Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Braincase of Siamraptor suwati and insights into the cranial anatomy of Carcharodontosauria

  • Soki Hattori ,

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Project administration, Visualization, Writing – original draft, Writing – review & editing

    s-hattori@g.fpu.ac.jp (SH); duangsuda.c@nrru.ac.th (DC)

    Affiliations Institute of Dinosaur Research, Fukui Prefectural University, Eiheiji, Fukui, Japan, Fukui Prefectural Dinosaur Museum, Katsuyama, Fukui, Japan

  • Duangsuda Chokchaloemwong ,

    Roles Conceptualization, Data curation, Funding acquisition, Investigation, Resources, Visualization, Writing – original draft, Writing – review & editing

    s-hattori@g.fpu.ac.jp (SH); duangsuda.c@nrru.ac.th (DC)

    Affiliations School of Science Education, Faculty of Science and Technology, Nakhon Ratchasima Rajabhat University, Nakhon Ratchasima, Thailand, Northeastern Research Institute of Petrified Wood and Mineral Resources, Nakhon Ratchasima Rajabhat University, Nakhon Ratchasima, Thailand

  • Soichiro Kawabe,

    Roles Data curation, Investigation, Visualization, Writing – review & editing

    Affiliations Institute of Dinosaur Research, Fukui Prefectural University, Eiheiji, Fukui, Japan, Fukui Prefectural Dinosaur Museum, Katsuyama, Fukui, Japan

  • Elena Cuesta,

    Roles Writing – review & editing

    Affiliation SNSB-Bavarian State Collection for Paleontology and Geology, Munich, Germany

  • Masateru Shibata,

    Roles Resources, Supervision, Visualization, Writing – review & editing

    Affiliations Institute of Dinosaur Research, Fukui Prefectural University, Eiheiji, Fukui, Japan, Fukui Prefectural Dinosaur Museum, Katsuyama, Fukui, Japan

  • Kazunori Miyata,

    Roles Investigation, Writing – review & editing

    Affiliation Fukui Prefectural Dinosaur Museum, Katsuyama, Fukui, Japan

  • Yoichi Azuma

    Roles Supervision, Writing – review & editing

    Affiliations Institute of Dinosaur Research, Fukui Prefectural University, Eiheiji, Fukui, Japan, Fukui Prefectural Dinosaur Museum, Katsuyama, Fukui, Japan

Abstract

Two partial theropod braincases, namely NRRU-F01020035 and F01020036, were recovered from the Lower Cretaceous Khok Kruat Formation in Nakhon Ratchasima, Thailand. Detailed observations on these two specimens revealed their carcharodontosaurian affinity based on the presence of several synapomorphies for the clade, such as partially-roofed anteromedial corner of the supratemporal fossa and a tall nuchal crest. In combination with the monospecific occurrence of Carcharodontosauria in the locality, both specimens can be attributed to Siamraptor suwati, although such braincase preserving the synapomorphic characters has been unknown in this taxon so far. In addition, the wedge-shaped frontoparietal suture and two deep pits on the lateral margin of the frontal shared in the two braincases, which have not been reported in any allosauroid, are shared by both specimens and thus can be regarded as additional autapomorphies of Siamraptor. The phylogenetic analysis incorporating information from the braincases confirmed the position of Siamraptor as an earliest-branching member of Carcharodontosauria as discussed in the original description, and further comparison with other allosauroid braincases revealed several aspects of the evolutionary history of carcharodontosaurian characters.

Introduction

Since 2007, the Fukui Prefectural Dinosaur Museum (FPDM) and the Northeastern Research Institute of Petrified Wood and Mineral Resources, Nakhon Ratchasima Rajabhat University (NRRU) have conducted joint paleontological research in Thailand through the Japan-Thailand Dinosaur Project (JTDP) [15]. By 2013, 22 specimens of large-bodied theropods had been recovered from an outcrop of the Lower Cretaceous Khok Kruat Formation at Ban Saphan Hin (Ban meaning “village”), the Suranaree Subdistrict, northwest of the Mueang Nakhon Ratchasima District, Nakhon Ratchasima Province (Fig 1), and were subsequently identified as the earliest-diverging carcharodontosaur Siamraptor suwati [5]. Carcharodontosauria is a clade comprising large-bodied theropods flourished as apex predators in all land masses in the Late Jurassic to the Late Cretaceous period, except for Antarctica [6]. However, Asian carcharodontosaurs had been poorly known, especially on the early members of the clade. Therefore, the presence of Siamraptor bears an important role to understand the carcharodontosaurian evolutionary history.

thumbnail
Fig 1. Locality map and stratigraphy of Khorat Group.

A, map of Nakhon Ratchasima Province, Thailand; B, distribution map of the Khok Kruat Formation in Nakhon Ratchasima Province; C, enlarged locality map of Suranaree and Khok Kruat subdistricts with the subdistrict boundaries (broken lines); D, photograph of the excavation site; E, stratigraphic column of the Khorat Group (modified from [7]). Maps in A–C are adapted from Chokchaloemwong et al. [5] under CC BY 4.0. Red stars in A–C indicate the locality described in this study and the grey lines in C indicate national highways.

https://doi.org/10.1371/journal.pone.0345155.g001

Prior to 2007, a large theropod braincase, NRRU-F01020035, was recovered from an outcrop of the Khok Kruat Formation in Nakhon Ratchasima Province by Prasart Anankasa and his team. The precise locality of this specimen has long been uncertain for a long time due to a lack of accompanying field documentation. Recently, however, it was confirmed that NRRU-F01020035 had been collected between 2003 and 2005 at the Ban Saphan Hin locality, based on a collation of a diary by Pratueng Jintasakul, the museum’s director at the time, and a conversation with Prasart Anankasa.

Following the description of Siamraptor in 2019, a partial skull roof of a large theropod (NRRU-F01020036) collected in 2007 from the same locality was recognized in the collection. This specimen was recovered within a limited excavation area of approximately 300 m², together with several elements previously referred to Siamraptor, including right and left maxillae (NRRU-F1020004, F1020005), a middle caudal vertebra (NRRU-F1020017), and a manual ungual (NRRU-F1020018).

At the time of the original description of Siamraptor, only 22 skeletal remains referable to Allosauroidea had been recovered from the bone-bearing bed at the locality, and all were consistently identified as Siamraptor. For example, all of four partial mandibles recovered from this locality were referred to this taxon, despite substantial variation in their sizes [5]. The elements comprising NRRU-F01020035 overlap and share anatomical features with NRRU-F01020036, as well as other braincases of allosauroids. NRRU-F01020035 and F01020036 represent the only additional skeletal remains referable to Allosauroidea currently known and are provisionally referred to Siamraptor, pending for further confirmation.

Among allosauroids, almost complete braincases are known in both the basal allosauroids such as Yangchuanosaurus [8,9], Sinraptor [1012] and Allosaurus [1316] and the late-diverging carcharodontosaurs such as Acrocanthosaurus [17,18], Tameryraptor [19], Carcharodontosaurus [2023], Meraxes [24], Giganotosaurus [2527], Shaochilong [28], Megaraptor [29] and Murusraptor [3032], although the latter three taxa have been sometimes considered as coelurosaurs rather than allosauroids [19,32,33]. In addition, a possible basal allosauroid Asfaltovenator also presents a well-preserved braincase [34]. Although there are such a handful of specimens of allosauroid braincases, information on those of basal carcharodontosaurs is limited. In this context, the probable braincase of the earliest-branching carcharodontosaur will provide important information to fill the gap in the evolutionary history of their braincase morphology.

