The Early Permian mesosaurids are the oldest known aquatic amniotes with an exclusively Gondwanan distribution. Although several hundred of complete skeletons have been discovered and intensively studied, the anatomy and taxonomic composition of the group, as well as its phylogenetic relationships remain controversial. Several well-preserved mesosaurid specimens found in Uruguay justify a new anatomical reconstruction of the skull of Mesosaurus tenuidens, differing from earlier ones especially in the presence of a lower temporal fenestra. The significance of this structure for the evolution of temporal fenestration in amniotes is evaluated according to the two most recent phylogenetic hypotheses, in which mesosaurids are basalmost sauropsids or basalmost parareptiles. A synapsid-like fenestration may be the primitive condition for Amniota, and it may be also a basal condition for parareptiles, because recent phylogenies suggest a basal position for mesosaurids and lanthanosuchoids within that group, and both possess a lower temporal fenestra. Our results also give a moderately strengthened support for diapsid affinities of turtles.
Les mésosauridés du Permien inférieur sont les plus anciens amniotes aquatiques connus, avec une distribution exclusivement Gondwanienne. Bien que plusieurs centaines de squelettes complets aient été découverts et étudiés intensivement, l’anatomie et la composition taxonomique du groupe, ainsi que ses relations phylogénétiques demeurent controversées. Plusieurs spécimens de mésosauridés bien conservés trouvés en Uruguay justifient une nouvelle reconstitution anatomique du crâne de Mesosaurus tenuidens, différant des plus anciennes, en particulier par la présence d’une fenêtre temporale inférieure. La signification de cette structure pour l’évolution de la fenestration temporale chez les amniotes est évaluée en fonction des deux hypothèses plylogénétiques les plus récentes, à savoir que les mésosauridés sont les sauropsidés les plus basaux ou les parareptiles les plus basaux. Une fenestration de type synapside peut être la condition primitive pour les amniotes et pour les parareptiles, car de récentes phylogénies suggèrent une position basale pour les mésosauridés et les lanthanosuchoïdés au sein de ce groupe et les deux possèdent une fenestration temporale inférieure. Nos résultats renforcent modérément le soutien pour une position des tortues au sein des diapsides.
Mesosauridae, Early Permian, Gondwana, Skull anatomy, Temporal fenestration, Taphonomy, AmniotesMesosauridae, Permien inférieur, Gondwana, Anatomie du crâne, Fenestration temporale, Taphonomie, AmniotespresentedPresented by Philippe TaquetIntroduction
Geologically younger than the earliest known amniotes, the aquatic Mesosauridae from the Early Permian of Gondwana have disputed phylogenetic affinities as the basalmost sauropsids (Laurin and Reisz, 1995) or as basal parareptiles (Gauthier et al., 1988, Modesto, 1999, Modesto, 2000, Modesto, 2006 and Modesto and Anderson, 2004). More recently, a Bayesian study of amniote phylogeny placed them in various positions, including as the basalmost Eureptilia (Müller and Reisz, 2006). However, even more controversial is the configuration of their temporal region, which has shifted over time from diapsid (MacGregor, 1908), to synapsid (von Huene, 1941) and more recently, to anapsid (Modesto, 1999 and Modesto, 2006). Such different interpretations are surprising, considering the large number of specimens discovered in South America (mostly in Brazil) and South Africa, including mostly articulated and almost complete skeletons (Araújo, 1976). Thus, the interpretation of the skull morphology is complicated by the quality of preservation or inadequate preparation of the specimens. The latter condition prevails in specimens preserved within a calcareous matrix, which is difficult to remove without serious damage to the bones. In most cases, these skull roof bones are fractured and overlap each other and part of the palatal elements, due to compression during fossilization. Mesosaurid skull roof bones are very thin and delicate, a condition that favored distortion and fragmentation. Other specimens are preserved as natural moulds, and latex-rubber casts were commonly used to study and reconstruct the mesosaurid skeleton (e.g. von Huene, 1941, MacGregor, 1908, Modesto, 1996, Modesto, 1999, Modesto, 2006 and Oelofsen, 1981). Although useful, the casts reproduce the distortion present in the specimens.
Uruguayan mesosaurids have been little studied until the 1990s and they were virtually absent from the literature. Preliminary studies by Piñeiro, 2002, Piñeiro, 2004 and Piñeiro, 2006 documented dozens of specimens (including cranial and postcranial elements) in the Uruguayan Early Permian (Artinskian) Mangrullo Formation. In the Uruguayan specimens, the cranial and postcranial bones can easily be removed from the sediments, and there are also very well-preserved natural moulds that show the bones in their original anatomical position, nearly without distortion. All the taphonomic categories described by Soares, 2003 from the Brazilian Iratí Formation at the Paso de São Borja (Rio Grande do Sul) locality are present in Uruguay. Nonetheless, the well-preserved moulds of isolated skulls, which can preserve a very thin layer of the original bone, are considered an additional taphonomic class (Morosi, 2011) that, although not described by Soares, 2003, is also present among the Brazilian specimens.
The first interesting clue to the presence of a lower temporal fenestra provided by Uruguayan mesosaurid specimens was the discovery of isolated temporal bones (Piñeiro, 2002, Piñeiro, 2004 and Piñeiro, 2006). As von Huene, 1941 noted, the triradiate morphology of the jugal and the excavated lateral area of the postorbital suggested the delimitation of a lower temporal fenestra. Later discoveries, including well-preserved natural moulds of articulated specimens showing the skull roof bones, part of the palate, as well as part of the braincase and the occipital region, allow a complete description of the mesosaurid skull morphology, particularly the temporal region. The study of those specimens is the subject of the present contribution. The implications of mesosaurid temporal fenestration for the evolution of this character within Amniota and by extension, the controversal position of turtles, are also discussed.
