Clevosaurus was a cosmopolitan rhynchocephalian genus, known from the Late Triassic to the Early Jurassic. In South America this genus is represented by C. brasiliensis, an important component of the Linha São Luiz taphocoenosis, on the top of the Norian Santa Maria 2 Sequence of Southern Brazil. The best preserved and most abundant bone elements of C. brasiliensis are dentaries, in which variations of shape and size are observed. The aim of this study is to describe and evaluate the variation, using geometric morphometrics methods. Geometric morphometric analysis of 10 specimens highlights variations in relative size of the dentary. Most of the variation observed for PC1 (83.3%) is likely related to ontogeny, and PC2 (10.0%) is likely related to taphonomic signatures. The development patterns observed, such as the growth of the dentary, consists of differential growth in length between the posterior portion of the dentary, that grows at a higher rate, regarding the anterior portion of the element. This allometric growth is similar to what is observed in other rhynchocephalians and is accompanied by the allometric skull growth, similar to the trend exhibited by clevosaurs. The taphocoenosis is bimodal (juveniles and adults) with a bias towards adult preservation. Some diagenetic influence is reflected in deformed skulls and this is observed in the tangent-plot. Finally, a strong correlation was detected between the taphonomic signatures and the PC2, regarding specially disarticulation and degree of fragmentation.
Citation: Romo de Vivar Martínez PR, Bento Soares M (2015) Dentary Morphological Variation in Clevosaurus brasiliensis (Rhynchocephalia, Clevosauridae) from the Upper Triassic of Rio Grande do Sul, Brazil. PLoS ONE 10(3): e0119307. https://doi.org/10.1371/journal.pone.0119307
Academic Editor: David Carrier, University of Utah, UNITED STATES
Received: February 24, 2014; Accepted: January 29, 2015; Published: March 20, 2015
Copyright: © 2015 Romo de Vivar Martínez, Bento Soares. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Funding: The authors thank CAPES (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior) for financial support (PRM) and CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico) for MBS financial support. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Rhynchocephalia, the monophyletic group encompassing Sphenodon and its fossil relatives , is the sister group of Squamata (lizards, snakes and amphisbaenas) and, together, they comprise the lepidosaurian reptiles (Lepidosauria). Squamata includes more than 9000 extant species, in contrast, the only living genus of Rhynchocephalia is Sphenodon from New Zealand [1,2 3]. The current diversity of this group does not reflect its past diversification, when its members were a common component of the continental vertebrate faunas from the Triassic to the Jurassic, with more than 40 fossil taxa. The rhynchocephalians ranged from small to large forms which lived in aquatic and terrestrial ecosystems and experienced different dietary habits such as insectivores, carnivores, omnivores and even herbivores [1,4]
Geometric morphometrics provide an efficient tool for quantitative biology in the study of shape variation and identification of the cause of it . Recently, the use of geometric morphometrics has expanded in vertebrate paleontological studies (e.g. Maxwell and Dececchi  in ichthyosaurs; Campione and Evans , Hedrick and Dodson , Foth and Rauhut  in dinosaurs; Fariña and Vízcaino , Meloro , in mammals).
Concerning Rhynchocephalia, there are only three works in the realm of geometric morphometrics, Jones , Meloro and Jones , and Humpries and Jones . Before this new approach, ontogenetic studies were carried out using Sphenodon and, in fact, some ontogenetic stages are well documented [14,15,16,17]. Regarding fossil Rhynchocephalia, there are some studies with brief notes about ontogeny, as those of Hoffstetter , Fraser [19,20], Renesto , Reynoso , Apesteguía et al.  and Rauhut et al. . However, the only study focusing primarily on the ontogeny of the fossil material was conducted by Reynoso , in which were described growth patterns and ontogenetic variation of the jaws and teeth in Cynosphenodon huizachalensis. Jones  used reconstructions of the following fossil taxa: Kuehneosarus latus (outgroup), Gephyrosaurus bridensis, Diphydontosaurus avonis, Planocephalosaurus robinsonae, Palaeopleurosaurus posidoniae, Pleurosaurus goldfussi, Brachyrhinodon taylori, Clevosaurus hudsoni, Clevosaurus bairdi, Priosphenodon avelasi, and 37 specimens of different ontogenetic stages of Sphenodon. The author noted that some of the evolutionary trends in fossil rhyncocephalian skulls are very similar to the ones found along the ontogeny of Sphenodon, in which the skull undergoes an allometric growth associated with possible by dietary changes. The sample of Meloro and Jones , is the same of Jones  with the addition of Clevosaurus brasiliensis, and the two stem lepidosaurs Sophieta and Marmoretta. The authors observed the relation between skull shape and dentition, with habitat and dietary preferences. Recently Apesteguía and Carballido  in their description of Priosphenodon minimus made brief comments on the ontogeny of Pr. minimus, such as: the differences in size of interpterygoid fenestra between juveniles and adults; the differences of tooth wear, which is stronger in subadults and adults; the form of supratemporal fenestra, which changes from oval to D-shape; the postorbitals, which expand anteroposteriorly during growth; the interpterygoid fenestra and the pterygoid central region narrow during growth; and the nasals overlap the prefrontal and form a hook-like structure in dorsal view that embraces the nasal posterior end, in juveniles specimens .
The aim of the current study is, with the help of geometric morphometrics, to understand the morphological variation observed in a sample of 17 specimens of Clevosaurus brasiliensis Bonaparte and Sues  recovered from the outcrop “Linha São Luiz”, assigned to the top of the Upper Triassic Santa Maria 2 Sequence  (Norian), and the potential causes of this variation, whether by ontogenetic processes, sexual dimorphism, taxonomic diversity, or even effect of taphonomy. C. brasiliensis was the first Rhyncocephalia recovered from the Triassic of South America. More recently, a new taxon, Sphenotitan leyesi from the Quebrada del Barro Formation, Northwestern Argentina, also Norian in age, was described .
The genus Clevosaurus was erected by Swinton (1939) and corresponds to a cosmopolitan genus recorded from the Upper Triassic of Belgium, Brazil, England, and the USA to the Lower Jurassic  of China, South Africa and Wales, with a possible occurrence in the Lower Jurassic of Mexico [20, 26, 30, 31, 32, 33]. The clade Clevosaurs is formed by Brachyrhinodon, Polysphenodon and Clevosaurus, and has been recognized in three different phylogenetic analysis: Wu , Reynoso  and Rauhut et al. . According to Reynoso  Clevosaurs is supported by the following characters: anterorbital region is less than a quarter of the length of the skull; maxillary teeth with a small medial flange. Bonaparte and Sues  formally proposed the family Clevosauridae and defined it as: comprising the last common ancestor of Brachyrhinodon, Polysphenodon and Clevosaurus and all of its descendants . In this sense, Clevosauridae is the formal taxonomic equivalent of the informal grouping Clevosaurs recognized in the phylogenetic analysis aforementioned. The diagnosis of Clevosauridae is as follows: length of anterorbital region one-quarter or less of total skull length; length of lower temporal fenestra more than one-quarter of skull length; anterior (premaxillary) process of maxilla small or absent; maxilla excluded from posterior margin of external naris in Clevosaurus; condition uncertain in Brachyrhinodon.
