Articles

https://doi.org/10.1038/s41559-018-0627-y

Caelestiventus hanseni gen. et sp. nov. extends

the desert-dwelling pterosaur record back
65 million years
Brooks B. Britt 1*, Fabio M. Dalla Vecchia 2, Daniel J. Chure3, George F. Engelmann4,
Michael F. Whiting5 and Rodney D. Scheetz1

Pterosaurs are the oldest known powered flying vertebrates. Originating in the Late Triassic, they thrived to the end of the
Cretaceous. Triassic pterosaurs are extraordinarily rare and all but one specimen come from marine deposits in the Alps.
A new comparatively large (wing span >​150 cm) pterosaur, Caelestiventus hanseni gen. et sp. nov., from Upper Triassic desert
deposits of western North America preserves delicate structural and pneumatic details not previously known in early ptero-
saurs, and allows a reinterpretation of crushed Triassic specimens. It shows that the earliest pterosaurs were geographically
widely distributed and ecologically diverse, even living in harsh desert environments. It is the only record of desert-dwelling
non-pterodactyloid pterosaurs and predates all known desert pterosaurs by more than 65 Myr. A phylogenetic analysis shows
it is closely allied with Dimorphodon macronyx from the Early Jurassic of Britain.

P
terosaurs are extinct archosaurs that ranged from the Late dune complex is divided into the Nugget Sandstone (primarily in
Triassic to the Late Cretaceous (around 215−​66 million years northeastern Utah and Wyoming) and its stratigraphic equivalents,
ago) and were the first vertebrates to attain powered flight. the Aztec Sandstone and Glen Canyon Group, to the south9,10 (Fig. 1)
Their extremely fragile skeletons are rare, and their record is strongly The Saints & Sinners Quarry is within the Nugget Sandstone,
biased towards marine and lacustrine depositional environments1–3. approximately 65 m above the top of the underlying Chinle
The unequivocal record of Triassic pterosaurs consists of fewer than Formation (Fig. 2a). The quarry is within interdunal sediments
30 specimens, including single bones3,4. They are known exclusively bounded above and below by aeolian sands (Fig. 2b). The bones are
from marine strata in the Alps (Italy, Austria and Switzerland), preserved in structureless, silty fine-grained sandstone deposited
except for Arcticodactylus cromptonellus from fluvial deposits in in the shallow, near-shore waters of an interdunal lake into which
Greenland3 (Fig. 1). Fragmentary remains and isolated teeth in dunes were migrating11. The quarry has yielded >​18,000 bones and
Upper Triassic rocks of the USA have been referred to pterosaurs bone fragments of 9 tetrapod taxa: the pterosaur described herein,
(see references within ref. 4, as well as refs 5,6; A. K. Behrensmeyer, a coelophysoid theropod dinosaur, a medium-sized theropod of
personal communication), but these referrals have been shown to unknown affinity, 2 crocodylomorphs, 2 sphenodontians, a drepa-
be unsubstantiated, are ambiguous, or are not yet adequately sup- nosaurid and a procolophonid.
ported or reported3,5. The high degree of homoplasy in multicusped The Nugget Sandstone spans the Triassic–Jurassic bound-
teeth among cynodonts, pterosaurs and tanystropheids3,4,7,8 makes ary but the boundary’s position within the formation is unre-
the referral of isolated teeth and fragmentary tooth-bearing ele- solved10–12. Drepanosauromorphs and procolophonids, however, are
ments to pterosaurs problematic. known exclusively from the Carnian to Rhaetian13 and Permian to
Here, we present unequivocal evidence of a Triassic pterosaur Rhaetian14, respectively. These taxa, and evidences in ref. 11, indicate
from the western hemisphere outside Greenland. The specimen, that: (1) the quarry horizon is latest Triassic (probably late Norian
BYU 20707, is from the Saints & Sinners Quarry in the Nugget or Rhaetian) in age; and (2) the Triassic–Jurassic boundary within
Sandstone of Utah (Fig. 1). Its significance lies in: (1) its excep- the Nugget Sandstone lies somewhere above the quarry.
tional state of preservation, which reveals details obscured in
other early pterosaurs; (2) its close phylogenetic relationship with Systematic palaeontology. Pterosauria Kaup, 1834
Dimorphodon macronyx from the lowermost Jurassic of the UK; Dimorphodontidae Seeley, 1870
and (3) its desert habitat. Caelestiventus hanseni gen. et sp. nov.

Results Etymology. C. hanseni comes from the Latin language: caelestis,

Palaeoenvironment and fauna. Arid conditions dominated what is ‘heavenly or divine’, and ventus, ‘wind’, referring to the volant nature
now western North America for up to around 45 Myr from the Late of pterosaurs, and ‘hanseni’, honouring Robin L. Hansen, Bureau
Triassic through the Middle Jurassic9, and desertification produced of Land Management (BLM) geologist, who facilitated work at the
a sand sea that covered some 2.2 million km2 (Fig. 1). This vast sand Saints & Sinners Quarry.

1
Museum of Paleontology, Department of Geological Sciences, Brigham Young University, Provo, UT, USA. 2Institut Català de Paleontologia Miquel
Crusafont, Sabadell, Spain. 3Dinosaur National Monument, National Park Service, Jensen, UT, USA. 4Department of Geography and Geology, University
of Nebraska at Omaha, Omaha, NE, USA. 5Department of Biology and Monte L. Bean Life Science Museum, Brigham Young University, Provo, UT, USA.
*e-mail: [email protected]

Nature Ecology & Evolution | www.nature.com/natecolevol

Articles Nature Ecology & Evolution

Arid a b c
Formation

Dunes
Tracks
Warm-temperate

Marine
Disarticulated

MJ
USA Greenland Boreotropical Carmel bones
Switzerland Austria Tropical Articulated

1m
Lower Jurassic

Structureless former dunes with

Nugget Sandstone
30° N
Nugget Salt flat

cross-bed relics
Lost to

Aeolian
erosion Italy
0° 3

200 m
U t a h
?
Glen

1–3
Canyon 30° S
Group Pterosaur

Interdune
Upper Triassic
Aztec 250 km 2

Bone beds
Lacustrine
Pterosaur
Fig. 1 | Locality map. Saints & Sinners Quarry in northeastern Utah is
1

