A new and unusual specimen of a probable azhdarchoid pterosaur is described for the Early Cretaceous (Albian) Romualdo Formation of Brazil. The specimen consists of a palate that, although fragmentary, has a unique morphology differing from all other known pterosaurs with preservation of palatal elements. The new specimen probably indicates the presence of a yet undescribed pterodactyloid taxon for Romualdo Formation and brings new information on pterosaur diversity of this sedimentary unity. Mainly due to the rarity of pterodactyloid specimens with palate preservation, this structure has been overlooked in this clade. Here, we reassess the palatal anatomy of Pterodactyloidea, revealing an intriguing variety of morphotypes and evolutionary trends, some of them described here for the first time. The morphological disparity displayed by different pterodactyloid taxa may be further evidence of the presence of diverse feeding strategies within the clade.
Citation: Pinheiro FL, Schultz CL (2012) An Unusual Pterosaur Specimen (Pterodactyloidea, ?Azhdarchoidea) from the Early Cretaceous Romualdo Formation of Brazil, and the Evolution of the Pterodactyloid Palate. PLoS ONE 7(11): e50088. doi:10.1371/journal.pone.0050088
Editor: Andrew A. Farke, Raymond M. Alf Museum of Paleontology, United States of America
Received: August 16, 2012; Accepted: October 16, 2012; Published: November 21, 2012
Copyright: © 2012 Pinheiro, Schultz. 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: Funding visit to AMNH was funded by a Collection Study Grant (Richard Gilder Graduate School - http://rggs.amnh.org/), whereas research at the Bayerische Staatssammlung für Paläontologie und Geologie was possible thanks to a Deutscher Akademischer Austauschdienst (http://www.daad.de/) scholarship. This work was also partially funded by the Conselho Nacional de Desenvolvimento Científico e Tecnológico (http://www.cnpq.br/) through scholarships granted to FLP and CLS. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. No additional external funding received for this study.
Competing interests: The authors have declared that no competing interests exist.
The fragile nature of pterosaur skeletons has had the effect of limiting superior preservation of their remains to isolated Lagerstätten throughout the world , . Even in these deposits, three-dimensional preservation rarely occurs. In most cases, pterosaur fossils are crushed, and important anatomical features are often obliterated. As a consequence, some details of pterosaur anatomy remain poorly known, which frequently leads to misinterpretations of structures. A good example of this is the pterosaur palate, because its study depends on either three-dimensionally preserved specimens or on exceptionally rare palatal views of compressed skulls. Principally because of this limitation, some bones and structures have been misidentified throughout the literature .
Only recently was a new interpretation of the pterosaur palate made , in a study that utilized the Extant Phylogenetic Bracket  to identify homologous structures in the palates of pterosaurs, birds and crocodiles. Although this new research, focusing on pterosaur palate anatomy, did indeed improve our understanding of this structure, it was focused primarily on non-pterodactyloids. Examining the palates of well-known pterodactyloid pterosaurs, in addition to those of still-unpublished specimens, led us to the conclusion that some anatomical features and evolutionary trends were not yet properly described for this clade. Therefore, a new examination of this subject is needed.
The Romualdo Formation of the Araripe Basin (Early Cretaceous of Northeastern Brazil) (Figure1) is probably the world’s most abundant source of three-dimensionally preserved pterosaur specimens, with some of the best pterodactyloid fossils with preservation of the palate, such as Anhanguera blittersdorfii , A. araripensis , Tapejara wellnhoferi , and Tropeognathus mesembrinus  having been found in its sediments. In fact, palatal features are often used in the diagnosis of pterosaur taxa from the Romualdo Formation, such as Thalassodromeus sethi , Tupuxuara leonardii , and Tropeognathus mesembrinus, among others. As will be discussed here, palatal morphology can be especially helpful in determining the taxonomy of azhdarchoid pterosaurs from this formation. The three-dimensionally preserved specimens from the Romualdo Formation, in addition to other specimens with exposed palates, can also assist in acquiring knowledge of the anatomy and evolution of this structure within the Pterodactyloidea.
Figure 1. Location map of the Araripe Basin, northeastern Brazil and simplified stratigraphic chart of the Santana Group.doi:10.1371/journal.pone.0050088.g001
We describe here a new and unusual pterodactyloid pterosaur specimen from the Romualdo Formation. The new material consists of a fragmentary palate and, although very incomplete, displays a combination of anatomical features so far never observed in other pterodactyloids. In addition, the anatomy and evolution of the pterodactyloid palate is reassessed, evidencing interesting morphologies and evolutionary trends within the clade.
Specimen MPSC R 859 was mechanically prepared by FLP at the Vertebrate Paleontology Laboratory of Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil. Most of the other specimens analyzed and described in this paper were first hand examined by FLP, whilst other data utilized for comparisons were obtained from the literature.
Institutional AbbreviationsAMNH: American Museum of Natural History, New York, New York, USA; BSP: Bayerische Staatssammlung für Paläontologie und Geologie, Munich, Germany; DGM: Museu de Ciências da Terra, Rio de Janeiro, Brazil; IMCF: Iwaki Coal and Fossil Museum, Iwaki, Japan; KUPV: Museum of Natural History, Univesity of Kansas, Lawrence, USA; MN: Museu Nacional, Rio de Janeiro, Brazil; MPSC: Museu de Paleontologia de Santana do Cariri, Santana do Cariri, Brazil; SMNK: Staatliches Museum für Naturkunde, Karlsruhe, Germany; TMM: Texas Memorial Museum, Austin, USA; UOSG/SÃO: Collection Oberli, St. Gallen, Switzerland; YPM: Peabody Museum of Natural History, New Haven, Connecticut, USA.
PTEROSAURIA Kaup 1834 .
PTERODACTYLOIDEA Plieninger 1901 .
Gen. et sp. indet.
A fragmentary palate, composed mainly by the maxillae, vomers and, probably, palatines (Figure 2). The specimen is housed at the Museu de Paleontologia de Santana do Cariri (Ceará, Brazil) under the collection number MPSC R 859.
Figure 2. Specimen MPSC R 859 in A, B, ventral; C, D, dorsal and E, left lateral views.
In F, the inferred position of the palatal fragment is demonstrated in a hypothetical azhdarchoid skull. Scale bar: 100 mm. ch, choanae; m, maxilla; p, palatine; sof, suborbital fenestra; v, vomers.doi:10.1371/journal.pone.0050088.g002
Locality and horizon.
The specimen comes from a calcareous concretion typical of the Romualdo Formation of Araripe Basin. Nevertheless, the exact locality is unknown. The Romualdo Formation, one of the formations that compose the Santana Group, crops out throughout the Araripe Plateau, close to the boundaries of Ceará, Pernambuco and Piauí States, Northeastern Brazil (Figure 1) and is usually dated as Albian. For further information on Romualdo Formation geology, paleoecology and age, see Mabesoone and Tinoco , Assine ,  and Martill .