The following characters are generally considered diagnostic of Clevosaurus by several authors such as Wu , Säilä , Fraser , Fraser and Walkeden , Robinson , Jones [29, 36] and Apesteguía and Novas , although some of them are found in other Rhynchocephalia: a dorsally expanded lateral process of the premaxilla that may exclude the maxilla from the external naris (this was considered an important diagnostic character of Clevosaurus, but not exclusive); the premaxillary process of the maxilla is very short or absent; a relatively short snout; a broad contact between the maxillary and jugal; process of the jugal extends posteriorly to meet the squamosal; presence of supratemporal bones; a complete lower temporal bar (although it is often only inferred); the relatively broad orbital portion of the maxilla; suborbital fenestra bounded solely by the lateral process of the ectopterygoid and palatine; a few (<5) large blade-like marginal sub-conical and conical teeth; a reduced number of teeth on the palate; two rows of pterygoid teeth; and the remarkable flanges on the teeth. Thus Clevosaurus is still a valid genus based on those characters.
Material and Methods
Concerning the item Observational and Field Studies, we inform that all fossil materials that were studied by us and presented in this manuscript, have already been collected in field trips between 2000 and 2009 (and housed the Paleovertebrate Collection of at the Universidade Federal do Rio Grande Sul, Brazil). Marina Bento Soares (co-author) is Curator of this collection. Therefore, none authorization was required to access this collection. The access to the other collection cited, FZBRS, was permitted by Dr. Ana Maria Ribeiro, as mentioned in the Acknowledgments.
Locality and horizon
The material used in this study proceeds from the outcrop Linha São Luiz (29°33ʹ;45”S; 53°26ʹ;48”W), which is located in the Faxinal do Soturno Municipality, Rio Grande do Sul State, Southern Brazil (Fig. 1). This locality is assigned to the top of the Santa Maria 2 Sequence  (Norian in age) (Fig. 2A). This sequence includes two systems: a transgressive system tract, composed of red, either laminated or massive mudstone on its base, and a low stand system tract dominantly composed by sandstones, located on its upper portion . This sequence is interpreted as a fluvial/deltaic depositional environment .
Location map of the Faxinal do Soturno municipality in the central region of Rio Grande do Sul State, Brazil. Modified from .
The Linha São Luiz outcrop assigned to the latter portion, is about 20m thick (Fig. 2B), and it is composed on its base by fine-grained and well selected medium-grained sandstone with cross-bedded, low angle stratification, followed by massive sandstones with cross-bedded stratification The middle portion is composed of mudstones, and the upper portion is composed by rhythmic sandstones and mudstones .
The vertebrate fossiliferous levels correspond to the basal portion of the outcrop, where the sandstones are dominant (Fig. 2B). The rhynchocephalian material, referred as Clevosaurus brasiliensis in , Arantes et al. , and Arantes (unpublished data), is associated with the non-mammaliaform cynodonts Riograndia guaibensis, Irajatherium hernandezi, Brasilodon quadrangularis, Brasilitherium riograndensis, Minicynodon maieri, the procolophonid Soturnia caliodon, the basal ornithodira Faxinalipterus minima, the non-rhynchocephalian lepidosauromorph Cargninia enigmatica, and the dinosaur Guaibasarus candelariensis, the latter being the largest tetrapod of the assemblage [41,42,43, 44, 45, 46, 47, 48, 49, 50]. The presence of R. guaibensis, B. riograndensis and G. candelariensis allows the correlation of the Linha São Luiz Fauna with other faunas from the top of the Santa Maria 2 Sequence. This faunistic association is individualized as Riograndia Assemblage Zone, as proposed by Soares et al. .
The total sample consists of 17 dentaries attributed to Clevosaurus brasiliensis [26, 40], Arantes (unpublished data). These dentaries were found in three different degrees of disarticulation: (1) isolated dentaries (Fig. 3: B, E, F, H, L, Q), (2) dentaries articulated with the angular and the articular complex (surangular, angular and prearticular), but disarticulated from the skull (Fig. 3: A, K, I, M), or (3) complete lower jaw articulated with skull (Fig. 3: C, D, G, J, N, O, P). Some of the specimens are included in the rock matrix and for this reason it was not possible to observe both sides. The specimens used in this study were:
- Isolated dentaries
- UFRGS-PV-0606-T: incomplete right dentary (lacking the posterior portion, and a small antero-ventral portion)
- UFRGS-PV-0738-T: incomplete right dentary (lacking the posterior portion)
- UFRGS-PV-0739-T incomplete right dentary (lacking the posterior portion, and a small antero-ventral portion)
- UFRGS-PV-0750-T: part of the right dentary (lacking the posterior portion)
- UFRGS-PV-0758-T: incomplete left dentary (lacking the posterior portion)
- UFRGS-PV-1153-T: two unarticulated dentaries (lacking the posterior portion)
- Dentaries articulated with the angular and the articular complex:
- MCN-PV-2852: right and left unarticulated lower jaws (associated to the skull but not articulated)
- UFRGS-PV-0752-T: left lower jaw (lacking the posterior portion)
- UFRGS-PV-0754-T: right and left unarticulated lower jaws
- UFRGS-PV-0848-T: left lower jaw
- Complete lower jaw articulated with skull:
- UFRGS-PV-0613-T: skull with right and left lower jaws, laterally compressed and lacking the anterior portion
- UFRGS-PV-0735-T: skull with right lower jaw (lacking the anterior portion)
- UFRGS-PV-0748-T: skull with right and left lower jaws in occlusion (dorsoventrally compressed)
- UFRGS-PV-0753-T: skull with lower jaw in occlusion (only the left part preserved)
- UFRGS-PV-0972-T: skull with right lower jaw (obliquely compressed, lacking the anterior portion)
- UFRGS-PV-0974-T: skull with right and left lower jaws in occlusion (laterally compressed)
- UFRGS-PV-1152-T: skull with right and left lower jaws in occlusion (dorsoventrally compressed)
Some of this material has been previously described. The type series of C. brasiliensis as described by Bonaparte and Sues  is composed by the specimens UFRGS-PV-0748-T (holotype) and UFRGS-PV-0613-T (referred specimen). The specimens UFRGS-PV-0735-T; UFRGS-PV-0753-T; UFRGS-PV-0972-T; UFRGS-PV-0974-T; UFRGS-PV-1152-T; UFRGS-PV-1153-T; UFRGS-PV-0972-T were described by Arantes (unpublished data), as C. brasiliensis. The specimen MCN-PV-2852 was described by Ferigolo [51, 52] but the author did not assign it to C. brasiliensis, proposing instead that this material could belong to a different taxon. This point of view was rejected by Arantes (unpublished data) who has considered MCN-PV-2852 as C. brasiliensis, a proposition followed in this study.
