LETTER
doi:10.1038/nature13866
A basal ichthyosauriform with a short snout from the Lower Triassic of China Ryosuke Motani1, Da-Yong Jiang2,3, Guan-Bao Chen4, Andrea Tintori5, Olivier Rieppel6, Cheng Ji7 & Jian-Dong Huang4 a 31°51′
117°12′
117°54′
10 km
31°24′
Lithological column
Beds 650
beds 611–999
633
Middle
beds 558–610 beds 508–557
621
611
Helongshan Formation
Argillaceous limestone
Nodular limestone
Columbites
Smi Dolostone
637
Lower
Spathian
Upper member
643
Subcolumbites
Dongmaanshan Formation
598
20 m
591
Procolumbites
Anisian
b
Cartorhynchus lenticarpus gen. et sp. nov. Etymology. kartos (Greek), meaning shortened; rgynxgos (Greek), meaning snout; lentus (Latin), meaning flexible; carpus (Latin), meaning wrist. Named after truncated snout and cartilaginous wrist. Holotype. Anhui Geological Museum AGB6257. Locality and horizon. From the second level of Majiashan Quarry (31u 379 260 N, 117u 499 190 E), near downtown Chaohu, Hefei City, Anhui Province, China (Fig. 1a). Bed 633, about 13 m above the bottom of the Upper Member of the Nanlinghu Formation (Fig. 1b), within the ammonite Subcolumbites zone, Spathian, Olenekian, Lower Triassic (Fig. 1b). Diagnosis. Autapomorphies are: snout only half as long as the rest of the skull; very large hyoid; forelimb strongly curved posteriorly; anteriorly curved hindlimb; ribs robust, with proximal intercostal space narrower
Chaohu
Chaohu Lake
Nanlinghu Formation
Ichthyosauriformes nov. Diagnosis. All ichthyosauromorphs more closely related to Ichthyosaurus communis than Hupehsuchus nanchangensis. Nasal extending anteriorly, well beyond external naris; scleral ring large, filling orbit; snout constricted in dorsal view; converging digits with limited interdigital space.
Majiashan Quarry
Olenekian
Diagnosis. The last common ancestor of Ichthyosaurus communis and Hupehsuchus nanchangensis, and all its descendants. Anterior flanges on humerus and radius present; ulna distal width equal to or greater than proximal width; forelimb longer than or almost equal to hindlimb; manus length at least about three-quarters the length of the stylopodium and zeugopodium combined; fibula extending further post-axially than femur; transverse process of neural arch extremely short or absent.
Chaohu Lake
M.T.
Reptilia Laurenti, 1768 Diapsida Osborn, 1903 Ichthyosauromorpha nov.
Anhui Prov.
Mainland China
Lower Triassic
The incompleteness of the fossil record obscures the origin of many of the more derived clades of vertebrates. One such group is the Ichthyopterygia, a clade of obligatory marine reptiles that appeared in the Early Triassic epoch, without any known intermediates1. Here we describe a basal ichthyosauriform from the upper Lower Triassic (about 248 million years ago) of China, whose primitive skeleton indicates possible amphibious habits. It is smaller than ichthyopterygians and had unusually large flippers that probably allowed limited terrestrial locomotion. It also retained characteristics of terrestrial diapsid reptiles, including a short snout and body trunk2. Unlike more-derived ichthyosauriforms3, it was probably a suction feeder. The new species supports the sister-group relationships between ichthyosauriforms and Hupehsuchia4, the two forming the Ichthyosauromorpha. Basal ichthyosauromorphs are known exclusively from south China, suggesting that the clade originated in the region, which formed a warm5,6 and humid7 tropical archipelago8 in the Early Triassic. The oldest unequivocal record of a sauropterygian is also from the same stratigraphic unit of the region9.
Hefei
5m
Marl/shale Marly limestone
Figure 1 | Locality and horizon of the new species. a, Map of Majiashan relative to Chaohu and Hefei. b, Stratigraphic columns of the relevant strata. The specimen is from bed 633 (red) within the Upper Member of the Nanlinghu Formation (yellow). Ichthyopterygian occurrences are indicated by silhouettes. M.T., Middle Triassic; Smi, Smithian. Stratigraphic column in b spans from 251.2 to 247.2 million years ago.
1
Department of Earth and Planetary Sciences, University of California, Davis, One Shields Avenue, Davis, California 95616, USA. 2Laboratory of Orogenic Belt and Crustal Evolution, Ministry of Education, and Department of Geology and Geological Museum, Peking University, Yiheyuan Street 5, Beijing 100871, China. 3State Key Laboratory of Palaeobiology and Stratigraphy (Nanjing Institute of Geology and Palaeontology, Chinese Academy of Science), Nanjing 210008, China. 4Department of Research, Anhui Geological Museum, Jiahe Road 999, Hefei, Anhui 230031, China. 5Dipartimento di Scienze della Terra, Universita` degli Studi di Milano, Via Mangiagalli, 34-20133 Milan, Italy. 6Center of Integrative Research, The Field Museum, Chicago, Illinois 60605-2496, USA. 7Key Laboratory of Economic Stratigraphy and Palaeogeography, Nanjing Institute of Geology and Palaeontology, Chinese Academy of Sciences, 39 East Beijing Road, Nanjing 210008, China. 0 0 M O N T H 2 0 1 4 | VO L 0 0 0 | N AT U R E | 1
©2014 Macmillan Publishers Limited. All rights reserved