Materials and methods
The examined specimens (FC-DPV 1732, 2061, 2209, 2242, 2389, 2395, 2446, 2454, 2457, 2468 and 2534) are isolated skulls housed in the Collection of Fossil Vertebrates (FC-DPV) of the Facultad de Ciencias, Montevideo, Uruguay. They are preserved as natural moulds that reproduce details of the finest structures. The elements are articulated in anatomical position, a condition rarely observed in mesosaurid skulls. Mostly, they consist of part and counterpart of the same individual, allowing the observation of the bone morphology in more than one plane. For instance, one part can show a plane closer to the dorsal surface, while the counterpart can reveal a plane closer to the ventral surface of the skull. This configuration, which is often very useful, is not always optimal. If only one part is found for several specimens, different interpretations of the same structures can result from taphonomic differences (consisting of various fracture planes).
The materials were collected from light and black shales of the Early Permian Mangrullo Formation, which contain a variable amount of organic matter. This unit outcrops in several localities in the north and northeast of the country, in Tacuarembó and Cerro Largo counties. They are associated with pygocephalomorph crustaceans and the trace fossil Chondrites (Piñeiro, 2006, Piñeiro, 2008, Piñeiro et al., in press and Piñeiro et al., 2012a). Binocular microscopes were used to study and photograph the specimens.
The evolution of temporal fenestration in amniotes was assessed through a parsimony optimization in Mesquite (Maddison and Maddison, 2011). The reference phylogeny, compiled from the literature (see figure legend for sources), was time-calibrated using Stratigraphic Tools (Josse et al., 2006). The phylogenetic signal of the characters was assessed through comparisons between the number of steps required by the reference trees and the same statistic in a population of 10,000 random trees produced by random taxon reshuffling (Laurin, 2004) and by a Yule process (Yule, 1925).
Systematic palaeontology
Amniota Haeckel, 1866
MesosauridaeBaur, 1889a
Mesosaurus tenuidensGervais, 1865
The assignment of the materials to Mesosauridae was based upon the elongate shape of the skull and its general morphology, including the long and slightly recurved teeth (von Huene, 1941, Modesto, 1999 and Modesto, 2006). Specific assignment to Mesosaurus tenuidens is supported by recent studies, which suggested that this taxon is the only mesosaurid present in Uruguay (Morosi, 2011), and by the lack of unambiguous characters that allow discrimination between this taxon and Stereosternum tumidum (Piñeiro et al., 2012b). According to previous authors, Mesosaurus appeared to differ from Stereosternum tumidum because the posterior projection of its frontal points laterally, whereas it is posteriorly directed in Stereosternum (Modesto, 1996, Modesto, 1999 and Modesto, 2006). However, the Uruguayan materials suggest that this character is taphonomic (see below). The other cranial character that was used to differentiate the two taxa is the length of the teeth, the largest ones in a series being equivalent to the length of three tooth positions in Stereosternum, and more than five positions in Mesosaurus (Modesto, 1996 and Modesto, 2006). Our measurements confirm this distinction, because none of the Uruguayan specimens have teeth as short as only three tooth positions. However, measurements of the largest teeth and tooth positions in our materials suggest that this character also exhibits intraspecific variation. In most specimens, the tooth size matches the length of five tooth positions, but we found also some that are equivalent to four, and others that were between four and five. We also found that the tooth position length can vary by several millimeters, even in the same tooth series. Thus, we are more inclined to think that intraspecific variation is common in this character and we have one single taxon in Uruguay, which according to previous studies is Mesosaurus. However, the value of this feature to recognize Stereosternum should be validated by studing the Brazilian and South African specimens assigned to this taxon. At the same time, postcranial characters used to distinguish Mesosaurus from Stereosternum may be strongly influenced by taphonomy, intraspecific ontogenetic variation, and sexual dimorphism (Piñeiro, 2002 and Piñeiro, 2004, Piñeiro et al., 2012b). Some characters, such as the skull-neck proportions (Araújo, 1976), may be subject to preservation problems that prevent accurate cervical and presacral vertebral counts. The different morphology of the interclavicle (Modesto, 1996 and Modesto, 1999) seems to be related to ontogenetic changes, as the diamond-like shape is more frequently present in juvenile and young adults, while large, mature individuals display a more triangular interclavicular shaft. The presence or absence of pachyostotic hemal arches (Araújo, 1976), is not a valid character, as both types can be observed on the tail of the same individual; the V-shaped, non-pachyostotic elements the posteriormost (Piñeiro, 2002). The supraneural process, a posterior projection of the neural canal observed in dorsal vertebrae assigned to Stereosternum (Modesto, 1996 and Modesto, 1999), was the only postcranial character that could not be checked because it is not present in the available materials from Uruguay. The status of those characters in Brazilosaurus sanpauloensis is difficult to establish because of the limited information about this taxon in the literature, and it seems not to be present in the Mangrullo Formation shales. However, the revision of some of the Brazilian specimens suggests that this taxon is a smaller mesosaurid, as suggested by the size of its teeth, which are invariably shorter than the length of two tooth positions. Moreover, all Brazilosaurus specimens have the centrae of anteriormost cervical vertebra (posterior to the atlas-axis complex) longer than the succeeding, producing a small increase of the length of the neck, feature that was considered to distinguish this species (Araújo, 1976 and Shikama and Ozaki, 1966).