Finally, the specimens UFRGS-PV-0606-T; UFRGS-PV-0738-T; UFRGS-PV-0739-T; UFRGS-PV-0750-T; UFRGS-PV-0752-T; UFRGS-PV-0754-T; UFRGS-PV-0758-T; UFRGS-PV-0848-T have not been described yet. In this study, we consider they belong to Clevosaurus brasiliensis due to the features they share with this taxon  and other Clevosaurus (Arantes, unpublished data). These features are: acrodont tooth implantation; additional teeth in the lower jaw, which will decrease gradually and disappear; the evident coronoid process (being half of the height of the lower jaw); absence of a ventral projection in the vertical dentary symphysis with a moderate dorsal development; the posterior process of the dentary expands to the level of the glenoid fossa; sub-conical and conical teeth [20, 26, 37].
Dentition is a relevant aspect that changes along the ontogeny. Thus, we judge important to make some briefly comments about it. The dentition in Clevosaurus brasiliensis is characterized by conical or sub-conical flanged teeth with the long basis mesiodistally directed. This was previously observed by Bonaparte and Sues  and Arantes (unpublished data), and this is also a general trend in Clevosaurus [20, 30, 35, 36, 53]. According to Bonaparte and Sues , in mature C. brasiliensis, two additional flanged teeth were found in the maxillary followed by one or two smaller, sub-conical teeth. In general we observed this, however, with respect to the additional flanged teeth, in UFRGS-PV-0735-T and UFRGS-PV-1152 we only observed one element; UFRGS-PV-0735-T was eroded and UFRGS-PV-1152 was in occlusion with skull, so it was not possible to observe the teeth. Regarding additional small teeth, we observed between two and four of them in the most posterior portion of the maxilla. In MCN-PV-2852 eight hatchling teeth (these teeth are only observed in medial view) were found. In C. brasiliensis, between one and three additional teeth are generally found in the dentary and between five and nine hatchling teeth. However, the hatchling teeth tend to be enclosed by the secondary bone in the mature individuals, Table 1.
Other features observed in C. brasiliensis dentition include two types of wearing in the dentary teeth: (1) Triangular score marks on the lateral side of the teeth, produced by the orthal occlusion with the maxillary teeth. These score marks are reported in other Clevosaurus as well [20, 30, 35]. This was also previously observed in UFRGS-PV-06013-T by Bonaparte and Sues , and it was observed by us in UFRGS-PV-0972-T, and in other specimens (MCN-PV-2852, UFRGS-PV-0606-T; UFRGS-PV-0752-T; UFRGS-PV-0754-T; UFRGS-PV-0758-T), but they comprised smoother marks (Fig. 4); (2) Marks in the occlusal surface of the crown produced by the occlusion with the “groove for the external flange” of the palatine (pers. Obs. PRM) and palatine teeth (Fig. 5)  (pers obs. PRM), with the more worn teeth placed in the anterior portion, some of them covered by the secondary bone, forming the edentulate portion of the beak-shaped dentary [20. 35]. This pattern of wearing in the teeth of Clevosaurus is well known [20, 30, 35]. We suspect that these distinct patterns of wearing among the individual of our sample are reflecting different ontogentic stages, as already observed by Fraser  and Fraser and Walkden , and in Priosphenodon minimus by Apesteguía and Carballido , that could be statistical correlated with the results from PCA, however, other authors mentioned that when acrodont teeth are susceptible to worn, and hence the extant of wear is not thought to correlate simply age or skull size[17, 29].
A) MCN-PV-2852; B) UFRGS-PV-0606-T; C) UFRGS-PV-0752-T; D) UFRGS-PV-0754-T; E) UFRGS-PV-0758-T; F) UFRGS-PV-0613-T; G) UFRGS-PV-0972-T; H) Scheme of teeth wear along ontogeny. WF: Wear Facet; htl T: hatchling tooth; S Add T: Small additional teeth; SB: secondary bone. Scale bar 0.5cm.
Linear measurements of the different structures were taken with a calliper (Fig. 6 and Table 1). Among the specimens that have two lower jaws, we selected the best preserved, in order to reduce possible taphonomic biases. The images were obtained by two different steps: (1) all specimens were photographed in lateral, medial or both views, when it was possible, with a digital camera Canon EOS Rebel T3i using macro lens Ef 35–70mm. Only the specimen MCN-PV-2852 was photographed using a Canon EOS Rebel XSi camera with macro lens Ef 50mm. After this, mirror images were obtained of the right dentaries, in lateral view, and of the left dentaries, in medial view, so that images of all the specimens were oriented in the same direction; (2) since not all dentaries are complete, a search for missing data was made (which will be explained in the next section), using the previously linear measures taken. The estimated values were employed for the reconstruction of the incomplete dentaries (Table 2).
I) lower jaw length; II) dentary length; III) length from the most anterior and ventral point of the secondary bone to mandibular fossa; IV) anterior height of the dentary; V) anterior height of the secondary bone; VI) height of the dentary in the inflexion point; VII) length from the anterior portion of the dentary to anterior portion of the coronoid process; VIII) length from the anterior portion of the dentary to the posterior portion of the laster additional tooth. IX) length from the anterior portion of the last additional tooth to the anterior portion of the coronoid process; X) length of the base of the last additional tooth; XI) height of the last additional tooth; XII) length from the coronoid process; XIII) height of the coronoid process.
A survey for the taphonomic signatures was performed as well as the establishment of the degree of tooth wear, which is related with ontogeny in the group of Clevosaurs . These data were condensed into a Table 3.
After linear measurements were obtained, this information was used to estimate the missing values through the method “Bayesiam PCA missing value estimator” by the function “bpca” of the package “pcaMethods” [54, 55, 56] of the software R vs. 3.0.0 . This method was originally created by  for estimated missing data in array of DNA. It consists of three elementary processes: a principal component (PC) regression; Bayesian estimation and; expectation-maximization (EM) . It is a highly complex method requiring multiple iterative matrix inversions and attempts to replace the missing value based via PCA by regressing the remaining values against the principal components for complete values . It has been used for missing data estimation in morphometrics , and according to Brown et al. , this method is recommendable because introducing the least amount of error when handling with missing data. Then, the same procedure was done with the method “correlated variable regression” using the function “best.reg” of the package “Lost”  of the software R . This method used the variable most highly correlated with the variable experiencing missing data .
Such estimation was repeated three times for each method. It was given priority to the first method because it is the most accurate , only in the case when the estimated value were incoherent, we used the estimates value thrown out by the second method. With the help of new measures, we performed reconstructions of the broken specimens and their location of the missing landmarks.
For the geometric morphometric analyses 9 landmarks were employed (Fig. 7 and Table 4). These landmarks were not influenced by the available view of each dentary (lateral or medial). Zeldicht et al.  recommended the optimal number of landmarks to be a third of the n size of the sample, however, while reviewing the literature, it was observed that some authors do not agree with this proportion, such as Jones (n = 49 using 54 landmarks), Campione and Evans  (n = 17 and use 13 landmarks) and Meloro and Jones  (n = 39 and using 54 landmarks), Drake and Klingenberg  (n = 47 and using 64 landmarks).