RESEARCH LETTER of the nasal and posterior relocation of the external naris preceded snout elongation in ichthyosauriforms. The short and constricted snout suggests that the gape of this species was much smaller than the mouth cavity, enabling pressure concentration for suction feeding3. The inference of suction feeding in Cartorhynchus is further supported by the presence of a large and robust hyobranchial element (Fig. 2a, d), and also by edentulism. There is, however, a dental groove at least on the right mandibular ramus, and this is also present in the edentulous Hupehsuchus13. The distance between the zeugopodium and metapodium is unusually large, suggesting the presence of cartilaginous centralia. These elements only occur in Chaohusaurus among ichthyopterygians14. The carpal gap in Cartorhynchus is even wider than in a specimen of Chaohusaurus in which both lateral and medial centralia are ossified14 (Fig. 3c). We reprepared both limbs twice but did not find any additional elements. The flippers may well have allowed limited terrestrial locomotion given their unusually large size (Fig. 3d). The carpus allowed dorso-ventral flexion of the flipper without a functional elbow, enabling seal-like flipper bending on land—flipper bendability is essential in terrestrial locomotion15. The cartilaginous forelimb may not have been as strong as fully ossified flippers but the mass it supported was small. The small body size, through scaling effects, provided a low body mass/flipper area ratio that was about one-fifth to one-twelfth of the values in Chaohusaurus, assuming similar body shapes (Fig. 3d, grey lines). Note that juvenile sea turtles, with highly cartilaginous wrist and ankle joints within the flipper, are able to undertake safe locomotion on land. Considering that the pachyostotic ribs would ballast the body in surging water near shores, Cartorhynchus may have been amphibious. Added mass from thick ribs may be disadvantageous for terrestrial locomotion but they are not as heavy as turtle shells. Moreover, even dugongs, which are heavily pachyostotic, are known to occasionally give birth on the beach16. A short trunk and
than ribs; scapula wider distally than proximally; autopodium with broadly spaced tiny ossifications; only three manual digits ossified; gastralia without median element. Other features: mandible deep; trunk shorter than in ichthyopterygians by at least five vertebrae; pineal foramen very large; interclavicle cruciform; parapophyses confluent with anterior vertebral margin. Cartorhynchus is the smallest ichthyosauriform to date. The preserved length of the specimen is 21.4 cm (Fig. 2a). Total body length is estimated to be about 40 cm, assuming tail proportions of basal ichthyopterygians. Of the 31 pre-sacral vertebrae, 5 seem to be cervical. Ichthyopterygians typically have an elongated body with 40–80 pre-sacral vertebrae, except for Chaohusaurus, which has about 36 (Fig. 3a). The pre-sacral vertebral count of extant terrestrial reptiles with well-developed limbs ranges from 16 to 36, with 24 being the norm10. Cartorhynchus is within this terrestrial range (Fig. 3a). The axial skeleton is heavily built in Cartorhynchus. The ribs are pachyostotic, limiting the intercostal space (Fig. 2b). In contrast, the proximal intercostal space is about twice as wide as the ribs in Chaohusaurus (Fig. 2c). However, pachyostosis in Cartorhynchus is not as pronounced as in the swollen ribs of pachypleurosaurs or sirenians. Osteosclerosis cannot be confirmed without damaging the only specimen. Pachyostosis is a common feature among basal members of secondarily aquatic reptiles11. Thickened ribs were lost in Ichthyopterygia but reappeared in the Middle Jurassic Mollesaurus12. The snout of Cartorhynchus is constricted but not elongated (Fig. 2a). Pre-orbital skull length relative to total skull length is shorter in Cartorhynchus than in most terrestrial diapsids (Fig. 3b). In contrast, ichthyopterygians uniformly show snout elongation (Fig. 3b). Despite the short snout, the premaxilla of Cartorhynchus is elongated, and the external naris is located posteriorly. The nasal of Cartorhynchus extends anteriorly to the tip of the snout unlike in most reptiles (Fig. 2d). Elongation a
d
b
f po st
sc
p
ptf
sp
scl
prf
p
f ptf
n
st sq sc a
qj
cl
pm
l
icl
c
pm
n
prf
i
r
i
u
m
po scl
iv
j
d
sp
sa q
R
U
f
op cl
H
a
iii
g ii
Fe Ti
Fi
ca
as ii i iv iii
cbi?
e
Figure 2 | The holotype of Cartorhynchus lenticarpus gen. et sp. nov. a, Whole specimen. b, Close-up of ribs. c, Ribs of Chaohusaurus (AGM CH-628-16) for comparison. d, Skull and shoulder elements. e, Skull of a newborn Chaohusaurus20 drawn to the same scale as d for comparison. f, Right forelimb. g, Right hindlimb. a, angular; as, astragalus; ca, calcaneum; cbi, first ceratobranchial; cl, clavicle; d, dentary; f, frontal; Fe, femur; Fi, fibula;
H, humerus; i, intermedium; icl, interclavicle; j, jugal; l, lacrimal; m, maxilla; n, nasal; op, opisthotic; p, parietal; pm, prefrontal; po, postorbital; prf, prefrontal; ptf, postfrontal; q, quadrate; qj, quadratojugal; R, radius; r, radiale; sa, surangular; sc, scapula; scl, scleral ossicles; sp, splenial; sq, squamosal; st, supratemporal; Ti, tibia; U, ulna; u, ulnare; i–v (in g), metapodials. Scale bars, 1 cm.
2 | N AT U R E | VO L 0 0 0 | 0 0 M O N T H 2 0 1 4
©2014 Macmillan Publishers Limited. All rights reserved
LETTER RESEARCH a
c
40 30
Carpus length (mm)
Terrestrial Sauropterygia Other marine Ichthyopterygia Cartorhynchus
20
10 9 8 7 6 5 4
Pre-sacral count
80 90 10 0 11 0
70
60
50
40
0 10
80
60
40
20
3
Forelimb length (mm)
b
d Forelimb length (mm)
Terrestrial Other marine Ichthyopterygia short Ichthyopterygia long Cartorhynchus
120 110 100 90 80 70 60 50 40
×1
30
8 9 10
7
5
4
3
3
2
5 2.
1.
5
1
0. 5
Snout/post-snout ratio
6
×12
×5
2nd Caudal length (mm)
Figure 3 | Quantitative comparisons of selected characteristics. a, Pre-sacral count. b, Snout/post-snout ratio of the skull. c, Carpus/whole-limb ratio of the forelimb. d, Forelimb length versus the second caudal centrum length, a proxy for body size. In c and d, triangles are Cartorhynchus, squares are Chaohusaurus geishanensis and diamonds are C. chaoxianensis. Dashed line in d represents the same forelimb/vertebra proportion as in Cartorhynchus. Grey lines in d are isoclines of the body mass/flipper area ratio, assuming that the cube of the second caudal centrum length is approximately proportional to body mass, and the length of the flipper is roughly proportional to its area. Symbols of box plots in a and b: circle, outlier; rectangle, 50 percentile; thick line, median; whisker, extreme point or, when there are outliers, data range times interquartile range. See refs 13 and 14 for data sources.
