sign in sign up
<<
Home , Brachypterygius, Chaohusaurus, Cryopterygius, Cymbospondylus, Grippia, Ichthyopterygia

Supplementary Information

Phylogeny of the incorporating the recent discoveries from South

CHENG JI,1,2 DA-YONG JIANG,*, 2, 3 RYOSUKE MOTANI,4 OLIVIER RIEPPEL,5 WEI-CHENG HAO,2 and ZUO-YU SUN2

1Key Laboratory of Economic Stratigraphy and Palaeogeography, Nanjing Institute of Geology and Palaeontology, Chinese Academy of Sciences, Nanjing 210008, People’s Republic of China, [email protected]; 2Laboratory of Orogenic Belt and Crustal Evolution, Ministry of Education; Department of Geology and Geological Museum, Peking University, Beijing 100871, People’s Republic of China, [email protected]; 3State Key Laboratory of Palaeobiology and Stratigraphy, Nanjing Institute of Geology and Palaeontology, Chinese Academy of Sciences, Nanjing 210008, People’s Republic of China; 4Department of Geology, University of California, Davis, California 95616, U.S.A., [email protected]; 5Section of Earth Sciences, Integrative Research Center, The Field Museum, Chicago, Illinois 60605-2496, U.S.A., [email protected]

RH: JI ET AL.—PHYLOGENY OF THE ICHTHYOPTERYGIA

*Corresponding author

INGROUP TAXA SELECTION

We aim at conducting a comprehensive phylogengetic analysis incorporating the recent discoveries during the past fifteen years. However, some genera are known from only partial skeleton and many characters are coded as “?” (Table 1S). Therefore, we provide two analyses in the text based on different combinations of ingroup taxa. In the first analysis, all the taxa in Table 1S are included and results in many polytomies. In the second analysis, the taxa which can not be coded for more than 1/3 (54.3 characters) are removed and the topology is highly resolved.

Table S1. Comparison on the numbers of characters that could be coded for each taxon. Abbreviations: Mix, . Pha, Phalarodon. Cym, . ‘C’, Callawayia. Sho, Shonisaurus. Qian, Qianichthyosaurus. Lep, . Pla, . Oph, Ophthalmosaurus. 155 Guanlingsaurus 128 48 Thadeosaurus 90 96 40 Claudiosaurus 141 Sho.popularis 88 43 Hovasaurus 116 Sho.sikanniensis 48 Athabascasaurus 61 120 Californosaurus 58 117 151 80 64 110 Qian.zhoui 127 Caypullisaurus 83 Gulosaurus 80 Qian.xingyiensis 101 Pla.australis 140 147 Callawayia 126 Pla.americanus 85 Mix.panxianensis 144 Hudsonelpidia 52 83 Mix.cornalianus 138 Macgowania 75 Oph.icenicus 156 Mix.kuhnschnyderi 23 Suevoleviathan 134 Oph.natans 110 Pha.atavus 104 Lep.tenuirostris 149 Malawania 40 Pha.fraasi 62 Lep.moorei 97 Chacaicosaurus 37 Pha.callawayi 49 Lep.solei 81 37 Cym.piscosus 132 118 Parvinatator 53 Cym.buchseri 62 Eurhinosaurus 145 Leninia 30 Cym.nichollsi 46 Hauffiopteryx 108 Thalattoarchon 41 Xinminosaurus 95 162 Palvennia 31 Besanosaurus 85 161 Sisteronia 20 Guizhouichthyosaurus 137 Temnodontosaurus 159 ‘C’.wolonggangensis 90 114

CHARACTER EVALUATION Most of the characters in Motani (1999) are followed or modified here. We try to avoid the gradually changed and proportional characters since the analyses require discrete characters. We also adopt most of the characters from Fischer et al. (2013) since it is so far the most complete phylogenetic analysis for the Thunnosauria. Some characters are removed or modified because they either already exist in our old matrix or appear to be inappropriate which will be explained below. Thus, each character is coded within four discrete character states. The followings are the characters that are modified or excluded from the present analysis with a variety of reasons. Motani (1999) char 5: this character describes the posterior extension of premaxilla dorsal process. In the post- , the premaxilla dorsal process is largely reduced and appears shorter than the ventral process, thus, does not separate the nasal from contacting the external naris. In most Triassic ichthyosaurs as Utatsusaurus, Grippia and Guanlingsaurus, the posterior process of premaxilla does not extend far posteriorly and the nasal contacts the external naris at least partially (Fig. 3A; Fig. 4A, F). Cymbospondylus piscosus has a relatively long premaxilla which forms the entire dorsal margin of the external naris, eliminating the nasal from the latter (Fig. 3C; Fig. 4C). However, in Guizhouichthyosaurus, it is quite obvious that the nasal is overlapped by the premaxilla posteriorly and the latter is not completely preserved (Maisch et al., 2006; Fig. 3E; Fig. 4E). Therefore, the premaxilla could have extended farther originally. In other words, this character is easily biased by the preservation especially on large-bodied ichthyosaurs and thus not adopted in this study. Motani (1999) char 15: the exposure of the frontal is related with the posterior extension of the nasal and the medially extension of postfrontal since the frontal is overlapped by the both. In Utatsusaurus, Grippia and Mixosaurus, the nasal does not extend far posteriorly and postfrontal is restricted along the lateral margin of the upper temporal fenestra (Fig. 4A, B). Thus, the frontal is exposed relatively large and the widest position is located on the posterior edge. The ichthyosaurs such as Guizhouichthyosaurus and Guanlingsaurus have a medially enlarged postfrontal and a deep V-shaped nasal frontal suture, leaving the frontal wider posteriorly than anteriorly (Fig. 4E, F; Maisch et al., 2006). Although they are both coded by the same state, the patterns are actually different. In Shastasaurus and Ichthyosaurus, the frontal exposure is restricted within the center of the skull roof due to the large postfrontal and the widest position is located anteriorly at the nasal suture (Fig. 4G). In ‘Callawayia’ wolonggangensis, Cymbospndylus piscosus and C. nichollsi, the frontal is exposed mostly lateral to the parietals due to the huge nasal extension and it is difficult to define the widest position (Fig. 4C, D). In Callawayia neoscapularis, Stenopterygius and Platypterygius, the frontal is overlapped by the nasal anteriorly and postfrontal postero-laterally, leaving the widest position in the middle rather than at the nasal or parietal