The description is based mainly upon specimens FC-DPV 2534 and FC-DPV 2061 (Fig. 1), both preserved as natural moulds of the skull roof, although a very thin and discontinuous layer of bone seems to have been present in some of the elements, recognizable for its different texture and color. Other more fragmentary materials (see Material and Methods) also preserved as natural moulds, were used as additional specimens to better document some of the characters that are not so well displayed by the more complete skulls. FC-DPV 2061 is just a single slab, and FC-DPV 2534 consists in part and counterpart of a single skull. Most of the skull roof bones are preserved in their correct anatomical position in both specimens, which correspond to adult mesosaurids, according to size comparisons with articulated skeletons from Uruguay and Brazil. One of the specimens used for comparative purposes is a Mesosaurus embryo (Piñeiro et al., 2012b), which displays a triradiate jugal and a temporal fenestra, suggesting that the temporal area changes little during ontogeny.
Detailed descriptions of the skull of Mesosaurus tenuidens were published by von Huene, 1941 and Modesto, 2006 and most of the arrangements depicted by these authors were found in our materials. Thus, we base our description on their work, and emphasize new data that allow us to discriminate between their alternative interpretations, as shown in the most spectacular way in the temporal region.
Antorbital region
In this area, a small foramen, the foramen nariale obturatum (fno), after von Huene, 1941 is found posterior to the naris (Fig. 2). This foramen is invariably present in all mesosaurids, and has been described as an indentation of the lacrimal bone, anterodorsally bounded by the nasal (Modesto, 2006), which forms a short suture with the maxilla between the septomaxilla and the fno (Piñeiro et al., in press). This arrangement is reminiscent of the anteorbital fenestra of most archosaurs, which is much larger but similarly opens between the orbit and the naris and is surrounded by lacrimal, nasal and maxilla (Piñeiro et al., in press and Romer, 1956).
The prefrontal is a large bone, which forms the greatest part of the anterior orbital margin (Fig. 1, Fig. 2 and Fig. 3). It is partially overlapped by the lacrimal laterally (Modesto, 2006) (Fig. 3A); and both bones extend posterolaterally towards the jugal. The lacrimal make contact with the maxilla at its more posterior section and the prefrontal meets the jugal in a short suture. This lateral extension of the prefrontal reduces the contribution of the lacrimal to the orbital margin, which is restricted to its anteroventral edge. However, in specimens that preserve the internal surface of the skull roof the lacrimal seems to not reach the anterior orbital edge (or barely), which is formed entirely by the prefrontal, or nearly so (Fig. 3). The lacrimal duct is exposed in most specimens. Its orbital foramen opens at the posterolateral end of the lacrimal; anteriorly, it opens into the foramen nariale obturatum (Modesto, 2006) at its medial margin (Piñeiro et al., in press), as shown by some specimens that preserve the course of the lacrimal duct between both foramina (Fig. 3 F-G).
Our interpretation of the nasal differs from recently published accounts. Instead of the relatively short bone described by Modesto, 2006, it is very long, almost as long as the premaxilla, extending anteriorly to the level of the anteriormost end of the maxilla, as suggested by von Huene, 1941 and more recently by Morosi, 2011. The nasal meets the premaxilla in a long lateral and a short, fairly straight anterior suture near the middle of the snout (Fig. 1, Fig. 2 and Fig. 4). However, the internal view of these bones shows an apparently shorter nasal because it is in part hidden by the extensions of the premaxilla (Fig. 4E). As also suggested by von Huene, 1941, the premaxilla sends posterior processes lateral to the nasal, which defines nearly the entire dorsomedial margin of the naris (Fig. 4 C-G). This configuration differs from that favored by Modesto, 2006, in which the anterior end of the nasal is placed at the level of the anterior narial margin, where the bone bears extensive longitudinal ornamentation (Fig. 4C,D).
Interorbital region
The mesosaurid frontal has been described as transversely broadened posteriorly and bearing processes directed either posteriorly (as in Stereosternum) or lateroposteriorly (as in Mesosaurus), a condition unique among early amniotes (Modesto, 1996, Modesto, 1999 and Modesto, 2006). The analyzed Uruguayan specimens suggest that the presence of long posterior processes of the frontal in mesosaurids may be a taphonomic artifact instead of an anatomical attribute (contra Modesto, 1996, Modesto, 1999 and Modesto, 2006 and Piñeiro, 2004), although as with all mesosaurid material, this interpretation has to be treated with caution. Several natural moulds of the dorsal skull roof, preserved as part and counterpart, suggest that the frontal has an almost conventional transverse suture with the parietal and the postfrontal, displaying only short projections medial to the latter (Fig. 1 and Fig. 5). Those projections are not visible on the internal surface of the skull roof, and the postfrontal appears to occupy the position of the lateral tongue-like processes figured by Modesto, 1996, Modesto, 1999 and Modesto, 2006 (Fig. 5 B).
Postorbital and temporal regions
Paired postparietals are almost always present in the studied materials (Fig. 1, Fig. 2 and Fig. 5). They most likely were confined entirely to the occipital region, even though their flattened preservation occasionally gives the impression that a dorsal exposure was present. Thus, it is possible that the smooth ventromedial concavity met the ridged dorsal surface of the supraoccipital, thus forming the junction between braincase and skull roof, as suggested by von Huene, 1941.
There is a historical debate about the nature of the mesosaurid temporal area; however, Uruguayan skull materials suggest that at least in Mesosaurus, a lower temporal opening was present, in the shape and configuration described by von Huene, 1941. The fenestra is relatively large and it is clearly delimited by the triradiate jugal and contributions of the postorbital, the squamosal and the quadratojugal (Fig. 1, Fig. 2 and Fig. 6), a configuration reminiscent of some early synapsids, namely caseids (e.g., Romer and Price, 1940 and Sigogneau-Russell and Russell, 1974) and varanodontines (except Elliotsmithia; e.g., Campione and Reisz, 2010), given that the quadratojugal is excluded from the lower temporal fenestra in other synapsids. This configuration is shared also with a few reptiles such as Petrolacosaurus (Reisz, 1981) and, more interestingly, Australothyris (Modesto et al., 2009), Millerosaurus nuffieldi (Gow, 1972), and Lanthanosuchus watsoni (deBraga and Reisz, 1996).
