
occur. Animals illustrated above chart are repre-
sentative taxa within each mass bin; species
names in italics at top indicate body masses of the
largest carnivores. See also Data S2.

100
00
The presence of countershading in a
80
Countershaded taxa (%)

large, heavily armored herbivorous dino-

saur provides a unique insight into the
predator-prey dynamic of the Cretaceous
60 Period. With an estimated length of 5.5 m
and a conservative body mass estimate
of 1,300 kg (similar to estimates for other
40 nodosaurids, i.e., Sauropelta [38], but
also see higher estimates [39, 40]), Borea-
lopelta is much larger than any modern
20
terrestrial animal exhibiting counter-
shading. Modern mammalian predators
0 0 do not represent a significant predatory
0
log10kg 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 risk to the largest mammalian herbivores
(>1,000 kg) [41, 42], and herbivores above
kg this size threshold generally do not exhibit
1 10 102 103 104 countershading or other types of camou-
flage (Figure 4). Additionally, herbivores
below this threshold that also possess
sheath-like, being thick, pointed, and extending significantly defensive weapons (e.g., horn, quills) experience lower preda-
beyond the bony core, substantially increasing the length of tion than those that do not [42]. Similarly, large primates are
the spine. The macroscopic structure and incorporation of less likely to be countershaded due to lower predation risk
hardening calcium phosphate into the spine-like epiosteoder- [43]. A parallel pattern is also seen in Testudinidae, where island
mal scales is broadly similar to the keratinized horn sheaths forms, which are free of significant adult predation, consistently
in Bovidae. become large [44, 45] and lose the cryptic pattern seen in smaller
Historically, the function of osteoderms within Ankylosauria relatives.
has largely been discussed within the context of anti-predator The terrestrial predator guild for most of the Mesozoic
defense, with the term ‘‘armor’’ used ubiquitously to describe was dominated by large non-avian theropod dinosaurs [46].
these dermal components of the skeleton [9, 33]. More recently, Although the present specimen is not preserved in the
other potential functions including thermoregulation [34–36] and context of its accompanying fauna, the consistent occurrence
intraspecific combat and display [37] have been proposed, often of large theropod footprints (e.g., Irenesauripis) in coeval
to augment the seemingly default function as defensive armor. and nearby formations in northern Alberta and British
Preserved evidence of countershading suggests that the preda- Columbia [47, 48], and allosauroid/carcharodontosaurid taxa
tion pressure on Borealopelta, even at large adult size, was (i.e., Acrocanthosaurus) in equivalent formations farther south
strong enough to select for camouflage from visual predators [49, 50], suggests that the ecosystem of which Borealopelta
(see also Supplemental Discussion of Countershading and was a part would have included one or more large, multi-ton
Body Mass and Data S2). This offers support to the idea that predatory theropods. Given modern birds’ possession of tri-
many of the osteoderms functioned in a defensive role, even in and tetrachromatic vision [51, 52] and the visual acuity of extant
adults of large species. In contrast, the distinct pigmentation crocodilians [53], theropod carnivores would have been highly
and enlarged keratinous sheath of the parascapular spine visual predators [54, 55]. This is in contrast to modern systems
suggest that this particular spine may have functioned more in which top predators of large herbivores are dichromatic
predominantly in display. When combined with data indicating mammalian carnivores [56]. Active predation on large and
highly species-specific morphology of the parascapular spines heavily armored herbivorous dinosaurs highlights how distinct
of other taxa, this suggests that the extensive elaboration of the predator-prey dynamics of the Cretaceous were from
these spines may be attributable to sociosexual display. those of today. The greater size and deep visual reliance of

6 Current Biology 27, 1–8, August 21, 2017

 Please cite this article in press as: Brown et al., An Exceptionally Preserved Three-Dimensional Armored Dinosaur Reveals Insights into Coloration and
Cretaceous Predator-Prey Dynamics, Current Biology (2017), http://dx.doi.org/10.1016/j.cub.2017.06.071