The landmarks were placed directly on the combination of the photos plus the reconstructions (Fig. 8) using the software TpsDig2 . Using the package “Geomorph”  of the software R vs 3.0.0  it was performed as General Procrustes Analysis (using the function “gpagen”) for removing non-shape variation (position, size and orientation) and to align specimens to a common coordinate system [60, 63, 64]. A Principal Analysis Components (PCA) was performed in the function “plotTangentSpace” to identify the main components of shape variation within the sample Thin-plate spline transformations were generated to help demonstrate the difference between the centre of morphospace (the mean shape) and other areas of morphospace e.g. along principle component axes [63, 64]. The geometric morphometric analyses were repeated twice with the aim of decreasing the chances of errors.
The principal component scores were provided by “Geomorph” twice and were compared among them and then with: (1) the taphonomic signatures: for this comparison, we created new scores of taphonomic signatures (AF) using PCA, so that the values were independent of each other; and (2) the degree of wear of the teeth (WT), feature that was used as a proxy of age. These comparisons were made using a Pearson’s test correlation with the software PASW v.18 .
One regression analysis was performed with the Centroid Size as the independent variable, and PC1 as dependent variable, because PC1 against CS is often used in an ontogenetic sample measuring allometry, once the PC1 is usually the allometric shape component .
Finally, seven new analyses were performed but with some modifications in the size of “N” using the specimens with less “missing data”; and also without landmark 2 and 9. This was done to detect any possible effect of the incomplete specimens or of these two landmarks (S1 Text). Because these new results have not modified the original ones, only the results of the analyses whit 10 most complete dentary (whit less missing data: MCN-PV-2852; UFRGS-PV-0606-T; UFRGS-PV-0752-T; UFRGS-PV-0753-T; UFRGS-PV-0754-T; UFRGS-PV-0848-T; UFRGS-PV-0972-T; UFRGS-PV-1153-T; UFRGS-PV-0613-T; UFRGS-PV-0748-T) and 7 landmarks (without landmark 2 and 9), are discussed. However, the respective tables and figures of all additional analyses are presented as supporting material (S1–S15 Tables, S1 Text, and S1 Fig.).
In the PCA, the most explicative variation was displayed by PC1 (83.3%) and PC2 (10.0%), representing 93.3% of the total variation (Table 5).
It can be observed (axis X) in the resulting plot for the PC1 versus PC2 in the tangent space, together with the visualization of the deformation grids for shapes along PC1 (Fig. 9), that as the dentary grows, the absolute value of PC1 also grows, but in a negative way. In the deformation grids we can also see that the “G Shape” is located in positive value in the axis X, and the “R shape” is located in the negative value for the axis X, and the difference between these shapes are: as the PC1 value moves to the left (negative values) in the plot, the proportion (rate) of the total length of the dentary relative to the tooth row portion (RD), decreases. On the other hand, the correlation between PC1 and the teeth wear was r = −0.640, also a strong correlations. But the correlations between PC1 and the scores of taphonomic signatures are weak (Table 6).
With thin plate of the analysis whit ten specimens and without landmark 2 and landmark 9. Note: in green color are the specimens without tooth information.
In relation to PC2, as shown in the plot (Fig. 9), the trend is that dorsoventrally compressed skulls are located in the highest positive values for PC2 (axis Y); whereas the ones closer to zero of that axis are the laterally compressed skulls; and generally, below zero (higher and negatives in Y), isolated lower jaws appears, and when their value increases, the specimens are more incomplete (the patterns are inverted, like mirror image, in the others analysis see S1 Fig.). In contrast with the results of the Pearson’s test between PC1 versus the teeth wear and the taphonomic signatures (AF), the correlation of the PC2 versus teeth wear (WT) is very weak; but when PC2 is compared with the scores 1 for taphonomic signatures, the result is stronger, with a value in the first analysis of r = 0.876 (Table 6).
In the allometric analyses, in the first regression (CS versus PC1) the result was p = 0.012, and this is statistically signifcant.But with the second regression (CS versus PC2), the result was not significant (Fig. 10).
Formula (A+ BX): 0.659−0.369(CS); R = −0.753; R square 0.567; P = 0.012.
It was detected that our sample comprises only juvenile and adult specimens, with absence of earlier ontogenetic stages, from T to T4 of Robinson  (these ontogenetic stages range from few weeks to few months). This asseveration is given as: 1) in all juvenile specimens are present additional and hatchling teeth; hatchling teeth alone occur exclusively in the hatchling individuals , and 2); in all specimens it is present a diastema between the last additional tooth of the dentary and the coronoid process; usually, this diastema is absent in hatchling Sphenodon, but in mature specimens, it is produced one diastema between the last additional tooth and the coronoid process .
Thus, the specimen UFRGS-PV-0848-T is recognized as a juvenile, and it is confirmed that UFRGS-PV-0613-T and UFRGS-PV-0972-T are also juveniles as proposed by Bonaparte and Sues  and Arantes (unpublished data). The specimen UFRGS-PV-0974-T is recognized as a young adult, and the remaining specimens of the sample are recognized as mature forms.
Principal Component 1 (83.3%) is significantly associated with differences in tooth wear and describes differences in the proportions of the coronoid process of the tooth row length relative to jaw length. Principal Component 2 (10.0%) is significantly associated with differences in taphonomic distortions.
Morphological differences are consistent with changes that might occur during ontogeny. However, the ontogenetic changes are not the only cause for these differences, once other processes related to the structure of population, such as sexual dimorphism, phenotypic variations of individuals, and taphonomic biases could help to explain the size variations inside the sample.
In juveniles specimens the hatchling teeth of the dentary are exposed (e.g. Fig. 4F), and in general they are not worn in the occlusal surface (e.g Fig. 4: F, G, H). These teeth show alternation in size which is a characteristic of the rhynchocephalia . It is possible to see the decrease of their number from juveniles to mature specimens, which is related to the secondary bone growth, as observed in other Clevosaurus . Three or four little additional teeth are placed in the most posterior part of the maxilla (e.g. Fig. 3D; 9: F, G). Strong score marks are observed in the lateral surface of the adding teeth of the dentary (e.g. Fig. 4: F, G, H). (Table 1)
In the mature specimens we can see high wear in the occlusal surface of the crown of the dentary teeth, and the lateral score marks are poorly visible or even absent (e.g. Fig. 4: A, B, C, D, E, H). Probably they have been enclosed by the secondary bone.