suture. Here, this character is replaced by char 14 in the present analysis because the nasal extension is hard to define accurately. Motani (1999) char 59, 60, 62, 63, 65, 68, 71, 92: Reduction pattern of the shaft in long bones is significant through ichthyopterygian evolution and has been proposed to be related with the swimming ability (Motani, 1997a). In Motani (1999), it was divided into three states: present and complete, notch or largely reduced and absent because notch is proposed to represent a middle state of the shaft reduction. In addition, notch has been reported to be correlated with ontogenetic stages (Caldwell, 1997). However, as a matter of fact, notch occurs more optionally and randomly that it can not be accurately predicted. For example, on three specimens of Guizhouichthyosaurus of similar size, radiale notch occurs on two individuals while distal carpal II and metacarpal II notch occurs on only one individual. Similar phenomenon happens on some post-Triassic groups, too. Moreover, the radius of Shastasaurus clearly shows a notch on the leading edge while the whole element appeared as a flat discoidal element, lacking a shaft, indicating that notched and shaft absent can not be distinctly separated. Another counter example is that the radius of Mixosaurus panxianensis has two notches on the leading edge and this feature is consistent on plenty of specimens. Therefore, the notches of long bones might not be strictly resulted from the shaft reduction. Besides, as the peripheral margins of the epipodial of Toretocnemus and Qianichthyosaurus show an intermediate state between a complete shaft and a notch as Motani (1999) mentioned, it is not easy and objective to separate the two states. Therefore, the shaft of long bone is described by only two states here including: “(0) complete” such as the radius of Utatsusaurus and Grippia and “(1) largely reduced or absent” such as the radius of Shastasaurus and post-Triassic ichthyosaurs. Some characters are proposed based on the new described taxa since the publication of Motani (1999) and Maisch and Matzke (2000) and personal observations on some unpublished specimens from south China. (8) shallow groove anterior to the external naris: (0) absent; (1) present. Anterior to the external naris, a shallow and long groove is present in “Callwayia” wolonggangensis and Guizhouichthyosaurus (Fig. 3D, E). It is much narrower than the external naris and becomes narrower and pointed anteriorly. The length of the groove is about 3 times of that of the external naris. In the of Guizhouichthyosaurus, it is possibly present on the right side of the skull based on the irregular suture anterior to the external naris while it is not so obvious on the left side. However, since this skull is dorsoventrally compressed, its presence is not unambiguous and we tentatively code it as “0/1”. This feature has not been found in other so far. (18) postfrontal medially extension: (0) not over the anterior margin of upper temporal fenestra; (1) over the anterior margin of upper temporal fenestra. In Grippia, Utatsusaurus and Mixosaurus, the postfrontal locates lateral to the upper temporal fenestra and does not extend medially remarkably while the frontal is relatively wide posteriorly (Fig. 4A, B). In Cympospondylus piscosus, Guizhouichthyosaurus, Guanlingsaurus, Shastasaurus, ‘Callawayia’ wolonggangensis and most post-Triassic ichthyosaurs, the postfrontal has extended over the anterior margin of the upper temporal fenestra and resulted in the frontal restricted to the medial part of the skull roof (Fig. 4C-F). (19) supratemporal antero-medial extension: (0) short; (1) long. In most Triassic ichthyosaurs as Mixosaurus and Guizhouichthyosaurus, the supertemporal is outside of the parietal posteriorly while it contacts the parietal along a suture in most post-Triassic species (Fig. 4B, E). In Cymbospondylus piscosus and C. nichollsi, the supratemporal has a long process lateral to the parietal and extends anteriorly along the medial line, forming the posterior margin of the upper temporal fenestra. Therefore, this feature is probably the of the Cymbospondylus (Fig. 4C). (23) frontal participation in utf: (0) absent；(1) present. In Utatsusaurus, Grippia and Mixosaurus, the frontal contributes to a large part of the anterior terrace of the upper temporal fenestra without entering it (Fig. 4A). In Cymbospondylus, ‘Callawayia’ wolonggangensis and Shastasaurus, the frontal forms a part of the anterior margin of the upper temporal fenestra as well as the anterior terrace (Fig. 4C, D). Guizhouichthyosaurus has a small anterior terrace formed by parietal and postfrontal, eliminating frontal from the upper temporal fenestra (Fig. 4E). In most post-Triassic ichthyosaurs, no anterior terrace is present and the frontal is restricted to the medial area of the skull roof and appears to be relatively small (Fig. 4G). (31) quadratojugal: (0) longer than tall; (1) taller than long. This character is to show the variation on the width of the postorbit region which is decided by both postorbital and quadratojugal. However, the postorbital is lunate in most ichthyosaurs and is hard to quantify the variation. Thus, the shape of the quadratojugal is proposed instead to represent the width of the postorbit region. In the Early and ichthyosaurs, the cheek is as wide as the orbit or even wider and the quadratojugal appears wider than tall (Fig. 3A-C, E). In Shastasaurus, Qianichthyosaurus zhoui, ‘Callawayia’ wolonggangensis and most post-Triassic ichthyosaurs as Leptonectes, the cheek is narrow and the quadratojugal is clearly taller than long (Fig. 3D, F). The orbit/postorbit ratio varies ontogenetically and the coding here is mainly based on adult individuals. (57) dentigerous region in adult: (0) complete; (1) largely reduced; (2) edentulous. Shonisaurus popularis has been reported to have teeth set in sockets without dental groove (Camp, 1980; McGowan and Motani, 1999) but the jaw fragments are much smaller compared with the size of adult individuals. In addition, according to