Cretaceous apex predators, compared to extant mammalian improved the paper. Funding was provided by Suncor Energy, the Royal
analogs, may have produced an evolutionary arms race result- Tyrrell Museum of Palaeontology, the Royal Tyrrell Museum Cooperating
Society, the National Geographic Society (10004-16), and the NASA Astro-
ing in a combination of armor, crypsis, and other visual defense
biology Institute (NNA13AA90A).
strategies in even the largest dinosaurs.
Received: March 6, 2017
STAR+METHODS Revised: May 16, 2017
Accepted: June 27, 2017
Detailed methods are provided in the online version of this paper Published: August 3, 2017
and include the following:
REFERENCES
d KEY RESOURCES TABLE
1. Vermeij, G.J. (1977). The Mesozoic marine revolution: evidence from
d CONTACT FOR REAGENT AND RESOURCE SHARING
snails, predators and grazers. Paleobiology 3, 245–258.
d EXPERIMENTAL MODELS AND SUBJECTS
2. Vermeij, G.J. (1993). Evolution and Escalation: An Ecological History of Life
d METHOD DETAILS (Princeton University Press).
B Computer Tomography Scanning
3. Vermeij, G.J. (1994). The evolutionary interaction among species: selec-
B Sampling for Scanning Electron Microscopy and tion, escalation, and coevolution. Annu. Rev. Ecol. Syst. 25, 219–236.
Geochemical Analysis 4. Allen, W.L., Baddeley, R., Cuthill, I.C., and Scott-Samuel, N.E. (2012). A
B Scanning Electron Microscopy quantitative test of the predicted relationship between countershading
B Time-of-Flight Secondary Ion Mass Spectroscopy and lighting environment. Am. Nat. 180, 762–776.
B Pyrolysis Gas Chromatography-Mass Spectrometry 5. Penacchio, O., Lovell, P.G., Cuthill, I.C., Ruxton, G.D., and Harris, J.M.
B Photography (2015). Three-dimensional camouflage: exploiting photons to conceal
d QUANTIFICATION AND STATISTICAL ANALYSIS form. Am. Nat. 186, 553–563.
B Phylogenetic Methods 6. Vinther, J., Nicholls, R., Lautenschlager, S., Pittman, M., Kaye, T.G.,
d DATA AND SOFTWARE AVAILABILITY Rayfield, E., Mayr, G., and Cuthill, I.C. (2016). 3D camouflage in an ornith-
ischian dinosaur. Curr. Biol. 26, 2456–2462.
B Specimen
B Nomenclature 7. Holtz, T.R. (2003). Dinosaur Predation. In Predator-Prey Interactions in
the Fossil Record, P.H. Kelley, M. Kowalewski, and T.A. Hansen, eds.
(Springer), pp. 325–340.
SUPPLEMENTAL INFORMATION
8. Alexander, R.M. (1997). Engineering a dinosaur. In The Complete
Supplemental Information includes four figures, supplemental text, and two Dinosaur, J.A. Farlow, and M.K. Brett-Surman, eds. (Indiana University
datasets and can be found with this article online at http://dx.doi.org/10. Press), pp. 414–425.
1016/j.cub.2017.06.071. 9. Matthews, W. (1922). A super-dreadnaught of the animal world: The
armored dinosaur Palaeoscincus. Nat. Hist. 22, 333–342.
AUTHOR CONTRIBUTIONS 10. Owen, R. (1842). Report on British fossil reptiles, part II. Rep. Br. Assoc.
Adv. Sci. 1841, 60–204.
D.M.H. was part of the team that excavated the specimen. C.M.B. and 11. Seeley, H.G. (1888). The classification of the Dinosauria. Rep. Br. Assoc.
D.M.H. performed the description and phylogenic analyses. J.V. performed Adv. Sci. 1887, 698–699.
the SEM and EDAX analyses and planned the strategy for chemical analyses.