Geological setting

The Khok Kruat Formation is the uppermost unit of the Khorat Group and consists of meandering river deposit in the Khorat Basin, the southern part of the Khorat Plateau in northeastern Thailand [35]. The bone-bearing bed at the Ban Saphan Hin locality consists of lithofacies representing channel and bar deposits, characterized by medium- to coarse-grained sandstones and conglomerates with planar, cross, and large scaled epsilon-type cross laminations, and containing numerous clasts of clay rip-up pebble and rounded calcareous nodule granules [36]. The thickness of this bed is mostly 1–2 m and does not exceed 3 m.

Although the presence of carbonate clasts suggest that Khok Kruat Formation was deposited under the semi-arid climate [7], Amiot et al. [37] suggest that the Khorat Basin was formed under a subtropical or tropical climate and a semi-humid condition. The geological age of the Khok Kruat Formation has been dated to the Aptian based on the palynological data, occurrences of a freshwater hybodont Thaiodus ruchae and a basal ceratopsian Psittacosaurus sattayaraki, and the Albian–Cenomanian age inferred for the overlying Maha Sarakham Formation [7,3841].

Materials and methods

An incomplete braincase (NRRU-F01020035) lacking at least both paroccipital processes and the cultriform process (Figs 24) and a partial skull roof (NRRU-F01020036) composed of an almost complete left frontal and a fragment of left parietal (Fig 5) are described in the present study. Both specimens are housed at a public and permanent repository in the collection of the Northeastern Research Institute of Petrified Wood and Mineral Resources, Nakhon Ratchasima Rajabhat University (NRRU), Thailand, and are accessible to all researchers. No permits were required for the described study, which complies with all the relevant regulations. The excavation and collection of fossil remains were agreed with the landowner and were officially reported to the Department of Mineral Resources, Thailand.

NRRU-F01020035 was analyzed using an industrial microfocus CT, TXS320- ACTIS (TESCO Co., Yokohama, Japan), at FPDM with the following parameters: voltage of 270 kV, current of 250 μA, and voxel size of 0.05 mm (x- and y-axes) and 0.05 mm (z-axis). Cranial endocast was reconstructed and segmented from the acquired CT images and exported as a polygon mesh to observe, measure and capture images for figures under the orthographic projection view using Amira (v 2020.1, Thermo Fisher Scientific; Waltham, USA).

thumbnail
Fig 2. NRRU-F01020035 in dorsal (A, B), left lateral (C, D) and ventral (E, F) views.

Red lines indicate sutures visible in the bone surface. Red and cyan characters indicate abbreviations of the skeletal elements and cranial nerves, respectively. Scale bar equals 50 mm. Abbreviations: adc, anterodorsal chamber of the rostral tympanic recess; aoc, anterolaterally oriented crest; bo, basioccipital; br, basisphenoid recess; cr, columeller recess; ctr, caudal tympanic recess; eo, exoccipital; f, foramen for some vein; fm, foramen magnum; fr, frontal; ic, internal carotid artery foramen; is, interorbital septum; ls-po, laterosphenoid-prootic complex; nc, nuchal crest; obc; olfactory bulbs contact; oc, occipital condyle; os, orbitosphenoid; ov, orbitocerebral vein; pa, parietal; pf, pituitary fossa; pc, prefrontal contact cavity; pi, pit; pc, posterior chamber for the caudal diverticulum of the rostral tympanic recess and the medial basioccipital diverticulum; sg, stapedial groove; so, supraoccipital; stf, supratemporal fossa; tlp, tongue-like process; II–XII, exits of each cranial nerve.

https://doi.org/10.1371/journal.pone.0345155.g002

thumbnail
Fig 3. NRRU-F01020035 in posterior (A, D), right lateral (B, E) and anterior (C, F) views, and its sagittal section in right lateral view (G).

Red lines indicate sutures visible in the bone surface in A–F and occipitofrontal angle in G. Red and cyan characters indicate abbreviations of the skeletal elements and cranial nerves, respectively. Scale bar equals 50 mm for A–F and 100 mm for G. Abbreviations: adc, anterodorsal chamber of rostral tympanic recess; aoc, anterolaterally oriented crest; ba, basioccipital apron; bo, basioccipital; cr, columellar recess; ct, crista tuberalis (metotic strut); cr, crack; ctr, caudal tympanic recess; eo, exoccipital; fm, foramen magnum; fr, frontal; ic, internal carotid artery foramen; is, interorbital septum; ls-po, laterosphenoid-prootic complex; nc, nuchal crest; oc, occipital condyle; os, orbitosphenoid; pa, parietal; pf, pituitary fossa; pc, prefrontal contact cavity; pap, paracondylar pocket; pc, posterior chamber for the caudal diverticulum of the rostral tympanic recess and the medial basioccipital diverticulum; pi, pit; so, supraoccipital; stf, supratemporal fossa; sr, subcondylar recess; I–VII, exits of respective cranial nerves.

https://doi.org/10.1371/journal.pone.0345155.g003

thumbnail
Fig 4. Exoccipital region in CT-based 3D rendered model of NRRU-F01020035 in posterolateral (left) and lateral slightly-posterior (right) views.

Abbreviations: ci, crista interfenestralis; ct, crista tuberalis (metotic strut); fm, foramen magnum; fo, fenestra ovalis; fp, fenestra pseudorotunda; oc, occipital condyle; pp, paracondylar pneumatopore; sr, subcondylar recess; IX–XII, exits of respective cranial nerves.

https://doi.org/10.1371/journal.pone.0345155.g004

thumbnail
Fig 5. NRRU-F01020036 in anterior (A, D), dorsal (B, C), lateral (E, F), posterior (G, J), medial (H, I) and ventral (K, L) views.

Red lines indicate sutures visible in the bone surface. Scale bar equals 50 mm. Abbreviations: alc, anterolaterally oriented crest; fr, frontal; pa, parietal; pc, prefrontal contact cavity; pi, pit; stf, supratemporal fossa.

https://doi.org/10.1371/journal.pone.0345155.g005

To assess the phylogenetic position of Siamraptor incorporating new anatomical features recognized in the present study, they are included as a single operational taxonomic unit (OTU) and added to the recently published phylogenetic dataset [42], which is one of the updated versions of the dataset by Carrano et al. [43] that has been revised several times [30,44,45]. In addition, some scores of other OTUs are modified based on a review of the published literature as follows: character 100 in Shaochilong and Giganotosaurus, from 0 to 1; character 100 in Carcharodontosaurus, from? to 1; character 100 in Acrocanthosaurus, from 0 to 0&1; character 103 in Murusraptor, from 1 to 0; character 104 in Carcharodontosaurus and Giganotosaurus, from 1 to 0; character 104 in Shaochilong, from unknown to 0; character 104 in Sinraptor, from 0 to 1. For reasons of the modification, see “Implications for the braincase morphology of Allosauroidea” in Discussion.