MPSC R 859 is a fragmentary pterosaur palate, consisting of a portion of the maxillae (primarily in the form of the palatal maxillary plates), the vomers and, most likely, the palatines. The specimen has 153 mm of preserved length and 46 mm of maximum width. The straight, unbroken dorsal margins suggest that the preserved portion of the maxillae were situated under anteroposteriorly extended nasoantorbital fenestrae. Although fragmentary, the specimen is very well preserved, presenting no signs of compression. The ventral surface was exposed on the outside of the calcareous concretion and is considerably weathered. The dorsal surface was only partially prepared, because the bone becomes very thin (less than 0.5 mm thick) and fragile at the posterior half of the specimen. The choanae and the suborbital fenestrae are partially preserved. The specimen is broken approximately 97 mm from the anterior margins of the choanae, whereas posteriorly, the specimen ends 23 mm from the anterior margins of the suborbital fenestrae. Perhaps due to the weathering, the suture lines are not distinguishable in the ventral view. However, in the dorsal view, a clear medial suture separates the maxillary palatal plates.
The holotype consists almost entirely of the two maxillae. These bones are medially fused, and a clear suture line can be visualized in the dorsal view, forming a very discrete ridge. Although ventrally, there are shallow grooves between the palatal plates of the maxillae and their lateral walls, the maxillae are continuous in their dorsal aspect, with no sign of division (such as grooves or sutures) between their two distinct components. From the dorsal view, the maxillae are concave, with the palatal plates curving gently into the lateral rims. The ventral grooves between the palatal plates and the lateral walls of the maxillae, which are often visible in pterodactyloid palates, have been interpreted by most authors as being the sutures between the maxillae and the palatines (see  for a revision).
The palatal maxillary plates form a flat ventral surface, with no sign of palatal ridges. The palate is very slightly depressed medially, marking the place where the two maxillae fuse, although no clear sign of a suture is visible. The maxillary palatal plates are relatively thick anteriorly and gradually reduce in thickness to an exceptionally thin bony sheet in a region close to the anterior margins of the choanae. The maxillae border the choanae anterolaterally and, likely, the suborbital fenestrae anteriorly. There is no discernible suture between the maxillae and palatines.
The lateral walls of the maxillae are very slender and shallow (108 mm in height). Although subparallel posteriorly, their lateral margins begin to converge, in dorsal view, at a region close to the rostral ending of the choanae. Throughout the entire specimen, the maxillary walls constrict dorsally into very thin bony blades, which border the nasoantorbital openings ventrally. The maxillae maintain their dorsoventral height over the complete length of the specimen, with no evidence of dorsal expansion, indicating that the entire preserved portion of MPSC R 859 was located under nasoantorbital fenestrae of large proportions. Ventrally, the maxillary walls display neither teeth nor empty alveoli. The sutures between the maxillae and the vomers and between the maxillae and the palatines are not visible.
The fused vomers form a slim triangular element that partially divides the choanae anteriorly. There is no sign of sutures between the two elements or between these and the maxillae. The vomers are most likely incomplete. Although elongated vomers completely dividing the choanae and contacting the medial processes of the pterygoids are visible in exceptionally well-preserved pterodactyloid specimens (e.g., Anhanguera araripensis), in most cases the fragility of these bones prevents complete preservation.
Although these bones cannot be individualized, they likely comprise the slender components that form the margins of the suborbital fenestrae medially and the choanae laterally.
Comparison and Taxonomic Assignment
As described above, the dorsal margins of the maxillae of MPSC R 859 are intact and remain straight, lacking any ascendant curvature along their entire preserved length. This indicates that all of the preserved elements were situated under nasoantorbital fenestrae of large proportions, extending well anterior of the rostral borders of the choanae.
The preserved maxillae of MPSC R 859 are edentulous throughout their length. Some toothed pterosaurs, such as members of the clades Archaeopterodactyloidea  (e.g., Gnathosaurus , Feilongus , Cycnorhamphus , Ctenochasma  and Moganopterus ) and Istiodactylidae , have teeth restricted to the anterior portion of the skull, rostral to the nasoantorbital fenestrae. However, the extension of the nasoantorbital openings in MPSC R 859 is incompatible with the clade Archaeopterodactyloidea. In the new specimen, as described above, the nasoantorbital openings extend substantially further from the anterior end of the choanae, suggesting a very large size for these fenestrae. This condition differs from the relatively short nasoantorbital openings observed in archaeopterodactyloid pterosaurs. Moganopterus zhuiana, referred to Boreopteridae by  shares some similarities with MPSC R 859. However, the two-dimensionally preserved holotype of the former prevents detailed comparisons. Although istiodactylids have exceptionally large nasoantorbital fenestrae, in these pterosaurs (at least in Istiodactylus latidens, the only one with three-dimensionally preserved cranial elements), the tip of the rostrum is remarkably blunt, but the maxillae converge in a higher angle than what is observed in MPSC R 859. Also, the skull of I. latidens is more robust, differing from the slender condition observed in the specimen we describe (see , ). Additionally, the posterior palatal anatomy of these pterosaurs remains unknown, and the extension of the choanae with respect to the nasoantorbital fenestrae cannot be determined. The palatal maxillary plates of I. latidens, although mainly planar, are slightly raised in a region close to the sagittal plane (Mark Witton, personal communication, 2012), also differing from the condition displayed by MPSC R 859.
While lacking elements comparable to MPSC R 859, the recently-described Unwindia trigonus , also from the Romualdo Formation, has teeth restricted to the rostral end of the skull, well anteriorly from the rostral margin of the nasoantorbital fenestrae. The incomplete nature of Unwindia’s holotype avoids an accurate determination of the nasoantorbital opening’s size for this taxon. Nevertheless, based on the general construction of its skull, it’s unlikely that Unwindia had nasoantorbital fenestrae comparable in size with what is inferred for MPSC R 859.
Because of the reasons cited above, the morphology of MPSC R 859 is more compatible with a few edentulous pterosaur taxa, so that it remains probable that the new specimen was completely toothless. Although the possibility that MPSC R 859 had teeth cannot be totally excluded, the combination of exceptionally large nasoantorbital openings, slender, anteriorly convergent maxillae and teeth restricted to the anterior end of the rostrum has never been observed in any known pterosaur taxon. Because of the probable absence of teeth in MPSC R 859, we’ll focus further comparisons of this specimen with edentulous pterosaurs (Nyctosaurus , Pteranodontidae, Tapejaridae and Azhdarchidae). However, it is worth noting that, considering the fragmentary nature of the new specimen, it is possible that more complete material of poorly known tooth-bearing taxa (such as Unwindia and Moganopterus) will, eventually, display similarities with MPSC R 859.
The palatal anatomy of Nyctosaurus and the pteranodontids can be reconstructed based on the few specimens preserved, at least partially, in a palatal view –. Both Nyctosaurus and Pteranodon  have comparatively short nasoantorbital fenestrae (with respect to the length of the choanae), with a very different configuration than that found in MPSC R 859, and can therefore be eliminated from the discussion. It is noteworthy that a nominal species of Nyctosaurus (N. lamegoi ) was proposed for the Late Cretaceous Gramame Formation of Northeastern Brazil. Nevertheless, the holotype – and only specimen known thus far – consists of a single fragmentary humerus, and its attribution to the genus can be regarded as tentative .