12. Osborn, H.F. (1923). Two Lower Cretaceous dinosaurs of Mongolia. Am.
I.F. and J.V. performed the TOF-SIMS analyses. A.S., J.H., and R.E.S. per-
Mus. Novit. 95, 1–10.
formed the pyrolysis GC-MS analyses. J.V., C.M.B., and D.M.H. analyzed
the ecological implications of coloration in B. markmitchelli. All authors 13. Marsh, O.C. (1890). Additional characters of the Ceratopsidae, with notice
contributed to the manuscript. of new Cretaceous dinosaurs. Am. J. Sci. 39, 418–429.
14. Arbour, V.M., Zanno, L.E., and Gates, T.A. (2016). Ankylosaurian dinosaur
ACKNOWLEDGMENTS palaeoenvironmental associations were influenced by extirpation,
sea-level fluctuation, and geodispersal. Palaeogeogr. Palaeoclimatol.
We thank Shawn Funk for the initial discovery of the specimen and Palaeoecol. 449, 289–299.
Mike Gratton, Doug Lacey, Bob Herron, Hans Heltke, Steve Hill, Ryan 15. Druckenmiller, P.S., and Maxwell, E.E. (2010). A new Lower Cretaceous
McPhearson, and many tens of Suncor Energy employees for assistance (lower Albian) ichthyosaur genus from the Clearwater Formation,
in excavation of the specimen. We recognize the technical skills of Darren Alberta, Canada. Can. J. Earth Sci. 47, 1037–1053.
Tanke (RTMP) that enabled the collection of the specimen. Jim McCabe 16. Maxwell, E.E., and Druckenmiller, P.S. (2011). A small ichthyosaur from the
(RTMP) delivered key supplies to the worksite. Mark Mitchell committed Clearwater Formation (Alberta, Canada) and a discussion of the taxonomic
nearly six years to the preparation of the specimen, and many RTMP prep- €ontologische Zeitschrift 85, 457–463.
utility of the pectoral girdle. Pala
aration, collection, and fabrication staff assisted with the project. Donna
17. Druckenmiller, P.S., and Russell, A.P. (2009). Earliest North American
Sloan produced scientific illustrations of the specimen (Figures 2B, 2C,
occurrence of Polycotylidae (Sauropterygia: Plesiosauria) from the
and S1). Sue Sabrowski, Corey Jaskolski, and Ann Jaskolski assisted
Lower Cretaceous (Albian) Clearwater Formation, Alberta, Canada.
with specimen photography. CT scans were performed at Western Veteri-
J. Paleontol. 83, 981–989.
nary Specialist and Emergency Centre, Calgary, with the assistance of Nic
Roussset, Joni Klaassen, and Cathy Gaviller. Xander Warren assisted with 18. Colleary, C., Dolocan, A., Gardner, J., Singh, S., Wuttke, M., Rabenstein,
SEM and EDAX analyses. TNT software was made available through the R., Habersetzer, J., Schaal, S., Feseha, M., Clemens, M., et al. (2015).
Willi Hennig Society. We also thank Don Brinkman, Michael Burns, David Chemical, experimental, and morphological evidence for diagenetically
Eberth, Aurora Elmore, Michael Habib, Michael Greshko, Carrie-Anne altered melanin in exceptionally preserved fossils. Proc. Natl. Acad. Sci.
Lunde, Mark Mitchell, Matthew Piscitelli, Patty Ralrick, and Anthony Russell USA 112, 12592–12597.
for discussions and assistance. Steve Brusatte, Florian Maderspacher (the 19. Lindgren, J., Sjövall, P., Carney, R.M., Uvdal, P., Gren, J.A., Dyke, G.,
scientific editor), and an anonymous reviewer provided comments that Schultz, B.P., Shawkey, M.D., Barnes, K.R., and Polcyn, M.J. (2014).