Nicholls and Manabe (2004), the articulated segments are not large individuals and none of the large specimens of Shonisaurus sikanniensis has teeth. Besides, no sockets or dental groove are found on the specimen skull and a segment which was referred to Shonisaurus comprises some teeth set in separate alveoli without a dental groove, similar to Shonisaurus popularis. Therefore, Nicholls and Manabe (2004) proposed that Shonisaurus might have teeth in juvenile but appear to be edentulous in adult. Sander et al. (2011) reported that Guanlingsaurus was toothless with a largely reduced snout. For the post-Triassic ichthyosaurs referred in the analysis, no taxa have been found to have reduced dentation. Ophthalmosaurus and Caypullisaurus of Late have been reported to have reduced and unfirmly attached teeth and the dental groove is also present so it may not be the same as Shonisaurus. (59) anterior sockets: (0) present; (1) absent. This character is coded separately for it is not correlated with the presence of dental groove. Motani (1997b) summarized the categories of the tooth implantation and defined different types by the presence of dental groove and sockets. Guizhouichthyosaurus and’Callawayia’ wolonggangensis have teeth set in separate sockets arranged at the bottom of a longitudinal dental groove and teeth are not tightly packed, which is subthecodont, different from that of the post-Triassic aulacodont. Qianichthyosaurus zhoui has relatively closely packed teeth which indicate that there may not be separate sockets there and the implantation could probably be aulacodont. The isolated jaw fragments of Shonisaurus popularis and S. sikanniensis both show deep sockets set on the jaws without a dental groove. In post-Triassic ichthyosaurs as Ichthyosaurus, teeth are closely packed in the dental groove leaving no space for the bony wall between two adjacent teeth. (78) coracoid foramen: (0) absent; (1) present. The coracoid of Cymbospondylus piscosus (Merriam, 1908) appears to have a small foramen enclosed near the anterior proximal margin (Merriam, 1908) which is a very primitive character among . The same structure were reported on C. buchseri and C. nichollsi (Sander, 1989; Fröbisch et al., 2006). No other ichthyosaurs show the same feature so far. (79) coracoid anterior notch or concavity: (0) absent; (1) present. (80) coracoid posterior notch or concavity: (0) absent: (1) present. In Utatsusaurus (Shikama et al, 1978), Grippia, Mixosaurus, Besanosaurus, Guizhouichthyosaurus and Ichthyosaurus, the coracoid is concave anteriorly and posteriorly and appears nearly symmetrical except in Mixosaurus. In Cymbospondylus piscosus and C. nichollsi, the coracoid has an anterior flange and appears convex anteriorly and concave posteriorly. In Californosaurus, Qianichthyosaurus zhoui, Toretocnemus and Stenopterygius, the coracoid has a posterior flange and appears convex posteriorly and concave anteriorly. (123) femur strongly constricted medially, forming a slender shaft region, proximal width remarkably larger than medial width: (0) absent; (1) present. Qianichthyosaurus (Li et al., 1999; Nicholls et al., 2003; Yang et al., 2013) and Toretocnemus (Merriam, 1903) both have strongly constricted femora, with the proximal width remarkably narrower than the distal width (Fig. 5K). In Callawayia neoscapularis, the femur of TMP 94.380.11 is slightly constricted which is not as typical as the former but more constricted than most Triassic species such as Guizhouichthyosaurus and Shastasaurus (Fig. 5F, H). In most Triassic genera as Mixosaurus and Shastasaurus, the femur is slightly constricted medially and expanded distally and the proximal width is much wider compared with Qianichthyosaurus (Fig. 5B, H). (124) femur antero-distal expansion, forming a distinctive structure at the distal end: (0) absent; (1) present. The femora of Toretocnemus and Qianichthyosaurus are narrow proximally and remarkably wider distally. The anterodistal and posterodistal margins are expanded, forming a thick and straight edge nearly parallel to the shaft of the femur (Fig. 5K). In other Triassic groups as Mixosaurus and Shastasaurus, the femora are gradually expanded distally, forming regularly concave anterior and posterior margins and the striations are parallel to the margins (Fig. 5B, H). The femur of Callawayia has a slight expansion on the anterodistal margin but not as typical as Qianichthyosaurus, still retaining the shape of most Triassic ichthyosaurs. (132) tibia antero-proximal end nearly rectangular, forming a deep notch on the anterior margin: (0) absent; (1) present. In Toretocnemus and Qianichthyosaurus, the anteroproximal margin of the tibia is not concave as in Shastasaurus but rather straight following with a deep notch distally on the anterior margin (Fig. 5K). Therefore, the anteroproximal margin appears shoulder-like and nearly perpendicular to the proximal facet. This is also seen in Callawayia neoscapularis (TMP 94.380.11) but not in other groups. (134) fibula contiguous margin: (0) concave; (1) nearly straight or convex. In Utatsusaurus (Shikama et al, 1978), Mixosaurus, Xinminosaurus (Jiang et al., 2008), Besanosaurus (Dal Sasso et Pinna, 1996), Guizhouichthyosaurus, Guanlingsaurus and Californosaurus (Merriam, 1902), the contiguous margin of the fibula is concave and the proximal facet is narrower than the distal. In most post-Triassic ichthyosaurs, the fibula is nearly round-shaped and the contiguous margin is either straight or convex (Fig. 5A-C, E-F, H-I). (135) fibula posterior flange: (0) absent; (1) present. Camp (1980) figured a fibula with posterior flange in Shonisaurus popularis which is different from the concaved posterior margin as most Triassic ichthyosaurs (Fig. 5J). A similar structure is also found in an additional specimen of Shonisaurus sikanniensis (RTMP 84.169.11). Nicholls and Manabe (2004) did not include this fibula in the paper while according to our personal observation, the size of the fibula matches with that of Shonisaurus because other ichthyosaur species reported from the same