The data matrices were managed using Mesquite 3.6 [46] and imported into TNT 1.5 [47] to perform a heuristic tree search and find the most parsimonious trees (MPTs). The heuristic tree search used the New Technology algorithms: sectorial searches, ratchet, tree-drifting, and tree fusing, using the default settings for all of them. These algorithms were applied to new searched trees using the driven search with a stabilization of the consensus twice with a factor of 25. Subsequently, the resulting most parsimonious trees (MPTs) were exposed to the branch-swapping algorithm of tree bisection reconnection (TBR). The characters were unordered. The MPTs found from both iterations were examined under the strict consensus, and the consistency index (CI) and retention index (RI) were obtained using the “stats.run” script of TNT. The branch support was tested using the methodology proposed by Goloboff et al. [48] to calculate Bremer Support values, retaining up to 99,999 suboptimal trees that are up to 5 steps longer than the MPTs.

Systematic paleontology

Dinosauria Owen, 1842 [49].

Theropoda Marsh, 1881 [50].

Tetanurae Gauthier, 1986 [51].

Allosauroidea Marsh, 1878 [52].

Carcharodontosauria Benson, Carrano and Brusatte, 2010 [53].

Siamraptor Chokchaloemwong, Hattori, Cuesta, Jintasakul, Shibata and Azuma, 2019 [5].

Type species: Siamraptor suwati Chokchaloemwong, Hattori, Cuesta, Jintasakul, Shibata and Azuma, 2019 [5].

Emended diagnosis

Allosauroid theropod with following characters autapomorphic among allosauroids: jugal with straight ventral margin, and dorsoventrally deep anterior process below the orbit; frontoparietal suture directed posterolaterally on the skull table; two deep pits on the lateral margin of the anterolateral corner of the supratemporal fossa of the frontal; surangular with a deep oval concavity at the posterior end of the lateral shelf and four posterior surangular foramina; long and narrow groove along the suture between surangular and prearticular; articular with a foramen at the notch of the suture with prearticular; anterior cervical vertebra with an additional pneumatic foramen excavating parapophysis; cervical and posterior dorsal vertebrae penetrated by a pair of small foramina bilaterally at the base of each neural spine.

Referred specimens

An incomplete braincase lacking at least both paroccipital processes and the cultriform process (NRRU-F01020035) and a partial skull roof composed of an almost complete left frontal and a fragment of left parietal (NRRU-F01020036).

Locality and horizon

The Lower Cretaceous (Aptian) Khok Kruat Formation of the Khorat Group cropping out in Ban Saphan Hin, Suranaree Subdistrict, Mueang Nakhon Ratchasima District, Nakhon Ratchasima Province, Thailand.

NRRU-F01020035 was analyzed using an industrial microfocus CT, TXS320-ACTIS (TESCO Co., Yokohama, Japan), at FPDM with the following parameters: voltage of 270 kV, current of 250 µA, and voxel size of 0.05 mm (x- and y-axes) and 0.05 mm (z-axis). Cranial endocast was reconstructed and segmented from the acquired CT images and exported as a polygon mesh to observe, measure and capture images for figures under the orthographic projection view using Amira (v 2020.1, Thermo Fisher Scientific; Waltham, USA).

Description

Frontal.

In NRRU-F01020035, both frontals are present as wedge-shaped plates as in other allosauroids, although the anterior and lateral margins are damaged (Figs 2, 3). In NRRU-F01020036, although the left frontal lacks its anteromedial and posterolateral parts, the anterolateral margin is better preserved than those of NRRU-F01020035 (Fig 5). The size and morphological features of the frontal are quite similar between these two specimens.

In comparison with other allosauroids, the frontal is lateromedially broad even in the anterior part (Fig 2A, B) as in Sinraptor hepingensis [12] and carcharodontosaurids [17,19,21,25,54] but unlike Allosaurus and Sinraptor dongi, in which the anterior part is much narrower than the posterior part in dorsal view [10,13,15]. The bone gradually tapers anteriorly in dorsal view to form a blunt but pointed end at its anteromedial margin. The dorsal surface of the frontal except for the supratemporal fossa forms an anterior part of the skull table as a flat plane like those of other allosauroids [10,12,13,15,17,19,21,25] but unlike the narrow one of Murusraptor [31] and the sagittal crest of Shaochilong [28]. The interfrontal suture is straight in dorsal and ventral views in its anterior two-thirds but becomes intricately sculptured in the posterior one-third. Both frontals are connected even at their anterior ends, unlike most allosauroids, in which both frontals diverge before reaching their anterior ends [10,13,30]. At the posterior end of the interfrontal suture, the frontoparietal suture emerges as a deeply interdigitating line that is directed posterolaterally on the skull table forming a slight dorsal swelling. Such direction of the frontoparietal suture accompanied with the wedge-shaped anterior margin of the parietals is unknown in other allosauroids, in which the suture is directed laterally even on the skull table [10,12,13,15,30,54]. There is a pair of small foramina possibly for the exits of some vascular components at the middle of the frontoparietal suture on the skull table (Fig 2B). The frontoparietal suture enters into the supratemporal fossa with a slight lateral swelling (frontoparietal process sensu [54]) that is less distinctive than those of Sinraptor [10] and Eocarcharia [54], and becomes a straight line running laterally on the bottom of the fossa as in other allosauroids. The frontoparietal suture is located more anteriorly on the skull table than in the supratemporal fossa as in Sinraptor (Fig 4 in [12]; Fig 7 in [10]) but unlike Allosaurus, in which the one is located more posteriorly on the skull table than in the supratemporal fossa (Fig 11 in [13]; Fig 2 in [15]). At the anteromedial corner of the supratemporal fossa, the dorsal rim of the fossa slightly protrudes posterolaterally to partially roof the fossa (Figs 2A, B, 5B, C) as in some carcharodontosaurids [43,54]. At the lateral end of this rim, the lateral margin of the bone is damaged but preserves two deep pits excavating the frontal medially (Figs 2AD, 3B, E, 5B, C, E, F). These pits probably related to the postorbital contact and a similar excavation is also seen in Eocarcharia [54] whereas the surface is rather flat or convex in other allosauroids [14,17,25,31,55]. These pits are separated by a thick lamina into the anterodorsal and posteroventral ones unlike an almost single slot of Eocarcharia [54]. In NRRU-F01020035, the ventral rim of the posteroventral pit protrudes somewhat ventrally from the remainder of the ventral surface of the bone, whereas this protrusion is broken in NRRU-F01020036.

On the ventral surface, there is an anterolaterally oriented crest aligned with the broken anterior margin of the orbitosphenoid (Figs 2EF, 5K, L). At the lateral end of the crest, the lateral margin of the frontal bears a deep pit penetrating the frontal posteromedially to receive the posterior apex of the prefrontal as in Allosaurus [13] and Carcharodontosaurus [21]. Anteromedial to the crest, the thickness of the frontal becomes about half of the posterolateral part of the bone and exhibits a shallowly-concave surface representing the area of the contact for the olfactory bulbs (Fig 2E, F). Posterior to the crest, the medial and posterior margins of the ventral surface of the frontal are covered by the orbitosphenoid and the laterosphenoid, respectively.

Parietal.

In NRRU-F01020035, both parietals are well preserved but lacking a large part of the transverse nuchal crest (Figs 2, 3). In NRRU-F01020036, only the anteriormost part of the left parietal adhered to the frontal is preserved (Fig 5). In contrast to the frontals, the interparietal suture is obliterated (Fig 2A, B) so that both parietals are completely fused as in carcharodontosaurids [17,19,21,25,28] but unlike basal allosauroids [10,13,15] and Murusraptor [30]. In dorsal view, the parietals extend anteriorly to form a wedge invading the interfrontal suture and laterally to meet the suture against laterosphenoids. In lateral view, the parietal-laterosphenoid suture is oriented almost horizontally (Figs 2C, D, 3B, E).