Current Biology 27, 1–8, August 21, 2017 7

 Please cite this article in press as: Brown et al., An Exceptionally Preserved Three-Dimensional Armored Dinosaur Reveals Insights into Coloration and
Cretaceous Predator-Prey Dynamics, Current Biology (2017), http://dx.doi.org/10.1016/j.cub.2017.06.071

Skin pigmentation provides evidence of convergent melanism in extinct lution indicate 170 million years of sustained ecological innovation on the
marine reptiles. Nature 506, 484–488. avian stem lineage. PLoS Biol. 12, e1001853.
20. Glass, K., Ito, S., Wilby, P.R., Sota, T., Nakamura, A., Bowers, C.R., 40. Paul, G. (1997). Dinosaur models: the good, the bad, and using them to es-
Vinther, J., Dutta, S., Summons, R., Briggs, D.E.G., et al. (2012). Direct timate the mass of dinosaurs. In Dinofest International: Proceedings of a
chemical evidence for eumelanin pigment from the Jurassic period. Symposium Sponsored By Arizona State University, D.L. Wolberg, E.
Proc. Natl. Acad. Sci. USA 109, 10218–10223. Stump, and G.D. Rosenberg, eds. (Academy of Natural Sciences),
21. Liu, Y., Hong, L., Wakamatsu, K., Ito, S., Adhyaru, B., Cheng, C.Y., pp. 39–45.
Bowers, C.R., and Simon, J.D. (2005). Comparison of structural and 41. Owen-Smith, N., and Mills, M.G.L. (2008). Predator-prey size relationships
chemical properties of black and red human hair melanosomes. in an African large-mammal food web. J. Anim. Ecol. 77, 173–183.
Photochem. Photobiol. 81, 135–144. 42. Hayward, M.W., and Kerley, G.I.H. (2005). Prey preferences of the lion
22. Pautard, F.G.E. (1964). Calcification of keratin. In Progress in the (Panthera leo). J. Zool. 267, 309–322.
Biological Sciences in Relation to Dermatology, Volume 2, A. Rook, and 43. Kamilar, J.M. (2009). Interspecific variation in primate countershading:
R.H. Champion, eds. (Cambridge University Press), pp. 227–239. effects of activity pattern, body mass, and phylogeny. Int. J. Primatol.
23. Mayr, G., Pittman, M., Saitta, E., Kaye, T.G., and Vinther, J. (2016). 30, 877.
Structure and homology of Psittacosaurus tail bristles. Palaeontology 59, 44. Arnold, E.N. (1979). Indian Ocean giant tortoises: their systematics and is-
793–802. land adaptations. Philos. Trans. R. Soc. Lond. B Biol. Sci. 286, 127–145.
24. Spearman, R.I.C., and Riley, P.A. (1969). A comparison of the epidermis 45. Jaffe, A.L., Slater, G.J., and Alfaro, M.E. (2011). The evolution of island
and pigment cells of the crocodile with those in two lizard species. Zool. gigantism and body size variation in tortoises and turtles. Biol. Lett. 7,
J. Linn. Soc. 48, 453–466. 558–561.
25. Reese, A.M. (1915). The Alligator and Its Allies (Knickerbocker Press). 46. Zanno, L.E., and Makovicky, P.J. (2013). Neovenatorid theropods are
26. Pan, Y., Zheng, W., Moyer, A.E., O’Connor, J.K., Wang, M., Zheng, X., apex predators in the Late Cretaceous of North America. Nat. Commun.
Wang, X., Schroeter, E.R., Zhou, Z., and Schweitzer, M.H. (2016). 4, 2827.
Molecular evidence of keratin and melanosomes in feathers of the Early 47. McCrea, R.T., and Currie, P.J. (1998). A preliminary report on dinosaur
Cretaceous bird Eoconfuciusornis. Proc. Natl. Acad. Sci. USA 113, tracksites in the Lower Cretaceous (Albian) Gates Formation near
E7900–E7907. Grande Cache, Alberta. N M Mus. Nat. Hist. Sci. Bull. 14, 155–162.
27. Moyer, A.E., Zheng, W., and Schweitzer, M.H. (2016). Keratin durability 48. Currie, P.J. (1989). Dinosaur footprints of Western Canada. In Dinosaur
has implications for the fossil record: results from a 10 year feather degra- Tracks and Traces, D.D. Gillette, and M.G. Lockley, eds. (Cambridge
dation experiment. PLoS ONE 11, e0157699. University Press), pp. 293–300.
28. Lingham-Soliar, T., Feduccia, A., and Wang, X. (2007). A new Chinese 49. Eddy, D.R., and Clarke, J.A. (2011). New information on the cranial anat-
specimen indicates that ‘protofeathers’ in the Early Cretaceous theropod omy of Acrocanthosaurus atokensis and its implications for the phylogeny