The fenestra is an elongate window, with its maximum height at mid-length. The postorbital and jugal define its anterior edge. The postorbital is roughly triangular, with an acute lateral end that fits into a notch in the dorsal process of the jugal (Fig. 1 A,B). The squamosal forms the dorsal and posteroventral margin of the fenestra; its edge is smoothly concave along almost all its extension, and curves ventrally and slightly anteriorly to meet the quadratojugal at the posterolateral corner. The quadratojugal probably contributed to a short portion of the ventral edge of the fenestra, between the squamosal and jugal (Fig. 6).
The mesosaurid tabular has been interpreted in most previous studies as a predominantly occipital bone, or was considered to be absent (Modesto, 2006). On the contrary, other workers argued for the plesiomorphic condition of a large tabular in mesosaurids (Laurin and Reisz, 1995). Nevertheless, Uruguayan materials suggest that the tabulars are present on the dorsal skull table as thin slivers of bone, positioned lateral to the postparietals and flooring the supratemporals (Fig. 1, Fig. 2, Fig. 5 and Fig. 6). The contribution of these bones to the occipital area is speculative at this stage, as we have no specimens that preserve this region completely.
Discussion and conclusions
The restoration of some areas in dorsal and lateral views of the Mesosaurus tenuidens skull presented herein suggests that the anatomy of this basal amniote is far from completely known. The interpretation of mesosaurid cranial anatomy is complicated by taphonomic factors, and the same problem was experienced during the study of the associated pygocephalomorph crustaceans, (Piñeiro et al., in press and Piñeiro et al., 2012a). Nevertheless, the availability of specimens preserved with little or no postmortem disruption, where bones remained in their anatomical disposition, allowed us to present a new reconstruction of Mesosaurus tenuidens, which confirms early interpretations by von Huene, 1941. The presence of fenestration in Brazilosaurus is difficult to assert, since we have no specimens assignable to this taxon among the well-preserved skulls yielded by the Mangrullo Formation.
Without doubt, the main contribution of the present work is the reconstruction of the temporal area in Mesosaurus, which suggests the presence of a lower temporal fenestra, because of the historical importance of this character in the taxonomy and nomenclature of amniotes. Although its function is controversial (Case, 1924 and Cisneros et al., 2004, and references cited therein), the presence/absence of temporal fenestration has been long been considered a taxonomically very important character, long used in the Amniota classification (Baur, 1889b, Günther, 1867 and Osborn, 1903). Indeed, the taxa Synapsida, Diapsida, Euryapsida, Parapsida and Anapsida were all named according to the type of temporal fenestration. However, the description of several parareptiles with temporal fenestration (Cisneros et al., 2004, deBraga and Reisz, 1996, Gow, 1972, Hamley and Thulborn, 1993 and Reisz et al., 2007, among others) has impacted the relevance of this character by showing that it exhibits more homoplasy than previously thought. Furthermore, the bones bordering the lower temporal fenestra are not the same in all taxa; this is not reflected in the character coding of most analyses (e.g. deBraga and Rieppel, 1997, Laurin and Reisz, 1995 and Müller and Reisz, 2006). Thus, the phylogenetic implications of the presence of a synapsid-like fenestra in Mesosaurus tenuidens needs to be investigated by adding the new anatomical information described above; this will be included in a forthcoming paper.
Here, we analyze only the impact that the presence of a synapsid-like fenestra in mesosaurids has on the inferred ancestral condition for Amniota, by considering the two most probable hypotheses of mesosaurid relationships. If mesosaurids are the basalmost sauropsids as suggested by Laurin and Reisz (1995) and by Müller and Reisz (2006: Fig. 3b–d), and if all mesosaurids have a lower temporal fenestra, the fenestra is most parsimoniously interpreted as primitively present in Amniota and Sauropsida (Fig. 7A). Thus, the absence of a lower temporal fenestra or the presence of a ventral emargination in some parareptiles and in turtles may be a secondary condition (Piñeiro, 2004). However, the earliest known turtles (e.g. Triassochelis or Proganochelis) suggest that the primitive condition for turtles is the absence of temporal fenestration and ventral emargination appears to be absent or weakly developed. Under the phylogeny of Laurin and Reisz (1995), if mesosaurids are scored as polymorphic for this character, the primitive condition for Amniota, Sauropsida, and various basal reptilian sub-clades becomes ambiguous (Fig. 7B). However, if mesosaurids were the basalmost parareptiles (Gauthier et al., 1988, Modesto, 1999, Modesto, 2000 and Modesto, 2006) the primitive condition for Amniota, Reptilia, and Eureptilia is ambiguous, whether mesosaurids are scored as having the fenestra or as polymorphic, although the exact scoring of mesosaurids impacts on the most parsimonious condition in Parareptilia (Fig. 7C–D). For all four cases illustrated in Fig. 7, the phylogenetic signal in the character is highly significant (p ≤ 0.0025), whether the random trees are generated by random reshuffling or by a Yule process. Thus, our parsimony optimizations should yield reliable data on ancestral states (Laurin, 2004). The slight differences in results depending on scoring of Mesosauridae and in the position of this taxon among amniotes clearly show that more investigations about these topics are warranted. Nevertheless, the common and long-held view that amniotes were primitively anapsid seems to rest on shaky grounds.