snout would have been advantageous in terrestrial locomotion. The forelimb flipper is strongly curved, with the phalangeal axis tilted about 50u post-axially relative to the zeugopodial axis. Both right and left limbs display similar angles, suggesting that the curvature is natural. The curvature allowed the tip of the flipper to be kept close to the body, raising the mechanical advantage of terrestrial body support by flippers. With a well-ossified axial skeleton and finished surfaces on many skull bones, the holotype appears to be mature, but its appendicular skeleton is poorly ossified. Given the large size of the forelimb, it seems likely that a
Hupehsuchia Cartorhynchus Chaohusaurus chaoxianensis Chaohusaurus geishanensis Chaohusaurus zhangjiawanensis Parvinatator Utatsusaurus ( Ichthyosauria Unnamed Ichthyopterygia
Figure 4 | Phylogenetic hypotheses of Cartorhynchus. a, Position of Cartorhynchus among Diapsida, with aquatic adaptations excluded. b, As in a but with aquatic adaptations included. c, Relationships of basal ichthyosauriforms. Analyses are based on most recent phylogenetic data sets
c
Ichthyosauriformes Ichthyosauromorpha
Araeoscelidia Orovenator Lanthanolania Hovasaurus Acerosodontosaurus Youngina Thadeosaurus Tangasaurus Claudiosaurus Kuehneosauridae Pamelina Sophineta Lepidosauria Coelurosauravus Helveticosaurus Choristodera Archosauromorpha Thalattosauriformes Ichthyopterygia Cartorhynchus Hupehsuchia Wumengosaurus Saurosphargidae Sauropterygia
(
Araeoscelidia b Orovenator Lanthanolania Tangasaurus Acerosodontosaurus Hovasaurus Youngina Thadeosaurus Claudiosaurus Thalattosauriformes Wumengosaurus Hupehsuchia Cartorhynchus Ichthyopterygia Coelurosauravus Sophineta Archosauromorpha Choristodera Saurosphargidae Sauropterygia Helveticosaurus Lepidosauria Kuehneosauridae Pamelina
its ossification was delayed through paedomorphosis, as is the case in many basal marine reptiles4,14,17–19. We consider the holotype to be almost, if not fully, mature. However, with only one specimen available, the possibility of immaturity cannot be completely rejected. It is unlikely that Cartorhynchus is a young individual of a known species. Phylogenetic characters eliminate non-ichthyosauriform diapsids from such a comparison (Fig. 4), while a list of features prevents it from being an immature ichthyopterygian. The snout of ichthyopterygians is already elongated in newborns20,21 (for example, Fig. 2e), unlike in Cartorhynchus (Fig. 2d). The pre-sacral vertebral count is not expected to increase after birth10. Furthermore, only adult ichthyopterygians have forelimbs that are almost as long as the skull, as in Cartorhynchus. Additional features may be compared to Chaohusaurus, the only ichthyopterygian genus that co-occurs with the new species. The mandible of Cartorhynchus (55.4 3 12.2 mm, length by depth) is distinctively deeper than that of AGM-CH-628-19 (103.4 3 7.0 mm), the largest individual of Chaohusaurus chaoxianensis. The ceratobranchial is very large (22.8 3 4.0 mm) unlike in AGMCH-628-19 (11.7 3 1.28 mm), and the radiale is already ossified, a bone that appears only in the largest individual of C. chaoxianensis14,19. The eye appears large relative to the skull largely because the snout is short; however, it is not proportionally larger than in some terrestrial reptiles, such as araeoscelidians. Three characteristics of the scleral ring are indicative of visual optics, namely the absolute size that determines the number of retinal cells, absolute aperture size that determines the ability to spot bioluminescent prey in the dark, and relative aperture size that is related to f numbers22, an optical index that describes relative brightness. For all three features Cartorhynchus has the smallest values of ichthyosauriforms, indicating a lesser degree of adaptation to underwater vision. The phylogeny shows Hupehsuchia13, Cartorhynchus and Chaohusaurus, all known exclusively from the south China block, to be the three most basal members of Ichthyosauromorpha (Fig. 4). The oldest ichthyosauromorph fossil is from the Procolumbites zone of the Spathian (Lower Triassic) of Chaohu20 (Fig. 1b), but many more are known from the overlying Subcolumbites zone at multiple locations around the Panthalassa1. Marine reptiles older than the Procolumbites zone are unknown despite pertinent outcrops around the world. Multiple exposures of relevant strata are accessible in Chaohu, yet persistent prospecting efforts have failed to locate a reptile fossil in underlying Spathian rocks, or in Smithian strata rich in fossil fishes (Fig. 1b). Furthermore, decades of limestone mining activities in the region did not uncover any older reptile fossils. Their absence coincides with the recent report that Chaohusaurus from the Subcolumbites zone still retained a terrestrial
Grippia Gulosaurus Cymbospondylidae
Mixosauridae
Shastasauria
Euichthyosauria
for marine reptile relationships4 and ichthyopterygian phylogeny (C. Ji et al., manuscript in preparation), respectively (Supplementary Information). All trees were abbreviated from more complete topologies (Extended Data Figs 1–3). 0 0 M O N T H 2 0 1 4 | VO L 0 0 0 | N AT U R E | 3
©2014 Macmillan Publishers Limited. All rights reserved
RESEARCH LETTER mode of reproduction that carried a risk in water20, suggesting that ichthyosauromorphs invaded the sea not much earlier than the appearance of Chaohusaurus. A recent statement that ichthyosauromorphs were present in the Smithian5 is a misunderstanding based on outdated references23,24 that have been updated25,26. The causes driving marine invasion could be multiple, including predation pressure and competition for food that may be lower in the sea than on adjacent land27,28. Notably, Ichthyosauromorpha was not the only marine reptile group to have emerged in the Early Triassic: the oldest unequivocal record of Sauropterygia is also from Chaohu9, 45 beds above the oldest ichthyopterygian fossil. The south China block was in the tropical latitudes at the time29, forming a warm5,6 and humid7 archipelago8. Future studies would be required to test if any climatic and geographic factors may have encouraged marine invasion. Online Content Methods, along with any additional Extended Data display items and Source Data, are available in the online version of the paper; references unique to these sections appear only in the online paper. Received 10 June; accepted 15 September 2014. Published online 5 November 2014. 1. 2. 3. 4. 5. 6. 7.
8. 9. 10. 11. 12.
13. 14.
McGowan, C. & Motani, R. Ichthyopterygia Vol. 8 (Verlag Dr. Friedrich Pfeil, 2003). Mu¨ller, J. et al. Homeotic effects, somitogenesis and the evolution of vertebral numbers in recent and fossil amniotes. Proc. Natl Acad. Sci. USA 107, 2118–2123 (2010). Motani, R. et al. Absence of suction feeding ichthyosaurs and its implications for Triassic Mesopelagic paleoecology. PLoS ONE 8, e66075 (2013). Chen, X.-h., Motani, R., Cheng, L., Jiang, D.-y. & Rieppel, O. The enigmatic marine reptile Nanchangosaurus from the Lower Triassic of Hubei, China and the phylogenetic affinity of Hupehsuchia. PLoS ONE 9, e102361 (2014). Sun, Y. D. et al. Lethally hot temperatures during the Early Triassic greenhouse. Science 338, 366–370 (2012). Goudemand, N., Romano, C., Brayard, A., Hochuli, P. A. & Bucher, H. Comment on ‘‘Lethally Hot Temperatures During the Early Triassic Greenhouse’’. Science 339, 1033 (2013). Pe´ron, S., Bourquin, S., Fluteau, F. & Guillocheau, F. Paleoenvironment reconstructions and climate simulations of the Early Triassic: impact of the water and sediment supply on the preservation of fluvial systems. Geodin. Acta 18, 431–446 (2005). Xiao, W. J. & He, H. Q. Early Mesozoic thrust tectonics of the northwest Zhejiang region (Southeast China). Geol. Soc. Am. Bull. 117, 945–961 (2005). Jiang, D. et al. Early Triassic eosauropterygian Majiashanosaurus discocoracoidis, gen. et sp. nov. (Reptilia, Sauropterygia) from Chaohu, Anhui Province, China. J. Vertebr. Paleontol. 34, 1044–1052 (2014). Hoffstetter, R. & Gasc, J.-P. in Biology of the Reptilia. Vol 1 (eds Gans, C., Bellairs, A. d’A. & Parsons, T. S.) 201–310 (Academic, 1969). Houssaye, A. ‘‘Pachyostosis’’ in aquatic amniotes: a review. Integr. Zool. 4, 325–340 (2009). Talevi, M. & Fernandez, M. S. Unexpected skeletal histology of an ichthyosaur from the Middle Jurassic of Patagonia: implications for evolution of bone microstructure among secondary aquatic tetrapods. Naturwissenschaften 99, 241–244 (2012). Carroll, R. L. & Dong, Z. Hupehsuchus, an enigmatic aquatic reptile from the Triassic of China, and the problem of establishing relationships. Phil. Trans. R. Soc. B 331, 131–153 (1991). Motani, R. et al. First evidence of centralia in Ichthyopterygia reiterating bias from paedomorphic characters on marine reptile phylogenetic reconstruction. J. Vertebr. Paleontol. (in the press).