dinosaur Sinosauropteryx are degraded collagen fibres. Proc. Biol. Sci. of Allosauroidea (Dinosauria: Theropoda). PLoS ONE 6, e17932.
274, 1823–1829. 50. D’Emic, M.D., Melstrom, K.M., and Eddy, D.R. (2012). Paleobiology and
29. Saitta, E.T., Rogers, C., Brooker, R.A., Abbott, G.D., Kumar, S., O’Reilly, geographic range of the large-bodied Cretaceous theropod dinosaur
S.S., Donohoe, P., Dutta, S., Summons, R.E., and Vinther, J. (2017). Acrocanthosaurus atokensis. Palaeogeogr. Palaeoclimatol. Palaeoecol.
Low fossilization potential of keratin protein revealed by experimental 333-334, 13–23.
taphonomy. Palaeontology 60, 547–556. 51. Vorobyev, M., Osorio, D., Bennett, A.T.D., Marshall, N.J., and Cuthill, I.C.
30. Arbour, V.M., Burns, M.E., Bell, P.R., and Currie, P.J. (2014). Epidermal (1998). Tetrachromacy, oil droplets and bird plumage colours. J. Comp.
and dermal integumentary structures of ankylosaurian dinosaurs. Physiol. A Neuroethol. Sens. Neural Behav. Physiol. 183, 621–633.
J. Morphol. 275, 39–50. 52. Cuthill, I.C., Partridge, J.C., Bennett, A.T.D., Chrurch, S.C., Hart, N.S., and
31. Vickaryous, M.K., and Sire, J.-Y. (2009). The integumentary skeleton of Hunt, S. (2000). Ultraviolet vision in birds. Adv. Stud. Behav. 29, 159–214.
tetrapods: origin, evolution, and development. J. Anat. 214, 441–464. 53. Sillman, A.J., Ronan, S.J., and Loew, E.R. (1991). Histology and micro-
32. Vickaryous, M.K., and Hall, B.K. (2008). Development of the dermal skel- spectrophotometry of the photoreceptors of a crocodilian, Alligator mis-
eton in Alligator mississippiensis (Archosauria, Crocodylia) with comments sissippiensis. Proc. R. Soc. Lond. B Biol. Sci. 243, 93–98.
on the homology of osteoderms. J. Morphol. 269, 398–422. 54. Stevens, K.A. (2006). Binocular vision in theropod dinosaurs. J. Vertebr.
33. Romer, A.S. (1945). Vertebrate Paleontology (University of Chicago Press). Paleontol. 26, 321–330.
34. Carpenter, K. (1982). Skeletal and dermal armor reconstruction of 55. Rowe, M.P. (2000). Inferring the retinal anatomy and visual capacities of
Euoplocephalus tutus (Ornithischia: Ankylosauridae) from the Late extinct vertebrates. Palaeontol. Electronica 3, 1–43.
Cretaceous Oldman Formation of Alberta. Can. J. Earth Sci. 19, 689–697. 56. Jacobs, G.H. (1993). The distribution and nature of colour vision among
35. Blows, W.T. (2001). Dermal armour of the polacanthine dinosaurs. In the mammals. Biol. Rev. Camb. Philos. Soc. 68, 413–471.
The Armored Dinosaurs, K. Carpenter, ed. (Indiana University Press), 57. O’Gorman, E.J., and Hone, D.W.E. (2013). Body size distribution of the
pp. 363–385. dinosaurs. PLoS ONE 7, e51925.
36. Hayashi, S., Carpenter, K., Scheyer, T.M., Watabe, M., and Suzuki, D. 58. Stoner, C.J., Caro, T.M., and Graham, C.M. (2003). Ecological and
(2010). Function and Evolution of Ankylosaur Dermal Armor. Acta behavioral correlates of coloration in artiodactyls: systematic analyses
Palaeontol. Pol. 55, 213–228. of conventional hypotheses. Behav. Ecol. 14, 823–840.
37. Padian, K., and Horner, J.R. (2011). The evolution of ‘bizarre structures’ in 59. Schwieters, J., Cramer, H.G., Heller, T., Jürgens, U., Niehuis, E.,
dinosaurs: biomechanics, sexual selection, social selection or species Zehnpfenning, J., and Benninghoven, A. (1991). High mass resolution
recognition? J. Zool. 283, 3–17. surface imaging with a time-of-flight secondary ion mass spectroscopy
38. Carpenter, K. (1984). Skeletal reconstruction and life restoration of scanning microprobe. J. Vac. Sci. Technol. A 9, 2864–2871.
Sauropelta (Ankylosauria: Nodosauridae) from the Cretaceous of North 60. Crabb, P. (2001). The use of polarised light in photography of macrofossils.
America. Can. J. Earth Sci. 21, 1491–1498. Palaeontology 44, 659–664.
39. Benson, R.B., Campione, N.E., Carrano, M.T., Mannion, P.D., Sullivan, C., 61. Goloboff, P.A., Farris, J.S., and Nixon, K. (2008). TNT, a free program for
Upchurch, P., and Evans, D.C. (2014). Rates of dinosaur body mass evo- phylogenetic analysis. Cladistics 24, 774–786.