Aeolian
indicated by a star. The locality details are on file with the BYU Museum

65 m

55 m
of Paleontology. The geographic extent of the Triassic–Jurassic erg is Pterosaur
represented by the dune sands of the Nugget Sandstone, Aztec Sandstone

Interdune
Fluvial-

0.25 m
Interdune
and Glen Canyon Group in the present-day western USA9,10. The Late eolian

flat
flat
Fluvial
Triassic (Carnian, Norian and Rhaetian) palaeogeography is shown, with Chinle
Mollweide projection50. Epicontinental seas are shown in light blue. Dots
indicate other Late Triassic pterosaur localities.
Fig. 2 | Stratigraphy and depositional environments of the Saints &
Sinners Quarry. a, Stratigraphically, the quarry is about 65 m above the
conformable Nugget Sandstone–Chinle Formation contact. The basal
Holotype. BYU 20707 (Fig. 3 and Supplementary Table 1), depos- Nugget Sandstone is transitional, with fluvial and aeolian sediments. The
ited in the Museum of Paleontology, Brigham Young University Upper Triassic–Lower Jurassic boundary lies somewhere in the Nugget
(BYU), Provo, Utah. It includes the left maxilla fused with the jugal, Sandstone above the quarry horizon. b, The quarry is near the base of a
the right maxilla, the right nasal, the fused frontoparietals, the right 10-m-thick set of interdunal units representing interdunal flat, lacustrine,
and left mandibular rami, the right terminal wing phalanx and three salt pan and bioturbated dune environments11. This wet-phase, interdunal
fragments of indeterminate bones (Supplementary Table 1). package is bounded above and below by cross-bedded aeolianites. c, The
quarry has three superimposed bone beds composed of silt to fine sands,
Locality, horizon and age. Bone bed 2 (Fig. 2c) of the Saints separated by sub-mm-thick clay laminae, all blown into the shallows
& Sinners Quarry (Fig. 1), in the Nugget Sandstone, northeast- of a lake11. The pterosaur (star) is from the middle bone bed. Tridactyl
ern Utah, USA. The age is latest Triassic, probably late Norian theropod tracks, assignable to Grallator sp., occur on silty sandstones with
or Rhaetian. salt pressure ridges and biofilm crenulations, representing a salt flat that
overlies the uppermost bone bed11. MJ, Middle Jurassic.
Diagnosis. Dimorphodontid pterosaur with the following unique
characters: square and tall jugal (ventral margin of the orbit at mid-
height of the antorbital fenestra) braced medially by a diagonal ridge no elements are duplicated, and no other pterosaur bones are
pierced by a large pneumatic foramen; comparatively small, antero- known from the quarry. They are largely uncrushed and undistorted
laterally facing orbit; frontoparietal that is short and broader than (Fig. 3), revealing details that cannot be discerned in the ­usually
long, with a transverse nuchal crest that is C-shaped in dorsal view, crushed and flattened skeletons of other early pterosaurs4.
and adductor muscle ridges curving posteriorly and meeting a low The triradiate maxilla is fused with the jugal (Fig. 3h–j). The
sagittal ridge to form a trifurcate structure; anterodorsally directed ascending process is straight, narrow, S-shaped in cross section
premaxillary process of the nasal; maxillary tooth crowns that are and angled 125–130° posteriorly. A dorsoventrally elongated neu-
leaf-shaped and labiolingually compressed with a median ridge on rovascular foramen enters the lateral side of the expanded base
the labial and lingual sides; interdental space between the maxillary of the ascending process, as in D. macronyx and Preondactylus
teeth that is much shorter than the mesiodistal crown length below buffarinii. The articular surface for the premaxilla is long, indi-
the ascending process and along the jugal process, and more widely cating that the premaxilla formed much of the ventral margin of
spaced along the premaxillary ramus; small elliptical Meckelian the external naris, as in D. macronyx15,16 and unlike in most other
fenestra at the mid-length of the mandibular ramus; large mandibu- non-monofenestratan pterosaurs. The shape of the maxilla and out-
lar teeth 1 and 2, followed by 38 small teeth with labiolingually com- line of the external naris and antorbital fenestra are very similar to
pressed crowns that are triangular and symmetrical and increase in D. macronyx15,16. The external naris is enormous, with an obtuse
size distally in the series. triangular outline, and larger than the antorbital fenestra whose
C. hanseni has a narrow remnant of an antorbital fossa in the outline is a right triangle (Figs. 3h,i and 4b). Medially, the maxilla
jugal (a structure possibly present in the eudimorphodontid exhibits numerous pneumatic features (Fig. 3i,j).
Carniadactylus rosenfeldi) and a furrow along the posterior mar- The body of the jugal is an unusually tall, square plate (Fig. 3h,i).
gin of the terminal wing phalanx (like the rhamphorhynchids It is relatively higher than in any other pterosaurs, including
Rhamphorhynchus muensteri and Nesodactylus hesperius). Parapsicephalus purdoni, for which a high jugal is a diagnostic fea-
ture17. The plate is strengthened by a thickened antorbital border
Description. The pterosaur bones are from a <​1 m2 area and a narrow, diagonal, medial ridge (Fig. 3i) with a lateral pneu-
(Supplementary Fig. 1) around 10 cm above the base of bone bed 2 matic foramen (Fig. 3h). There is a narrow, deep, antorbital fossa
of the Saints & Sinners Quarry (Fig. 2c). The bones (Supplementary along the anterior margin of the jugal (Fig. 3h). The base of the
Table 1) pertain to a single individual as they were closely ­associated, orbit is shallowly concave and level with the middle of the ascending

Nature Ecology & Evolution | www.nature.com/natecolevol

Nature Ecology & Evolution Articles
Broken Sandp Teeth dia al2
rp Glenoid Surangular al1
Teeth
Dentary