Toothless pterosaurs with proportionately large nasoantorbital openings are thus far restricted to the Azhdarchoidea (Azhdarchidae, Tapejaridae and Chaoyangopteridae sensu Lü et al. , but see ). Although azhdarchid pterosaurs once had a world-wide distribution, their remains are, in most cases, restricted to fragmentary postcranial bones . Fairly complete skulls are known only for Zhejiangopterus linhaiensis  and Quetzalcoatlus . Also, an incomplete skull (TMM 42489-2) from the Maastrichtian Javelina Formation (United States), sometimes attributed to Tapejaridae, may also be referred to this clade (Mark Witton, personal communication, 2012). As is common in pterosaur preservation, known Z. linhaiensis skulls are laterally compressed  and information regarding their palatal morphology is unavailable. However, this pterosaur had very large nasoantorbital fenestrae, and it is possible that the rostral margin of this opening was situated at a considerable distance from the anterior margins of the choanae. Nevertheless, direct comparisons between this species and MPSC R 859 cannot be made until more information regarding the two taxa is available.
Although badly crushed, specimens attributed to Quetzalcoatlus sp. with partial preservation of palatal bones were described by Kellner and Langston . Similar to Zhejiangopterus linhaiensis, Quetzalcoatlus also presents large nasoantorbital fenestrae. According to Kellner and Langston , the choanae are incomplete in all of the specimens. However, in the best- preserved one (TMM 41961-1), these openings occupy approximately 20% of the inferred skull length. Although this measurement is an approximation, the length of the choanae with respect to the size of the nasoantorbital fenestrae in Quetzalcoatlus (after Kellner and Langston , one-third of the total skull length) seems to be incompatible with the condition observed in MPSC R 859. Nevertheless, the fragmentary nature of the latter avoids more accurate comparisons. Additionally, the palatal plates of the maxillae in Quetzalcoatlus (described as palatines by ) are flattened anteriorly, and they gradually become convex posteriorly. This contrasts with the flat maxillary plates of MPSC R 859.
The morphology of MPSC R 859 compares more favorably with that observed in members of Tapejaridae sensu Pinheiro et al. 2011  (i.e., Tapejaridae sensu Kellner and Campos, 2007  and Chaoyangopteridae sensu Lü et al. 2008 ). Although tapejarinid tapejarids, such as Tapejara, Tupandactylus  and Sinopterus , are characterized by “short-faced” skulls (at least when compared with thalassodrominid tapejarids or azhdarchids), all tapejarids have exceptionally long nasoantorbital fenestrae and lack teeth. In contrast with azhdarchids, all unambiguous species in the family Tapejaridae described thus far preserve cranial material. Additionally, Romualdo Formation Tapejaridae (Thalassodromeus, Tupuxuara and Tapejara) are known from three-dimensionally preserved specimens, from which palatal morphology can be assessed. Indeed, palatal characters are often used in the diagnosis of nominal tapejarid species from this formation.
Tapejarid pterosaurs have thus far been confidently recorded for the Crato and Romualdo Formations (Aptian/Albian) of the Araripe Basin (Northeastern Brazil), the Jiufotang Formation (Aptian) of Liaoning province (Northeastern China) and the La Huérguina Formation (Barremian) of Las Hoyas, Spain , , . In addition, some fragmentary specimens from the Kem Kem beds (Cenomanian) of Morocco  and the Javelina Formation (Maastrichtian) of the United States ,  may also be attributable to the Tapejaridae (but see above), indicating a worldwide distribution of this taxon during the Cretaceous.
The monophyly of Tapejaridae is still debated, with some authors supporting it , , , , –, while others regard the taxon as paraphyletic with respect to Azhdarchidae , , , . Although further discussions regarding the phylogeny of this taxon are beyond the scope of the present paper, a monophyletic Tapejaridae is supported herein (see  for a recent discussion of this issue).
The Chaoyangopteridae sensu Lü et al. 2008  are a group of edentulous pterosaurs from the Yixian and Jiufotang Formations (Early Cretaceous of China). Lacusovagus magnificens , from the Crato Formation of the Araripe Basin, is also tentatively referred in this taxon. The taxonomic position of the Chaoyangopteridae sensu Lü et al. 2008  is uncertain, largely due to the scarcity of information and the brief descriptions of existing specimens. Although some authors regard this taxon as closely related to the Azhdarchidae , , a recent phylogenetic analysis reclassified the group as a clade within the Tapejaridae and renamed it as Chaoyangopterinae . Nevertheless, due to the scarcity of data, both positions are still disputable. In any case, all chaoyangopterinids with preserved cranial elements evidence large nasoantorbital fenestrae, in a condition similar to what is observed in other tapejarids and azhdarchoids. Unfortunately, further comparisons between MPSC R 859 and chaoyangopterinids are impossible due to the lack of preserved palatal elements in the chaoyangopterinid specimens thus far described.
Brazilian tapejarids are represented by Tupandactylus (T. imperator and T. navigans ) from the Crato Formation (?Aptian) and Tupuxuara (T. longicristatus , T. leonardii and T. deliradamus ), Tapejara wellnhoferi and Thalassodromeus sethi from the younger Romualdo Formation (Albian) of the Araripe Basin. As is usual in Crato Formation fossils, the specimens referred to T. imperator are laterally compressed, with no information whatsoever on palatal anatomy. The same can be stated for a number of recently described tapejarinid tapejarids from the Jiufotang Formation (northeastern China), such as Sinopterus and “Huaxiapterus” .
Although not mentioned in the original description of the species , the two specimens thus far attributed to the Brazilian tapejarinid taxon Tupandactylus navigans (SMNK PAL 2344 and SMNK PAL 2343) do have preserved palatal elements. Although these materials are also laterally compressed, the manner in which the bones are preserved suggests that the palate of T. navigans was convex in the region where the palatal openings are located.
In spite of the fact that the palatal anatomy of Tupandactylus remains poorly known, this genus is closely related to Tapejara wellnhoferi, whose palate can be assessed. Tapejara wellnhoferi is the best known tapejarid from the Romualdo Formation, with several specimens having been formally described , –. Most of these specimens consist of skulls with palatal components.
As observed in the holotype (MN 6595 V) and in the specimens AMNH 2440 (Figure 3A) and UOSG 12891 that were referenced, the anteriormost region of the palatal surface of T. wellnhoferi bears a shallow concavity. In this region, the premaxillomaxilla is inclined downwards at an angle of approximately 25° with respect to the posterior ventral border of the maxillae , . Following this depression, where the maxillae become abruptly horizontal, specimens AMNH 2440 and UOSG 12891 show a pronounced convexity (Figure 3A). In the holotype, this region is poorly preserved: the extremely thin bone layer collapsed, creating an artificially flat surface (FLP, personal observation). However, when the palate is intact, the choanae of T. wellnhoferi, closely followed by the narrow suborbital fenestrae, are located in a strong convexity, and the suborbital fenestrae can easily be observed in lateral aspect. The palatal surface of T. wellnhoferi bears a well-developed medial foramen, identified as a probable foramen incisivum by Ösi et al. . Additionally, between the foramen incisivum and the choanae, the holotype (MN 6595 V) has two foramina that may correspond to the aperturae maxillo-premaxillaris. If this identification is correct, then the contact between the maxillae and the premaxillae of T. wellnhoferi is located in this region . The palatal anatomy of T. wellnhoferi contrasts sharply with that observed in MPSC R 859, since the palatal surface of the latter is flat throughout its entire preserved length. Although it is possible that a convexity or concavity develops anteriorly (in the missing area of the palate), the region bearing the palatal openings is remarkably planar. Both the choanae and the suborbital fenestrae are ventrally oriented and cannot be properly seen unless in ventral aspect. Other differences between MPSC R 859 and the monospecific genus Tapejara include the great distance between the anterior borders of the choanae and the suborbital fenestrae (larger than the maximum width of the choanae), as well as the absence of foramina on the palatal surface in the new specimen.