These different patterns of wear in the teeth can be possibly correlated with the differences in the way of type food between adults and juveniles, as previously proposed for Clevosaurus by Fraser  and Fraser and Walkden , and also occurring in Sphenodon [2, 17,67, 68]. According to these authors, the juveniles were insectivores, whereas the mature specimens were omnivores, but Clevosaurs feeding habits are still in discussion from facultative herbivory to carnivory [12, 30, 35, 53]. In the particular case of C. brasilieinsis, it is supported the hypothesis that juveniles have been restricted to an insectivorous diet, as proposed for the juveniles of all Triassic rhynchocephalian, since the features of their dentition are well adapted for trapping small insects . It has been suggested that the adult Clevosaurs and other taxa with six distinct rows of palatal teeth also possessed strong rounded snouts and robust palatal bones, and the palatal teeth themselves in these taxa are stout and suggestive of a triturating surface . Added to this, we observed in C. brasiliensis an apparent tendency to develop a greater amount of secondary bone, and consequently a greater number of teeth hidden by it than other Clevosaurus. In order to check this observation, some additional measurements would have to be made but the available published data on Clevosaurus do not allow this. This greater amount of secondary bone produces a sharp cutting edge in the anterior region [2, 35], larger than in other known species of Clevosaurus.
As aforementioned, the mature individuals of C. brasiliensis do not exhibit strong score marks in the lateral teeth surface, as in other mature individuals of the same genus, as, for example, C. hudsoni (pers. obs. PRM). It is possible that mature individuals of C. brasiliensis had more interaction between additional teeth of the dentary with the “groove” of palatine and palatine teeth, than juveniles and even mature individuals of other Clevosaurus. These features are probably indicative of an omnivorous diet for C. brasiliensis, different from what was proposed for C. hudsoni, which would be mainly carnivorous according to Jones [2, 36] and Apesteguía (unpublished data).
In Sphenodon, “the dentary increases in length posteriorly, and may continue to do so after the implantation of the last member of the additional series, producing an diastema between the last additional tooth and the coronoid process of the dentary” , and the same was inferred for Cynosphenodon huizachalensis . In the thin plate (Fig. 9, and S1 Fig.), we see that the dentary is gracile in the “G shape” (juveniles), and that in the “R shape” (adults) the dentary is more robust. Also, the tooth row is relatively longer in juvenile compared to mature specimens (Fig. 11). Because in Clevosaurus there are no successional teeth, the only reference point are the last additional teeth, and it is not possible to compare the other points of growth in the anterior part with Sphenodon and Cynosphenodon huizachalensis.
A) Juvenile individual; B) Mature individual. The tricolor, bar represents the regions of dentary: Blue = the tooth row; Black = diastema; Green = the posterior region; j = the regions in juvenile individual, and m = regions in mature individual Scale bars 1 cm.
The rate of dentary growth observed along the thin plate (Figs. 9, 11 and S1 Fig.) is congruent with the result of the allometric analysis within the PC1 with Centroid size (CS), whose value of p (0.012) was significant (Fig. 10). It is noteworthy that, these methods (allometric analysis and Thin plate), provide complementary visualizations of the allometric and ontogenetic patterns . The result of the regression of value for R square is 0.567, which statistically means that 56.7% of the variations of PC1 can be explained by allometric process during ontogeny as observed in the Thin Plate. This would be reflected in the difference of the growth rate of the posterior portion of the dentary with respect to its anterior portion.
The number of teeth is associated with the skull shape in Rhynchocephalia . When a taxon has a smaller number of teeth, like Sphenodon and Clevosaurus, the tendency is to possess a skull more robust skull with greater room for the adductor muscles . Thus, fewer teeth mean a smaller surface area of initial contact with food items, and greater loading from bite forces [2, 12, 36]. In the ontogeny of Clevosaurus, the tendency is to reduce the number of teeth. We also observed another difference between dentaries of juveniles and mature specimens. The former exhibit gracile dentaries, whereas the latter robust ones, which could support the idea of differences in feeding habits among juveniles and mature specimens.
The evolution of Clevosaurs is characterized by a shortening of the snout, so that the nares become more upright, and elongation of the posterior jugal process . Therefore, it is possible that this trend is reflected in the ontogenetic development of Clevosaurus. Particularly, in C. brasiliensis, it was observed an allometric growth with respect to the anterior and posterior portion of the dentary, wherein the posterior portion has a higher growth. This allometric growth, together with the development of a more robust dentary, is probably accompanied by the allometric growth of the skull, similar to the trend exhibited by the group.
In the analysed sample there are specimens with the same shape, but that display different length (e.g. MCN-PV-2852, UFRGS-PV-0750-T, UFRGS-PV-0752-T, UFRGS-PV-0754-T, UFRGS-PV-0758-T, see S1 Fig.). This may be due to sexual dimorphism, as it is well known in Lepidosauria. In the case of lizards, sexual dimorphism is common with respect to the head size, and associated with bite force . These also occur in the majority of Squamata and in Sphenodon , in which the head is typically male-biased [68, 69, 70]. In both cases, this can be related with different food sources, and with male-male interactions, because a stronger bite force is advantageous in acquiring and maintaining territories, being determinant for fitness in male Sphenodon [68, 70].
In addition to these, the different degrees of teeth wear, besides related with the ontogenetic stage, also may be related with sexual dimorphism (stronger bite force in males) and therefore with the foraging habits. Finally, part of the wearing of the teeth may be product of taphonomic processes, as discussed in the following section.
As it has already been described, the PC2 represents 10.0% of the variation of the form, and this is influenced by taphonomic processes. The deformation produced by taphonomic biases can result in morphological interpretations that could lead to the equivocal creation of two or more species. In fact, there are studies using geometric morphometrics that have succeeded in elucidating deformation patterns caused by taphonomic process, and they have detected taxa in synonymy (i.e. Hedrick and Dodson ). In our monotypic sample, it is only possible to observe some diagenetic influence on the morphology, reflected in deformed skulls (see ). In the plot (Fig. 8) the trend is shown between the PC2 and the type of compressed skull, so that the dorsoventrally compressed skulls are located in the highest positives values, and the laterally compressed skulls are located closer to the zero. Unfortunately, the number of skulls in the sample is not enough to confirm the hypothesis that one of the processes that had a substantial influence on the shape variation was the fossil diagenesis.
Regarding biostratinomy, we find a stronger correlation within the taphonomic signatures and the PC2 especially for disarticulation (A) and degree of fragmentation (B1 and B2). These signatures were most influenced in the AF1, the score that had the higher correlation with the PC2 (Table 6). Finally, we do not rule out the possibility that the PC2 can be related to a lack of data, since the signatures are related to loss of structures and therefore this may reflect some degree of error due to the missing data and the reconstructions.
Concerning the temporal resolution, the time-averaging involved in the generation of fluvial and deltaic environments generally takes from 101 to 104 yrs [72, 73]. So it is very likely that the bed containing C. brasiliensis can encompass such temporal resolution, implying time enough for the succession of one or more populations, and consequently the development of the small morphological differences. Thus, it is important to remember that usually the most recent populations are better represented in number of individuals , this could have occurred in our sample. Unfortunately, given the nature of the rock where the fossils are included, a massive sandstone, it was not possible to obtain a finer stratigraphic control for a more accurately time-resolution. Finally, regarding the structure of the population, it is probable that the thanatocoenosis is bimodal, generated by selective death. However, there is a bias towards the mature individuals. This may be due to the fact that generally immature organisms are weaker and less ossified, and therefore, they possess smaller chances of preservation.