15. Mazouchova, N., Umbanhowar, P. B. & Goldman, D. I. Flipper-driven terrestrial locomotion of a sea turtle-inspired robot. Bioinspir. Biomim. 8, 026007 (2013). 16. Marsh, H., Heinsohn, G. E. & Marsh, L. M. Breeding cycle life history and population dynamics of the Dugong Dugong-Dugon Sirenia Dugongidae. Aust. J. Zool. 32, 767–788 (1984). 17. Rieppel, O. Helveticosaurus zollingeri Peyer (Reptilia Diapsida) skeletal pedomorphosis, functional anatomy and systematic affinities. Palaeontographica Abteilung A 208, 123–152 (1989). 18. Caldwell, M. W. Limb ontogeny, evolution and aquatic adaptation in lepidosauromorph diapsids. J. Vertebr. Paleontol. 14, 19A (1994). 19. Motani, R. et al. Status of Chaohusaurus chaoxianensis (Young and Dong, 1972). J. Vertebr. Paleontol. (in the press). 20. Motani, R., Jiang, D., Tintori, A., Rieppel, O. & Chen, G. B. Terrestrial origin of viviparity indicated by the oldest embryonic fossil of Mesozoic marine reptiles. PLoS One 9, e8B640 (2014). 21. Boettcher, R. New information on the reproductive biology of Ichthyosaurs (Reptilia). Stuttgarter Beitraege zur Naturkunde Serie B 164, 1–51 (1990). 22. Motani, R., Rothschild, B. M. & Wahl, W. Large eyeballs in diving ichthyosaurs. Nature 402, 747 (1999). 23. Callaway, J. M. & Brinkman, D. B. Ichthyosaurs (Reptilia, Ichthyosauria) from the Lower and Middle Triassic Sulfur Mountain Formation, Wapiti Lake Area, British-Columbia, Canada. Can. J. Earth Sci. 26, 1491–1500 (1989). 24. Cox, C. B. & Smith, D. G. Review of Triassic vertebrate faunas of Svalbard. Geol. Mag. 110, 405–418 (1973). 25. Nicholls, E. L. & Brinkman, D. B. in Vertebrate Fossils and the Evolution of Scientific Concepts (ed. Sarjeant, W. A. S.) 521–535 (Gordon and Breach, 1995). 26. Harland, W. B. The geology of svalbard. Mem. Geol. Soc. Lond. 17, 1–521 (1997). 27. Carroll, R. L. Evolutionary constraints in aquatic diapsid reptiles. Spec. Pap. Palaeontol. 33, 145–155 (1985). 28. Vermeij, G. J. & Dudley, R. Why are there so few evolutionary transitions between aquatic and terrestrial ecosystems? Biol. J. Linn. Soc. 70, 541–554 (2000). 29. Sun, Z. M. et al. Magnetostratigraphy of the Lower Triassic beds from Chaohu (China) and its implications for the Induan-Olenekian stage boundary. Earth Planet. Sci. Lett. 279, 350–361 (2009). Supplementary Information is available in the online version of the paper. Acknowledgements We thank T. Sato for the excellent preparation of the holotype. The study was enabled by grants from the National Geographic Society Committee for Research and Exploration (8669-09) to R.M., Project 40920124002 and 41372016 from the National Natural Science Foundation of China to D.-Y.J., Project 123102 from State Key Laboratory of Palaeobiology and Stratigraphy (Nanjing Institute of Geology and Palaeontology, CAS) to D.-Y.J., Project 20120001110072 from the Research Fund for the Doctoral Program of Higher Education to D.-Y.J., and a Project of Protection for Geological Heritage from Department of Land and Resource of Anhui Province to G.-B.C. Author Contributions R.M. conceived the study, participated in the relevant fossil excavations, ran all analyses, drew all figures except Fig. 1b, and wrote the manuscript. D.-Y.J. conceived the study, supervised the relevant fossil excavations and preparations, drew Fig. 1b, and revised the manuscript; G.-B.C. conceived the study and supervised off-season fossil collections; A.T. conceived the study, participated in the relevant fossil excavations, and revised the manuscript; O.R. conceived the study, participated in the relevant fossil excavations, and revised the manuscript; C.J. provided an unpublished data matrix, participated in the relevant fossil excavations, and revised the manuscript; J.-D.H. helped supervise off-season fossil collections. Author Information Reprints and permissions information is available at www.nature.com/reprints. The authors declare no competing financial interests. Readers are welcome to comment on the online version of the paper. Correspondence and requests for materials should be addressed to R.M. ([email protected]) or D.-Y.J. ([email protected]).
4 | N AT U R E | VO L 0 0 0 | 0 0 M O N T H 2 0 1 4
©2014 Macmillan Publishers Limited. All rights reserved
LETTER RESEARCH
Extended Data Figure 1 | Phylogenetic hypothesis of diapsid relationships when aquatic adaptations are recoded as ambiguous. Strict consensus of two most parsimonious trees (tree length (TL) 5 805, consistency index (CI) 5 0.317, retention index (RI) 5 0.587) obtained by a heuristic search in
PAUP* 4b10 (hold 5 10, nreps 5 100, addseq 5 random, swap 5 tbr). Numbers associated with clades are Bremer support values calculated in TNT 1.1. Original tree of Fig. 4a. Phylogenetic tree is based on the data described in Supplementary Information.
©2014 Macmillan Publishers Limited. All rights reserved
RESEARCH LETTER
Extended Data Figure 2 | Phylogenetic hypothesis of diapsid relationships when aquatic adaptations are recoded normally. Strict consensus of 20 most parsimonious trees (TL 5 839, CI 5 0.311, RI 5 0.612) obtained by a heuristic search in PAUP* 4b10 (hold 5 10, nreps 5 100, addseq 5 random,
swap 5 tbr). Numbers associated with clades are Bremer support values calculated in TNT 1.1. Original tree of Fig. 4b. Phylogenetic tree is based on the data described in Supplementary Information.
©2014 Macmillan Publishers Limited. All rights reserved
LETTER RESEARCH
Extended Data Figure 3 | Phylogenetic hypothesis of Ichthyosauriformes. Strict consensus of 82 most parsimonious trees (TL 5 527, CI 5 0.423, RI 5 0.796) obtained by a heuristic search in PAUP* 4b10 (hold 5 10,
nreps 5 100, addseq 5 random, swap 5 tbr). Numbers associated with clades are Bremer support values calculated in TNT 1.1. Original tree of Fig. 4c. Phylogenetic tree is based on the data described in Supplementary Information.