Our new findings reduce somewhat the apparent conflict between morphological and molecular data concerning the position of turtles. Many (Laurin and Reisz, 1995, Lyson et al., 2010 and Modesto, 1999;) but not all (deBraga and Rieppel, 1997, Hill, 2005 and Li et al., 2008) morphological and some total evidence (Lee, 2001) analyses place turtles within parareptiles, a placement that was usually interpreted as implying that the absence of a lower temporal fenestra in turtles was primitive. On the contrary, most (e.g. Hugall et al., 2007 and Iwabe et al., 2005) but not all (Frost et al., 2006) molecular phylogenies place turtles within diapsids, which implies that the absence of temporal fenestrae in turtles is secondary rather than primitive. The position of turtles remains a difficult problem, as shown by some recent analyses (e.g. Becker et al., 2011). Our results raise the possibility that the lower temporal fenestra was indeed present in some ancestors of turtles, but this does not explain the absence of an upper temporal fenestra, if turtles are placed among diapsids. The upper temporal fenesta seems to have appeared only once and to have been lost only very slowly, according to the fossil record (Carroll, 1988). In that respect, it differs strongly from the lower temporal fenestra, whose evolutionary rate is apparently very high (Fig. 7), with multiple appearances and losses. Also, the temporal fenestration in recent morphological data matrices plays a more minor role than in early 20th century works; the position of turtles among parareptiles is supported by several osteological characters, most of which are not linked with temporal fenestration (Laurin and Reisz, 1995 and Lyson et al., 2010). Thus, our findings strengthen only moderately support for the position of turtles within Diapsida, but much work will be needed to resolve this outstanding problem.
Acknowledgements
We thank Elizabeth Morosi, Freddie Kuczera, César Goso and Daniel Picchi for assistance in the fieldworks and Melitta Meneghel for valuable discussions on the subject of this paper. Inés Castiglioni drew the reconstructions of the mesosaurid skull, taking into account all the anatomical novelties described in this paper. Alvaro Mones provided us valuable bibliography. We thank very much reviewers Jocelyn Falconnet, Kevin Padian, and Marcello Ruta for their useful comments. This work was financially supported by Grants from ANII FCE2007-96 to GP and CNRS (UMR 7207) and a CNRS/CONICET grant 22977 (ML).
AraújoD.Taxonomia e relações dos Proganosauria da Bacia do ParanáAn. Acad. Bras. Ciênc.48197691–116BaurG.Palaeohatteria Credner, and the ProganosauriaAm. J. Sci.37Third Series1889310–313BaurG.On the morphology of the vertebrate-skullJ. Morphol.31889467–474BeckerR.E.ValverdeR.A.CrotherB.I.Proopiomelanocortin (POMC) and testing the phylogenetic position of turtles (Testudines)J. Zoo. Syst. Evol. Res.492011148–159CampioneN.E.ReiszR.R.Varanops brevirostris (Eupelycosauria: Varanopidae) from the Lower Permian of Texas, with discussion of varanopid morphology and interrelationshipsJ. Vert. Paleontol.3032010724–746CarrollR.L.Vertebrate Paleontology and Evolution1988W.H. FreemanNew York698 pCaseE.C.A possible explanation of fenestration in the primitive reptilian skull, with notes on the temporal region of the genus DimetrodonContr. Mus. Geol. Univ. Mich.219241–12CisnerosJ.C.DamianiR.SchultzC.da RosaA.SchwankeC.NetoL.W.A procolophonoid reptile with temporal fenestration from the Middle Triassic of BrazilProc. R. Soc. Lond. B.27120041541–1546deBragaM.ReiszR.R.The Early Permian reptile Acleistorhinus pteroticus and its phylogenetic positionJ. Vert. Paleontol.161996384–395deBragaM.RieppelO.Reptile phylogeny and the interrelationships of turtlesZool. J. Linn. Soc.1201997281–354FrostD.R.GrantT.FaivovichJ.BainR.H.HaasA.HaddadC.F.B.The amphibian tree of lifeBull. Am. Mus. Nat. Hist.29720061–370GauthierJ.A.KlugeA.G.RoweT.The early evolution of the Amniota. The Phylogeny and Classification of the TetrapodsBentonM.J.Volume 1: Amphibians, Reptiles, Birds. Systematics Association Special Volume 35A.1988ClarendonOxford103–155GervaisP.Description du Mesosaurus tenuidens, reptile fossile de l’Afrique australeAcadémie des Sciences et Lettres de Montpellier Mémoires de la Section des SciencesTome Sixième1865Boehm et FilsMontpellier169–175GowC.E.The osteology and relationships of the Millerettidae (Reptilia: Cotylosauria)J. Zool. (Lond.)1671972219–264GüntherA.Contribution to the anatomy of Hatteria (Rhynchocephalus Owen)Philos. Trans. R. Soc. Lond.1571867595–629HamleyT.ThulbornT.Temporal fenestration in the primitive Triassic reptile Procolophon. In: Lucas S.G., Morales M. The Nonmarine TriassicN. M. Mus. Nat. Hist. Sci.31993171–174HaeckelE.Generelle Morphologie Der Organismen1866ReimerBerlinHillR.V.Integration of morphological data sets for phylogenetic analysis of Amniota: the importance of integumentary characters and increased taxonomic samplingSyst. Biol.542005530–547von HueneF.Osteologie und systematische Stellung von MesosaurusPalaeontogr. Abt. A.92194145–58HugallA.F.FosterR.LeeM.S.Y.Calibration choice, rate smoothing, and the pattern of tetrapod diversification according to the long nuclear gene RAG-1Syst. Biol.562007543–563IwabeN.HaraY.KumazawaY.ShibamotoK.SaitoY.MiyataT.Sister group relationship of turtles to the bird-crocodilian clade revealed by nuclear DNA–coded proteinsMol. Biol. Evol.222005810–813Josse, S., Moreau, T., Laurin, M., 2006. Stratigraphic tools for Mesquite. http://mesquiteproject.org/packages/stratigraphicTools/LaurinM.The evolution of body size. Cope's rule and the origin of amniotesSyst. Biol.532004594–622LaurinM.ReiszR.R.A reevaluation of early amniote phylogenyZool. J. Linn. Soc.11321995165–223LeeM.S.Y.Molecules, morphology, and the monophyly of diapsid reptilesContrib. Zool.7020011–18LiC.WuX.