The morphological differences among the specimens of the sample can be assigned to characteristics that are often influenced by changes during the ontogenetic development, like tooth pattern and wear, dentary growth and secondary bone growth. In C. brasiliensis the lateral wear is stronger in the juvenile specimens, being even absent in some mature specimens. In contrast, the wear in the occlusal surface is stronger in the mature specimens and smooth or absent in the juveniles ones. The dentary in juveniles is slender, bearing more hatchling teeth and the dentary carrier portion of teeth with respect to its length is greater than in mature individuals. In the latter, the dentary is more robust and the hatchling teeth are enclosing by the secondary bone.
We believe that there are differences in the way of processing food between adults and juveniles, being juveniles restricted to an insectivorous diet [20,35]. We hold that mature specimens of C. brasiliensis have an omnivorous habit, and it is probable that they have consumed more plant items, compared with to other Clevosaurus. This is supported by the fact that C. brasiliensis has a greater development of the secondary bone, which creates a bigger edentulate portion of the beak-shaped dentary in comparison with other Clevosaurus, and that their teeth do not resemble the mammalian carnassials as in Clevosaurus hudsoni [2,12]. This edentuale portion is similar to the “beak” of Uromastix lizard and it is used for cropping plant [17, 35]. The habit of herbivoy was previously proposed for the rhynchocephalian Sigmala Sigmala, Pelecymala robusta, Sphenotitan as well as the facultative herbivory for Clevosaurus hudsoni, due to the presence of a “beak”[19, 20, 27, 28, 35]. Finally the group that was distinguished for having successflly explored herbivory was the Eilenodontines of America (Late Jurassic to Late Cretaceous) [25, and Apesteguía (unpublished data)].
In the case of rhynchocephalians, some important diagnostic features change during the development, such as the proportion of the orbit with respect to the skull size , besides the other teatures discussed above. The geometric morphometrics is an effective tool for showing these kinds of biases and for elucidating the shape variation as well as its causes. In the specific case of Clevosaurus brasiliensis, due to the absence of caniniform teeth, which could help to draw reference lines to detect the dentary growth (as in Sphenodon and Cynosphenodon huizachalensis), the geometric morphometry was an alternative tool that helped to understand the dentary growth pattern.
S1 Fig. Plot PCA with thin Plate of the all specimens.
Note in green color are the specimens without tooth information.
S1 Table. Importance of components (all specimens).
S2 Table. Pearson Correlation (all specimens).
S3 Table. Importance of components (without UFRGS-PV-0972-T).
S4 Table. Pearson Correlation (without UFRGS-PV-0972-T).
S9 Table. Importance of components (without LM 2 and LM 9).
S10 Table. Pearson Correlation (without LM 2 and LM 9).
S11 Table. Importance of components (without LM 2 and LM 9 and UFRGS-PV-0972-T).
S12 Table. Pearson Correlation (without LM 2 and LM 9 and UFRGS-PV-0972-T.
S13 Table. Importance of components (without LM 2 and LM 9 and N = 8).
S14 Table. Pearson Correlation (without LM 2 and LM 9 and N = 8).
We thank Luiz Flávio Lopes for the photographs; Dr. Ana Maria Ribeiro for the permission to access the MCN-FZB collection; Dr. Sidia M. Callegari-Jacques for statistical analysis; Dr. Anne Hsiou by valuable comments in the early versions of the manuscript. We also thank CAPES (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior) for financial support (PRM) and CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico) for MBS financial support. We also thank curators who send us photos for comparison: Dr. Martha Richter and Ms. Sandra Champman of the Natural History Museum of London; Dr. Roger M. H. Smith of the Iziko South African Museum, of South Africa and Dr. Zheng Fang of the Institute of Vertebrate Paleontology and Paleoanthropology of China.
Conceived and designed the experiments: PRM. Performed the experiments: PRM. Analyzed the data: PRM MBS. Contributed reagents/materials/analysis tools: PRM. Wrote the paper: PRM MBS.
- 1. Evans SE, Jones MEH (2010) The origin, early history and diversification of Lepidosauromorph reptiles. In Bandyopadhyay S ed. New Aspects of Mesozoic Biodiversity, Lecture Notes in Earth Sciences 123. Springer-Verlag Berlin Heidelberg: pp 27–43
- 2. Jones MEH (2008) Skull shape and feeding strategy in Sphenodon and other Rhynchocephalia (Diapsida: Lepidosauria). Journal of Morphology 269: 945–966. pmid:18512698
- 3. Uetz P (editor) The Reptile Database. Available: http://www.reptile-database. Accessed Nov 12, 2014.
- 4. Rauhut OWM, Heying AM, López-Arbarello A, Hecker A (2012) A new Rhynchocephalian from the Late Jurassic of Germany with a dentition that is unique amongst tetrapods. PLoS ONE 7:e46839. pmid:23118861
- 5. Adams DC, Rohlf FJ, Slice DE (2013) A field comes of age: geometric morphometrics in the 21st century. Hystrix, the Italian Journal of Mammalogy 24:7–14
- 6. Maxwell EE, Dececchi TA (2012) Ontogenetic and stratigraphic influence on observed phenotypic integration in the limb skeleton of a fossil tetrapod. Paleobiology 39: 123–134
- 7. Campione NE, Evans DC (2011) Cranial growth and variation in edmontosaurs (Dinosauria: Hadrosauridae): Implication for Latest Cretaceous megaherbivore diversity in North America. PLoS ONE 6:e25186. pmid:21969872
- 8. Hedrick BP, Dodson P (2013) Lujiatun psittacosaurids: Understanding individual and taphonomic variation using 3D geometric morphometircs. PloS ONE 8: e69265. pmid:23950887
- 9. Foth C, Rauhut OWN (2013) The good, the bad, and the ugly: The influence of skull reconstructions and intraspecific variability in studies of cranial morphometrics in theropods and basal saurischians. Plos ONE 8: e72007. pmid:23951279
- 10. Fariña RA, Vizcaino SF (2001) Carved teeth and strange jaws: How glyptodonts masticated. Acta Palaeontologica Polonica 46: 219–234
- 11. Meloro C (2012) Mandibular shape correlates of tooth fracture in extant Carnivora: implications to inferring feeding behaviour of Pleistoncene predators. Biological Journal of the Linnean Society 106: 70–80.
- 12. Meloro C, Jones MEH (2012) Tooth and cranial disparity in the fossil relatives of Sphenodon (Rhynchocephalia) dispute the persistent 'living fossil' label. Journal of Evolutionary Biology 25: 2194–2209 pmid:22905810
- 13. Humphries ED, Jones MEH (2010) Geographic variation in the jaws of Holocene Sphenodon (Lepidosauria: Rhynchocephalia) demonstrated by landmark analysis. Journal of Vertebrate Paleontology, SVP Program and Abstracts Book 2010: 108.