Angular
al2 Keel Glenoid rp
al1 Sandp
dia vf
mdf al Dentary al rim Tooth

pf
Adductor
b fossa
Keel
al2 Prearticular
Cross section c mdf
al vf Splenial Cross section d al2
Tooth Ascending mf dia
dia
Medial

process al1

pnc
f e popj
Medial

vf
Tooth Keel
Splenial pnc Lap Orbit
al rim Sulcus Orbit pf Adductor ridge
Jugal
Sagittal crest
Ir
Keel

mdf aof ltf

g nvf
d Dentary nf
c en
pmxpm jaof Nuchal
mn
crest
Frontal
h
n Parietal
Ascending qjp utf
process jp pf
qjf
Tooth 8 pf
popf
Ridge Cerebellum
popj Ridge Cerebrum Frontal
m Ridge
utf
ba
Orbit Lap Optic
lobe
k
Parietal
Jugal
Ridge ba Cerebral
ltf Nasal process
Sulcus Cerebrum fossa
en
aof
p
pfo l pfo Olfactory bulb
pf pmxpm
ba
Orbit Premaxillary
pf
qjp Frontal o process
process
i jp Ascending Palatal plate q
pfo en
Tooth 11 al Tooth 5 process Tooth 3 pmxpm
Tooth 8
qjf Jugal al Maxillary
Ridge process
qjp
jp
al dia al dia r Frontal
al en
j process

Maxillary Premaxillary
process
s Furrow process

Fig. 3 | C. hanseni gen. et sp. nov., a Triassic dimorphodontid from the Nugget Sandstone near Vernal, Utah. BYU 20707 holotype images from micro-
computed tomography scans. a,b, Right mandibular ramus in lateral (a) and medial views (b). c,d, Transverse sections from rostral (c) and posterior (d)
portions of the right mandibular ramus, with positions indicated in b. e, Rostral half of the left mandibular ramus in dorsomedial view. f, Occlusal view of a
distal tooth and adjacent alveolus of the left mandibular ramus. g, Lingual view of the tooth. h–j, Fused left maxillojugal lateral (h), medial (i) and ventral
views (j). k,l, Maxillary tooth 8 in labial (k) and lingual views (l). m, Maxillary tooth 11 in labial view. n,o, Frontoparietals in dorsal (n) and ventral views
(o). p, Cranial endocast in left anterolateral view. q,r, Right nasal in ventrolateral (q) and dorsolateral views (r). s, Right wing phalanx 4. Scale bars: 10 mm,
except in c, d and k–m (5 mm), and f and g (2 mm). Dashed lines indicate portions known from opposite dentary, or impressions in rock for the phalanx.
al, alveolus; aof, antorbital fenestra; ba, bifid apex; dia, diastema; en, external naris; jaof, antorbital fossa of jugal; jp, jugal process; lap, lacrimal process; lr,
longitudinal ridge; ltf, lower temporal fenestra; mdf, medially deflected flange; mf, Meckelian fenestra; mn, median notch; nf, nasal facet; nvf, neurovascular
foramen; pf, pneumatic foramen; pfo, pneumatic fossa; pmxpm, premaxillary process of maxilla; pnc, pneumatic chamber; popf, postorbital process of the
frontal; popj, postorbital process of jugal; qjf, quadratojugal facet of jugal; qjp, quadratojugal process of jugal; rp, retroarticular process; sandp, surangular
dorsal process; utf, upper temporal fenestra; vf, vertical flange.

­ rocess of the maxilla, resulting in a high, round orbit, while in

p corner of the external naris. The orientation of the premaxil-
D. macronyx the orbit is vertically elongate with an acute basal angle. lary process suggests that the premaxilla was arched only above
The nasal (Fig. 3q,r) is elongate and slender, with a long the naris, rather than over the entire rostrum, as
frontal process. The bifid rostral end defines the posterodorsal in D. macronyx.