-C.RieppelO.WangL.-T.ZhaoL.-J.An ancestral turtle from the Late Triassic of southwestern ChinaNature4562008497–501LiuJ.AitchisonJ.C.SunY.-Y.ZhangQ.-Y.ZhouC.-Y.LvT.New mixosaurid ichthyosaur specimen from the Middle Triassic of SW China: further evidence for the diapsid origin of ichthyosaursJ. Paleontol.85201132–36LysonT.R.BeverG.S.BhullarB.-A.S.JoyceW.G.GauthierJ.A.Transitional fossils and the origin of turtlesBiol. Lett.62010830–833MacGregorJ.H.On Mesosaurus brasiliensis nov. sp. from the Permian of BrazilWhiteI.C.Final report presented to H. Ex. Dr. Làuró Seveŕiàno Müller Minister of Industry, Highways and Public Works. Imprensa Nacional, Rio de Janeiro1908301–336Maddison, W.P., Maddison, D.R., 2011. Mesquite: a modular system for evolutionary analysis. Version 2.75. http://mesquiteproject.orgModestoS.P.The anatomy, relationships, and palaeoecology of Mesosaurus tenuidens and Stereosternum tumidum (Amniota: Mesosauridae) from the Lower Permian of GondwanaXVIII1996University of TorontoToronto279 pModestoS.P.Observations on the structure of the Early Permian reptile Stereosternum tumidum CopePalaeontol. Afr.3519997–19ModestoS.P.Eunotosaurus africanus and the Gondwanan ancestry of anapsid reptilesPalaeontol. Afr.36200015–20ModestoS.P.The cranial skeleton of the Early Permian aquatic reptile Mesosaurus tenuidens: implications for relationships and palaeobiologyZool. J. Linn. Soc.1462006345–368ModestoS.P.AndersonJ.The phylogenetic definition of ReptiliaSyst. Biol.532004815–821ModestoS.P.ScottD.M.ReiszR.R.A new parareptile with temporal fenestration from the Middle Permian of South AfricaCan. J. Earth. Sci.46120099–20MorosiE.Estudio comparativo del cráneo en Mesosauridae de la Formación Mangrullo (Pérmico Temprano) de Uruguay2011Facultad de Ciencias. Universidad de la República, MontevideoUruguay137 pMüllerJ.ReiszR.R.The phylogeny of early eureptiles: comparing parsimony and Bayesian approaches in the investigation of a basal fossil cladeSyst. Biol.552006503–511OelofsenB.An anatomical and systematic study of the family Mesosauridae (Reptilia, Proganosauria) with special reference to its associated fauna and palaeoecological environment in the Whitehill Sea1981Ph.D. thesis, University of StellenboschSouth Africa250 pOsbornH.F.On the primary divison of the Reptilia into two sub classesSynapsida and Diapsida Science171903275–276PiñeiroG.Paleofaunas del Pérmico-Eotriásico de Uruguay. Tesis de Maestría2002PEDECIBA, Universidad de la República, MontevideoUruguay208 p. +28 FigsPiñeiroG.Paleofaunas del Pérmico y Permo-Triásico de Uruguay. Bioestratigrafía, Paleobiogeografía y sistemática.2004Universidad de la República MontevideoUruguay206 pPiñeiroG.Nuevos aportes a la Paleontología del Pérmico de UruguayVeroslavskyG.UbillaM.MartínezS.Cuencas Sedimentarias de Uruguay: Geología, Paleontología y Recursos Minerales, Paleozoico. DIRAC2006Facultad de Ciencias257–79PiñeiroG.Los mesosaurios y otros fósiles de fines del PaleozoicoPereaD.Fósiles de Uruguay. DIRAC. Facultad de Ciencias2008179–205Piñeiro, G, Ramos, A, Goso, C, Scarabino, F, Laurin, M., in press. Unusual environmental conditions preserve a Permian mesosaur-bearing Konservat-Lagerstätte from Uruguay. Acta. Palaeontol. Pol., doi:org/10.4202/app.2010.0113.PiñeiroG.MorosiE.RamosA.ScarabinoF.Pygocephalomorph crustaceans from the Early Permian of Uruguay: constraints on taxonomyRev. Bras. Paleontol.15201233–48PiñeiroG.FerigoloJ.MeneghelM.LaurinM.The oldest known amniotic embryos suggest viviparity in mesosaursHistorical Biology2012doi:org/10.1080/08912963.2012.662230ReiszR.R.A diapsid reptile from the Pennsylvanian of Kansas. University of Kansas. Museum of Natural HistorySpec. Publ.719811–74ReiszR.R.LaurinM.Owenetta and the origin of turtlesNature3491991324–326ReiszR.R.MüllerJ.TsujiL.ScottD.The cranial osteology of Belebey vegrandis (Parareptilia: Bolosauridae), from the Middle Permian of Russia, and its bearing on reptilian evolutionZool. J. Linn. Soc.1512007191–214RomerA.S.Osteology of the Reptiles1956University of Chicago PressChicago772 pRomerA.S.PriceL.I.Review of the Pelycosauria. Geological Society of America Special Paper 28Periodical monographs19401–538ShikamaT.OzakiH.On a reptilian skeleton from the Paleozoic formation of São PauloBrazil. Transaction Proceedings of Paleontological Society of Japan641966351–358Sigogneau-RussellD.RussellD.E.Étude du premier caséidé (Reptilia Pelycosauria) d’Europe occidentaleBulletin du Muséum national d’Histoire naturelle230Série 31974145–215SoaresM.B.A taphonomic model for the Mesosauride assemblage of the Irati Formation (Paraná Basin Brazil)Geologica. Acta.142003349–361YuleG.U.A mathematical theory of evolution, based on the conclusions of Dr J. C. Willis, F. R. S.PhilTrans. R. Soc. B.213192521–87
Mesosaurus tenuidens from the Early Permian Mangrullo Formation of Uruguay. Skull morphology. A. FC-DPV 2534, photograph of an almost complete natural mould of the skull roof, showing most bones in their anatomical position. B. Interpretive drawing of A. C, Photograph of the counterpart of FC-DPV 2534. D. Interpretive drawing of C. E. FC-DPV 2061, photograph of an almost complete skull showing areas where a very thin sheet of the original bones was preserved (dark areas). F. Interpretive drawing of E. Scale bar: 10 mm. Anatomical abbreviations: en, external naris; f, frontal;?hy, probable fragmentary hyoid; j, jugal; l, lacrimal; ld, lacrimal duct; ltf, lower temporal fenestra; mx, maxilla ; n, nasal; fno, foramen nariale obturatum; p, parietal; pf, parietal foramen; pmx, premaxilla; po, postorbital; pof, postfrontal; pp, postparietal; prf, prefrontal; pt, pterygoid; q, quadrate; qj, quadratojugal; sm, septomaxilla; sq, squamosal; sr, sclerotic ring; st, supratemporal; t, tabular.