In dorsal view, the parietals are largely occupied by the supratemporal fossae. The dorsal surface between supratemporal fossae forms a supratemporal crest with a flat plane as a posterior part of the skull table (Figs 2A, B, 5B, C). The lateral width of this crest is similar to that of Allosaurus (Plate 2 in [13]), which is narrower than those of Acrocanthosaurus, Carcharodontosaurus and Giganotosaurus [17] but wider than those of Shaochilong [28] and Murusraptor [30]. Posteriorly, the parietals are projected dorsally with the skull table gradually becomes facing anterodorsally and spread laterally to form a prominent dorsal margin of the nuchal crest (Figs 2AD, 3B, C, E, F). In anterior view, the center of the dorsal margin is concave as in Allosaurus (Fig 13 in [13]) but unlike the prominent top of the crest seen in Sinraptor [10] and Giganotosaurus [25]. The dorsal margin of the nuchal crest does not extend much above the supraoccipital (Figs 2C, D, 3B, E) as in Sinraptor [10] and is tall as in some abelisauroids and carcharodontosaurs [43]. In dorsal view, there is a small tongue-like process on the midline partially overlapping the dorsal surface of the supraoccipital as in Sinraptor [10] and Allosaurus [15] but is much smaller than that of Giganotosaurus [25].

Remainder of the braincase.

The remainder of the braincase is represented only by NRRU-F01020035 lacking at least opisthotics and the parasphenoid. The foramen magnum is circular, which is 3 cm wide and 2 cm high in posterior view (Fig 3A, D). The occipital condyle is 5 cm wide and 4 cm high in posterior view. In lateral view, the angle between the skull roof and the occiput (occipitofrontal angle) is obtuse (Fig 3G) as in other allosauroids except for Allosaurus and Acrocanthosaurus [25,31,43].

The orbitosphenoids are the anteriormost elements of the braincase except for the frontals (Figs 2CF, 3B, E). The orbitosphenoids are completely fused with each other to form a low median ridge, which probably correspond to the interorbital septum seen in Acrocanthosaurus [17], Giganotosaurus and Carcharodontosaurus [25]. The anterior part of the orbitosphenoid contacts the frontal, as in carcharodontosaurids [17,21,28,54], but unlike Sinraptor and Allosaurus, in which the one is bordered only by the laterosphenoid [10,14,56]. Although the anterior part of the orbitosphenoid may pertain to the sphenethmoid, as suggested for Carcharodontosaurus iguidensis [21] and possibly for Acrocanthosaurus [57], there is no obvious evidence of its ossification in NRRU-F1020035. In ventral view, the anterior margin of the left orbitosphenoid is aligned with the anterolaterally oriented crest (Figs 3E, F), as in Sinraptor (Fig 7E in [10]) and Allosaurus (FPDM-V-9672). However, this margin is somewhat broken in both orbitosphenoids along the same straight line. Therefore, the orbitosphenoid might extend slightly more anteriorly as in Acrocanthosaurus (Fig 16 in [17]). A sizable opening for the exit of the cranial nerve (CN) I (olfactory nerve) excavates the medial parts of the orbitosphenoids anteroposteriorly. The mesethmoid is absent in this specimen, resulting in a single opening for CN I, as in Sinraptor [10] and multiple specimens of Allosaurus [13,16,17].

Posterior to the orbitosphenoids, there is a large circular opening corresponds to the exit of CN II (optic nerve) with the path of the pituitary (Fig 3F). The flange of bone overlying and separating them in other theropods (e.g., [17,25,58]) is absent. It is uncertain whether the flange was not ossified or eroded by the weathering. Such a large opening is similar to the exit of CN II of Allosaurus [14]. However, the opening of NRRU-F01020035 is only anteriorly bordered by the orbitosphenoids, whereas is completely surrounded by the orbitosphenoids in Allosaurus [14]. The exit of CN III (oculomotor nerve) lies adjacent to the lateral margin of the exit of CN II and, being bordered anteromedially by the orbitosphenoid and posterolaterally by the laterosphenoid, marks the approximate posterior end of the suture between these elements.

The laterosphenoid is covered dorsally by the frontal and the parietal (Figs 2D, 3E) as in other allosauroids. While the laterosphenoid-prootic suture is not obvious, the probable remnant of the laterosphenoid-orbitosphenoid suture is partially visible. The laterosphenoid has a barlike lateral extension but its distal end, namely the distal cotylus that contacts the postorbital, is broken (Fig 2F). The anteromedial part of the left laterosphenoid has a pair of small foramina. The more anteromedial one corresponds to the exit of CN IV (trochlear nerve) because it is present on the remnant of the orbitosphenoid-laterosphenoid suture along with the exit of CN III as in most theropods [14,27,59]. The other, more posterolateral one probably corresponds to the orbitocerebral vein. These foramina seem to be merged in the right laterosphenoid as in Murusraptor [31] because there is only a single, slightly larger foramen that intercepts the remnant of the orbitosphenoid-laterosphenoid suture.

The pituitary (hypophyseal) fossa is situated posteroventral to the exit of CN II and excavates the braincase posteriorly (Fig 2D, F). This fossa is possibly composed of the complex of basisphenoid and parasphenoid, but there is no obvious suture line to distinguish them. Left and right exits of CN VI (abducens nerves) are visible as a pair of small foramina within the anterior part of the fossa as in Shaochilong [28], Carcharodontosaurus [23] and Giganotosaurus [25,26] unlike those on the outside of the pituitary fossa and separated by a median ridge seen in Acrocanthosaurus [18] and Shaochilong [31]. The deepest part of the pituitary fossa has a pair of exits that correspond to the internal carotid artery foramina connected to the rostral (anterior) tympanic recess as in Shaochilong [28].

The prootic is completely fused with the laterosphenoid as the suture between them is not visible (Figs 2, 3). The posteroventral region of this laterosphenoid-prootic complex is perforated by exits of CNs V and VII (trigeminal and facial nerves). There are four trigeminal foramina (exits of CN V) penetrating medially into the braincase on the right side while only three on the left side (Fig 2D, F). Among the trigeminal foramina, the anterior one is separated from the posterior foramina as in Allosaurus [14,16] and Murusraptor [30], despite the posterior foramina are merged into one in the latter two taxa. Therefore, the area anterior to the posterior foramina is probably occupied by the laterosphenoid as in these two taxa [14,16,31]. The anterior foramen is the exit of CN V1 (ophthalmic branch of CN V), which is situated posterolateral to the exits of CNs II and III and penetrates posteromedially into the braincase. Somewhat posterior to the exit of CN V1, there are three openings for CN V2,3 (maxillary and mandibular branches of CN V) on the right side, whereas only two are on the left side. Based on their position, the two smaller openings on the right correspond to the smaller opening on the left because both are dorsal to the largest opening (Figs 2D, 3E). Each of these two sections possibly corresponds to each branch but cannot be determined here. Posterior to these trigeminal foramina, there is a small opening for the exit of CN VII penetrating the prootic region. When the frontal is horizontal, the exit of CN V1 is located ventral to the nuchal crest, whereas those of CNs V2,3 and VII are located more posteriorly. Posterodorsal to the exit of CN VII, there is a large foramen that corresponds to the columellar recess [56]. Ventral to the exit of CN VII, the rostral tympanic recess excavates the braincase anteromedially, which seems to be covered laterally by the basisphenoid if preserved. The anterodorsal end of the rostral tympanic recess is connected to an oval chamber excavating the region ventromedial to the exit of CN V1. This chamber is almost completely exposed on the left side due to the absence of its lateral wall, which is partially preserved in the right side. Ventral to the entrance to this anterodorsal chamber is the internal carotid artery foramen as a junction to the pituitary fossa. Posterior to this foramen, there is an entrance for an internal chamber invading the braincase toward the neck of the occipital condyle (see “Endocranial cavities” for details).