Nature Ecology & Evolution | www.nature.com/natecolevol

Articles Nature Ecology & Evolution

as in Eudimorphodon ranzii and D. macronyx and is blunt. In dorsal

n fr pa view, the ramus is straight except for the posterior quarter, which
curves slightly medially. The Meckelian fenestra is an elliptical
opening at the rostral end of the splenial.
Orbit A very thin, blade-like structure extends the full length of the
External
naris ventral dentary margin. Rostrally, the blade is vertical and forms a
a semicircular keel (the margin is damaged) that tapers to a position
aof j
nvf below the first small mandibular tooth. The blade remains a verti-
mx sa cal flange for the first half of the dentary before deflecting ventro-
medially along the remainder of the dentary (Fig. 3b–d). A similar,
d blade-like structure with a rostral keel exists in D. macronyx24, but
a the post-keel flange was misinterpreted because it is crushed into
b
Keel the same plane as the keel. In C. hanseni, there is a thick, longitudi-
vf mdf fr pa utf
nal ridge (Fig. 3d) along the dorsomedial side of the ventral flange.
This ridge (the ‘ventral ridge’ of ref. 16) is well exposed in D. ­macronyx
n l specimens due to crushing and flattening of the flange.
Orbit
External
The angular (Fig. 3a) has a large opening laterally, but jagged
naris margins and bone fragments within the mandible indicate that the
aof
opening is a syndepositional break, not an external mandibular
nvf j
pmx fenestra. The glenoid is well above the tooth row, as in D. ­macronyx
mx and unlike most pterosaurs. The pointed dorsal process of the
surangular is the anterior end of a blade-like ridge extending along
the dorsal margin of the surangular above the adductor fossa. The
d
retroarticular process is dorsally concave and pierced by large pneu-
c Keel matic foramina (Fig. 3a,b).
vf mdf
The maxilla bore 12 teeth, represented in the left maxilla by 8
Fig. 4 | Reconstructions of C. hanseni and D. macronyx. a, Schematic alveoli and 4 teeth; teeth 3 and 5 are partially erupted (Fig. 3h–j).
silhouette of a dimorphodontid pterosaur in dorsal view. b, Preserved skull There is an edentulous gap between the premaxillary and max-
and mandible elements of C. hanseni. The left maxillojugal is completed illary teeth, as in Raeticodactylus filisurensis25 and D. macronyx16.
with the premaxillary process from the right maxilla. The right mandibular The last tooth position is slightly posterior to the posterior border
ramus is mirrored and is completed with the rostral end from the left of the antorbital fenestra. Maxillary teeth are widely spaced along
ramus. Missing teeth have been reconstructed. The right nasal and the premaxillary ramus but closely spaced along the remainder of
frontoparietal are mirrored. c, Skull and mandible of the largest D. macronyx the element. There is no regular posterior decrease in tooth size
reconstructed from ref. 16 and modified incorporating our observations. as in D. macronyx16, P. buffarinii and Austriadactylus cristatus4,
Scale bars: 0.5 m in a and 10 mm in b and c. a, angular; aof, antorbital but the last three teeth are smaller than the other teeth. Maxillary
fenestra; d, dentary; fr, frontal; j, jugal; l, lacrimal; mx, maxilla; mdf, medially teeth below the ascending process are not much larger than the
deflected flange; n, nasal; nvf, neurovascular foramen; pa, parital; pmx, others, unlike E. ranzii, P. buffarinii and A. cristatus4. Fully erupted
premaxilla; sa, surangular; utf, upper temporal fenestra; vf, ventral flange. tooth crowns 8 and 11 (7.60 and 4.40 mm high, respectively) are
broad, leaf shaped (Fig. 3k–m) and labiolingually thin (Fig. 3j).
Mesiodistal cutting edges lack cusps and serrations. There is a low
The co-ossified frontals and parietals (Fig. 3n,o) are anteropos- and wide median, basoapical ridge on the labial and lingual sur-
teriorly short, hexagonal in outline and domed over the cerebral faces of the crowns (Fig. 3k–m). The apices of all preserved maxil-
area, with a low, thin sagittal ridge. The anterior margin has a deep, lary teeth are bifid (Fig. 3k–m), as in some dentary tooth crowns
median, V-shaped notch for the premaxillae. The transverse nuchal of D. macronyx16.
crest is high and concave posteriorly (C-shaped in dorsal view). The Each dentary has 40 teeth, most represented by empty alveoli.
anteromedial rim of each mandibular adductor muscle origin is a The first tooth was procumbent and the second more erect. Their
ridge. These ridges converge posteromedially with the sagittal ridge. large (3−​4 mm mesiodistally), ovate alveoli indicate they were fang-
The large pneumatic foramina posterior to the anterior terminus of like. These fangs are followed by 38 small, closely spaced teeth with
each adductor ridge communicates with a small pneumatic cham- a crown height range of 1.12−​2.24 mm. There is a general mesial-to-
ber. Allkaruen koi18 and R. muensteri19 have similarly positioned distal trend of increasing size in mandibular teeth 3−​40 (Fig. 3a,b)
pneumatic foramina. The orbital margin of the frontal are short and they are labiolingually compressed (Fig. 3f). There are long dia-
and, in contrast with other pterosaurs, face anterolaterally1,4,18–22. stemata between teeth 1 and 2 and between tooth 2 and the follow-
The cranial endocast (Fig. 3p), digitally rendered, is the only ing small tooth. This second diastema is 17 mm long and concave.
instance from a Triassic pterosaur. The olfactory lobes are min- The same tooth pattern and diastemata occur in D. macronyx16,24.
ute and the cerebral lobes large. The low-relief bulges of the optic The small, evenly spaced mandibular teeth have triangular, sym-
lobes are posterior and at the same level as the cerebral lobes, as metrical, monocuspid crowns that are higher than long in mesial
expected in a basal pterosaur where there is no pronounced flexure teeth and longer than high in distal teeth (a similar change in crown
of the brain that displaces the optic lobes ventrally18. The cerebel- proportions occurs in Peteinosaurus zambellii4).
lum has a posterodorsal expansion related to the nuchal crest, as in Excluding teeth 1 and 2, dentary teeth are much smaller than
Allkaruen18. In Rhamphorhynchus, the cerebellum is not expanded maxillary teeth, as in P. buffarinii, A. cristatus and D. macronyx, and
dorsally, and the optic lobes are comparatively smaller and anteri- differ in shape.
orly bent below the level of the cerebral lobes23. The only known postcranial element is the terminal right wing
The mandibular rami (Fig. 3a–e) are slender (length: mid-ramus phalanx (140 mm long, missing the distal tip) (Fig. 3s). Its concave
height =​ 14.8). The short symphysis is unfused. The rami have thin posterior margin has a distinct longitudinal furrow bordered by a
walls and a large internal pneumatic chamber (Fig. 3c,d). The den- sharp ridge representing the insertion zone of the retrophalangeal
tary forms 80% of each ramus’ length. The rostral tip is downturned wedge of the brachiopatagium26.

Nature Ecology & Evolution | www.nature.com/natecolevol

Nature Ecology & Evolution Articles
Phylogenic analysis. The paper-thin compacta, maxillary and man- Macrocnemus bassanii
dibular morphology (which is close to that of D. macronyx; Fig. 4c) Postosuchus kirkpatricki
and wing phalanx support the referral of BYU 20707 to Pterosauria. Herrerasaurus ischigualastensis

To determine the phylogenetic position of C. hanseni within Preondactylus buffarinii