Mesosaurus tenuidens de la Formation Margrullo, Permien inférieur d’Uruguay. Morphologie du crâne:A. FC-DPV 2534, photographie d’un moulage naturel presque complet du toit du crâne montrant la plupart des os dans leur position anatomique. B. Dessin interprétatif de A. C. Photographie de la contre-partie de FC-DPV 2534. D. Dessin interprétatif de C. E. FC-DPV 2061, photographie d’un crâne presque complet montrant des zones où une très mince couche d’os original a été préservée (zones foncées). F. Schéma interprétatif de E. Barre d’échelle : 10 mm. Abréviations anatomiques : en, narine externe; f : frontal; hy : probablement hyoïde fragmentaire ; j : jugal ; l : lacrimal, ld : conduit lacrimal ; ltf : fenêtre temporale inférieure ; mx : maxillaire ; n : nasal ; fno : foramen nariale obturatum ; p : pariétal ; pf : foramen pariétal ; pmx : prémaxillaire ; po : postorbitaire ; pof : postfrontal ; pp : postpariétal ; prf : préfrontal ; pt : ptérygoïde ; q : carré ; qj : quadratojugal ; sm : septomaxillaire ; sq : squamosal ; sr : anneau sclérotique ; st : supratemporal ; t :tabulaire.
Mesosaurus tenuidens from the Lower Permian Mangrullo Formation of Uruguay. Anatomical reconstruction. A. Reconstruction of the dorsal view of the skull based mostly on FC-DPV 2534 (part and counterpart) and FC-DPV 2061. FC-DPV 2109, 2377 and 2395 were used for the reconstruction of the snout. B. Black and white version. C. Lateral reconstruction. D. Black and white version. Colour code and anatomical abbreviations as for Fig. 1.
Mesosaurus tenuidens de la Formation Margrullo, Permien inférieur d’Uruguay. Reconstitution anatomique. A. Reconstitution de la vue dorsale du crâne, fondée pour la plus grande partie sur FC-DPV 2534 (partie et contre-partie) et FC-DPV 2061. FC-DPV 2109, 2377 et 2395 ont été utilisés pour la reconstitution du museau B. Version noir et blanc. C. Reconstitution latérale. D. Version noir et blanc. Le code de couleur et les abréviations anatomiques sont les mêmes que pour la Fig. 1.
Mesosaurus tenuidens from the Lower Permian Mangrullo Formation of Uruguay. Lacrimal/prefrontal configuration. A and B. Part and counterpart of specimen FC-DPV 2389 respectively, a partial anterior skull, showing apparently different configurations of the lacrimal-prefrontal contact and of their contribution to the anterior orbital margin (because of the different planes). C. Composite interpretive drawing of A and B. D. FC-DPV 2446, a partial anterior fragment of the skull roof displaying an internal view of the lacrimal and prefrontal. E. Interpretive drawing of D. Scale bars: 10 mm. Anatomical abbreviations as for Fig. 1.
Mesosaurus tenuidens de la Formation Margrullo, Permien inférieur d’Uruguay. Configuration lacrimale/préfrontale. A et B. Partie et contre-partie du spécimen FC-DPV 2389 respectivement, un crâne antérieur partiel montrant apparemment différentes configurations du contact entre lacrimal et préfrontal et de leur contribution à la marge orbitale antérieure (à cause des différents plans de fracture). C. Dessin interprétatif composite de A et B. D. FC-DPV 2446, fragment antérieur partiel de la calotte crânienne montrant une vue interne du lacrimal et du préfrontal. Dessin interprétatif de D. Barres d’échelle : 10 mm. Abréviations anatomiques comme pour la Fig. 1.
Mesosaurus tenuidens from the Lower Permian Mangrullo Formation of Uruguay. Nasal/premaxilla configuration. A. FC, DPV 2395, a fragment of the anterior region of the snout, showing the anterior extension of the nasal and its contact with the premaxilla. B. Interpretive drawing of A. C. FC-DPV 1732, a complete, nearly isolated long nasal bone, preserving the related lateral posterior process of the premaxilla. D. Interpretive drawing of C. E. FC-DPV 2468, a fragment of the narial region and incomplete snout preserving the outline of the nasal and premaxilla in internal view. F and G. FC-DPV 2454 and FC-DPV 2209 respectively, a complete nasal and a fragment of the anterior skull roof with the nasal completely preserved. Both specimens display the same configuration as in A-E. Scale bars: 10 mm. Anatomical abbreviations as for Fig. 1.