Below the laterosphenoid- prootic region, the basisphenoid and parasphenoid are not preserved or are extremely poorly preserved (Figs 2D, 3E). Although the pituitary fossa could be interpreted as a remnant of the basisphenoid, this interpretation is uncertain, and the posterior part of the fossa may instead be formed by the laterosphenoid, as in Murusraptor (Fig S2 of [31]). The dorsal part of the basisphenoid recess is formed by the basioccipital as a large opening directed posteroventrally anterior to the broken basal tubera (Fig 2F). The basisphenoid recess is divided anteroposteriorly and the posterior part is further divided transversely by thin septa. The anterior part is a fossa with a faint connection to an internal chamber invading the braincase anterodorsally, toward the region just posterior to the pituitary fossa (Figs 6, 7; see “Endocranial cavities” for details). However, a large crack just across this connection indicates that the wall separating the chamber from the fossa has been lost by fracture (Fig 3G). The posterior parts are connected to the two larger chambers mentioned above, which communicate with the rostral tympanic recesses and extend into the neck of the occipital condyle.

thumbnail
Fig 6. Cranial endocast of NRRU-F01020035 in right lateral (A), left lateral (B), dorsal (C), ventral (D), anterior (E), and posterior (F) views.

Scale bar equals 100 mm. Abbreviations: ac, subsidiary diverticulum occupying the anterior chamber; asc, anterior semicircular canal; cc, common crus; cer, cerebral hemisphere; dp, dural peak; floc, floccular recess; fm, foramen magnum; lag, lagena; lsc, lateral semicircular canal; ob, olfactory bulb; ot, olfactory tract; ov, orbitocerebral vein; pit, pituitary; pc, complex of the caudal diverticulum of the rostral tympanic recess and the medial basioccipital diverticulum occupying the posterior chamber; psc, posterior semicircular canal; vc, venous canal; II–XII, cranial nerves.

https://doi.org/10.1371/journal.pone.0345155.g006

thumbnail
Fig 7. Cranial endocast with transparent braincase of NRRU-F01020035 in ventral (A) and left lateral (B) views.

Scale bar equals 100 mm. Abbreviations: bo, basioccipital; eo, exoccipital; fr, frontal; ls-po, laterosphenoid-prootic complex; ob, olfactory bulb; os, orbitosphenoid; ov, orbitocerebral vein; pa, parietal; pc, complex of the caudal diverticulum of the rostral tympanic recess and the medial basioccipital diverticulum occupying the posterior chamber; pit, pituitary; so, supraoccipital; II-XII, cranial nerves.

https://doi.org/10.1371/journal.pone.0345155.g007

The posterodorsal region of the braincase is composed of the supraoccipital, which contacts the parietal anteriorly and the exoccipital-opisthotic complexes lateroventrally (Figs 2B, D, 3D, E). The lateral parts of the supraoccipital are broken. In dorsal view, the suture line between the supraoccipital and the parietal is invaded by the tongue-like process of the parietal (Fig 2B) as in other allosauroids [10,15,25]. There is a dorsal projection of the supraoccipital (supraoccipital knob) along with the median ridge on the posterior surface (supraoccipital crest) as in other allosauroids [10,21,24,25,28] except for Acrocanthosaurus and some specimens of Allosaurus, in which the knob is divided into two by a midline fold [17]. The width of the supraoccipital knob (39.60 mm) is larger than that of the foramen magnum (29.93 mm) as in other allosauroids [43]. Although the suture line against the exoccipitals is indistinct near the dorsal margin of the foramen magnum, the remaining part of the suture indicates that the contribution of the supraoccipital to the foramen magnum is very narrow or absent (Fig 3D).

In posterior view, the exoccipitals form lateral edges of the foramen magnum (Fig 3D). Whether or not the exoccipitals are separated by the supraoccipital on the dorsal edge of the foramen magnum is ambiguous because the suture against supraoccipital is invisible in that part. In contrast, the exoccipitals are clearly separated by the basioccipital on the ventral edge of the foramen magnum. At the midheight of the foramen magnum, a horizontal suture-like line is present as in Murusraptor (Fig 6E’ in [30]). In dorsal view, sutures against the basioccipital are present just anterodorsal to the condylar surface (Fig 2B), thus exoccipitals have only a little contribution to the occipital condyle unlike Allosaurus, in which the exoccipitals occupy a conspicuous part of the condyle [13]. Dorsolateral to this suture, a large pneumatic cavity probably correspond to the caudal tympanic recess invades the base of the paroccipital process (Fig 3D, E) as in Sinraptor and many other theropods [56].

The basioccipital composes most of the occipital condyle and makes a small contribution to the floor of the foramen magnum (Fig 2B). Just below the occipital condyle, the basioccipital forms a broad, plate-like surface (“basioccipital apron” sensu [60,61]) that faces posterodorsally and is slightly wider than the occipital condyle in posterior view (Fig 3D). The basioccipital apron continues posteroventrally with forming a rough and concave posterodorsal surface toward the basal tubera, whereas the tuber itself is not preserved in this specimen. The concave posterodorsal surface becomes a deep fossa at the dorsal margin of the basioccipital apron excavating the base of the neck of occipital condyle anterodorsally (Fig 4A). This fossa probably corresponds to the subcondylar recess but is not connected with any internal pneumatic cavities unlike those of Shaochilong [28] and Murusraptor [30]. The lateral width of this fossa (21 mm) is much narrower than that of the occipital condyle (51 mm) as in most theropods [43].

Anterolateral to the occipital condyle and posteroventral to the caudal tympanic recess, there is a large fossa correspond to the paracondylar pocket [28,62] or paracondylar recess ([56,58,63]; Fig 3D, E). The dorsal part of this fossa bears two foramina separated dorsoventrally by a septum. The dorsal foramen represents the exit of the posterior branch of CN XII (hypoglossal nerve). The ventral foramen is further separated mediolaterally by a vertical septum (Fig 4A) as in Giganotosaurus [27,64]. The medial, smaller one corresponds to the anterior branch of CN XII, whereas the lateral, larger foramen is interpreted as the vagal foramen, representing the caudal opening of a subdivided metotic fissure and transmitting CNs X and XI (vagus and accessory nerves; Fig 2D). In total, there are three foramina as in Giganotosaurus [18], Shaochilong [28] and Murusraptor [31] on the right side, unlike Sinraptor and Allosaurus, in which there are only two foramina because CN XII exits only through one foramen [16,56]. The remaining ventral part of the paracondylar pocket probably corresponds to the paracondylar pneumatopore, although which is more deeply excavated in carcharodontosaurids to lead into an extensive recess below the endocranial cavity [21,25,28,43]. The anterior margins of these foramina and a fossa is formed by a thin bony web as a remnant of the crista tuberalis (metotic strut). Anterior to this bony web, there is a large triangular cavity representing the columellar recess excavating the braincase anteromedially.