- 14. Dendy A (1899) Outlines of the development of the Tuatara, Sphenodon (Hatteria) punctatus. Quarterly Journal of Microscopical Sciences. 42: 1–72
- 15. Harrison HS (1901) Hatteria punctata its dentitions and its incubation period. Anatomizcher Anzeiger 20: 145–158
- 16. Howes GB, Swinnerton HH (1901) On the development of the skeleton of the Tuatara, Sphenodon punctatus; with remarks on the egg, hatching, and on the hatched young. Transactions of the Zoological Society of Lonton. 16 (Pt. 1):1–74
- 17. Robinson PL (1976) How Sphenodon and Uromastyx grow their teeth and use them. In Bellairs Ad'A, Cox CB ed. Morphololgy and biology of Reptiles. Academic Press. London. pp. 43–64
- 18. Hoffstetter R (1953) Caractères dentaires chez des Rhynchocéphales jurassiques, Theretairus et Leptosauruss. Bulletin du Muséum, 2e. Série, 25: 145–158
- 19. Fraser NC (1986) New Triassic sphenodontids from South-West England and a review of their classification. Palaeontology 29: 165–186
- 20. Fraser NC (1988) The osteology and relationships of Clevosaurus (Reptilia: Sphenodontia). Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences. 321: 125–178.
- 21. Renesto S (1955) A sphenodontid from the Norian (Late Triassic) of Lombardy (Northern Italy): a preliminary note. Modern Geology 20: 149–158
- 22. Reynoso VH (1997) A “beaded” sphenodontian (Diapsida: Lepidosauria) from the Early Cretaceous of Central Mexico. Journal of Vertebrate Paleontology 17: 52–59
- 23. Apesteguía S, Gómez RO, Rougier GW (2012) A basal sphenodontian (Lepidosauria) from the Jurassic of Patagonia: new insights on the phylogeny and biogeography of gondwanan rhynchocephalians. Zoological journal of the Linnean Society. 166: 342–360
- 24. Reynoso VH (2003) Growth patterns and ontogenetic variation of the teeth and jaws of the Middle Jurassic sphendontian Cynosphenodon huizachalensis (Reptilia: Rhynchocephalia) Can. J. Earth Sci. 40: 609–619.
- 25. Apesteguía S, Carballido JL (2014) A new Eilenodontine (Lepidosauria, Sphenodontidae) from the Lower Creaceous of Central Patagonia. Journal of Vertebrate Paleontology 34(2): 303–317.
- 26. Bonaparte JF, Sues HD (2006) A new species of Clevosaurus (Lepidosauria: Rhynchocephalia) from the Upper Triassic of Rio Grande Do Sul, Brazil. Paleontology 49: 917–923
- 27. Zerfass H, Lavina EL, Schultz CL, Vasconcelos-Garcia AJ, Faccini UF, et al. (2003) Sequence stratigraphy of continental Triassic strata of Southernmost Brazil: a contribution to southwestern Gondwana palaeogeography and palaeoclimate. Sedimentary Geology. 161: 85–105.
- 28. Martínez RN, Apaldetti C, Colombi CE, Praderio A, Fernandez E, et al. (2013) A new sphenodontian (Lepidosauria: Rhynchocephalia) from the Late Triassic of Argentina and the early orign of the herbivore opisthodontians. Proceedings of The Royal Society B 280: 20132057 http://dx.doi.org/10.1098/rspb.2013.2057] pmid:24132307
- 29. Jones MEH (2006) The Early Jurassic Clevosaurs from China (Diapsida:Lepidosauria). In Harris JD, Lucas SG, Spilmann JA, Lockley MG, Milner ARC, Kirkland JI ed. The Triassic-Jurassic Terrestrial Transition. New Mexico Museum of Natural History and Science Bulletin. 37: 548–562
- 30. Säilä LK (2005) A new species of the sphenodontian reptile Clevosaurus from the Lower Jurassic of South Wales. Palaeontology 48: 817–831
- 31. Wu XC (1999) Late Triassic-Early Jurassic sphenodontians from China and the phylogeny of the sphenodontia. In Fraser NC, Sues AD ed In the shadow of the dinosaurs, Early Meosozoic tetrapods.Cambridge University Press NY, USA: 38–69
- 32. Sues HD, Reisz RR (1995) First record of the Early Mesozoic sphenodontian Clevosaurus (Lepidosauria: Rhynchocephalia) from the southern hemisphere. Journal of Paleontology 69: 123–126
- 33. Reynoso VH, Cruz JA (2014) Mesozoic Lepidosauromorphs of Mexico: a review and discussion of taxonomic assignments. In Rivera-Sylva HE, Carpenter K,Frey E. ed. Dinosaurs and Other Reptiles from the Mesozoic of Mexico. 4–44
- 34. Reynoso VH (1996) A Middle Jurassic Sphenodon-like sphenodontian (Diapsida: Lepidosauria) from Huizachal canyon, Tamaulipas, México. Journal of Vertebrate Paleontology. 16:210–221.
- 35. Fraser NC, Walkeden GM (1983) The ecology of a Late Triassic reptile assemblage from Gloucestershire, England. Palaeogeography, Palaeoclimatology, Palaeoecology. 42: 341–365.
- 36. Jones MEH (2009) Dentary tooth shape in Sphenodon and its fossil Relatives (Diapsida:Lepidosauria: Rhynchocepahlia). In Koppe T, Meyer G, Alt KW ed. Comparative dental morphology, front Oral Biol. Basel, Karger. 13:9–15. pmid:20014504
- 37. Apesteguía S, Novas FE (2003) Large Cretaceous sphenodontian from Patagonia provides insight into lepidosaur evolution in Gondwana. Nature. 425: 609–612. pmid:14534584
- 38. Zerfass H, Chemale F Jr, Schultz CL, Lavina E (2004) Tectonics and sedimentation in Southern South America during Triassic. Sedimentology Geology. 166: 265–292
- 39. Costa da Silva R, Barboni R, Dutra T, Marques Godoy M, Barros Binotto R. (2012) Footpints of large theropod dionsaurs and implications on the age of Triassic biotas from Southerm Brazil. Journal of South American Earth Sciences. 39:16–23.
- 40. Arantes BA, Soares MB, Schultz CL (2009) Clevosaurus brasiliensis (Lepidosauria, Sphenodontia) do Triássico Superior do Rio Grande do Sul: Anatomia pós-craniana e relações filogenéticas. Revista Brasileira de paleontologia 12: 43–54
- 41. Bonaparte JF, Ferigolo J, Ribeiro AM (1999) A new Early Late Triassic saurischian dinosaur from Rio Grande do Sul State, Brazil. National Sciences Museum Monographs 15:89–109.
- 42. Bonaparte JF, Ferigolo J, Ribeiro AM (2001) A primitive Late Triassic “Ictidosaur” from Rio Grande do Sul, Brazil. Palaeontology 44: 623–635
- 43. Bonaparte JF, Martinelli AG, Schultz AG, Rubert R (2003) The sister group of mammals: small cynodonts from the Late Triassic of Southern Brazil. Revista Brasileira de Paleontologia 5: 5–7
- 44. Bonaparte JF, Martinelli AG, Schultz CL, Rubert R (2005) New information on Brasilodon and Brasilitherium (Cynodontia, Probainognathia) from the Late Triassic, Southern Brazil. Revista Brasileira de Paleontologia 8:25–46.