Austriadactylus cristatus
Pterosauria, a cladistic analysis by parsimony was performed
Austriadraco dallavecchiai
using PAUP 4.0a161 (ref. 27) with an updated version of the data-
Arcticodactylus cromptonellus
set in ref. 28 (see Methods). C. hanseni was recovered as the sister
Peteinosaurus zambellii
taxon of D. macronyx (Fig. 5). ‘Dimorphodon’ weintraubi from the Pterosauria
Dimorphodon macronyx
Lower or Middle Jurassic of Mexico29,30—the only other purported Caelestiventus hanseni
Dimorphodontidae
Dimorphodon species—is the sister taxon to Anurognathidae sensu Eudimorphodon ranzii
ref. 31, and thus falls outside Dimorphodontidae (Fig. 5), which is Carniadactylus rosenfeldi
here defined as the least inclusive clade containing D. macronyx Caviramus schesaplanensis
Eudimorpho-
Buckland 182932 and C. hanseni (this paper). Macronychoptera
dontidae Raeticodactylus filisurensis
C. hanseni uniquely shares the following derived features with Campylognathoides species
D. macronyx (Fig. 4b,c): a ventral blade along the dentary that forms Lonchognatha Dorygnathus banthensis
a rostral keel and becomes a flange distally; a diastema between the Rhamphorhynchus muensteri
second large mandibular tooth and the following smaller teeth; the Scaphognathus crassirostris
overall morphology of the maxilla; the shape of the external naris Novialoidea Sordes pilosus
and antorbital fenestra; the external naris by far the largest skull Darwinopterus species
opening; the orbit smaller than the antorbital fenestra; and teeth Pterodactylus antiquus
with bicuspid apices. Pteranodon longiceps
Caelidracones
C. hanseni differs from D. macronyx in numerous ways (Fig. 4b,c). Anhanguera species
Monofenestrata ‘Dimorphodon’ weintraubi
In D. macronyx: (1) the jugal process of the maxilla ends anterior
Batrachognathus volans
to the posterior margin of the antorbital fenestra, and the suture Pterodactyloidea
Anurognathus ammoni
between the maxilla and jugal is vertical and sigmoid16; (2) the jugal Anurognathidae Dendrorhynchoides species
does not contribute to the antorbital fossa16,24,33; (3) the orbit faces
Jeholopterus ningchengensis
laterally and deeply notches the jugal16; (4) the orbit shape is an
inverted tear drop16; (5) the frontoparietal is more elongated than Fig. 5 | Phylogenetic relationships of C. hanseni gen. et sp. nov. Strict
in C. hanseni16; (6) the nasal is comparatively shorter and triangular, consensus tree of the six most parsimonious trees as computed in
with an anteriorly directed premaxillary process16; (7) the suran- PAUP 4.0a161 (ref. 27) with the wildcard taxon MCNSB 8950 removed
gular dorsal process is a small, blunt point (BMNH 43486); and (length =​281, consistency index =​0.583, retention index =​ 0.703) (see
(8) there is no longitudinal furrow along the posterior margin of the Methods and Supplementary Information).
wing phalanges34.
Both C. hanseni and D. macronyx are heterodont, with the max-
illary teeth much larger than the mandibular teeth, but the tooth The skull was narrow, as indicated by the slight medial curva-
crowns are markedly different. The maxillary teeth of D. macronyx: ture of the mandibular rami and, especially, frontoparietal width.
(1) are fewer in number; (2) are more widely spaced below the The function of the medially deflected mandibular flange is
ascending process and along the jugal ramus; (3) have a last ­puzzling. Its shape and position suggest it may mark the attach-
tooth that is more anteriorly positioned; and (4) are conical, labi- ment of gular skin. Gular skin is preserved in Rhamphorhynchus
oligually thick, and lack wide basoapical ridges16. The crowns and Pterodactylus36,37. Such a structure could have been a throat
of the small mandibular teeth of D. macronyx are relatively pouch used to store prey, as in some birds38. A gular pouch may
taller apicobasally and semi-conical (moderately labiolingually also function in visual display and vocal communication, as in
compressed)16 compared with those of C. hanseni, which are­ male frigatebirds39 or the flying lizard Draco40.
blade-like (labioligually thin). Pneumatic features in most skull and mandibular elements
of C. hanseni show the degree to which pneumaticity was devel-
Discussion oped in some Triassic pterosaurs (discussed in the Supplementary
The mandible indicates that the skull of C. hanseni was around Information).
180 mm long, which is 82% of the skull length of the largest Tetrapod bones are rare in the Upper Triassic–Lower Jurassic
D. ­macronyx specimen16. The wing phalanx 4 of C. hanseni is the aeolian deposits of the western USA, and worldwide pterosaur fos-
longest of any non-pterodactyloid pterosaur35. Scaling wing pha- sils from palaeo desert settings are extraordinarily rare. A tapeja-
lanx 4 of a small D. macronyx specimen (GSM 1546) to the size rid (Caiuajara dobruskii) bone bed occurs in an Upper Cretaceous
of the largest specimen, the terminal wing phalanx of the largest interdunal lake deposit of Brazil41, and ctenochasmatid remains no
specimen would be 96.8 mm—only 69% of the preserved length in older than the Early Cretaceous (Hauterivian; around 132 million
C. hanseni. This means that the Utah taxon could have: (1) a pro- years ago (Ma)42) occur in desert floodplain deposits in Chile43,44.
portionally smaller head than D. macronyx; (2) proportionally lon- Both occurrences preserve numerous individuals and may repre-
ger wings; or (3) just a proportionally longer terminal wing phalanx. sent breeding colonies41,43. This contrasts with the Saints & Sinners
The relationship between the length of the terminal wing phalanges Quarry, where only a single pterosaur has been identified out of
and wing span in other non-pterodactyloid pterosaurs34,35 suggests >​18,000 bones.
a wing span greater than 1.5 m for the new taxon. Previously, the
largest Triassic pterosaur specimens were the holotypes of E. ­ranzii, Conclusion
R. filisurensis and A. cristatus, with 90-, 95- and 110-mm-long C. hanseni is the first record of a desert-dwelling, non-pterodacty-
skulls, respectively4. Based on skull length and estimated wing span, loid pterosaur, predating by >​65 Ma all known desert occurrences
C. hanseni is one of the largest Triassic pterosaurs, if not the largest. of pterosaurs. It shows that in the Late Triassic—early in their
Unfused skull and mandibular elements suggest that BYU 20707 evolution and diversification—pterosaurs were widely distributed
was skeletally immature or C. hanseni had indeterminate growth24 and lived in a broad range of habitats, including inland deserts far
(discussed in the Supplementary Information). (>​800 km) from the sea. Despite their morphological similarity,