The dorsal and anteroventral rims of the columellar recess are formed by a thin ventral wall of the caudal tympanic recess and a thicker lamina, respectively (Figs 2D, 3D, E). The columellar recess accomodates two foramina separated by a vertical septum, namely the crista interfenestralis (Fig 4). The anterolateral division is the fenestra ovalis (or vestibularis) excavating the braincase anterodorsally to connect with the cochlear duct. The posteromedial division is the fenestra pseudorotunda (or fenestra cochleae). The anterior end of the dorsomedial corner is the deepest part that is excavated by an opening for CN VIII (vestibulocochlear nerve). Posteriorly, the dorsomedial corner has two additional openings of the vagal canal transmitting CNs X and XI in the right side, while there is only one opening in the left side. The area ventral to them is excavated by an opening for CN IX (glossopharyngeal nerve).

Cranial endocast.

The digital cranial endocast reconstructed from the well-preserved braincase (NRRU-F01020035) is nearly complete and consists of the cavities for the brain and inner ears, canals for cranial nerves and blood vessels, and pneumatic cavities (Figs 6, 7). The endocast preserves the olfactory bulbs and cerebral hemispheres, as well as the midbrain, hindbrain, and pituitary. Some dural sinuses are visible on the endocast, but cranial nerves and blood vessels themselves cannot be observed in the CT slices and were therefore not reconstructed. The endocast measures 175 mm long from the anterior end of the olfactory bulbs to the foramen magnum and has a maximum width of 42 mm at the lateral processes of the cerebral hemispheres. The volume of the endocast is approximately 124 ml, excluding canals for cranial nerves and blood vessels.

Overall, the cranial endocast is dorsoventrally low and mediolaterally narrow, with bulges indicative of forebrain, midbrain, and hindbrain structures. The cephalic flexure (angle between the forebrain and the midbrain) and the pontine flexure (angle between the midbrain and hindbrain) are at 130 and 150 degrees, respectively, which makes the forebrain slightly upward to the hindbrain, but the two are approximately parallel. The angles of the cephalic and pontine flexures are similar to those of Sinraptor [56], Tameryraptor [19], Carcharodontosaurus [23] and Giganotosaurus [25,26]. This makes the endocast elongate anteroposteriorly, unlike those of Allosaurus, Ceratosaurus, Majungasaurus and Viavenator, which are dorsoventrally high [16,6567]. There are no traces of sutures between the bones of the braincase on the surface of the endocast, confirming that the braincase belongs to a subadult or adult specimen.

Forebrain.

The anteriormost portion of the cranial endocast represents the olfactory bulbs and tract. Only the dorsal margin of the olfactory bulbs is visible as impressions on the ventral surface of the frontal as in other theropods [16]. Due to the absence of bony septum, the olfactory bulbs are reconstructed as a single structure expanded anterolaterally, but in dorsal view, it shows a heart shape, with its midline slightly depressed along the antroposterior axis. The olfactory tract is triangular in its cross-section with a flat dorsal surface and relatively short in comparison to these of Carcharodontosaurus [23] and Giganotosaurus [25,26]. The ventral apex of this tract becomes rounded posteriorly to merge with the cerebral hemispheres, while the flat dorsal margin continues posteriorly to the approximate midpoint of the cerebral hemispheres. The lateral process on the anterior part of the cerebral hemisphere marks the part of the endocast with the greatest mediolateral width. The venous canal, reconstructed on the right side only, extends anterolaterally from the anterior dorsolateral margin of the cerebral hemisphere as in Majungasaurus [58] and Allosaurus [16,66]. Ventral to the cerebral hemisphere is the large circular exit of CN II (optic nerve) with the path of the pituitary (Fig 7A). The dorsalmost part of the endocast, the dural peak, is at the same anteroposterior level as the caudal margins of the CN II exit and the cerebral hemispheres.

Midbrain.

There is no obvious representation of the optic lobes (optic tecta) and CNs III (oculomotor nerve) and IV (trochlear nerve). The midbrain is inclined posteroventrally along with the occiput (occipital plate) as in Carcharodontosaurus [23] and Giganotosaurus [25,26]. Posterior to the cerebrum, the dorsal surface of the endocast above the hindbrain is marked by a pronounced ridge that continues caudally from the dural peak.

Hindbrain.

Boundaries between the cerebellum, medulla, and pons of the hindbrain are not clearly defined in the cranial endocast. However, there is a distinct floccular process on each side (Figs 6, 7C), which in life housed the floccular lobe of the cerebellum, as in other theropods [10,27,66,68,69]. The floccular process extends posterolaterally to enter the region surrounded by the anterior semicircular canal. The floccular process is elongated dorsoventrally and slopes anteroventrally along the long axis of the anterior semicircular canal. The subtle swelling dorsal to the floccular process probably corresponds to a vascular component (Fig 6A, B) such as the vena capitis dorsalis of Carcharodontosaurus [23] and the caudal middle cerebral vein of Murusraptor [31] and Giganotosaurus [27]. Anteroventral to the floccular process, CN V (trigeminal nerve) is represented by a single root on the lateroventral margin of the hindbrain as in other allosauroids [16,18,23,27,31,56]. CN V extends anteroposteriorly to form an anterior exit for CN V1 (ophthalmic division) and two or three posterior exits for CN V2,3 (maxillary and mandibular divisions). An enlarged region at the base of these divisions indicates the location of the trigeminal ganglion [16,31]. A similar pattern, in which the ophthalmic, maxillary and mandibular branches diverge from a shared exit region, is also common in other tetanurans (e.g., Allosaurus, Struthiomimus, Dromaeosaurus, Troodon), although tyrannosaurids have ophthalmic and maxillomandibular canals that branch separately from the endocranial cavity [67]. CN VI (abducens nerve) projects anteroventrally from the ventral part of the hindbrain and enters into the pituitary fossa. CN VII (facial nerve) emerges posterior to the exits of CN V2,3 and extends lateroventrally. CN VIII (vestibulocochlear nerve) emerges ventral to the endosseous labyrinth and anterodorsal to the passage of CN IX (glossopharyngeal nerve). CN IX has its own passage completely separated from those of CNs X and XI (vagus and accessory nerves) as in Carcharodontosaurus [23], unlike that of Allosaurus conjoined only with CN X [16] nor with CNs X and XI as in Acrocanthosaurus [18], Giganotosaurus [27] and Murusraptor [31]. CNs X and XI shares the passage as the vagal canal directed posterolaterally, although its middle part is bifurcated in the left side. The endocast has a distinct lateral process dorsomedial to the metotic fissure. There are two passages for CN XII (hypoglossal nerve), which arise from different roots and extend posterolaterally from the posteroventral aspect of the hindbrain.

Diverticulum.

In the cranial endocast, three pneumatic chambers for the diverticula located in the basicranium ventral to the hindbrain are reconstructed (Figs 6, 7). These are the anterior one and the posterior two chambers.