Mesosaurus tenuidens de la Formation Margrullo, Permien inférieur d’Uruguay. Configuration nasale/prémaxillaire. A. FC-DPV 2395, fragment de la région antérieure du museau montrant l’extension antérieure du nasal et son contact avec le prémaxillaire. B. Dessin interprétatif de A. C. FC-DPV 1732, os nasal long, complet, presque isolé, préservant le processus postérieur latéral du prémaxillaire qui lui est lié. D. Dessin interprétatif de C. E. FC-DPV 2468, fragment de la région nasale et museau incomplet préservant les contours du nasal et du prémaxillaire en vue interne. F et G. FC-DPV 2454 et FC-DPV 2209 respectivement un nasal complet et un fragment de la partie antérieure de la calotte crânienne totalement conservé. Les deux spécimens montrent la même configuration qu’en A-E. Barres d’échelle : 10 mm. Abréviations anatomiques comme pour la Fig. 1.
Mesosaurus tenuidens from the Lower Permian Mangrullo Formation of Uruguay. Paired postparietals. A. FC-DPV 2457, photograph of the part of a well preserved posterior fragment of the skull showing the skull table, along with some of the elements from the occiput, such as the paired postparietals and the tabulars. B. Interpretive drawing of A. C. Photograph of the counterpart of the specimen. D. Interpretive drawing of C. Scale bar: 10 mm. Anatomical abbreviations in Fig. 1.
Mesosaurus tenuidens de la formation de Margrullo, Permien inférieur d’Uruguay. Paires de post-pariétaux. A. FC-DPV 2457, photographie de la partie d’un fragment de crâne bien préservé montrant le plateau du crâne, ainsi que certains éléments de l’occiput, tels que la paire de pariétaux et les tabulaires. B. Dessin interprétatif de A. C. Photographie de la contre-partie du spécimen. D. Dessin interprétatif de C. Barre d’échelle : 10 mm. Abréviations anatomiques comme pour la Fig. 1.
Mesosaurus tenuidens from the Lower Permian Mangrullo Formation of Uruguay. Temporal fenestration. A. FC-DPV 2242, photograph of a fragmentary skull of a very small individual preserved as part and counterpart showing the configuration of the temporal area and the presence of a relatively large lower temporal fenestra, delimited by postorbital, squamosal, jugal and quadratojugal. The apparent lack of the postorbital contribution is due to the disarticulation and posterior disruption of the jugal from its original anatomical position. B. Counterpart of A. C. Composited interpretive drawing of A and B. Scale bar: 10 mm. Anatomical abbreviations in Fig. 1.
Mesosaurus tenuidens de la Formation Margrullo, Permien inférieur d’Uruguay. Fenestration temporale. A FC-DPV 2242, photographie d’un crâne fragmentaire d’un très petit individu, préservé en tant que partie et contre-partie montrant la configuration de la zone temporale et la présence d’une fenêtre inférieure relativement grande, délimitée par l’ensemble postorbitaire, squamosal, jugal et quadratojugal. L’apparent manque de contribution du postorbitaire est dû à la désarticulation et à la rupture postérieure du jugal à partir de sa position anatomique originelle. B. Contre-partie de A. C. Dessin interprétatif composite de A et B. Barre d’échelle : 10 mm. Abréviations anatomiques comme pour la Fig. 1.
Parsimony optimization of the lower temporal fenestra (white, absent; black, present; green, ventral emargination) on a reference phylogeny of amniotes. This ternary character is considered unordered for optimization and phylogenetic signal purposes as there is evidence of changes from the fenestrate condition to the emargination (as in squamates and ichthyosaurs) and of appearance of an emargination directly from the anapsid condition, as in Owenetta rubidgei (Reisz and Laurin, 1991 and this figure). Topology from Laurin and Reisz, 1995 for basal amniotes, Reisz et al., 2007 for parareptiles, and Laurin, 2004 for other taxa, for A and B; from Modesto, 1999 for the position of mesosaurids, for C and D. Temporal fenestration is present in Ankyramorpha (deBraga and Reisz, 1996). To save space, the geological timescale is restricted to the Devonian-Triassic interval. Extant taxa are in bold type. For all taxa, we attempted to score the primitive condition. For Ichthyosauria, polymorphism follows Liu et al., 2011.
Optimisation par parcimonie de l’évolution de la fenêtre temporale inférieure (blanc, absente ; noire, présente ; verte, émargination ventrale) sur une phylogénie de référence des amniotes. Ce caractère ternaire est considéré comme non ordonné pour l’optimisation et le signal phylogénétique car les données suggèrent des transitions entre fenêtre et émargination (comme dans les squamates et les ichtyosaures) ainsi que l’apparition d’une émargination directement à partir de la condition d’anapside, comme chez Owenetta rubidgei (Reisz and Laurin, 1991 et Fig. 7). Topologie d’après Laurin and Reisz, 1995 pour les Amniotes basaux, d’après Reisz et al., 2007 pour les parareptiles et d’après Laurin, 2004 pour les autres taxons, pour A et B ; d’après Modesto, 1999 pour la position des Mésosauridés, pour C et D. La fenestration temporale est présente chez les Ankyramorpha (deBraga and Reisz, 1996). Pour économiser de l’espace, l’échelle des temps géologiques est cantonnée à l’intervalle Dévonien-Trias. Les taxons actuels sont indiqués en gras. Pour tous les taxons, les auteurs ont tenté de coder la condition primitive. Pour les Ichthyosauriens, le polymorphisme suit Liu et al., 2011.