Nature Ecology & Evolution | www.nature.com/natecolevol

Articles Nature Ecology & Evolution

C. hanseni and D. macronyx lived in very different environments. 10. Sprinkel, D. A., Kowallis, B. J. & Jensen, P. H. in Sevier Thrust Belt—Northern
Dimorphodon was an island dweller45 in a humid climate46 and and Central Utah and Adjacent Areas (eds Sprinkel, D. A., Yonkee, W. A. &
Chidsey, T. C. Jr) 131–149 (Utah Geological Association, Salt Lake City, 2011).
was preserved in the marine Blue Lias (Hettangian–Sinemurian) 11. Britt, B. B., Chure, D. J., Engelmann, G. F. & Shumway, J. D. Rise of the Erg:
of southern England16. C. hanseni indicates that dimorphodontids Paleontology and Paleoenvironments of the Triassic–Jurassic Transition in
originated by the Late Triassic and survived the end-Triassic extinc- Northeastern Utah (Utah Geological Association, Salt Lake City, 2016).
tion event. 12. Irmis, R. B., Chure, D. J., Engelmann, G. F., Wiersma, J. P. & Lindström, S. in
The Uinta Basin and Uinta Mountains (eds Vanden Berg, M. D., Ressetar, R. &
Birgenheier, L. P.) 13–48 (Utah Geological Association, Salt Lake City, 2015).
Methods 13. Renesto, S., Spielmann, J. A., Lucas, S. G. & Tarditi Spagnoli, G. The Taxonomy
Specimen preparation. Upper surfaces of the bones were mechanically prepared and Paleobiology of the Late Triassic (Carnian–Norian: Adamanian–Apachean)
using needles under a binocular microscope. Bones were digitally segmented from Drepanosaurs (Diapsida: Archosauromorpha: Drepanosauromorpha)
the sandstone using micro-computed tomography images obtained from a Siemens (New Mexico Museum of Natural History and Science, Albuquerque, 2010).
Inveon PET/SPECT/CT scanner at the Small Animal Imaging Core Facility, 14. Cisneros, J. C. Phylogenetic relationships of procolophonid parareptiles with
University of Utah and on an NSI scanner with a Fein Focus High Power source at remarks on their geological record. J. Syst. Palaeontol. 6, 345–366 (2008).
the High-Resolution X-ray CT facility at the University of Texas. Segmentation and 15. Padian, K. Osteology and functional morphology of Dimorphodon macronyx
wireframe rendering was conducted in ORS Visual SI 1.8. Lighting was applied in (Buckland) (Pterosauria: Rhamphorhynchoidea) based on new material in the
Blender 2.77. Yale Peabody Museum. Postilla 189, 1–43 (1983).
16. Sangster, S. The Anatomy, Functional Morphology and Systematics of
Phylogenetic analysis. The phylogenetic analysis used an updated data Dimorphodon macronyx (Diapsida: Pterosauria). PhD thesis, Univ.
matrix from ref. 28. The original matrix was modified by adding 31 characters, Cambridge (2003).
substantially revising 22 characters and eliminating 11 characters. The suggestions 17. O’Sullivan, M. & Martill, D. M. The taxonomy and systematics of
from ref. 47 were followed as much as possible in character revision. Our data Parapsicephalus purdoni (Reptilia: Pterosauria) from the Lower Jurassic
matrix consists of 93 characters coded across 29 taxa (3 outgroup and 26 ingroup). Whitby Mudstone Formation, Whitby, U.K. Hist. Biol. 29, 1009–1018 (2017).
Ticinosuchus ferox was replaced by Postosuchus kirkpatricki as an outgroup 18. Codorniú, L., Paulina-Carabajal, A., Pol, D., Unwin, D. & Rauhut, O. W. M.
taxon. The following tree ingroup taxa were added: C. hanseni gen. et sp. nov., A Jurassic pterosaur from Patagonia and the origin of the pterodactyloid
Darwinopterus (Darwinopterus modularis, Darwinopterus robustodens and neurocranium. PeerJ 4, e2311 (2016).
Darwinopterus linglongtaensis were coded together) and ‘Dimorphodon’ weintraubi. 19. Hone, D. W. E., Habib, M. B. & Lamanna, M. C. An annotated and illustrated
Dendrorhynchoides curvidentatus was coded together with Dendrorhynchoides catalogue of Solnhofen (Upper Jurassic, Germany) pterosaur specimens at
mutouangensis. The names of the operational taxonomic units ‘Eudimorphodon’ Carnegie Museum of Natural History. Ann. Carne. Mus. 82, 165–191 (2013).
cromptonellus and BSP 1994 I 51 were updated according to ref. 48. Missing data 20. Gasparini, Z., Fernández, M. & de la Fuente, M. A new pterosaur from the
were coded as ‘?’ and inapplicable data were coded as gaps, although gaps were Jurassic of Cuba. Palaeontology 47, 919–927 (2004).
treated as missing in the phylogenetic analysis. Three multistate characters were 21. Padian, K. The Early Jurassic pterosaur Dorygnathus banthensis (Theodori,
treated as ordered (character numbers 63, 75 and 92); the remaining characters 1830). Spec. Pap. Palaeontol. 80, 1–64 (2008).
were treated as unordered. However, treating the ordered characters as unordered 22. Padian, K. The Early Jurassic pterosaur Campylognathoides Strand, 1928.
did not influence the final topology, and only character number 92, when treated Spec. Pap. Palaeontol. 80, 65–107 (2008).
as ordered, provided one additional step to the final topology. Trees were rooted 23. Witmer, L. M., Chatterjee, S., Franzosa, J. & Rowe, T. Neuroanatomy of flying