The anterior chamber extends anterodorsally and expands anterolaterally. There are sagittal grooves on the dorsal and anteroventral surfaces, indicating a subtle division of the anterior chamber into left and right. Although the anterior chamber has a small opening in the anterior part of the basisphenoid recess, a large crack just across this opening indicates that the wall separating two chambers has been lost by fracture (Fig 3G). Instead of the external opening, the anterior chamber has a broad communication with the posterior chambers. Therefore, the anterior chamber probably housed the subsidiary diverticulum derived from the one occupying the posterior chamber.

The posterior chambers are posteriorly elongated and posterolaterally dilated. The relative position of the posterior chamber is similar to that of the caudal diverticulum of the rostral tympanic recess (cRTr) and the medial basioccipital diverticulum (MBd) in other avetheropods [70]. In addition, the posterior chamber communicates with the posteroventral part of the rostral tympanic recess as in cRTr, and with the dorsal apex of the basisphenoid recess as in MBd (Figs 2F, 7A). Each posterior chamber is elongated posteriorly as in cRTr [70]. However, the posterior chamber of Siamraptor is oval and expanded posterolaterally, unlike the narrow form of cRTr [70]. The inflation of this chamber renders the sagittal septum separating the posterior chambers incomplete, allowing a broad confluence between them (Fig 6D). Given this unusual inflation and the topographic similarity to both cRTr and MBd, the posterior chamber of Siamraptor appears to be formed by the fusion of chambers occupied by cRTr and MBd in other avetheropods.

Inner ear (endosseous labyrinth).

The cranial endocast also preserves the semicircular canals, cochleae, and vestibule of the inner ear. The inner ear measures approximately 112.8 mm dorsoventrally and 83.5 mm anteroposteriorly. The region outlined by semicircular canals is strongly triangular in lateral view (Fig 6A, B) as in Carcharodontosaurus [23] and Giganotosaurus [27]. The anterior and posterior semicircular canals are tall, slender and approximately triangular in posterolateral and anterolateral views, respectively. The anterior semicircular canal extends dorsally above the level of the posterior semicircular canal. The lateral semicircular canal is oval in dorsal view, with its main axis oriented anteroposteriorly.

Discussion

Identification as braincase and skull roof of Siamraptor

As described above, NRRU-F01020035 and F01020036 share an important feature supporting their affinity to Carcharodontosauria: the anteromedial corner of the supratemporal fossa excavating the frontal deeply, which is known as a possible synapomorphy of Carcharodontosauria or Carcharodontosauridae [43]. Furthermore, other carcharodontosaurian features can be observed in the part of NRRU-F01020035 that is not preserved in NRRU-F01020036: the supraoccipital knob substantially wider than the foramen magnum (an unambiguous synapomorphy of Allosauroidea [43]); fully split trigeminal foramen (a possible synapomorphy of Allosauroidea or Allosauria [43]); the internal carotid pneumatization developed as an opening (a possible synapomorphy of Allosauria [43]); tall nuchal crest (a possible synapomorphy of Carcharodontosauria [30,43] or late-diverging carcharodontosaurids [43]).

In addition to synapomorphies of the clades within Allosauroidea noted above, there are two features shared between in NRRU-F01020035 and F01020036 unique among allosauroids: the posterolaterally-directed frontoparietal suture on the skull table accompanied with the wedge-shaped anterior margin of the parietals; two deep pits on the lateral margin of the frontal at the anterior margin of the supratemporal fossa, despite their presence might be somewhat affected by the damage on the lateral margin of the frontal. These unique features indicate the closest affinity between NRRU-F01020035 and F01020036. In combination with the monospecific occurrence of Carcharodontosauria in the Ban Saphan Hin locality, both specimens can be identified as new materials of Siamraptor suwati. Nevertheless, given the current limitations of the fossil record, the presence of more than one closely related species of the same clade in the area cannot be entirely excluded.

Phylogenetic analysis

The heuristic tree search using the data matrix composed of 367 characters (with character 201 inactivated following Coria and Currie [42]) distributed among 68 theropod taxa and incorporating the new materials described above in Siamraptor resulted in a total of 32,076 MPTs with 1,152 steps (CI = 0.370, RI = 0.661). The topology of the strict consensus tree of these MPTs is basically the same as that provided by Coria and Currie [42], except for the placement of Siamraptor at the basalmost position within Carcharodontosauria (Fig 8), which is consistent with the phylogenetic position suggested in the original description of Siamraptor [5]. Although the phylogenetic placement of megaraptorans remains debated and may vary depending on the dataset employed, they are conservatively treated here as members of Carcharodontosauria in accordance with the results of the present analysis. This treatment allows their inclusion as an important comparative taxon, while acknowledging that alternative affinities, such as with Tyrannosauroidea or Coelurosauria [19,32,33], cannot be entirely ruled out.

thumbnail
Fig 8. Strict consensus tree of theropods including Siamraptor with Bremer support values on each node.

A, Neotheropoda; B, Averostra; C, Ceratosauria; D, Tetanurae; E, Orionidae; F, Megalosauroidea; G, Avetheropoda; H, Coelurosauria; I, Allosauroidea; J, Metriacanthosauridae; K, Allosauria; L, Allosauridae; M, Carcharodontosauria; N, Carcharodontosauridae; O, Neovenatoridae. Numbers at the top left of each node represent the Bremer support value.

https://doi.org/10.1371/journal.pone.0345155.g008

This analysis revealed that the braincase of Siamraptor exhibits the wide supraoccipital knob as an allosauroid synapomorphy [character 88 (state 1); reversed in Murusraptor], the opening of the internal carotid pneumatization as an allosaurian synapomorphy [ch. 107 (2); convergent with Eustreptospondylus] and the tall nuchal crest as a carcharodontosaurian synapomorphy [ch. 70 (1); convergent with Majungasaurus], in addition to known other synapomorphies seen in the remaining cranial and postcranial elements [5]. Furthermore, this analysis presented three diagnostic features for Siamraptor in the braincase: the shallow ventral extension of the subcondylar recess [ch. 101 (1); convergent with some ceratosaurs and some megalosauroids], the obtuse angle between the skull roof and the occiput in lateral view [ch. 103 (1); convergent with Sinraptor and the clade comprising Shaochilong and Carcharodontosaurinae] and the absence of the median ridge separating left and right exits of CN VI [ch. 105 (1); convergent with Giganotosaurus and Carcharodontosaurus].

Implications for the braincase morphology of Allosauroidea

Siamraptor shares several characters with other carcharodontosaurs that are known to be possible synapomorphies of Carcharodontosauria, and some of these are also supported as synapomorphies in the present phylogenetic analysis. On the frontal, the anteromedial corner of the supratemporal fossa partially roofed by the pronounced dorsal rim [ch. 71 (1)] has been observed only in some late-diverging carcharodontosaurs, so that has been regarded as a possible synapomorphy of Carcharodontosauria or Carcharodontosauridae [43]. Although its absence in a late-diverging neovenatorid Murusraptor [31] may supports the latter hypothesis, its presence in Siamraptor rather supports the former hypothesis; a synapomorphy of Carcharodontosauria lost in Neovenatoridae. Lack of enough fossil records on neovenatorid frontals makes it difficult to when and how this character state was lost in this lineage, so that further discoveries on neovenatorid frontals are required to reveal this point. The function of this fossa has recently been hypothesized to house blood vessels that perform a thermoregulatory function to moderate eye and/or brain temperatures [71]. This suggests that at least some carcharodontosaurs may have had a specialized thermoregulatory function in this region compared to other theropods.