Figure 3. Tapejarid specimens with preservation of the palate.
A, Tapejara wellnhoferi (AMNH 24440) showing an anterior concavity (white arrow) and a strong convexity (dark arrow) on the palatal surface; B, Tupuxuara longicristatus (holotype – MN 6591 V) showing the strongly convex palate that is typical of the genus. Scale bars: 30 mm in A and 20 mm in B.doi:10.1371/journal.pone.0050088.g003
The tapejarinid Europejara olcadesorum , recently described for the Lower Cretaceous of Spain, preserves some palatal elements. Nevertheless, the holotype is badly crushed and the original three-dimensional shape of the bones cannot be assessed, limiting comparisons with MPSC R 859.
The genus Tupuxuara, thus far composed of three nominal species (T. longicristatus, T. leonardii and T. deliradamus), is characterized by a strongly convex palate. As can be observed in the holotypes of T. longicristatus and T. leonardii, a median keel (which is much more developed in T. leonardii) originates at the anterior part of the rostrum and broadens posteriorly, where the palate becomes increasingly convex ,  (Figure 3, B). The specimen IMCF 1052, illustrated in lateral aspect by Veldmeijer  and Witton , demonstrates that the palate of T. leonardii remains convex throughout its entire length, with the suborbital and subtemporal fenestrae being easily distinguishable in lateral view. This is also the condition described by Witton  for T. deliradamus. Although the morphology of the palate where the palatal openings are located is still unknown for T. longicristatus, the holotype (MN 6591 V) shows strongly convex maxillary palatal plates below the nasoantorbital fenestrae (Figure 3, B), making it likely that the condition in T. longicristatus was similar to that observed in other Tupuxuara species. Specimen MPSC R 859, therefore, differs from the genus Tupuxuara in having a flat palate, with no evidence of palatal ridges or convexities.
Specimen MPSC R 859 also differs from Thalassodromeus, the other taxon of azhdarchoid pterosaur from the Romualdo Formation. Thus far, this genus is composed of a single species, T. sethi, represented by the holotype (DGM 1476 R), an almost complete skull , and a fragmentary mandibular symphysis, which was referred (SAO 251093) . Thalassodromeus is unique for its singular palatal configuration. The anteriormost portion of the premaxillomaxilla is convex, forming a sharp blade. Posteriorly, below the nasoantorbital fenestrae, the maxillary palatal plates become abruptly concave, with well-developed rims , , and the palate remains concave throughout its posterior length (FLP, personal observation). The absence of ventral maxillary rims in MPSC R 859 is sufficient for distinguishing the new specimen from Thalassodromeus.
As demonstrated above, the three genera of Romualdo Formation tapejarids can easily be distinguished from one another by their singular palatal morphologies. This strongly indicates that different feeding strategies were employed by closely related tapejarid taxa. Additionally, palatal anatomy can be a reliable source of information for the diagnosis of genera and nominal species in this clade.
As discussed, MPSC R 859 has a unique palatal morphology, different from all other pterosaurs with preserved palatal bones. Although the new specimen lacks unambiguous diagnostic characters of any known pterodactyloid clade, we tentatively attribute it to Azhdarchoidea because it probably lacks teeth and has nasoantorbital fenestrae of unusually large proportions (albeit neither of these two features are unique to the clade, this combination has thus far only been observed in azhdarchoids). A more accurate attribution of MPSC R 859 to any clade within the Azhdarchoidea is more challenging. The rarity of azhdarchid skulls, combined with the laterally compressed preservation of the majority of them, prevents a reliable reconstruction of azhdarchid palatal morphology, the same being true for chaoyangopterinids and Chinese tapejarinids. Nevertheless, as discussed above, the general construction of the Quetzalcoatlus skull and the inferred proportions of the nasoantorbital fenestrae with respect to the length of the choanae in this taxon is different from that found in MPSC R 859. In this respect, the new specimen is more compatible with the Tapejaridae, especially with the long-snouted thalassodrominid morphotypes.
Due to the scarcity of information concerning azhdarchid cranial morphology, it seems unwise to regard the condition observed in Quetzalcoatlus as the standard for Azhdarchidae, and MPSC R 859 also cannot be excluded from this clade with certainty. Nevertheless, taking into account the relative abundance of tapejarid pterosaurs in the Romualdo Formation, the absence of azhdarchids in this sedimentary unit thus far, and the general morphology of MPSC R 859 (more similar to what is currently observed in thalassodrominid tapejarids), it is also likely that the new specimen was a tapejarid.
Albeit, as discussed, MPSC R 859 has a unique palatal configuration, the new specimen is fragmentary to the extent that avoids the recognition of unambiguous diagnostic features. MPSC R 859, however, may indicate the presence of a yet undescribed azhdarchoid taxon in Romualdo Formation.
Evolution of the Pterodactyloid Palate
As mentioned above, the study of the pterosaur palate depends upon the rare specimens in which this structure is preserved, either three-dimensionally or as an uncommon palatal view of a crushed skull. Furthermore, the high degree of bone fusion, commonly observed in pterosaur skulls, can make the delimitation of palatal elements difficult . Although the absolute number of known pterosaur specimens has increased substantially during the last few decades (mainly due to the discovery of previously unknown pterosaur-bearing strata, such as the Romualdo and Crato formations in Brazil and the Jiufotang and Yixian formations in China), the relative number of skulls with preserved palates is still small. Among Pterodactyloidea, informative palatal preservation was reported or illustrated for the genera Anhanguera, Ctenochasma, Dsungaripterus , Europejara, Gnathosaurus, Quetzalcoatlus, Nyctosaurus, Pteranodon, Pterodactylus , Tapejara, Thalassodromeus, Tropeognathus and Tupuxuara –, –, , , , , –. See also the revision provided by Ösi et al. .
A major reinterpretation of pterosaur palatal anatomy was made by Ösi et al.  in a study that recognized crucial misinterpretations of bones and structures that were often repeated throughout the literature. The best example is the identification by most authors as “palatines” of what turned out to be palatal plates of the maxillae. The conclusions of Ösi et al.  are supported by topological correspondence, within an evolutionary framework provided by the Extant Phylogenetic Bracket . Our observations of pterosaur specimens with preserved palates are, thus far, in agreement with the new interpretations, and the model of Ösi et al.  is herein supported.