locality have much smaller sizes. Besides, Guanlingsaurus also has this flange on the fibula of the type specimen and an additional juvenile skeleton (Sander et al., 2011; Fig. 5G). No other ichthyosaurs have been reported to have similar feature. (158) last caudal rib facet reaching the tailbend area: (0) absent; (1) present. The primitive form is only found in Utatsusaurus where the caudal rib facets do not continue until the flexural area but disappear a few centra anteriorly. The last rib facet of Mixosaurus has reached the first articlined neural spine which is the same as Qianichthyosaurus, Guanlingsaurus and post-Triassic ichthyosaurs. (159) pre-flexural wedge-shaped centra: (0) absent; (1) present. The ichthyosaur tailbend is consisted by a series of wedge-shaped centra (McGowan, 1989). The anterior wedge-shaped centra are defined as the pre-flexural centra which are wedge-shaped from the lateral view. Nearly all the Triassic ichthyosaurs have already attained post-flexural wedge-shaped centra. Most Triassic ichthyosaurs as Utatsusaurus, Mixosaurus and Guanlingsaurus do not have the anterior wedge-shaped centra or not remarkable. In post-Triassic ichthyosaurs as Stenopterygius, the anterior wedge-shaped centra are typical and much shorter than the previous ones, contributing to the deep downward tailbend. (156) anterior caudal vertebral size about 1/2 of the largest dorsal vertebrae or less: (0) false; (1) true. In Utatsusaurus the size of the anterior caudal vertebrae do not vary greatly and remain similar sized as the middle dorsal vertebrae. In Chaohusaurus, they might be even slightly bigger than the middle dorsal centra. In Guanlingsaurus and Qianichthyosaurus, the size of the anterior caudal vertebrae has gradually reduced and is nearly half of the largest dorsal vertebrae at the caudal peak. In post-Triassic as Temnodontosaurus and Stenopterygius, the size of the vertebrae at the caudal peak is further reduced to only 1/3 of the largest dorsal vertebrae. (160) count between sacral and apical about: (0) 1/2 of the prepelvic count; (1) 2/3 of the prepelvic count or more. The presacral count of Chaohusaurus is about 38 and the preflexural count is about 54, suggesting that the vertebra count between sacral and apical is nearly half of the presacral count. Similar ratios are present in Mixosaurus, Xinminosaurus, Guanlingsaurus and Qianichthyosaurus. In Leptonectes tenuirostris, the presacral count is less than 50 while the preflexural count is over 80, indicating that the caudal region anterior to the peak is relatively longer. Similar ratios are also found in the other post-Triassic ichthyosaurs as Temnodontosaurus, Excalibosaurus, Eurhinosaurus, Ichthyosaurus and Stenopterygius. The characters for all Triassic ichthyopterygian taxa are coded based on personal observation of specimens of at least one of the authors except the taxa published from Middle Jurassic to Late during the last decade which have been followed here mainly based on literature. Most of the characters from Fischer et al. (2013) has been adopted here since it is so far the most complete phylogenetic analysis for the Thunnosauria. Some characters are removed because they either already exist in the matrix or appear inappropriate when applied on the Triassic taxa. The followings are the explanations for the characters of Fischer et al. (2013) which are not adopted in the current analysis. Fischer et al. (2013) char 1: crown striations: (0) presence of deep longitudinal ridges; (1) crown enamel subtly ridged or smooth. Here the author did not provide a quantitative definition and the two states of this character could be easily biased by the preservation. In addition, the presence of longitudinal ridges is ontogeny related at least in some groups such as Ichthyosaurus communis. Fischer et al. (2013) char 2: base of enamel layer: (0) poorly defined, invisible; (1) well defined, precise. Similar to the previous one, this character could easily be biased by the preservation condition. Fischer et al. (2013) char 6: processus postpalatinis pterygoidei: (0) absent; (1) present. This character was removed here because in most taxa the pterygoid was not preserved or not fully exposed. Even in Fischer et al. (2013)’s matrix, it was only coded for 8 taxa out of 25, which is less than 1/3. Fischer et al. (2013) char 27: tail fin centra; (0) strongly laterally compressed; (1) as wide as high. The shape of caudal centra varies anteroposteriorly. Anteriorly it appears closer to posterior dorsal centra and does not appear lateral compressed (e.g. Ophthalmosaurus in McGowan and Motani, 2003: fig. 18) while posteriorly the centra become compressed laterally in the apical region and postflexural segments. Fischer et al. (2013) did not specify which part of the tail, making this character impossible to be applied for the other taxa. Fischer et al. (2013) char 31: rib cross-section at mid-shaft: (0) rounded; (1) 8-shaped. This character appears to vary among different parts of the body. For example, within the same individual of Chaohusaurus, the rib cross-section is 8-shaped at the anterior dorsal part of the body while clearly rounded near the pelvic. It should also be specified that which part of the body is described before applied here. Fischer et al. (2013) char 32: (0) ossified haemapophyses: (0) present; (1) absent. The state ‘1’ was coded for only one taxon (Hudsonelpidia), suggesting it an autapomorphy of the genus. Fischer et al. (2013) char 51: longipinnate or lapinnate forefin architecture: (0) one; (1) two digit(s) directly supported by the intermedium. The intermedium does not give rise to any digit (Motani, 1999b: fig. 6) and ‘support’ is an ambiguous description of the contact between the intermedium and distal carpals. In most ichthyosaurs, the intermedium usually contacts both distal carpal III and IV and even distal carpal II (in Mixosaurus). Fischer et al. (2013) char 52: (0) zeugo-to autopodial elements: (0) flattened and plate-like; (1) strongly thickened. This character is highly preservational and size-related.