©2014 Macmillan Publishers Limited. All rights reserved
doi:10.1038/nature13866
A basal ichthyosauriform with a short snout from the Lower Triassic of China Ryosuke Motani1, Da-Yong Jiang2,3, Guan-Bao Chen4, Andrea Tintori5, Olivier Rieppel6, Cheng Ji7 & Jian-Dong Huang4 a 31°51′
117°12′
117°54′
10 km
31°24′
Lithological column
Beds 650
beds 611–999
633
Middle
beds 558–610 beds 508–557
621
611
Helongshan Formation
Argillaceous limestone
Nodular limestone
Columbites
Smi Dolostone
637
Lower
Spathian
Upper member
643
Subcolumbites
Dongmaanshan Formation
598
20 m
591
Procolumbites
Anisian
b
Cartorhynchus lenticarpus gen. et sp. nov. Etymology. kartos (Greek), meaning shortened; rgynxgos (Greek), meaning snout; lentus (Latin), meaning flexible; carpus (Latin), meaning wrist. Named after truncated snout and cartilaginous wrist. Holotype. Anhui Geological Museum AGB6257. Locality and horizon. From the second level of Majiashan Quarry (31u 379 260 N, 117u 499 190 E), near downtown Chaohu, Hefei City, Anhui Province, China (Fig. 1a). Bed 633, about 13 m above the bottom of the Upper Member of the Nanlinghu Formation (Fig. 1b), within the ammonite Subcolumbites zone, Spathian, Olenekian, Lower Triassic (Fig. 1b). Diagnosis. Autapomorphies are: snout only half as long as the rest of the skull; very large hyoid; forelimb strongly curved posteriorly; anteriorly curved hindlimb; ribs robust, with proximal intercostal space narrower
Chaohu
Chaohu Lake
Nanlinghu Formation
Ichthyosauriformes nov. Diagnosis. All ichthyosauromorphs more closely related to Ichthyosaurus communis than Hupehsuchus nanchangensis. Nasal extending anteriorly, well beyond external naris; scleral ring large, filling orbit; snout constricted in dorsal view; converging digits with limited interdigital space.
Majiashan Quarry
Olenekian
Diagnosis. The last common ancestor of Ichthyosaurus communis and Hupehsuchus nanchangensis, and all its descendants. Anterior flanges on humerus and radius present; ulna distal width equal to or greater than proximal width; forelimb longer than or almost equal to hindlimb; manus length at least about three-quarters the length of the stylopodium and zeugopodium combined; fibula extending further post-axially than femur; transverse process of neural arch extremely short or absent.
Chaohu Lake
M.T.
Reptilia Laurenti, 1768 Diapsida Osborn, 1903 Ichthyosauromorpha nov.
Anhui Prov.
Mainland China
Lower Triassic
The incompleteness of the fossil record obscures the origin of many of the more derived clades of vertebrates. One such group is the Ichthyopterygia, a clade of obligatory marine reptiles that appeared in the Early Triassic epoch, without any known intermediates1. Here we describe a basal ichthyosauriform from the upper Lower Triassic (about 248 million years ago) of China, whose primitive skeleton indicates possible amphibious habits. It is smaller than ichthyopterygians and had unusually large flippers that probably allowed limited terrestrial locomotion. It also retained characteristics of terrestrial diapsid reptiles, including a short snout and body trunk2. Unlike more-derived ichthyosauriforms3, it was probably a suction feeder. The new species supports the sister-group relationships between ichthyosauriforms and Hupehsuchia4, the two forming the Ichthyosauromorpha. Basal ichthyosauromorphs are known exclusively from south China, suggesting that the clade originated in the region, which formed a warm5,6 and humid7 tropical archipelago8 in the Early Triassic. The oldest unequivocal record of a sauropterygian is also from the same stratigraphic unit of the region9.
Hefei
5m
Marl/shale Marly limestone
Figure 1 | Locality and horizon of the new species. a, Map of Majiashan relative to Chaohu and Hefei. b, Stratigraphic columns of the relevant strata. The specimen is from bed 633 (red) within the Upper Member of the Nanlinghu Formation (yellow). Ichthyopterygian occurrences are indicated by silhouettes. M.T., Middle Triassic; Smi, Smithian. Stratigraphic column in b spans from 251.2 to 247.2 million years ago.
1
Department of Earth and Planetary Sciences, University of California, Davis, One Shields Avenue, Davis, California 95616, USA. 2Laboratory of Orogenic Belt and Crustal Evolution, Ministry of Education, and Department of Geology and Geological Museum, Peking University, Yiheyuan Street 5, Beijing 100871, China. 3State Key Laboratory of Palaeobiology and Stratigraphy (Nanjing Institute of Geology and Palaeontology, Chinese Academy of Science), Nanjing 210008, China. 4Department of Research, Anhui Geological Museum, Jiahe Road 999, Hefei, Anhui 230031, China. 5Dipartimento di Scienze della Terra, Universita` degli Studi di Milano, Via Mangiagalli, 34-20133 Milan, Italy. 6Center of Integrative Research, The Field Museum, Chicago, Illinois 60605-2496, USA. 7Key Laboratory of Economic Stratigraphy and Palaeogeography, Nanjing Institute of Geology and Palaeontology, Chinese Academy of Sciences, 39 East Beijing Road, Nanjing 210008, China. 0 0 M O N T H 2 0 1 4 | VO L 0 0 0 | N AT U R E | 1
©2014 Macmillan Publishers Limited. All rights reserved