Although the evolution of the pterosaur palate, culminating in the condition observed in generalized pterodactyloids, is discussed by Ösi et al. , this study focused primarily on non-pterodactyloid pterosaurs, especially Dorygnathus . Our reexamination of previously described specimens, combined with published data and some as yet unpublished material, revealed that the palatal anatomy within the Pterodactyloidea is complex and cannot be generalized by a single model. Additionally, the evolution of this structure within the group shows interesting patterns, which will be discussed here.
Four major evolutionary trends were identified by Ösi et al.  for the palate of pterosaurs: 1) an enlargement of the choanae, following the elongation of the rostrum and the shortening of the medial processes of the pterygoids; 2) a decrease in the size of the interpterygoid vacuity; 3) an enlargement of the rostral processes of the pterygoids relative to the length of the medial processes of the same bones; and 4) a loss of the lateral processes of the pterygoids, which, in basal pterosaurs, divide the subtemporal fenestrae in two, creating the paired pterygo-ectopterygoid fenestrae. An increase in the size of the subtemporal fenestrae through their confluence with the pterygo-ectopterygoid openings would be a consequence of a more developed adductor musculature, in response to the larger jaws of pterodactyloids .
The assumption made by Ösi et al.  that their model for the palatal evolution of the Pterodactyloidea is valid for all known taxa with palatal preservation, however, proved to be false. Confluent subtemporal and pterygo-ectopterygoid fenestrae are, indeed, observable in some forms. Nevertheless, as will be demonstrated below, lateral processes of the pterygoids are secondarily developed in some taxa, while the ectopterygoids are reduced to vestigial elements in others. We describe, below, the palatal anatomy of some representative pterodactyloid pterosaurs, demonstrating the morphological diversity within the group.
The palate of Pterodactylus-like pterosaurs is, generally, inferred based on BSP 1936 I 50, a specimen attributed to “Pterodactylus” micronyx  (Figure 4A, B). The reconstruction provided by Wellnhofer  follows the earlier conceptualization of the pterosaur palate, with the choanae limited anteriorly by the palatines and the maxillae restricted to the dental margin. According to this author, the ectopterygoid is a well-developed element, with an anteriorly directed process that borders the choanae laterally and the “postpalatine-fenestra” medially. However, a reassessment of the specimen, under the new anatomical paradigm, showed that what was interpreted by Wellnhofer  as an anterior process of the ectopterygoid is, probably, the palatine. Both the choanae and the interpterygoid vacuities are relatively large and appear to be confluent, i.e., the medial processes of the pterygoids do not contact each other. Nevertheless, the skull is distorted and the pterygoids are not in their natural positions, impeding an accurate estimation of the lengths of the openings. The unfused nature of the pterygoids could also be related to the ontogenetic stage of the specimen (Attila Ösi, personal communication, 2012). Interestingly, the pterygoids of “Pterodactylus” micronyx show laterally directed processes that do not reach the jugals, as if this taxon were transitional between a typical non-pterodactyloid palate (as seen in Dorygnathus and Rhamphorhynchus ) (Figure 5A) and the pterodactyloid model proposed by Ösi et al.  (Figure 5, B). Although the Gnathosaurus palate was observed only as a cast (BSP 1964 I 94), this pterosaur, closely related to Pterodactylus, demonstrates a similar morphology in what appears to be the primitive condition for Pterodactyloidea.
Figure 4. Photographs and reconstructions of representative specimens, showing palatal morphological variation among pterodactyloids.
A and B, Pterodactylus micronyx (BSP 1936 I 50); C and D, Anhanguera araripensis (BSP 1982 I 89); E and F, Pteranodon; G and H, Tupuxuara (IMCF 1052). E, modified from ; G, photo by André Veldmeijer, courtesy of the Iwaki Coal and Fossil Museum, Japan. Scale bars: 5 mm in A and 50 mm in C, E and G.doi:10.1371/journal.pone.0050088.g004
Figure 5. Palatal reconstructions of representative pterosaurs.
A, non-pterodactyloid condition; B, primitive pterodactyloid morphology; 1, Pterosauria; 2, Pterodactyloidea; 3, Archaeopterodactyloidea; 4, Ornithocheiroidea; 5, Pteranodontoidea; 6, Tapejaroidea. A, B and Dorygnathus palate: redrawn from Ösi et al. ; Pteranodon: redrawn from Bennett . Not to scale. The phylogenetic relationships follow the topology proposed by Kellner  and, more recently, Wang et al. .doi:10.1371/journal.pone.0050088.g005
Palate preservation in Pteranodon is rather rare, but reconstructions were made based on specimens such as KUPV 976, 2212 and YPM 1177 ,  (Figure 4E, F). This pterosaur shows a peculiar variation of the primitive pterodactyloid palate: the exceptionally well-developed ectopterygoids laterally contacted the maxillae, dorsally crossed the rostral processes of the pterygoids and contacted the fused medial processes of the latter, close to the sagittal plane of the skull. Provided that the reconstructions of Eaton  and Bennett  are accurate, there are no laterally directed processes on the pterygoids and the subtemporal openings are large. In contrast, the suborbital fenestrae are almost vestigial, constricted between the palatal plates of the maxillae anteriorly and the diagonally oriented ectopterygoids posteriorly.
An even more singular condition is observed in the Anhangueridae . This taxon shares a common ancestor with Pteranodon at the base of the clade Pteranodontoidea  and is well represented by several specimens with superb palatal preservation, such as the holotypes of Anhanguera blittersdorfii (MN 4805 V), Anhanguera araripensis (BSP 1982 I 89) (Figure 4C, D) and Tropeognathus mesembrinus (BSP 1987 I 46). Nevertheless, the high level of bone fusion and the obliteration of the sutures in these specimens make the interpretation of the bony elements exceptionally difficult. Thus far, all anhanguerids with good palatal preservation demonstrate a small paired bony element contacting the median processes of the pterygoids laterally. These bones were ignored when A. blittersdorffi  and T. mesembrinus  were first described but were later identified as ectopterygoids in the original description of A. araripensis by Wellnhofer . Actually, this author identifies two very distinct elements as ectopterygoids: the small bones laterally fused to the median processes of the pterygoids and the bony bridges that divide the subtemporal fenestrae from what Wellnhofer  called “fenestrae postpalatinalis”. Specimen comparisons revealed that a contact between the ectopterygoids and the median processes of the pterygoids is present in at least two other taxa of derived pterodactyloids – Pteranodon and Tupuxuara. The topological correspondence led us to conclude that the small elements described here for Anhanguera and Tropeognathus are vestigial ectopterygoids, partially agreeing with the identification by Wellnhofer . Further corroboration of this hypothesis lies in the fact that, in A. araripensis, the distal extremities of these elements seem to lie on the dorsal surface of the rostral processes of the pterygoids, in the way that would be expected if the bridge-like ectopterygoids of Pteranodon, which dorsally surpass the rostral processes of the pterygoids, were reduced to their proximal ends. One implication of this interpretation is that the bony division between the lateral palatal openings is, in fact, a secondarily developed lateral process of the pterygoid (contra Wellnhofer ), and the opening identified by Wellnhofer  as the “fenestra postpalatinalis” is, in fact, a confluence between two distinct openings, topologically analogous to the suborbital and pterygo-ectopterygoid fenestrae of non-pterodactyloids (for practical reasons, we propose that this opening continues to be labeled as the suborbital fenestra in the Anhangueridae). In addition, the pterygoids are preserved in the dorsal aspect in the holotype of Anhanguera santanae , showing a continuity between the main portions of these bones and their lateral processes. Therefore, in the Anhangueridae, the “pterodactyloid model” of two paired sets of lateral fenestrae is maintained, although this is acquired by a “reversion” to a primitive condition – the presence of lateral processes on the pterygoids.