CHARACTER LIST

(1) premaxilla dorsal process: (0) long (1) short (modified from Maisch and Matzke, 2000:10) (2) premaxilla ventral process: (0) long (1) short (modified from Maisch and Matzke, 2000:9) (3) Maxilla anterior process: (0) reduced; (1) extending anteriorly as far as nasal or further anteriorly (Fischer et al., 2011:7) (4) maxilla dorsal lamina: (0) absent; (1) present (Motani, 1999:2) (5) maxilla prefrontal contact: (0) absent; (1) present (Maisch and Matzke, 2000:12) (6) maxilla external naris contact: (0) present; (1) absent (Motani, 1999:3) (7) external naris orientation: (0) lateral; (1) dorsal (Motani, 1999:4) (8) shallow groove anterior to the exn: (0) absent; (1) present (9) narial shelf: (0) absent; (1) present (Jiang et al., 2006:4) (10) nasal anteriorly extending beyond exn: (0) false; (1) true (11) nasal parietal contact lateral to frontal: (0) absent; (1) present (Motani, 1999:7) (12) nasal postfrontal contact: (0) no contact; (1) contact extensive, posterior extension of nasal separates frontal from prefrontal in dorsal view; (2) eliminated by prefrontal medially extension (Modified from Motani, 1999:6) (13) descending process of the nasal on the dorsal border of the nares: (0) absent; (1) present (Fernández, 2007:2) (14) processus narialis of prefrontal: (0) absent; (1) present (Fischer et al., 2011:11) (15) supraorbital crest on prefrontal and postfrontal: (0) absent; (1) present (Maisch and Matzke, 2000:22) (16) prefrontal–postfrontal contact: (0) absent; (1) present (Motani, 1999:8) (17) anterior orbital margin: (0) of regular rounded shape; (1) irregular (Maisch and Matzke, 2000:23) (18) postfrontal medially extension: (0) over the anteromost margin of utf; (1) not over the anteromost margin of utf (19) supratemperal antero-medial extension: (0) short; (1) long (20) supratemporal-postorbital contact: (0) absent; (1) present (Fischer et al., 2013:15) (21) sagittal eminence: (0) absent; (1) present but small, involving only the parietal; (2) present and large, involving the parietal, frontal and nasal (Motani, 1999:16) (22) frontal dorsal exposure: (0) clearly present; (1) nearly absent (23) frontal participation in utf: (0) absent; (1) present (24) squamosal triangular shape: (0) false; (1) true; (2) squamosal absent (Fischer et al., 2013:16) (25) squamosal participation in utf: (0) present; (1) absent; (2) squamosal absent (Motani, 1999:13) (26) postorbital postero-dorsal corner: (0) narrow, giving triradiate shape; (1) broad and triangular; (2) absent or round (27) postorbital participation in utf: (0) present; (1) absent (Motani, 1999:12) (28) jugal anterior margin: (0) tapering, between lacrimal and maxilla; (1) broad and fan-like, covering ma (29) jugal/quadratojugal lateral contact: (0) present; (1) absent (Motani, 1999:23) (30) lower temporal arch between jugal and quadratojugal: (0) present; (1) lost (Fisher et al., 2013a:18) (31) quadratojugal: (0) longer than tall; (1) taller than long. (32) quadratojugal exposure: (0) quadratojugal small; (1) extensive; (2) small, largely covered by squamosal and postorbital (Fischer et al., 2013:17) (33) parietal ridge: (0) absent; (1) present (Motani, 1999:17) (34) parietal supratemporal process: (0) short; (1) long (Motani, 1999:18) (35) parietals’ anterior processes: (0) contacting each other anteriorly, eliminating frontal from pineal foramen; (1) narrowly separated anteriorly, forming parietal fork, and frontal dorsally visible along the pineal foramen; (2) widely open, resulting in absence of clear fork (Motani, 1999:19) (36) anterior terrace of utf: (0) absent; (1) present but small; (2) present and large (Motani, 1999:14) (37) basioccipital peg: (0) clearly present; (1) absent or extremely reduced (Motani, 1999:29) (38) basioccipital extracondylar area: (0) wide; (1) reduced to a narrow band of concavity (Motani, 1999:30) (39) basioccipital/atlas articulation convexity: (0) flat or anterior; (1) posterior (40) ventral notch in the extracondylar area of the basioccipital: (0) present; (1) absent (Fischer et al., 2012:19) (41) pterygoid, transverse flange: (0) antero-lateral; (1) not well defined; (2) postero-lateral (Motani, 1999:26) (42) basipterygoid processes: (0) short, giving basisphenoid a square outline in dorsal view; (1) markedly expanded laterally, being wing-like, giving basisphenoid a marked pentagonal shape in dorsal view (Fischer et al., 2011:18) (43) interpterygoid vacuity: (1) present; (0) absent or extremely reduced (Maisch and Matzke, 1997:22) (44) ectopterygoid: (0) present; (1) absent (Callaway, 1989:9) (45) shape of the paroccipital process of the opisthotic: (0) short and robust; (1) elongated and slender (Fischer et al., 2012:20) (46) stapes proximal head: (0) slender, much smaller than opisthotic proximal head; (1) massive, as large or larger than opisthotic (Fischer et al., 2013:24) (47) angular lateral exposure at its maximum depth: (0) semi-equal to surangular; (1) clearly shallower than surangular; (2) much deeper than surangular (Motani, 1999:32) (48) coronoid region: (0) slightly elevated or high; (1) flat (Jiang et al., 2006:13) (49) cheek orientation: (0) lateral; (1) posterior (Motani, 1999:25) (50) overbite: (0) absent or very slightly; (1) present (Motani, 1999:33) (51) snout extremely slender: (0) no; (1) yes (Motani, 1999:34) (52) root striations: (0) present; (1) absent (Fischer et al., 2013:4) (53) plicidentine: (0) absent; (1) at least partly present (Motani, 1999:36) (54) bony fixation of teeth: (0) present; (1) absent (Motani, 1999:43) (55) tooth horizontal section: (0) circular; (1) disto-medially compressed; (2) lateral compressed (Motani, 1999:37) (56) tooth size relative to the skull width: (0) over 0.1; (1) below 0.05 (Motani, 1999:39) (57) dentigerous region in adult: (0) complete; (1) largely reduced; (2) edentulous (58) dental groove: (0) present throughout jaw margin; (1) only present anteriorly; (2) absent (Motani, 1999:41, 42) (59) anterior sockets: (0) present; (1) absent (60) maxilla multiple tooth row: (0) absent; (1) present (Motani, 1999:40) (61) dentary labial shelf: (0) present; (1) absent (Jiang et al., 2006: char14) (62) posterior tooth crown: (0) conical; (1) rounded; (2) flat (Motani, 1999:38) (63) tooth crown surface of at least one maxillary tooth with mesiodistal ridge: (0) false; (1) true (Jiang et al., 2006:18) (64) ossified sternum: (0) absent; (1) present (Motani, 1999:50) (65) ossified cleithrum: (0) present; (1) absent (Motani, 1999:51) (66) clavicle orientation at proximal end: (0) oblique to sagittal plane; (1) transverse (67) clavicle length beyond distal to main body: (0) short; (1) long (68) interclavicle anterior process separating clavicles: (0) present; (1) absent (69) interclavicle posterior process: (0) rod-like; (1) triangular; (2) absent (70) scapular blade shaft: (0) absent; (1) present at learst proximally (Motani, 1999:47) (71) scapula anterior flange: (0) complete; (1) emarginated; (2) absent (modified from Maisch and Matzke, 2000:70) (72) scapula antero-proximal extension toward clavicle: (0) absent: (1) present 73. prominent acromion process of scapula: (0) absent; (1) present (Fischer et al., 2011:28) (74) scapula posterior extension: (0) present; (1) absent (modified from Maisch and Matzke, 2000:71) (75) scapular axis and glenoid facet orientations: (0) nearly parallel; (1) at 60 degrees or more (Motani, 1999:48) (76) coracoid facet on scapula: (0) fused scapulocoracoid; (1) absent; (2) equal or smaller than glenoid facet of scapula; (3) twice as large as glenoid facet (Motani, 1999:49) (77) coracoid parasagittal length vs. transverse width: (0) semi-equal or longer than wide; (1) clearly wider than long (78) coracoid foramen: (0) present; (1) absent (79) coracoid anterior notch or concavity: (0) absent; (1) present (80) coracoid posterior notch or concavity: (0) absent: (1) present (81) intercoracoid facet: (0) short, medial margin round shaped; (1) long, medial margin relatively straight (modified from Maisch and Matzke, 2000:73) (82) anterior flange: (0) absent; (1) present and complete; (2) present but reduced proximally, leaving leading edge tuberosity (Callaway, 1989:29) (83) plate-like dorsal ridge on humerus: (0) absent; (1) present (Motani, 1999:56) (84) protruding triangular deltopectoral crest on humerus: (0) present but small; (1) present and very large, bordered by concave areas (Fischer et al., 2013:39) (85) humerus distal proximal width ratio: (0) nearly equal; (1) distal wider (Motani, 1999:55) (86) humerus with posterodistally deflected ulnar facet and distally facing radial facet: (0) ulnar facet contains convexity in lateral view; (1) absent; (2) present (modified from Fischer et al., 2013:42). (87) humerus distal articular facets: (0) not terminal; (1) radial facet larger than facet; (2) two facet nearly equal; (3) three facets (Motani, 1999:52) (88) humerus anterodistal facet for accessory zeugopodial element anterior to radius: (0) absent; (1) present (Fischer et al., 2013:41) (89) humerus/intermedium contact: (0) absent; (1) present (Fernández, 2007:15) (90) radius peripheral shaft: (0) complete or nearly complete; (1) notch or absent (modified from Motani, 1999:59) please see the text for detailed discussion. (91) radius contiguous shaft: (0) about half or more of radial length; (1) notch or absent (modified from Motani, 1999:60) please see the text for detailed discussion. (92) ulna peripheral shaft: (0) complete or nearly complete; (1) notch or absent (modified from Motani, 1999:62) please see the text for detailed discussion (93) ulna contiguous shaft: (0) complete or nearly complete; (1) notch or absent (modified from Motani, 1999:63) please see the text for detailed discussion (94) shape of the posterior surface of the ulna: (0) radius not discoidal; (1) rounded or straight and nearly as thick as the rest of the element; (2) concave with a thin, blade-like margin (Fischer et al., 2012:36) (95) radius/ulna relative size: (0) nearly equal; (1) radius much larger than ulna; (2) ulna larger than radius (Motani, 1999:64) (96) radio-ulnar foramen: (0) present; (1) absent (Fischer et al., 2013:46) (97) radiale, anterior notch: (0) absent; (1) present (Motani, 1999:65) (98) manual pisiform: (0) present; (1) absent (Motani, 1999:67) (99) manual pisiform 2 (neomorph): (0) absent; (1) present (Jiang et al., 2006:20) (100) intermedium: (0) longer than wide; (1) as wide as long or wider than long (Jiang et al.,2006) (101) interdigital separation: (0) present; (1) absent (Motani, 1999:78) (102) mc I peripheral shaft: (0) complete; (1) notch or largely reduced; (2) absent; (3) mc 1 not ossified (Motani, 1999:68) (103) mc III shaft: (0) present; (1) absent (Motani, 1999:69) (104) mc V: (0) present; (1) not ossified (Motani, 1999:70) (105) manual digit 2 distal elements peripheral shaft: (0) complete; (1) notch or absent; (Motani, 1999:71) (106) maximum phalangeal count: (0) five or less; (1) seven or more (Motani, 1999:77) (107) notching of anterior facet of leading edge elements of forefin in adults: (0) elements not discoidal; (1) present; (2) absent (Fischer et al., 2013:48) (108) preaxial accessory digits on forefin: (0) absent; (1) present (Maisch and Matzke, 2000:91) (109) postaxial accessory digit on forefin: (0) absent; (1) only one; (2) more than one (Motani, 1999:72) (110) proximal manual phalanges proximo-distal packing: (0) well-packed; (1) not packed (111) manual digit S4-5: (0) absent; (1) present (Motani, 1999:73) (112) propodial + epipodial versus manus length: (0) propodial + epipodial longer; (1) manus longer (Motani, 1999:58) (113) /hindlimb ratio: (0) nearly equal or hind