RESEARCH LETTER of the nasal and posterior relocation of the external naris preceded snout elongation in ichthyosauriforms. The short and constricted snout suggests that the gape of this species was much smaller than the mouth cavity, enabling pressure concentration for suction feeding3. The inference of suction feeding in Cartorhynchus is further supported by the presence of a large and robust hyobranchial element (Fig. 2a, d), and also by edentulism. There is, however, a dental groove at least on the right mandibular ramus, and this is also present in the edentulous Hupehsuchus13. The distance between the zeugopodium and metapodium is unusually large, suggesting the presence of cartilaginous centralia. These elements only occur in Chaohusaurus among ichthyopterygians14. The carpal gap in Cartorhynchus is even wider than in a specimen of Chaohusaurus in which both lateral and medial centralia are ossified14 (Fig. 3c). We reprepared both limbs twice but did not find any additional elements. The flippers may well have allowed limited terrestrial locomotion given their unusually large size (Fig. 3d). The carpus allowed dorso-ventral flexion of the flipper without a functional elbow, enabling seal-like flipper bending on land—flipper bendability is essential in terrestrial locomotion15. The cartilaginous forelimb may not have been as strong as fully ossified flippers but the mass it supported was small. The small body size, through scaling effects, provided a low body mass/flipper area ratio that was about one-fifth to one-twelfth of the values in Chaohusaurus, assuming similar body shapes (Fig. 3d, grey lines). Note that juvenile sea turtles, with highly cartilaginous wrist and ankle joints within the flipper, are able to undertake safe locomotion on land. Considering that the pachyostotic ribs would ballast the body in surging water near shores, Cartorhynchus may have been amphibious. Added mass from thick ribs may be disadvantageous for terrestrial locomotion but they are not as heavy as turtle shells. Moreover, even dugongs, which are heavily pachyostotic, are known to occasionally give birth on the beach16. A short trunk and
than ribs; scapula wider distally than proximally; autopodium with broadly spaced tiny ossifications; only three manual digits ossified; gastralia without median element. Other features: mandible deep; trunk shorter than in ichthyopterygians by at least five vertebrae; pineal foramen very large; interclavicle cruciform; parapophyses confluent with anterior vertebral margin. Cartorhynchus is the smallest ichthyosauriform to date. The preserved length of the specimen is 21.4 cm (Fig. 2a). Total body length is estimated to be about 40 cm, assuming tail proportions of basal ichthyopterygians. Of the 31 pre-sacral vertebrae, 5 seem to be cervical. Ichthyopterygians typically have an elongated body with 40–80 pre-sacral vertebrae, except for Chaohusaurus, which has about 36 (Fig. 3a). The pre-sacral vertebral count of extant terrestrial reptiles with well-developed limbs ranges from 16 to 36, with 24 being the norm10. Cartorhynchus is within this terrestrial range (Fig. 3a). The axial skeleton is heavily built in Cartorhynchus. The ribs are pachyostotic, limiting the intercostal space (Fig. 2b). In contrast, the proximal intercostal space is about twice as wide as the ribs in Chaohusaurus (Fig. 2c). However, pachyostosis in Cartorhynchus is not as pronounced as in the swollen ribs of pachypleurosaurs or sirenians. Osteosclerosis cannot be confirmed without damaging the only specimen. Pachyostosis is a common feature among basal members of secondarily aquatic reptiles11. Thickened ribs were lost in Ichthyopterygia but reappeared in the Middle Jurassic Mollesaurus12. The snout of Cartorhynchus is constricted but not elongated (Fig. 2a). Pre-orbital skull length relative to total skull length is shorter in Cartorhynchus than in most terrestrial diapsids (Fig. 3b). In contrast, ichthyopterygians uniformly show snout elongation (Fig. 3b). Despite the short snout, the premaxilla of Cartorhynchus is elongated, and the external naris is located posteriorly. The nasal of Cartorhynchus extends anteriorly to the tip of the snout unlike in most reptiles (Fig. 2d). Elongation a
d
b
f po st
sc
p
ptf
sp
scl
prf
p
f ptf
n
st sq sc a
qj
cl
pm
l
icl
c
pm
n
prf
i
r
i
u
m
po scl
iv
j
d
sp
sa q
R
U
f
op cl
H
a
iii
g ii
Fe Ti
Fi
ca
as ii i iv iii
cbi?
e
Figure 2 | The holotype of Cartorhynchus lenticarpus gen. et sp. nov. a, Whole specimen. b, Close-up of ribs. c, Ribs of Chaohusaurus (AGM CH-628-16) for comparison. d, Skull and shoulder elements. e, Skull of a newborn Chaohusaurus20 drawn to the same scale as d for comparison. f, Right forelimb. g, Right hindlimb. a, angular; as, astragalus; ca, calcaneum; cbi, first ceratobranchial; cl, clavicle; d, dentary; f, frontal; Fe, femur; Fi, fibula;
H, humerus; i, intermedium; icl, interclavicle; j, jugal; l, lacrimal; m, maxilla; n, nasal; op, opisthotic; p, parietal; pm, prefrontal; po, postorbital; prf, prefrontal; ptf, postfrontal; q, quadrate; qj, quadratojugal; R, radius; r, radiale; sa, surangular; sc, scapula; scl, scleral ossicles; sp, splenial; sq, squamosal; st, supratemporal; Ti, tibia; U, ulna; u, ulnare; i–v (in g), metapodials. Scale bars, 1 cm.
2 | N AT U R E | VO L 0 0 0 | 0 0 M O N T H 2 0 1 4
©2014 Macmillan Publishers Limited. All rights reserved
LETTER RESEARCH a
c
40 30
Carpus length (mm)
Terrestrial Sauropterygia Other marine Ichthyopterygia Cartorhynchus
20
10 9 8 7 6 5 4
Pre-sacral count
80 90 10 0 11 0
70
60
50
40
0 10
80
60
40
20
3
Forelimb length (mm)
b
d Forelimb length (mm)
Terrestrial Other marine Ichthyopterygia short Ichthyopterygia long Cartorhynchus
120 110 100 90 80 70 60 50 40
×1
30
8 9 10
7
5
4
3
3
2
5 2.
1.
5
1
0. 5
Snout/post-snout ratio
6
×12
×5
2nd Caudal length (mm)
Figure 3 | Quantitative comparisons of selected characteristics. a, Pre-sacral count. b, Snout/post-snout ratio of the skull. c, Carpus/whole-limb ratio of the forelimb. d, Forelimb length versus the second caudal centrum length, a proxy for body size. In c and d, triangles are Cartorhynchus, squares are Chaohusaurus geishanensis and diamonds are C. chaoxianensis. Dashed line in d represents the same forelimb/vertebra proportion as in Cartorhynchus. Grey lines in d are isoclines of the body mass/flipper area ratio, assuming that the cube of the second caudal centrum length is approximately proportional to body mass, and the length of the flipper is roughly proportional to its area. Symbols of box plots in a and b: circle, outlier; rectangle, 50 percentile; thick line, median; whisker, extreme point or, when there are outliers, data range times interquartile range. See refs 13 and 14 for data sources.