As discussed, palatal anatomy can be of special importance in the taxonomy of azhdarchoid pterosaurs, notably the tapejarids. However, in these pterosaurs, palatal characters with taxonomic relevance are thus far restricted to the region anterior to the choanae, especially with respect to the presence or absence of palatal ridges and the general morphology of the maxillary palatal plates. Few azhdarchoid specimens possess complete palates, and the posterior region of this structure is poorly known in this lineage. Nevertheless, in specimens such as IMCF 1052, attributed to Tupuxuara leonardii (illustrated by Veldmeijer  and Witton ), the palatal morphology can be fully assessed. IMCF 1052 displays three pairs of lateral palatal fenestrae, in a pattern that closely resembles the non-pterodactyloid condition (Figure 4G, H). As in Pteranodon, the ectopterygoids are exceptionally well-developed and contact the fused medial processes of the pterygoids, crossing the rostral rami of the pterygoids dorsally. Additionally, lateral processes are present on the pterygoids, dividing the subtemporal fenestrae and creating secondary pterygo-ectopterygoid fenestrae.
Although it would be unwise to state that the highly specialized morphology observed in Tupuxuara is the standard for the Azhdarchoidea, this genus shares some similarities with what was described by Kellner and Langston  for Quetzalcoatlus, indicating that, for some aspects, a certain level of conservativeness should be expected. As in Tupuxuara, Quetzalcoatlus shows lateral processes on the pterygoids that, as observed by Kellner and Langston , probably divided the subtemporal fenestrae. In the same way, according to the authors, the ectopterygoid of Quetzalcoatlus, although incomplete on the specimen studied, extends diagonally above the pterygoid. Thus, it is likely that the conditions in this azhdarchid and in Tupuxuara were similar. Despite the fact that the palate of Thalassodromeus sethi was only superficially described by Kellner and Campos , this pterosaur also had three pairs of lateral palatal fenestrae, although the ectopterygoids are much broader and it is unlikely that they reached the median processes of the pterygoids (FLP, personal observation). The condition in Tapejara is currently unknown.
Piscivory is generally assumed to have been the feeding habit for most pterosaurs , , and it is indeed likely that a large number of known taxa preyed on fishes. As a matter of fact, most of the taxa studied directly herein are thought to be, at least partially, piscivorous , , , , . However, studies of pterosaur feeding strategies are scarce, and conclusions are often based on superficial anatomical observations rather than on comprehensive studies of functional morphology. It is also important to observe that our knowledge of pterosaurs is remarkably biased by a “Lagerstätten effect”: preservation of their remains often depends upon special environmental conditions, and our understanding of pterosaur diversity is probably strongly influenced by a concentration of informative specimens in a few deposits , , .
The study of palatal anatomy, as well as other aspects of the feeding apparatus, is of great relevance for a better understanding of pterosaur feeding habits. The diversity of palatal morphologies described, for the first time, herein may suggest that pterodactyloids displayed complex and diversified feeding strategies, in a way analogous with was already proposed for non-pterodactyloid stem-groups . The anatomical disparity between supposedly piscivorous forms, demonstrated here by several different palatal morphologies, could be evidence that piscivory emerged secondarily in a number of lineages. Nevertheless, more data are needed to test this hypothesis, mainly because, as highlighted above, most inferences about pterosaur feeding strategies are based on superficial anatomical characters. Regardless, it is likely that pterodactyloid feeding habits were much more diverse than the fossil record has led us to believe.
Fragmentary remains can, sometimes, provide relevant information about fossil taxa. MPSC R 859, although very incomplete, has a unique morphology and increases our knowledge of the Romualdo Formation pterosaur fauna. The new specimen makes it clear that palatal features can be of great relevance in diagnosing azhdarchoid pterosaurs and that the variation is probably related to the development of a diversity of feeding habits among the members of this clade.
Mainly because palatal anatomy is difficult to assess in most pterosaur specimens, this region has been overlooked, with very few palatal characters being used in pterosaur phylogenetic analyses. However, as demonstrated here, the Pterodactyloidea show considerable variation in palatal morphotypes, and this diversity seems to be congruent with the proposed phylogenetic relationships of the clade. Thus, with more information available, palatal anatomy can be significant for achieving a better resolution of pterosaur phylogeny. In the same way, a better understanding of pterodactyloid palatal anatomy is of crucial relevance in accessing the feeding habits and, in a broader sense, the ecology of this clade.
The authors would like to thank the following people for granting access to specimens under their care and, also, for kind help during visits: Álamo Saraiva and Joäo Kerensky (MPSC); Oliver Rauhut and Markus Moser (BSP); Mark Norell and Carl Mehling (AMNH); Alexander Kellner and Helder Silva (MN); Eberhard Frey (SMNK), Rainer Schoch (SMNS), Lorna Steel (BMNH) and Matt Riley (CAMSM). Thanks are given to André Veldmeijer, Yukimitsu Tomida and Marcel Lacerda, who significantly helped the research here presented. This paper was improved to a great extent after the revisions provided by Attila Ösi and Mark Witton, who were also kindly helpful during discussions about the theme here addressed. Also, the authors are indebted with Mr. Satoru Nabana for allowing publication of the Iwaki Tupuxuara specimen photographs. Richard Butler and Marcos A. F. Sales made relevant suggestions after critical reading of an early version of the manuscript.
Conceived and designed the experiments: FLP. Analyzed the data: FLP CLS. Wrote the paper: FLP.
- 1. Wellnhofer P (1991) The illustrated encyclopedia of pterosaurs. London: Salamander Books Ltd 1991: 192.
- 2. Buffetaut E (1995) The importance of “Lagerstätten” for our understanding of the evolutionary history of certain groups of organisms: the case of pterosaurs. In: II International Symposium on Lithographic Limestones. Extended Abstracts. Ediciones de la Universidade Autónoma de Madrid. 49–52.
- 3. Ösi A, Prondvai E, Frey E, Pohl B (2010) New interpretation of the palate of pterosaurs. The Anat Rec 293: 243–258.
- 4. Witmer LM (1995) The extant phylogenetic bracket and the importance of reconstructing soft tissues in Fossils. In: Thomason JJ ed. Functional morphology in vertebrate paleontology. New York: Cambridge University Press. 19–33.
- 5. Campos DA, Kellner AWA (1985) Panorama of the flying reptiles study in Brazil and South America. An Acad Bras Cienc 57: 454–466.
- 6. Wellnhofer P (1985) Neue Pterosaurier aus der Santana-Formation (Apt) der Chapada do Araripe, Brasilien. Palaeontographica Abt. A. 187: 105–182.
- 7. Kellner AWA (1989) A new edentate pterosaur of the Lower Cretaceous from the Araripe Basin, northeastern Brazil. An Acad Bras Cienc 61: 439–445.