longer; (1) forelimb longer but less than twice as hindlimb; (2) forelimb longer twice as much as hindlimb (Motani, 1999:79) (114) iliac blade shape: (0) with thick shaft; (1) plate-like; (2) narrow and styloidal (Motani, 1999:80) (115) iliac antero-medial prominence: (0) absent; (1) present (Motani, 1999:81) (116) pubis, styloidal or plate-like: (0) plate-like; (1) styloidal (Motani, 1999:85) (117) pubis obturator foramen: (0) completely enclosed in pubis; (1) mostly in pubis but open on one side; (2) part of obturator fossa (Motani, 1999:84) (118) ischium, styloidal or plate-like: (0) plate-like, longer sagittally than wide transversely; (1) plate-like, wider transversely than long sagittally; (2) styloidal (modified from Motani, 1999:87) (119) pubis and ischium median symphysis: (0) present; (1) not well defined (modified from Maisch and Matzke, 2000:108) (120) pubis and ischium fused in adult: (0) complete; (1) absent; (2) present only medially; (3) present medially and distally (Motani, 1999:83) (121) pubis ischium relative size: (0) nearly equal or ischium slightly larger; (1) pubis twice as large as ischium (Motani, 1999:86) (122) thyroid fenestra: (0) absent; (1) one median opening; (2) two openings, being medially separated (Motani, 1999:82) (123) femur strongly constricted medially, forming a slender shaft region, proximal width remarkably larger than medial width: (0) present; (1) absent (124) femur antero-distal expansion, forming a distinctive structure at the distal end: (0) absent; (1) present. (125) prominent, ridge-like dorsal and ventral processes demarcated from the head of the femur and extending up to mid-shaft: (0) absent; (1) present (Fischer et al., 2011:46) (126) wide distal femur blade: (0) absent, the proximal and distal extremity of the femur being sub-equal in dorsal view; (1) present (Fischer et al., 2013:61) (127) femur distal facets: (0) two; (1) three (Maxwell, 2010:32) (128) astragalus/femoral contact: (0) absent; (1) present (Maxwell, 2010:33) (129) femur anterodistal facet for accessory zeugopodial element anterior to tibia: (0) absent; (1) present (Fischer et al., 2011:48) (130) tibia contiguous shaft: (0) complete or nearly complete; (1) notch or absent (Motani, 1999:91) (131) tibia peripheral shaft: (0) complete or nearly complete; (1) notch or absent (modified from Motani, 1999:92) (132) tibia antero-proximal end nearly rectangular, forming a deep notch on the anterior margin: (0) absent; (1) present (133) fibula posterior extent: (0) not fixed, fibula being mobile relative to femur; (1) much posterior to femur; (2) about the same level as femur (Motani, 1999:93) (134) fibula contiguous margin: (0) concave; (1) nearly straight or convex (135) fibula posterior flange: (0) absent; (1) present (136) tibia and fibula: (0) in contact or closely placed with each other; (1) widely separated from each other (Motani, 1999:88) (137) spatium interosseum between tibia and fibula: (0) present; (1) absent (Fischer et al., 2013:64) (138) hind fin leading edge element in adults: (0) elements not discoidal; (1) notched; (2) straight (Fischer et al., 2013:65) (139) postaxial accessory digit: (0) absent; (1) present (Fischer et al., 2011:50) (140) pes digit 1: (0) present; (1) absent (Motani, 1999:89) (141) dt 2 in line with dt 3 and 4: (0) true; (1) false, dt 2 in line with calcaneum (142) atlas/axis fusion: (0) absent; (1) present (Motani, 1999:94) (143) presacral account: (0) less than 30; (1) between 40-52; (2) 55 or more (Motani, 1999:95) (144) posterior dorsal centra shape: (0) cylindrical; (1) discoidal (Motani, 1999:97) (145) posterior dorsal/anterior caudal centra: (0) 3.5 times or less as high as long; (1) four times or more as high as long (Fischer et al., 2013:26) (146) cervical bicipital rib facet: (0) absent; (1) present (Motani, 1999:99) (147) rib articulation in thoracic region: (0) predominantly unicapitate; (1) exclusively bicapitate (Maisch and Matzke, 2000:53) (148) antero-dorsal rib facets: (0) confluent with anterior facet in at least near pelvic girdle; (1) not confluent in any of the centra (Motani, 1999:101) (149) posterior-dorsal bicipital rib facet: (0) absent; (1) present, might be resulted from the split of diapophysis (Motani, 1999:100) (150) sacral ribs: (0) at least two, distinguishable; (1) absent (Motani, 1999:104) (151) anterior dorsal neural spine: (0) normal; (1) narrow (Motani, 1999:102) (152) neural spines of atlas-axis: (0) functionally separate, never fused; (1) completely overlapping, may be fused (Druckenmiller and Maxwell, 2010:26) (153) neural spine anticlination in tail: (0) absent; (1) present (Motani, 1999: 103) (154) posterior gastralia: (0) present; (1) absent (Motani, 1999:105) (155) caudal peak: (0) absent; (1) present (Motani, 1999:96) (156) anterior caudal vertebral size about 1/2 of the largest dorsal vertebrae or less: (0) false; (1) true (157) mid-caudal vertebrae height change: (0) gradually decrease; (1) increase; (2) sudden decrease (Motani, 1999:98) (158) last caudal rib facet reaching the caudal peak area: (0) false; (1) true (159) pre-flexural wedge-shaped centra: (0) absent; (1) present (160) vertebra count between sacral and apical about: (0) tail stem not well defined; (1) 1/2 of the prepelvic count; (2) 2/3 of the prepelvic count or more (161) tail: (0) as long or longer than the rest of the body; (1) distinctly shorter (modified from Maisch and Matzke, 2000:65) (162) lunate tailfin: (0) absent; (1) well developed lunate tailfin (modified from Maisch and Matzke, 2000:66) (163) chevrons in apical region: (0) present; (1) absent (Sander et al., 2000:72)

LITERATURE CITED