8 Current Biology 27, 1–8, August 21, 2017

 Please cite this article in press as: Brown et al., An Exceptionally Preserved Three-Dimensional Armored Dinosaur Reveals Insights into Coloration and
Cretaceous Predator-Prey Dynamics, Current Biology (2017), http://dx.doi.org/10.1016/j.cub.2017.06.071

STAR+METHODS

KEY RESOURCES TABLE

REAGENT or RESOURCE SOURCE IDENTIFIER

Biological Samples
Borealopelta markmitchelli holotype specimen This paper TMP 2011.033.0001
Deposited Data
Character/taxon matrix for cladistic analysis This paper (modified from [14]) Data S1
Mammal body mass/coloration data compiled from [57] and [58] Data S2
Software and Algorithms
TNT software Willi Hennig Society, https://cladistics.org/tnt/ N/A
R software https://www.r-project.org/ N/A

CONTACT FOR REAGENT AND RESOURCE SHARING

Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Caleb
Brown ([email protected]).

EXPERIMENTAL MODELS AND SUBJECTS

The experimental subject is the fossilized holotype of the nodosaurid dinosaur Borealopelta markmitchelli (TMP 2011.033.0001),
curated at the Royal Tyrrell Museum of Palaeontology, Drumheller, Alberta, Canada. The sex and developmental state are unknown,
although an adult (or nearly adult age) is inferred.

METHOD DETAILS

Computer Tomography Scanning

The skull was CT scanned in an attempt to reveal its internal structure. CT scans were performed at Western Veterinary Specialist and
Emergency Centre in Calgary, Alberta, using a Toshiba Aquilion Scanner. A variety scanning parameters were tested, verifying the
voltage and current (initial settings: 120 kV, 185 mA), and slice thickness (0.5-2 mm). Across all tested scanning parameters, the
massive and radio-opaque nature of the host rock did not allow for extraction of useful data.

Sampling for Scanning Electron Microscopy and Geochemical Analysis

The specimen was sampled extensively (total of 147 samples) across varying regions and tissues, largely concentrating on epidermal
structures, but also including rock matrix, dermal bone (osteoderm), endochondral bone, gut contents, and fossil wood (Figure S3i).
Samples we taken at three different intervals between Dec. 2016 and April 2017. The majority of the specimen was treated with ad-
hesives (Paraloid B-72 and Paleobond penetrate stabilizer) during the preparation process, but several sections of the sacral region
were collected as split blocks and were kept untreated. 103 samples were taken for electron microscopy from across the body, while
44 samples were taken for geochemical analyses. The geochemical samples largely concentrating on the untreated sacral region, but
did also include samples from the main body of the animal. Samples for both analyses were taken with a scalpel by chipping/prying
off small fragments. Those destined for SEM were carefully transferred over with a moistened tip onto an aluminum 12 mm SEM stub
with double sided carbon tape, while geochemical samples were transferred on a clean scalpel blade onto sterilized aluminum foil.
Samples were kept in plastic boxes for transport.