- 8. Wellnhofer P (1987) New crested pterosaurs from the Lower Cretaceous of Brazil. Mitt. Bayer Staatsslg Paläont hist Geol 27: 175–186.
- 9. Kellner AWA, Campos DA (2002) The function of the cranial crest and jaws of a unique pterosaur from the Early Cretaceous of Brazil. Science 297: 389–392.
- 10. Kellner AWA, Campos DA (1994) A new species of Tupuxuara (Pterosauria, Tapejaridae) from the Early Cretaceous of Brazil. An Acad Bras Cienc 66: 467–473.
- 11. Kaup JJ (1834) Versuch einer Eintheilung der Saugethiere in 6 Stämme und der Amphibien in 6 Ordnungen. Isis 3: 311–315.
- 12. Plieninger F (1901) Beiträge zur Kenntnis der Flugsaurier. Paläontogr 48: 65–90.
- 13. Nessov LA (1984) Upper Cretaceous pterosaurs and birds from Central Asia. Paleontol Zh 1: 47–57.
- 14. Unwin DM (2003) On the phylogeny and evolutionary history of pterosaurs. In: Buffetaut E, Mazin J-;M, eds. Evolution and paleobiology of pterosaurs. London: Geological Society. 139–199.
- 15. Mabesoone JM, Tinoco IM (1973) Paleoecology of the Aptian Santana Formation (northeastern Brazil). Palaeogeogr Palaeoclimatol Palaeoecol 14: 97–118.
- 16. Assine ML (1992) Análise estratigráfica da bacia do Araripe, Nordeste do Brasil. Revista Brasileira de Geociências 22: 289–300.
- 17. Assine ML (2007) Bacia do Araripe. Boletim de Geociências da Petrobrás 15: 371–389.
- 18. Martill DM (2007) The age of the Cretaceous Santana Formation fossil Konservat Lagerstätte of north-east Brazil: a historical review and an appraisal of the biochronostratigraphic utility of its paleobiota. Cretaceous Res 28: 895–920.
- 19. Kellner AWA (1996) Description of new material of Tapejaridae and Anhangueridae (Pterosauria, Pterodactyloidea) and discussion of pterosaur phylogeny. Unpublished doctoral thesis. 347 p.
- 20. Meyer HV (1834) Gnathosaurus subulatus, ein Saurus aus dem lithographischen Schiefer vom Solnhofen. Museum Senckenbergianum 1: 3–7.
- 21. Wang X, Kellner AWA, Zhou Z, Campos DA (2005) Pterosaur diversity and faunal turnover in Cretaceous terrestrial ecosystems in China. Nature 437: 875–879.
- 22. Quenstedt FA (1853) Über Pterodactylus suevicus im lithographischen Schiefer Württembergs. Tübingen. 52 p.
- 23. Meyer HV (1852) Ctenochasma romeri. Palaeontographica 2: 82–84.
- 24. Lü J, Pu H, Xu L, Wu Y, Wei X (2012) Largest toothed pterosaur skull from the Early Cretaceous Yixian Formation of Western Liaoning, China, with comments on the family Boreopteridae. Acta Geol Sin 86: 287–293.
- 25. Howse SCB, Milner AR, Martill DM (2001) Pterosaurs. In: Martill DM, Naish D eds. Dinosaurs of the Isle of Wight, Palaeontological Association, Field Guide to Fossils 10. Pp 324–335.
- 26. Witton MP (2012) New insights into the skull of Istiodactylus latidens (Ornithocheiroidea, Pterodactyloidea). PLOS ONE 7(3): e33170 DOI: 10.1371/journal.pone.0033170.
- 27. Martill DM (2011) A new pterodactyloid pterosaur from the Santana Formation (Cretaceous) of Brazil. Cretaceous Res 32: 236–243.
- 28. Marsh OC (1876) Principal characters of American pterodactyls. Amer J Sci Arts 12: 479.
- 29. Williston SW (1902) On the skull of Nyctodactylus, an Upper Cretaceous pterodactyl. J Geol 10: 520–531.
- 30. Wellnhofer P (1978) Pterosauria. Stuttgart: Gustav Fisher. 82 p.
- 31. Bennett SC (2001) The osteology and functional morphology of the Late Cretaceous pterosaur Pteranodon. Palaeontographica Abt. A 260: 1–153.
- 32. Price LI (1953) A presence de Pterosauria no Cretáceo Superior do estado da Paraíba. Notas Preliminares e Estudos. 71: 1–10.
- 33. Lü J, Unwin DM, Xu L (2008) A new azhdarchoid pterosaur from the Lower Cretaceous of China and its implications for pterosaur phylogeny and evolution. Naturwissenschaften 95: 891–897.
- 34. Pinheiro FL, Fortier DC, Schultz CL, De Andrade JAFG, Bantim RAM (2011) New information on the pterosaur Tupandactylus imperator, with comments on the relationships of Tapejaridae. Acta Pal Polon 56: 567–580.
- 35. Averianov AO (2010) The osteology of Azhdarcho lancicollis Nessov, 1984 (Pterosauria, Azhdarchidae) from the Late Cretaceous of Uzbekistan. Proceedings of the Zoological Institute RAS 314: 264–317.
- 36. Cai Z, Wei F (1994) Zhejiangopterus linhaiensis (Pterosauria) from the Upper Cretaceous of Linhai, Zhejiang, China. Vertebrata Palasiatica 32: 181–194.
- 37. Lawson DA (1975) Could pterosaurs fly? Science 188: 676–677.
- 38. Unwin DM, Lü J (1997) On Zhejiangopterus and the relationships of pterodactyloid pterosaurs. Hist Biol 12: 199–210.
- 39. Kellner AWA, Langston W (1996) Cranial remains of Quetzalcoatlus (Pterosauria, Azhdarchidae) from Late Cretaceous sediments of Big Bend National park, Texas. J Vert Pal 16: 222–231.
- 40. Kellner AWA, Campos DA (2007) Short note on the ingroup relationships of the Tapejaridae (Pterosauria, Pterodactyloidea) Boletim do Museu Nacional. 75: 1–14.
- 41. Campos DA, Kellner AWA (1997) Short note on the first occurrence of Tapejaridae in the Crato Member (Aptian), Santana Formation, Araripe Basin, Northeast Brazil. An Acad Bras Cienc 69: 83–87.
- 42. Wang X, Zhou Z (2003) A new pterosaur (Pterodactyloidea, Tapejaridae) from the Early Cretaceous Jiufotanf Formation of western Liaoning, China and its implications for biostratigraphy. Chi Sci Bull 48: 16–23.
- 43. Vullo R, Marugán-Lobón J, Kellner AWA, Buscalioni AD, Gomez B, et al. (2012) A new crested pterosaur from the Early Cretaceous of Spain: the first european tapejarid (Pterodactyloidea: Azhdarchoidea). PLOS ONE 7(7): e38900 doi:10.1371/journal.pone.0038900.
- 44. Wellnhofer P, Buffetaut E (1999) Pterosaur remains from the Cretaceous of Morocco. Paläontologische Zeitschrift 73: 133–142.