snout would have been advantageous in terrestrial locomotion. The forelimb flipper is strongly curved, with the phalangeal axis tilted about 50u post-axially relative to the zeugopodial axis. Both right and left limbs display similar angles, suggesting that the curvature is natural. The curvature allowed the tip of the flipper to be kept close to the body, raising the mechanical advantage of terrestrial body support by flippers. With a well-ossified axial skeleton and finished surfaces on many skull bones, the holotype appears to be mature, but its appendicular skeleton is poorly ossified. Given the large size of the forelimb, it seems likely that a
Hupehsuchia Cartorhynchus Chaohusaurus chaoxianensis Chaohusaurus geishanensis Chaohusaurus zhangjiawanensis Parvinatator Utatsusaurus ( Ichthyosauria Unnamed Ichthyopterygia
Figure 4 | Phylogenetic hypotheses of Cartorhynchus. a, Position of Cartorhynchus among Diapsida, with aquatic adaptations excluded. b, As in a but with aquatic adaptations included. c, Relationships of basal ichthyosauriforms. Analyses are based on most recent phylogenetic data sets
c
Ichthyosauriformes Ichthyosauromorpha
Araeoscelidia Orovenator Lanthanolania Hovasaurus Acerosodontosaurus Youngina Thadeosaurus Tangasaurus Claudiosaurus Kuehneosauridae Pamelina Sophineta Lepidosauria Coelurosauravus Helveticosaurus Choristodera Archosauromorpha Thalattosauriformes Ichthyopterygia Cartorhynchus Hupehsuchia Wumengosaurus Saurosphargidae Sauropterygia
(
Araeoscelidia b Orovenator Lanthanolania Tangasaurus Acerosodontosaurus Hovasaurus Youngina Thadeosaurus Claudiosaurus Thalattosauriformes Wumengosaurus Hupehsuchia Cartorhynchus Ichthyopterygia Coelurosauravus Sophineta Archosauromorpha Choristodera Saurosphargidae Sauropterygia Helveticosaurus Lepidosauria Kuehneosauridae Pamelina
its ossification was delayed through paedomorphosis, as is the case in many basal marine reptiles4,14,17–19. We consider the holotype to be almost, if not fully, mature. However, with only one specimen available, the possibility of immaturity cannot be completely rejected. It is unlikely that Cartorhynchus is a young individual of a known species. Phylogenetic characters eliminate non-ichthyosauriform diapsids from such a comparison (Fig. 4), while a list of features prevents it from being an immature ichthyopterygian. The snout of ichthyopterygians is already elongated in newborns20,21 (for example, Fig. 2e), unlike in Cartorhynchus (Fig. 2d). The pre-sacral vertebral count is not expected to increase after birth10. Furthermore, only adult ichthyopterygians have forelimbs that are almost as long as the skull, as in Cartorhynchus. Additional features may be compared to Chaohusaurus, the only ichthyopterygian genus that co-occurs with the new species. The mandible of Cartorhynchus (55.4 3 12.2 mm, length by depth) is distinctively deeper than that of AGM-CH-628-19 (103.4 3 7.0 mm), the largest individual of Chaohusaurus chaoxianensis. The ceratobranchial is very large (22.8 3 4.0 mm) unlike in AGMCH-628-19 (11.7 3 1.28 mm), and the radiale is already ossified, a bone that appears only in the largest individual of C. chaoxianensis14,19. The eye appears large relative to the skull largely because the snout is short; however, it is not proportionally larger than in some terrestrial reptiles, such as araeoscelidians. Three characteristics of the scleral ring are indicative of visual optics, namely the absolute size that determines the number of retinal cells, absolute aperture size that determines the ability to spot bioluminescent prey in the dark, and relative aperture size that is related to f numbers22, an optical index that describes relative brightness. For all three features Cartorhynchus has the smallest values of ichthyosauriforms, indicating a lesser degree of adaptation to underwater vision. The phylogeny shows Hupehsuchia13, Cartorhynchus and Chaohusaurus, all known exclusively from the south China block, to be the three most basal members of Ichthyosauromorpha (Fig. 4). The oldest ichthyosauromorph fossil is from the Procolumbites zone of the Spathian (Lower Triassic) of Chaohu20 (Fig. 1b), but many more are known from the overlying Subcolumbites zone at multiple locations around the Panthalassa1. Marine reptiles older than the Procolumbites zone are unknown despite pertinent outcrops around the world. Multiple exposures of relevant strata are accessible in Chaohu, yet persistent prospecting efforts have failed to locate a reptile fossil in underlying Spathian rocks, or in Smithian strata rich in fossil fishes (Fig. 1b). Furthermore, decades of limestone mining activities in the region did not uncover any older reptile fossils. Their absence coincides with the recent report that Chaohusaurus from the Subcolumbites zone still retained a terrestrial
Grippia Gulosaurus Cymbospondylidae
Mixosauridae
Shastasauria
Euichthyosauria
for marine reptile relationships4 and ichthyopterygian phylogeny (C. Ji et al., manuscript in preparation), respectively (Supplementary Information). All trees were abbreviated from more complete topologies (Extended Data Figs 1–3). 0 0 M O N T H 2 0 1 4 | VO L 0 0 0 | N AT U R E | 3
©2014 Macmillan Publishers Limited. All rights reserved
RESEARCH LETTER mode of reproduction that carried a risk in water20, suggesting that ichthyosauromorphs invaded the sea not much earlier than the appearance of Chaohusaurus. A recent statement that ichthyosauromorphs were present in the Smithian5 is a misunderstanding based on outdated references23,24 that have been updated25,26. The causes driving marine invasion could be multiple, including predation pressure and competition for food that may be lower in the sea than on adjacent land27,28. Notably, Ichthyosauromorpha was not the only marine reptile group to have emerged in the Early Triassic: the oldest unequivocal record of Sauropterygia is also from Chaohu9, 45 beds above the oldest ichthyopterygian fossil. The south China block was in the tropical latitudes at the time29, forming a warm5,6 and humid7 archipelago8. Future studies would be required to test if any climatic and geographic factors may have encouraged marine invasion. Online Content Methods, along with any additional Extended Data display items and Source Data, are available in the online version of the paper; references unique to these sections appear only in the online paper. Received 10 June; accepted 15 September 2014. Published online 5 November 2014. 1. 2. 3. 4. 5. 6. 7.
8. 9. 10. 11. 12.
13. 14.
McGowan, C. & Motani, R. Ichthyopterygia Vol. 8 (Verlag Dr. Friedrich Pfeil, 2003). Mu¨ller, J. et al. Homeotic effects, somitogenesis and the evolution of vertebral numbers in recent and fossil amniotes. Proc. Natl Acad. Sci. USA 107, 2118–2123 (2010). Motani, R. et al. Absence of suction feeding ichthyosaurs and its implications for Triassic Mesopelagic paleoecology. PLoS ONE 8, e66075 (2013). Chen, X.-h., Motani, R., Cheng, L., Jiang, D.-y. & Rieppel, O. The enigmatic marine reptile Nanchangosaurus from the Lower Triassic of Hubei, China and the phylogenetic affinity of Hupehsuchia. PLoS ONE 9, e102361 (2014). Sun, Y. D. et al. Lethally hot temperatures during the Early Triassic greenhouse. Science 338, 366–370 (2012). Goudemand, N., Romano, C., Brayard, A., Hochuli, P. A. & Bucher, H. Comment on ‘‘Lethally Hot Temperatures During the Early Triassic Greenhouse’’. Science 339, 1033 (2013). Pe´ron, S., Bourquin, S., Fluteau, F. & Guillocheau, F. Paleoenvironment reconstructions and climate simulations of the Early Triassic: impact of the water and sediment supply on the preservation of fluvial systems. Geodin. Acta 18, 431–446 (2005). Xiao, W. J. & He, H. Q. Early Mesozoic thrust tectonics of the northwest Zhejiang region (Southeast China). Geol. Soc. Am. Bull. 117, 945–961 (2005). Jiang, D. et al. Early Triassic eosauropterygian Majiashanosaurus discocoracoidis, gen. et sp. nov. (Reptilia, Sauropterygia) from Chaohu, Anhui Province, China. J. Vertebr. Paleontol. 34, 1044–1052 (2014). Hoffstetter, R. & Gasc, J.-P. in Biology of the Reptilia. Vol 1 (eds Gans, C., Bellairs, A. d’A. & Parsons, T. S.) 201–310 (Academic, 1969). Houssaye, A. ‘‘Pachyostosis’’ in aquatic amniotes: a review. Integr. Zool. 4, 325–340 (2009). Talevi, M. & Fernandez, M. S. Unexpected skeletal histology of an ichthyosaur from the Middle Jurassic of Patagonia: implications for evolution of bone microstructure among secondary aquatic tetrapods. Naturwissenschaften 99, 241–244 (2012). Carroll, R. L. & Dong, Z. Hupehsuchus, an enigmatic aquatic reptile from the Triassic of China, and the problem of establishing relationships. Phil. Trans. R. Soc. B 331, 131–153 (1991). Motani, R. et al. First evidence of centralia in Ichthyopterygia reiterating bias from paedomorphic characters on marine reptile phylogenetic reconstruction. J. Vertebr. Paleontol. (in the press).