to the outgroups with PAUP and to Macrocnemus bassanii as the most distant reptiles and implications for flight, posture and behaviour. Nature 425,
outgroup with TNT49. They were computed under maximum parsimony. 950–953 (2003).
Characters, character states, scorings and the matrix data files are given in the 24. Bennett, S. C. An external mandibular fenestra and other archosauriform
Supplementary Information. characters in basal pterosaurs re-examined. Hist. Biol. 27, 796–814 (2015).
Trees were computed in PAUP 4.0a build 161 (ref. 27) using 10,000 random- 25. Stecher, R. A new Triassic pterosaur from Switzerland (Central Australpine,
addition sequences and tree bisection and reconnection branch swapping. Trees Grisons), Raeticodactylus filisurensis gen. et sp. nov. Swiss J. Geosci. 101,
were also computed in TNT version 1.5 (ref. 49) using the commands ‘hold 1000; 185–201 (2008).
mult =​ replic 10000 hold 10’. 26. Bennett, S. C. New interpretation of the wings of the pterosaur
Rhamphorhynchus muensteri based on the Zittel and Marsh specimens.
Reporting Summary. Further information on experimental design is available in J. Paleontol. 89, 845–869 (2016).
the Nature Research Reporting Summary linked to this article. 27. Swofford, D. L. PAUP. Phylogenetic Analysis Using Parsimony (and Other
Methods) (Sinauer Associates, Sunderland, 2002).
Data availability. The specimen (BYU 20707) is curated in the Museum of 28. Dalla Vecchia, F. M. Austriadactylus (Diapsida, Pterosauria) from the Norian
Paleontology at Brigham Young University. Taxonomic data are deposited in (Upper Triassic) of northeastern Italy. Riv. Ital. Paleontol. Stratigr. 115,
ZooBank under Life Science Identifier urn:lsid:zoobank.org:pub:783A84EB-80A0- 291–304 (2009).
431F-A908-5C57C476FAAE. The authors declare that all other data supporting 29. Clark, J. M., Montellano, M., Hopson, J. A., Hernandez, R. R. & Fastovsky, D. E.
the findings of this study are available within the paper and its Supplementary in In the Shadow of the Dinosaurs: Early Mesozoic Tetrapods (eds Fraser, N. C.
Information. & Sues, H.-D.) 295−​302 (Cambridge Univ. Press, Cambridge, 1994).
30. Clark, J. M., Hopson, J. A., Hernandez, R. R., Fastovsky, D. E. & Montellano,
Received: 16 February 2018; Accepted: 28 June 2018; M. Foot posture in a primitive pterosaur. Nature 391, 886–889 (1998).
Published: xx xx xxxx 31. Unwin, D. M. in Evolution and Palaeobiology of Pterosaurs (eds Buffetaut, E.
& Mazin, J.-M.) 139–190 (Geological Society of London, London, 2003).
References 32. Buckland, W. On the discovery of a new species of pterodactyle in the Lias at
1. Wellnhofer, P. Pterosauria. Handbuch der Paläoherpetologie Vol. 19 (Gustav Lyme Regis. Trans. Geol. Soc. Lond. 2, 217–222 (1829).
Fischer, Stuttgart, 1978). 33. Nesbitt, S. J. & Hone, D. W. E. An external mandibular fenestra and other
2. Kellner, A. W. A. Remarks on pterosaur taphonomy and paleoecology. Acta archosauriform character states in basal pterosaurs. Palaeodiversity 3,
Geol. Leopold. 17, 175–189 (1994). 225–233 (2010).
3. Dalla Vecchia, F. M. in Anatomy, Phylogeny and Palaeobiology of Early 34. Unwin, D. M. New remains of the pterosaur Dimorphodon (Pterosauria:
Archosaurs and Their Kin: Triassic Pterosaurs (eds Nesbitt, S. J., Desojo, J. B. Rhamphorhynchoidea) and the terrestrial ability of early pterosaurs.
& Irmis, R. B.) 119−​155 (Geological Society of London, London, 2013). Mod. Geol. 13, 57–68 (1998).
4. Dalla Vecchia, F. M. Gli Pterosauri Triassici (Museo Friulano di Storia 35. Dalla Vecchia, F. M. A wing phalanx of a large basal pterosaur (Diapsida,
Naturale, Udine, 2014). Pterosauria) from the Norian (Late Triassic) of NE Italy. Boll. Soc. Paleontol.
5. Andres, B. The earliest pterosaurs. J. Vertebr. Paleontol. 26, 37A (2006). Ital. 39, 229–234 (2000).
6. Andres, B. & Myers, T. S. Lone star pterosaurs. Earth Env. Sci. Trans. R. Soc. 36. Frey, E. & Martill, D. M. Soft tissue preservation in a specimen of Pterodactylus
Edinb. 103, 1–16 (2013). kochi (Wagner) from the Upper Jurassic of Germany. Neues Jahrb. Geol.
7. Hahn, G., Lepage, J. C. & Wouters, G. Cynodontier-zähne aus der Ober-Trias Paläontol. 210, 421–441 (1998).
von Medernach. Bull. Soc. Belge. Geol. 93, 357–373 (1984). 37. Frey, E., Tischlinger, H., Buchy, M. C. & Martill, D. M. in Evolution and in
8. Fröbisch, N. B. & Fröbisch, J. A new basal pterosaur genus from the Upper Palaeobiology of Pterosaurs (eds Buffetaut, E. & Mazin, J.-M.) 233–266
Triassic of Northern Calcareous Alps of Switzerland. Palaeontology 49, (Geological Society of London, London, 2003).
1081–1090 (2006). 38. Zweers, G. A., Berkhoudt, H. & Vanden Berge, J. C. in Advances in Comparative
9. Hintze, L. F. & Kowallis, B. J. Geologic History of Utah: A Field Guide to Utah’s and Environmental Physiology Vol. 18 (eds Bels V. L., Chardon M. & Vandewalle P.)
Rocks (Brigham Young Univ., Provo, 2009). 242−​279 (Springer, Berlin & Heidelberg, 1994).