- 45. Kellner AWA (2004) New information on the Tapejaridae (Pterosauria, Pterodactyloidea) and discussion of the relationships of this clade. Ameghiniana 41: 521–534.
- 46. Kellner AWA (2003) Pterosaur phylogeny and comments on the evolutionary history of the group. In: Buffetaut E, Mazin J-;M, eds. Evolution and paleobiology of pterosaurs. London: Geological Society. 105–137.
- 47. Wang X, Kellner AWA, Zhou Z, Campos DA (2008) Discovery of a rare arboreal forest-dwelling flyng reptile (Pterosauria, Pterodactyloidea) from China. Proc Natl Acad Sci U S A 105: 1983–1987.
- 48. Wang X, Kellner AWA, Jiang S, Meng X (2009) An unusual long-tailed pterosaur with elongated neck from western Liaoning of China. An Acad Bras Cienc 81: 793–812.
- 49. Andres B, Ji Q (2008) A new pterosaur from the Liaoning province of China, the philogeny of the Pterodactyloidea, and convergence in their cervical vertebrae. Palaeontology 51: 453–469.
- 50. Martill DM, Naish D (2006) Cranial crest development in the azhdarchoid pterosaur Tupuxuara, with a review of the genus and tapejarid monophyly. Palaeontology 49: 925–941.
- 51. Lü J, Jin X, Unwin DM, Zhao L, Azuma Y, Ji Q (2006) A new species of Huaxiapterus (Pterosauria: Pterodactyloidea) from the Lower Cretaceous of Western Liaoning, China with comments on the Systematics of Tapejarid Pterosaurs. Acta Geol Sin 80: 315–326.
- 52. Witton MP (2008) A new azhdarchoid pterosaur from the Crato Formation (Lower Cretaceous, Aptian?) of Brazil. Palaeontology 51: 1289–1300.
- 53. Frey E, Martill DM, Buchy M-C (2003) A new species of tapejarid pterosaur with soft-tissue head crest. In: Buffetaut E, Mazin J-;M, eds. Evolution and paleobiology of pterosaurs. London: Geological Society. 65–72.
- 54. Kellner AWA, Campos DA (1988) Sobre um novo pterossauro com crista sagittal da Bacia do Araripe, Cretáceo Inferior do Nordeste do Brasil. An Acad Bras. Cienc 60: 459–469.
- 55. Witton MP (2009) A new species of Tupuxuara (Thalassodromidae, Azhdarchoidea) from the Lower Cretaceous Santana Formation of Brazil, with a note on the nomenclature of Thalassodromidae. Cretaceous Res 30: 1293–1300.
- 56. Lü J, Yuan C (2005) New tapejarid pterosaur from western Liaoning, China. Acta Geol Sin 79: 453–458.
- 57. Wellnhofer P, Kellner AWA (1991) The skull of Tapejara wellnhoferi Kellner (Reptilia, Pterosauria) from the Lower Cretaceous Santana Formation of the Araripe Basin, northeastern Brazil. Mitt. Bayer. Staatsslg Paläont hist Geol 31: 89–106.
- 58. Elgin and Campos (2011) A new specimen of the azhdarchoid pterosaur Tapejara wellnhoferi Hist Biol DOI: 10.1080/08912963.2011.613467.
- 59. Eck C, Elgin RA, Frey E (2011) On the osteology of Tapejara wellnhoferi Kellner 1989 and the first occurrence of a multiple specimen assemblage from the Santana Formation, Araripe Basin, NE-Brazil. Swiss J Palaeontol 130: 277–296.
- 60. Veldmeijer A (2006) Toothed pterosaurs from the Santana Formation (Cretaceous; Aptian-Albian) of northeastern Brazil. A reappraisal on the basis of newly described material. Unpublished doctoral thesis. 269 p.
- 61. Veldmeijer AJ, Signore M, Meijer HJM (2005) Description of two pterosaur (Pterodactyloidea) mandibles from the Lower Cretaceous Santana Formation, Brazil. Deinsea 11: 67–86.
- 62. Young CC (1964) On a new pterosaurian from Sinkiang, China. Vertebr Palasiatica 8: 221–256.
- 63. Cuvier G (1809) Mémoire sur le squelette fossile d’un Reptil Volant des environs d’Aichstedt, que quelques naturalists ont pris pour un oiseau, et donc nous formons un genre de Sauriens, sous le nom Ptero-Dactyle. Ann Mus Hist natur 13, 1–424.
- 64. Wellnhofer P (1970) Die Pterodactyloidea (Pterosauria) der Oberjura-Plattenkalke Süddeutschlands. Abh Bayer Akad Wiss, mat-naturwiss Kl, NF 141: 1–133.
- 65. Young CC (1973) Wuerho pterosaurs. Special Publication of the Institute of Vertebrate Paleontology and Paleoanthropology Academia Sinica 11: 18–34.
- 66. Veldmeijer AJ (2003) Description of Coloborhynchus spielbergi sp. Nov. (Pterodactyloidea) from the Albian (Lower Cretaceous) of Brazil. Scripta Geologica 125: 35–139.
- 67. Bennett SC (2007) A review of the pterosaur Ctenochasma: taxonomy and ontogeny. Neues Jahrb Geol Palaont Abh 245: 23–31.
- 68. Theodori C (1830) Knochen vom Pterodactylus aus der Liasformation von Banz. Frorieps Notizen für Natur- und Heilkunde 632: 1–101.
- 69. Meyer HV (1856) Letter on various fossil vertebrates. Neues Jahrbuch für Mineralogier, Geognomie Geologie Petrefakt 826.
- 70. Meyer HV (1847) Homoeosaurus maximiliani und Rhamphorhynchus (Pterodactylus) longicaudus, zwei fossile Reptilien aus dem Kalkschiefer von Solenhofen. Frankfurt (Schmerber). 22 p.
- 71. Eaton GF (1910) Osteology of Pteranodon. Mem. Connecticut Acad Arts Sci 2: 1–38.
- 72. Wang X, Kellner AWA, Jiang S, Cheng X (2012) New toothed flying reptile from Asia: close similarities between early Cretaceous pterosaur faunas from China and Brazil. Naturwissenschaften DOI 10.1007/s00114-012-0889-1.
- 73. Unwin DM (2005) The pterosaurs: from deep time. New York: Pi Press. 352 p.
- 74. Kellner AWA, Tomida Y (2001) Description of a new species of Anhangueridae (Pterodactyloidea) with comments on the pterosaur fauna from the Santana Formation (Aptian-Albian), northeastern Brazil. National Science Museum Monographs, Tokyo. 135 p.
- 75. Buffetaut E, Clarke JB, Le Loeuff J (1996) A terminal Cretaceous pterosaur from the Corbières (southern France) and the problem of pterosaur extinction. Bull Soc Geol France 167 753–759.
- 76. Butler RJ, Barrett PM, Nowbath S, Upchurch P (2009) Estimating the effects of sampling biases on pterosaur diversity patterns: implications for hypotheses of bird/pterosaur competitive replacement. Paleobiology 35: 432–446.
- 77. Ösi A (2011) Feeding-related characters in basal pterosaurs : implications for jaw mechanism, dental funcion and diet. Lethaia 44 146–152.