15. Mazouchova, N., Umbanhowar, P. B. & Goldman, D. I. Flipper-driven terrestrial locomotion of a sea turtle-inspired robot. Bioinspir. Biomim. 8, 026007 (2013). 16. Marsh, H., Heinsohn, G. E. & Marsh, L. M. Breeding cycle life history and population dynamics of the Dugong Dugong-Dugon Sirenia Dugongidae. Aust. J. Zool. 32, 767–788 (1984). 17. Rieppel, O. Helveticosaurus zollingeri Peyer (Reptilia Diapsida) skeletal pedomorphosis, functional anatomy and systematic affinities. Palaeontographica Abteilung A 208, 123–152 (1989). 18. Caldwell, M. W. Limb ontogeny, evolution and aquatic adaptation in lepidosauromorph diapsids. J. Vertebr. Paleontol. 14, 19A (1994). 19. Motani, R. et al. Status of Chaohusaurus chaoxianensis (Young and Dong, 1972). J. Vertebr. Paleontol. (in the press). 20. Motani, R., Jiang, D., Tintori, A., Rieppel, O. & Chen, G. B. Terrestrial origin of viviparity indicated by the oldest embryonic fossil of Mesozoic marine reptiles. PLoS One 9, e8B640 (2014). 21. Boettcher, R. New information on the reproductive biology of Ichthyosaurs (Reptilia). Stuttgarter Beitraege zur Naturkunde Serie B 164, 1–51 (1990). 22. Motani, R., Rothschild, B. M. & Wahl, W. Large eyeballs in diving ichthyosaurs. Nature 402, 747 (1999). 23. Callaway, J. M. & Brinkman, D. B. Ichthyosaurs (Reptilia, Ichthyosauria) from the Lower and Middle Triassic Sulfur Mountain Formation, Wapiti Lake Area, British-Columbia, Canada. Can. J. Earth Sci. 26, 1491–1500 (1989). 24. Cox, C. B. & Smith, D. G. Review of Triassic vertebrate faunas of Svalbard. Geol. Mag. 110, 405–418 (1973). 25. Nicholls, E. L. & Brinkman, D. B. in Vertebrate Fossils and the Evolution of Scientific Concepts (ed. Sarjeant, W. A. S.) 521–535 (Gordon and Breach, 1995). 26. Harland, W. B. The geology of svalbard. Mem. Geol. Soc. Lond. 17, 1–521 (1997). 27. Carroll, R. L. Evolutionary constraints in aquatic diapsid reptiles. Spec. Pap. Palaeontol. 33, 145–155 (1985). 28. Vermeij, G. J. & Dudley, R. Why are there so few evolutionary transitions between aquatic and terrestrial ecosystems? Biol. J. Linn. Soc. 70, 541–554 (2000). 29. Sun, Z. M. et al. Magnetostratigraphy of the Lower Triassic beds from Chaohu (China) and its implications for the Induan-Olenekian stage boundary. Earth Planet. Sci. Lett. 279, 350–361 (2009). Supplementary Information is available in the online version of the paper. Acknowledgements We thank T. Sato for the excellent preparation of the holotype. The study was enabled by grants from the National Geographic Society Committee for Research and Exploration (8669-09) to R.M., Project 40920124002 and 41372016 from the National Natural Science Foundation of China to D.-Y.J., Project 123102 from State Key Laboratory of Palaeobiology and Stratigraphy (Nanjing Institute of Geology and Palaeontology, CAS) to D.-Y.J., Project 20120001110072 from the Research Fund for the Doctoral Program of Higher Education to D.-Y.J., and a Project of Protection for Geological Heritage from Department of Land and Resource of Anhui Province to G.-B.C. Author Contributions R.M. conceived the study, participated in the relevant fossil excavations, ran all analyses, drew all figures except Fig. 1b, and wrote the manuscript. D.-Y.J. conceived the study, supervised the relevant fossil excavations and preparations, drew Fig. 1b, and revised the manuscript; G.-B.C. conceived the study and supervised off-season fossil collections; A.T. conceived the study, participated in the relevant fossil excavations, and revised the manuscript; O.R. conceived the study, participated in the relevant fossil excavations, and revised the manuscript; C.J. provided an unpublished data matrix, participated in the relevant fossil excavations, and revised the manuscript; J.-D.H. helped supervise off-season fossil collections. Author Information Reprints and permissions information is available at www.nature.com/reprints. The authors declare no competing financial interests. Readers are welcome to comment on the online version of the paper. Correspondence and requests for materials should be addressed to R.M. ([email protected]) or D.-Y.J. ([email protected]).
4 | N AT U R E | VO L 0 0 0 | 0 0 M O N T H 2 0 1 4
©2014 Macmillan Publishers Limited. All rights reserved
LETTER RESEARCH
Extended Data Figure 1 | Phylogenetic hypothesis of diapsid relationships when aquatic adaptations are recoded as ambiguous. Strict consensus of two most parsimonious trees (tree length (TL) 5 805, consistency index (CI) 5 0.317, retention index (RI) 5 0.587) obtained by a heuristic search in
PAUP* 4b10 (hold 5 10, nreps 5 100, addseq 5 random, swap 5 tbr). Numbers associated with clades are Bremer support values calculated in TNT 1.1. Original tree of Fig. 4a. Phylogenetic tree is based on the data described in Supplementary Information.
©2014 Macmillan Publishers Limited. All rights reserved
RESEARCH LETTER
Extended Data Figure 2 | Phylogenetic hypothesis of diapsid relationships when aquatic adaptations are recoded normally. Strict consensus of 20 most parsimonious trees (TL 5 839, CI 5 0.311, RI 5 0.612) obtained by a heuristic search in PAUP* 4b10 (hold 5 10, nreps 5 100, addseq 5 random,
swap 5 tbr). Numbers associated with clades are Bremer support values calculated in TNT 1.1. Original tree of Fig. 4b. Phylogenetic tree is based on the data described in Supplementary Information.
©2014 Macmillan Publishers Limited. All rights reserved
LETTER RESEARCH
Extended Data Figure 3 | Phylogenetic hypothesis of Ichthyosauriformes. Strict consensus of 82 most parsimonious trees (TL 5 527, CI 5 0.423, RI 5 0.796) obtained by a heuristic search in PAUP* 4b10 (hold 5 10,
nreps 5 100, addseq 5 random, swap 5 tbr). Numbers associated with clades are Bremer support values calculated in TNT 1.1. Original tree of Fig. 4c. Phylogenetic tree is based on the data described in Supplementary Information.
©2014 Macmillan Publishers Limited. All rights reserved