Nature Ecology & Evolution | www.nature.com/natecolevol

Nature Ecology & Evolution Articles
39. Diamond, A. W. Notes on the breeding biology and behavior of the Acknowledgements
magnificent frigatebird. Condor 75, 200–209 (1973). R. L. Hansen (US BLM’s Vernal Field Office) facilitated access under BLM permit
40. Musters, C. J. M. Taxonomy of the genus Draco L. (Agamidae, Lacertilia, UT08-025E. Funding and other support was provided by the BYU Geological Sciences
Reptilia). Zool. Verh. 199, 1–126 (1983). Department and Museum of Paleontology, Dinosaur National Monument (D.J.C.) and
41. Manzig, P. C. et al. Discovery of a rare pterosaur bone bed in a Cretaceous University of Nebraska at Omaha (G.F.E.). N. Fraser and J. Clark provided information
desert with insights on ontogeny and behavior of flying reptiles. PLoS ONE 9, on the procolophonid and ‘Dimorphodon’ weintraubi, respectively. Segmentation was
e100005 (2014). performed by A. J. Matthews and M. Colbert. M. Gonzales-Whittaker skilfully assisted
42. Charrier, R., Pinto, L. & Rodríguez, M. P. in The Geology of Chile (eds Moreno, T. with illustrations. Manual preparation was performed by S. Meek and C. Thelin. Thanks
& Gibbon, W.) 21−​114 (Geological Society of London, London, 2007). to J. Wilson for recommendations that improved an early version of the manuscript.
43. Bell, C. M. & Padian, K. Pterosaur fossils from the Cretaceous of Chile: evidence
for a pterosaur colony on an inland desert plain. Geol. Mag. 132, 31–38 (1995).
44. Martill, D. M., Frey, E., Bell, C. M. & Chong Diaz, G. Ctenochasmatid pterosaurs Author contributions
from Early Cretaceous deposits in Chile. Cretac. Res. 27, 603–610 (2006). B.B.B., F.M.D.V., D.J.C. and G.F.E. designed the research project. B.B.B., F.M.D.V.,
45. Martill, D. M., Vidovic, S. U., Howells, C. & Nudds, J. R. The oldest Jurassic D.J.C., G.F.E., R.D.S. and M.F.W. wrote the manuscript. B.B.B., D.J.C., G.F.E. and R.D.S.
dinosaur: a basal neotheropod from the Hettangian of Great Britain. conducted the field work. F.M.D.V. coded the characters. F.M.D.V. and M.F.W. performed
PLoS ONE 11, e0154352 (2016). the phylogenetic analysis.
46. Weedon, G. P., Jenkyns, H. C. & Page, K. N. Combined sea-level and climate
controls on limestone formation, hiatuses and ammonite preservation in the Competing interests
Blue Lias Formation, South Britain (uppermost Triassic–Lower Jurassic). The authors declare no competing interests.
Geol. Mag. 155, 1117–1149 (2018).
47. Brazeau, M. D. Problematic character coding methods in morphology and
their effects. Biol. J. Linn. Soc. 104, 489–498 (2011). Additional information
48. Kellner, A. W. A. Comments on Triassic pterosaurs with discussion about Supplementary information is available for this paper at https://doi.org/10.1038/
ontogeny and description of new taxa. An. Acad. Bras. Ciênc 87, 669–689 (2015). s41559-018-0627-y.
49. Goloboff, P. & Catalano, S. TNT version 1.5, including a full implementation Reprints and permissions information is available at www.nature.com/reprints.
of phylogenetic morphometrics. Cladistics 32, 221–238 (2016).
50. Boucot, J., Xu, C. & Scotese, C. R. Phanerozoic Paleoclimate: An Atlas of Correspondence and requests for materials should be addressed to B.B.B.
Lithologic Indicators of Climate (Concepts in Sedimentology and Paleontology Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in
Vol. 11, Society for Sedimentary Geology, Tulsa, 2013). published maps and institutional affiliations.

Nature Ecology & Evolution | www.nature.com/natecolevol

 nature research | reporting summary
Corresponding author(s): Brooks B. Britt

Reporting Summary
Nature Research wishes to improve the reproducibility of the work that we publish. This form provides structure for consistency and transparency
in reporting. For further information on Nature Research policies, see Authors & Referees and the Editorial Policy Checklist.

Statistical parameters
When statistical analyses are reported, confirm that the following items are present in the relevant location (e.g. figure legend, table legend, main
text, or Methods section).
n/a Confirmed
The exact sample size (n) for each experimental group/condition, given as a discrete number and unit of measurement
An indication of whether measurements were taken from distinct samples or whether the same sample was measured repeatedly
The statistical test(s) used AND whether they are one- or two-sided
Only common tests should be described solely by name; describe more complex techniques in the Methods section.

A description of all covariates tested

A description of any assumptions or corrections, such as tests of normality and adjustment for multiple comparisons
A full description of the statistics including central tendency (e.g. means) or other basic estimates (e.g. regression coefficient) AND
variation (e.g. standard deviation) or associated estimates of uncertainty (e.g. confidence intervals)

For null hypothesis testing, the test statistic (e.g. F, t, r) with confidence intervals, effect sizes, degrees of freedom and P value noted
Give P values as exact values whenever suitable.

For Bayesian analysis, information on the choice of priors and Markov chain Monte Carlo settings
For hierarchical and complex designs, identification of the appropriate level for tests and full reporting of outcomes
Estimates of effect sizes (e.g. Cohen's d, Pearson's r), indicating how they were calculated

Clearly defined error bars

State explicitly what error bars represent (e.g. SD, SE, CI)

Our web collection on statistics for biologists may be useful.

Software and code

Policy information about availability of computer code
Data collection Segmentation and wireframe rendering was conducted in ORS Visual Si 1.8. Lighting of the renderings was conducted in Blender 2.77
The study was largely based on these images.

Data analysis Phylogenetic methods utilized are described under Methods in the paper. Parsimony analyses were conducted using PAUP 4.0a build
161 and TNT vers. 1.5.
For manuscripts utilizing custom algorithms or software that are central to the research but not yet described in published literature, software must be made available to editors/reviewers
upon request. We strongly encourage code deposition in a community repository (e.g. GitHub). See the Nature Research guidelines for submitting code & software for further information.

Data
April 2018

Policy information about availability of data

All manuscripts must include a data availability statement. This statement should provide the following information, where applicable:
- Accession codes, unique identifiers, or web links for publicly available datasets
- A list of figures that have associated raw data
- A description of any restrictions on data availability
Provide your data availability statement here.

1
 nature research | reporting summary
Field-specific reporting
Please select the best fit for your research. If you are not sure, read the appropriate sections before making your selection.
Life sciences Behavioural & social sciences Ecological, evolutionary & environmental sciences
For a reference copy of the document with all sections, see nature.com/authors/policies/ReportingSummary-flat.pdf

Life sciences study design

All studies must disclose on these points even when the disclosure is negative.
Sample size The study is based on the partial remains of a single individual, BYU 20707.

Data exclusions No data were exluded.

Replication No physical experiments were performed.

Randomization The study was based on a singe specimen so no specimen data were randomized.

Blinding There was no group allocation because the study is based on a single specimen.

Reporting for specific materials, systems and methods

Materials & experimental systems Methods

n/a Involved in the study n/a Involved in the study
Unique biological materials ChIP-seq
Antibodies Flow cytometry
Eukaryotic cell lines MRI-based neuroimaging
Palaeontology
Animals and other organisms
Human research participants

Palaeontology
Specimen provenance The specimen, BYU 20707, is from the Saints & Sinners Quarry which is located in northeastern Utah, USA. Locality details are on
file with the Museum of Paleontology of Brigham Young University as Locality 1442. The specimen was collected under Bureau of
Land Management permit UT08-025E which was issued on 1 January 2009.

Specimen deposition The specimen, BYU 20707, is deposited in the collections of Brigham Young University's Museum of Paleontology and is available
for study to researchers.

Dating methods No radiometric dates are provided. The relative age of Norian or Rhetian presented in the paper is based on biostratigraphy and
constrained by superposition.

Tick this box to confirm that the raw and calibrated dates are available in the paper or in Supplementary Information.
April 2018