Scanning Electron Microscopy

Samples were coated with gold and analyzed in a Zeiss Sigma HD VP field emission SEM with an EDAX EBSD attached. Secondary
Electron images were taken off the horn sheath and skin from a representative set of samples to search for melanosomes and to
investigate the mode of preservation and the composition of the rock matrix (See also Figure S3ii and Supplemental Observations
Under the Electron Microscope).

Time-of-Flight Secondary Ion Mass Spectroscopy

The samples for ToFSIMS analysis were mounted directly onto a sample holder using double-sided carbon tape or clean stainless
steel screws and clips as appropriate.

Current Biology 27, 1–8.e1–e3, August 21, 2017 e1

 Please cite this article in press as: Brown et al., An Exceptionally Preserved Three-Dimensional Armored Dinosaur Reveals Insights into Coloration and
Cretaceous Predator-Prey Dynamics, Current Biology (2017), http://dx.doi.org/10.1016/j.cub.2017.06.071

Static SIMS analyses were carried out using an ION-TOF ‘TOF-SIMS IV – 200’ instrument (ION-TOF GmbH, Münster, Germany) of
single-stage reflectron design [59]. Positive and negative ion spectra were obtained using a Bi3+ focused liquid metal ion gun at 25 keV
energy, incident at 45 to the surface normal and operated in ‘bunched’ mode for high mass resolution. This mode used 20ns wide ion
pulses at 10kHz repetition rate. Charge compensation was effected by low-energy (ca. 20 eV) electrons provided by a flood gun. The
total ion dose density was 5 3 1016 ions m-2. The topography of the sample surface and the ion gun mode of operation limited the mass
resolution in this work to ca. m/Dm = 2000. The spatial resolution was limited by the primary ion beam diameter to ca. 4 mm.
Positive and negative ion static SIMS spectra were recorded from the outermost ca. 1nm of the sample surface at room temper-
ature. Raw data containing the secondary ions recorded at each pixel was acquired with a 128 3 128 pixel raster and a field of view of
50 mm 3 50 mm (see also Figure S3iii and Supplemental Results from the TOF SIMS).

Pyrolysis Gas Chromatography-Mass Spectrometry

Pyrolysis experiments were performed using a CDS Analytical 5250 pyroprobe equipped with an autosampler and coupled directly to
an Agilent 6890 N gas chromatograph interfaced with a Waters Micromass Autospec-Ultima mass spectrometer. Approximately 300-
700 mg of each sample was placed in a CDS quartz sample tube packed with a quartz wool plug followed by a quartz filler rod and a
second plug of quartz wool. Internal standards comprising 50ng each of decafluorobiphenyl (DFB) and pyrene in pentane were then
added to the samples immediately prior to analysis. Once loaded in the autosampler, the quartz tubes were dropped into the pyrolysis
chamber where they were kept at 50 C for 5 s and then heated at 10 C/ms to 600 C and held for 20 s. The valve oven and transfer line
were held at 300 C throughout the analysis. The thermal degradation products were transferred directly to a GC column held at 40 C.
The GC was equipped with a DB-5MS column (60 m x 0.25 mm x 0.25 mm) and was operated in split mode with a 5:1 split and 1 mL/min
He flow rate. The GC was held at 40 C for 2 min and then programmed to 310 C at 4 C/min where it was held for 20 min. The MS was
operated in electron impact mode at 70 eV and scanned m/z 50–600 once per second. Analyses were conducted in duplicate with
regularly interspersed blanks. Mass Lynx mass spectrometry software version 4.0 was used for data collection and processing.
The identification of pyrolysis products was based on comparison of their mass spectra with library spectra (NIST MS Search 2.0)
and by comparisons to data gathered previously [20] (see also Figure S3iv and Supplemental Results of Pyrolysis GC-MS).