- 45. Bonaparte JF, Schultz CL, Soares MB, Martinelli AG (2010) La fauna local de Faxinal do Soturno, Triásico Tardío de Rio Grande do Sul, Brasil. Revista Brasileira de Paleontologia. 13(3): 233–246.
- 46. Martinelli AG, Bonaparte JF, Schultz CL, Rubert R (2005) A new tritheledontid (Therapsida, Eucynodontia) from the Late Triassic of Rio Grande do Sul (Brazil) and its phylogenetic relationships among carnivorous non-mammalian eucynodonts. Ameghiniana 42: 191–208
- 47. Cisneros JC, Schultz CL (2003) Soturnia caliodon n.g.n sp., a procolophonid reptile from the Upper Triassic of Southern Brazil. Neus Jahrbuch für Geologie und Paläontologie, Abhandlungen 227: 365–380
- 48. Bonaparte JF, Brea G, Schultz CL, Martinelli AG (2007) A new specimen of Guaibasaurus candelariensis (basal Saurischia) from the Late Triassic Caturrita Formation of Southern Brazil. Historical Biology, 19: 73–82
- 49. Soares MB, Schultz CL, Horn BLD (2011) New information on Riograndia guaibensis Bonaparte, Ferigolo & Ribeiro, 2001 (Eucynodontia, Tritheledontidae) from the Late Triassic of Southern Brazil: anatomical and biostratigraphic implications. Anais da Academia Brasileira de Ciências. 83: 329–354.
- 50. Soares MB, Dalla Vechia FM, Schultz CL, Kellner AW (2013) On the supposed pterosaurian nature of Faxinalipterus minima Bonaparte et al,.(2010) from the Upper Triassic of Rio Grande do Sul, Brazil. In: International Symposium on Pterosaurs, Short Comunications, rio de Janeiro, Universidade Federal do Rio de Janeiro, Museu Nacional, 2013. p. 95–98
- 51. Ferigolo J (2000) Esfenodontídeos do Neo-Triássico/?Jurássico do Estado do Rio Grande do Sul, Brasil. In: Holz M, De Ros LF ed. Paleontologia do Rio Grande do Sul: CIGO/UFRGS pp.236–245
- 52. Ferigolo J (2009) Esfenodontes: os répteis mais primitivos do mundo no Rio Grande do Sul. In: Da-Rosa AAS ed. Vertebrados Fósseis de Santa Maria e Região. Santa Maria, RS, Brazil: Pallotti pp. 89–105 https://doi.org/10.1371/journal.pone.0116716 pmid:25635840
- 53. Jones MEH (2006). Tooth diversity and function in the Rhynchocephalia (Diapsida:Lepidosauria). In Barrett PM, Evans SE ed. Ninth International Symphosium on Mesozoic Terrestrial Ecosystems and Biota, Abstracts and Proceedings.pp 55–58.
- 54. Oba S, Sato M, Takemasa I, Monden M, Matsubara K, et al. (2003) A Bayesian missing value estimation method for gene expression profile data. Bioinformatics 19: 2088–2096 pmid:14594714
- 55. Stacklies W, Redestig H, Scholz M, Walther D, Selbig J (2007) pcaMethods-a bioconductor package providing PCA methods for incomplete data. Bioinformatics 23: 1164–1167. pmid:17344241
- 56. Stacklies W, Henning R (2013) pca Methods. Available: http://www.bioconductor.org/packages/release/bioc/html/pcaMethods.html
- 57. R-Development-Core-Team (2013) R: A Language and Environment for Statistical Computing 3.0.0 ed. Vienna, Austria: R Foundation for Statistical Computing. Available: http://www.r-project.org/.
- 58. Brown CM, Arbour JH, Jackson DA (2012) Testing of the effect of missing data estimation and distribution in morphometric. Systematic Biology 61:941–954 pmid:22511121
- 59. Arbour J, Brown C (2013) LOST 1.0. Available: http://cran.r-project.org/web/packages/LOST/index.html.
- 60. Zeldicht ML, Swiderski DL, Sheets HD, Fink WL (2004) Geometric morphometrics for biologists: A primer. Elservier/Academic Press, Amsterdam. 443p.
- 61. Drake AG, Klingenberg CP (2008) The pace of morphological change: historical transformation of skull shape in St Bernard dogs. Proceedings of The Roayal Society B 275: 71–76 pmid:17956847
- 62. Rohlf FJ (2013) TpsDig. 2.17 ed. Stony Brook, N.Y: Published by the Author. Available: http://life.bio.sunysb.edu/morph/.
- 63. Adams DC, Otárola-Castillo E (2013) geomorph: an R package for the collection and analysis of geometric morphometric shape data. Methods in Ecology and Evolution 4: 393–399
- 64. Adams DC, Rohlf FJ, Sllice DE (2013) A field comes of age: geometric morphometrics in the 21st century. Hystrix, the Italian journal of Mammalogy. 24: 7–14
- 65. PASW v.18 Available: http://www.spss.com.hk/statistics/
- 66. Mitteroecker P, Gunz P, Bernhard M, Schaefer K, Bookstein FL (2004) Comparison of cranial ontogenetic trajectories among great apes and humans. Journal of Human Evolution 46:679–698 pmid:15183670
- 67. Jones MEH, Anderson CL, Hipsley CA, Müller , Evans SE, Schoch RR (2013) Intergration of molecules and new fossils supports a Triassic origin for Lepidosauria (lizards, snakes, and tuatara). BMC Evolutionary biology 13: 208 pmid:24063680
- 68. Jones MEH, Lappin AK (2009) Bite-force performance of the last rhynchocephalian (Lepidosauria: Sphenodon). Journal of the Royal Society of New Zealand 39: 71–83
- 69. Vincent SE, Herrel A (2007) Functional and ecological correlates of ecologically-based dimorphism in squamate reptiles. Integrative and Comparative Biology 47:172–188 pmid:21672829
- 70. Herrel A, Schaerlaeken V, Moravec J, Rosss CF (2009) Sexual shape dimorphism in Tuatara. Copeia 4:727–731
- 71. Holz M, Schultz CL (1998) Taphonomy of the south Brazilian Triassic herpetofauna: fossilization mode and implications for morphological studies. Lethaia 31:335–345
- 72. Behrensmeyer AK, Hook RW (1992) Paleoenvironmental contexts and taphonomic modes. In: Behrensmeyer AK, Damuth JD, DiMichele WA, Potts R, Sues HD, Wing SL, editors. Terrestrial Ecosystems through Time; Evolutionary Paleoecology of Terrestrial Plants and Animals. The University of Chicago Press pp. 15–136
- 73. Kidwell SM, Holland SM (2002) The quality of the fossil record: Implications for evolutionary analyses. Annual Review of Ecology and Systematics 33:561–588
- 74. Gradstein FM, Ogg JG, Schmitz MD, Ogg G. ed (2012) The Geologic Time Scale 2012. Boston, USA, Elsevier.