Caldwell, M. W. 1997. Modified perichondral ossification and the evolution of paddle-like limbs in ichthyosaurs and plesiosaurs. Journal of Vertebrate 17:534–547. Callaway, J.M. 1989. Systematics, phylogeny, and ancestry of Triassic Ichthyosaurs (Reptilia, Ichthyosauria). Ph.D. dissertation, University of Rochester: 203 pp. Camp, C. L. 1980. Large ichthyosaurs from the Upper Triassic of Nevada. Palaeontographica A 170:139–200. Dal Sasso, C., and G. Pinna. 1996. Besanosaurus leptorhynchus n. gen. n. sp., a new shastasaurid ichthyosaur from the Middle Triassic of Besano (Lombardy, N. Italy). Paleontologia Lombarda, Nouva Serie 4:3–23. Druckenmiller, P. S., and E. E. Maxwell. 2010. A new Lower Cretaceous (lower ) ichthyosaur genus from the Clearwater Formation, Alberta, Canada. Canadian Journal of Earth Sciences 47:1037–1053. Fernández, M. S. 2007. Redescription and phylogenetic position of Caypullisaurus (Ichthyosauria: ). Journal of Paleontology 81:368–375. Fröbisch, N. B., P. M. Sander, O. Rieppel. 2006. A new species of Cymbospondylus (Diapsida, ichthyosauria) from the Middle Triassic of Nevada and a re-evaluation of the skull osteology of the genus. Zoological Journal of the Linnean Society 147:515–538. Fischer, V. E., Masure, M. S. Arkhangelsky and P. Godefroit. 2011. A new Barremian () ichthyosaur from western Russia. Journal of Vertebrate Paleontology 31:1010–1025. Fischer, V. E., M. W. Maisch, D. Naish, R. Kosma, J. Liston, U. Joger, F. J. Krüger, J. P. Pérez, J. Tainsh and R. M. Appleby. 2012. New Ophthaomosaurid ichthyosaurs from the European Lower Cretaceous demonstrate extensive ichthyosaur survival across the Jurassic-Cretaceous Boundary. PLoS ONE 7(1):e29234. doi:10.1371/journal.pone.0029234 Fischer, V. E., R. M. Appleby, D. Naish, J. Liston, J. B. Riding, S. Brindley, and P. Godefroit. 2013. A basal thunnosaurian from Iraq reveals disparate phylogenetic origins for Cretaceous ichthyosaurs. Biology letters, 9(4):1–6. Jiang, D., L. Schmitz, W. Hao, Y. Sun. 2006. A new mixosaurid ichthyosaur form the Middle Triassic of China. Journal of Vertebrate Paleontology 26:60–69. Jiang D., R. Motani, W. Hao, L. Schmitz, O. Rieppel, Y. Sun and Z. Sun. 2008. New primitive ichthyosaurian (Reptilia, Diapsida) from the Middle Triassic of Panxian (Guizhou, southwestern China) and its position in the Triassic Biotic Recovery. Progress in Natural Science, 18:1315–1319. Li, C. 1999. Ichthyosaur from Guizhou, China. Chinese Science Bulletin 44:1318–1321. Maisch, M. W., and A. T. Matzke. 1997. Mikadocephalus gracilirostris n. gen., n. sp., a new ichthyosaur from the Grenzbitumenzone (-) of Monte San Giorgio (). Paläontologische Zeitschrift 71:267–289. Maisch, M. W., and A. T. Matzke. 2000. The Ichthyosauria. Stuttgarter Beiträge zur Naturkunde, Serie B 298:1–159. Maisch, M. W., X. Pan, Z. Sun, T. Cai, D. Zhang and J. Xie. 2006. Cranial osteology of Guizhouichthyosaurus tangae (Reptilia: Ichthyosauria) from the Upper Triassic of China. Journal of Vertebrate Paleontology 26:588–597. Maxwell, E. E. 2010. Generic reassignment of an ichthyosaur from the Queen Elizabeth Islands, Northwestern Territories, Canada. Journal of Vertebrate Paleontology 30:403–415. McGowan, C. 1989. The ichthyosaurian tailbend: A verification problem facilitated by computed tomography. Paleobiology 15:429–436. McGowan, C., and R. Motani. 1999. A reinterpretation of the Upper Triassic ichthyosaur Shonisaurus. Journal of Vertebrate Paleontology 19:42–49. McGowan, C., and R. Motani. 2003. Ichthyopterygia. Handbook of Paleoherpetology, Part 8. Verlag Dr. Friedrich Pfeil, Munich, 175 pp. Merriam, J. C. 1902. Triassic Ichthyopterygia from California and Nevada. University of California Publications, Bulletin of the Department of Geology 3:63–108. Merriam, J. C. 1903. New Ichthyosauria from the Upper Triassic of California. University of California Publications, Bulletin of the Department of Geology 3:249–263. Merriam, J. C. 1908. Triassic Ichthyosauria, with special reference to the American forms. Memoirs of the University of California 1:1–196. Motani, R. 1997a. Phylogeny of the Ichthyosauria (Amniota: Reptilia) with special reference to Triassic form. Ph.D. dissertation, University of Toronto, 384 pp. Motani, R. 1997b. Temporal and spatial distribution of tooth implantation in ichthyosaurs; pp. 81–103 in J. M. Callaway and E. L. Nicholls (eds.), Ancient Marine . Academic Press, London and New York. Motani, R. 1999. Phylogeny of Ichthyopterygia. Journal of Vertebrate Paleontology 19:473–496. Nicholls, E. L., C. Wei, and M. Manabe. 2003. New material of Qianichthyosaurus Li, 1999 (Reptilia, Ichthyosauria) from the Late Triassic of southern China, and implications for the distribution of Triassic ichthyosaurs. Journal of Vertebrate Paleontology 22:759–765. Nicholls, E. L., and M. Manabe. 2004. Giant ichthyosaurs of the Triassic: A new species of Shonisaurus from the Pardonet Formation (: Late Triassic) of British Columbia. Journal of Vertebrate Paleontology 24:838–849. Sander, R. M. 1989. The large ichthyosaur Cymbospondylus buchseri, sp. nov., from the Middle Triassic of Monte San Giorgio (Switzerland), with a survey of the genus in . Journal of Vertebrate Paleontology 9:163–173. Sander, P. M. 2000. Ichthyosauria: their diversity, distribution and phylogeny. Paläontologische Zeitschrift 74:1–35. Sander, P. M., X.-H. Chen and L. Cheng. 2011. Short-snouted toothless ichthyosaur from China suggests Late Triassic diversification of suction feeding ichthyosaurs. PLoS ONE 6(5):e19480. doi:10.1371/journal.pone.0019480 Shikama, T., T., Kamei and M. Murata. 1978. Ichthyosaurs, Utatsusauurs hataii gen. et sp. nov., from the Kitakami Massif, northeast . Science Reports of the Tohoku University, Second Series (Geology) 48:77–97. Yang, P., C. Ji, D. Jiang, R. Motani, A. Tintori, Y. Sun and Z. Sun. 2013. A new species of Qianichthyosaurus (Reptilia: Ichthyosauria) from Xingyi Fauna (Ladinian, Middle Triassic) of Guizhou. Acta Scientiarum Naturalium Universitatis Pekinensis 49:1002–1008.