Photography
Cross-Polarized Light Photography
In order to enhance color contrast for mapping the preservation and distribution of soft tissues across the specimen, a Lowell totalight
equipped with a 750W tungsten bulb was set up with a polarizing gel filter in front [60]. The specimen was then wetted with water and
photographed with a Nikon d800 dSLR camera and a 60mm f 2.8 Macro Nikkor lens with a polarizing filter mounted. Photos were then
taken so that the polarizing filter is perpendicular to the orientation of the polarizing gel. This configuration ensures the filtering of any
light that bounced off the surface of the specimen in resulting glare (see also Figures S4i and ii).
UV Fluorescence Photography
To identify any fluorescing minerals, such as calcium phosphate deposits, from the degraded keratin matrix the specimen was photo-
graphed illuminated with a short wave Superbright 2 (UV systems inc.) UV lamp using the same camera as for the cross-polarized light
photography. Due to the faintness of the fluorescence, photos were taken with a long exposure - between 20-30 s (see also Figure S4iii).

QUANTIFICATION AND STATISTICAL ANALYSIS

Phylogenetic Methods
To determine the phylogenetic position of Borealopelta markmitchelli, it was scored into the morphological character/taxon matrix of
Arbour et al. [14]. Although B. markmitchelli is known from a remarkably complete specimen, the majority of both cranial characters
and postcranial endoskeletal characters are obscured by preserved integument. As a result, only 41/91 (45%) cranial and 4/62 (6%)
postcranial could be coded along with 13/24 (54%) osteoderm characters, for a total of 58/177 (33%).
Several of the more incomplete and unstable, taxa included in the previous analysis [14] were removed: Ahshislepelta minor, Alec-
topelta coombsi, ‘‘Argentinian ankylosaur,’’ Hoplitosaurus marshi, ‘‘Paw paw scuteling,’’ Struthiosaurus languedocensis, Struthio-
saurus transylvanicus, Taohelong jinchengensis, Texasestes pleurohalio, Zarapelta sanjuanensis, and Zhejiangosaurus luoyangensis.
The resulting taxon character matrix consisted of 47 OTUs and 177 characters (Datafile S1 and S2). The character/taxon matrix was
analyses in TNT (version 1.1) [61]. All characters we treated as unordered and equally weighted. The traditional search option with one
random seed and 1000 replicates and tree bisection algorithm (TBR) was used to identify most parsimonious trees. A new technology
search was also performed. Bootstrap analysis (1000 replicates) using TBR, and Bremer support values (from suboptimal trees) were
also obtained to indicate clade support (see also Supplemental Phylogenetic Results and Figures 3 and S2).

DATA AND SOFTWARE AVAILABILITY

Specimen
The holotype of Borealopelta is deposited at the Royal Tyrrell Museum of Palaeontology, Drumheller, Alberta, under the accession
number: TMP 2011.033.0001. Samples for SEM analyses and TOF SIMS are held at the TMP under the same accession number.
Samples for py-GC-MS were consumed.

e2 Current Biology 27, 1–8.e1–e3, August 21, 2017

 Please cite this article in press as: Brown et al., An Exceptionally Preserved Three-Dimensional Armored Dinosaur Reveals Insights into Coloration and
Cretaceous Predator-Prey Dynamics, Current Biology (2017), http://dx.doi.org/10.1016/j.cub.2017.06.071

Nomenclature
This published work and the nomenclatural acts it contains have been registered in ZooBank, the online registration system for the
International Commission on Zoological Nomenclature. ZooBank LSIDs (Life Science Identifiers) can be viewed by appending the
LSID to the prefix http://zoobank.org/. The LSID for this publication is urn:lsid:zoobank.org:pub:39B2603A-CEF7-4657-B5D8-
71CE295AA2E4. The electronic edition of this work was published in a journal with an ISSN and will be archived and made available
from the following digital repository: CLOCKSS (http://www.clockss.org/clockss/Home).

Current Biology 27, 1–8.e1–e3, August 21, 2017 e3