In the skull roof, both the frontals and the parietals are unfused in early-diverging allosauroids such as Sinraptor [10,11] and Allosaurus [13,15] as well as a neovenatorid Murusraptor [31], but are fused in carcharodontosaurids such as Acrocanthosaurus [15], Carcharodontosaurus [22], Meraxes [24] and Giganotosaurus [25]. Although the unfused conditions of some allosauroids may be due to their ontogenetic stages [10,32], at least parietals are usually fused during early ontogenetic stage in theropods [31,72], so that its presence or absence may have some phylogenetic implications, although this feature was not coded as a phylogenetic character in the present analysis. In this case, the median fusion of the skull roof is at first glance a synapomorphy of Carcharodontosauridae. However, the combination of fused parietals and unfused frontals in a non-carcharodontosaurid carcharodontosaur Siamraptor indicates that the fused parietals is a synapomorphy of Carcharodontosauria, and the median fusion was extended to the frontals in Carcharodontosauridae to make the skull roof more robust, whereas the fusion was reduced in Neovenatoridae.

The obtuse occipitofrontal angle [ch. 103 (1)] is widely distributed within allosauroid lineage as seen in Sinraptor, Shaochilong, Carcharodontosaurus, Meraxes and Giganotosaurus, but not in Allosaurus and Acrocanthosaurus [24,25,31,43]. In Murusraptor, although this angle has been regarded as obtuse [30], another research focused on its braincase revealed that the angle is rather acute [31] as scored in the present phylogenetic analysis. Because of such a distribution pattern, this character state has been interpreted as a synapomorphy of late-diverging carcharodontosaurids convergent with Metriacanthosauridae [43]. The presence of this character state in a basal carcharodontosaur Siamraptor increases the possibilities of two other scenarios as follows: (1) a synapomorphy of Carcharodontosauria convergent with Metriacanthosauridae but lost in Acrocanthosaurus and Neovenatoridae; (2) a synapomorphy of Allosauroidea independently lost in Allosauridae, Acrocanthosaurus and Neovenatoridae. However, the original hypothesis is more parsimonious than the two alternatives, as it requires fewer evolutionary changes, even if the obtuse occipitofrontal angle was independently acquired in Siamraptor. The obtuse occipitofrontal angle was regarded as an indicator of a horizontal craniocervical posture and a slightly upturned skull posture in Ceratosaurus [73]. Furthermore, the skull roof of Siamraptor forms the angle of about 15 degrees with the plane of lateral semicircular canal, indicating that the skull is slightly upturned when the lateral semicircular canal is parallel to the ground surface as in Ceratosaurus [73] but unlike carcharodontosaurids, in which the skull roof is oriented horizontally (Fig 1 in [74]) or slightly anteroventrally [65]. Although the orientation of lateral semicircular canal does not absolutely indicate the natural head posture [75], in combination with the obtuse occipitoparietal angle, the head of Siamraptor can be regarded as relatively upturned than other basal allosauroids.

The tall nuchal crest [ch. 70 (1)], which has been recognized as a possible synapomorphy of Carcharodontosauria [30,43], is recovered as an unambiguous synapomorphy in the present analysis due to its presence in the earliest-diverging carcharodontosaur, Siamraptor. Although a tall nuchal crest is also present in Abelisauridae, the contribution of the squamosal to the crest is dorsoventrally deep in that clade, whereas there is little or no such contribution in other theropods [58,76,77]. The present character (ch. 70) refers to the relative height of the nuchal crest and does not capture differences in its osteological composition. Therefore, a tall nuchal crest with little or no squamosal contribution is possibly a feature unique to Carcharodontosauria. Such nuchal crests can provide large areas and lever arms for cervical axial muscles supporting dorsiflexion at the craniocervical joint, possibly as a response to the greater mass of their heads [7880].

The presence of fully split trigeminal foramina [ch. 104 (2)] has been regarded as a possible synapomorphy of Avetheropoda or Allosauroidea convergent with Megalosauria [43], as seen in allosauroids Allosaurus [15,16] and Murusraptor [30] as well as a megalosauroid Piveteausaurus [81]. However, the trigeminal foramina are merged into a single opening in Acrocanthosaurus as in most other theropods [18,28,31]. In addition, although the conditions of Sinraptor and Giganotosaurus have been scored as fully split [30,4245], they have been reported as having partially split [10,56] and single [25] openings, respectively, and were scored accordingly in the present phylogenetic analysis. Similarly, despite Carcharodontosaurus having been regarded as fully split [30,4245] as seen in C. iguidensis [21], only a single opening has been reported in C. saharicus [21,23] so that this genus was scored as having both states in the present analysis. Furthermore, although the scoring of Shaochilong for this character has remained unknown in previsou studies [30,4245], the presence of a single trigeminal foramen has been reported [28], and this taxon was therefore scored as exhibiting the primitive condition in the present study. The differences in reported conditions discussed above primarily reflect revisions of previous scorings based on re-examination of the literature, rather than direct evidence for biological or infraspecific variation. The presence of fully split trigeminal foramina in Siamraptor in combination with the distribution pattern revised above suggests that the fully split trigeminal foramina may represent a synapomorphy of Allosauria, while also indicating a complex evolutionary history within the group.

When the frontal is horizontal, the positions of the trigeminal foramina are anterior to the level of the nuchal crest in most theropods including Sinraptor and Allosaurus [25]. The position becomes more posterior in carcharodontosaurs to some degree, at least at the level of the crest as in Carcharodontosaurus iguidensis and Acrocanthosaurus [21], or posterior to the level as in Carcharodontosaurus saharicus and Giganotosaurus [25]. The trigeminal foramina of Siamraptor, with the one located at the level and the others located posterior to the level, represent an intermediate condition among these carcharodontosaurs.

The exit of CN VII splits into two in Acrocanthosaurus and possibly in Giganotosaurus [17,18] and thus it has been regarded as a character indicating carcharodontosaurid affinity [28]. Therefore, the absence of this character in Siamraptor supports its phylogenetic position as not a carcharodontosaurid. However, the presence of the partially-splitting condition represented by an hourglass-shaped opening in a neovenatorid Murusraptor [31] indicates that at least the intermediate condition was acquired immediately before the divergence of Carcharodontosauridae and Neovenatoridae, probably after the divergence of Siamraptor.

The exits of CNs X–XII are described as present lateroventral to the occipital condyle in the original phylogenetic dataset, whether the condition is primitive or derived [30,4245]. However, these exits are usually situated anterolateral to the occipital condyle and lateroventral to the foramen magnum in both theropods with the primitive (e.g., [10,14,58,73,8284]) and derived (e.g., [18,8588]) conditions. Therefore, the description of this character state is revised in the present study. Although the presence of the three foramina for CNs X–XII has been reported in Shaochilong [28] and Giganotosaurus [64], they had been scored as having only two foramina [30,4245]. Considering the lack of justifications about these conflicts in the latter studies [30,4245], these two taxa are regarded as having three foramina for CNs X–XII in the present study. Similarly, although Carcharodontosaurus has been scored as unknown for this character [30,4245], the presence of three foramina was reported previously [21], so that has been re-coded in the present study. Acrocanthosaurus has three foramina within the paracondylar pocket on the left side while having only two on the right side [18]. These three were identified as CNs IX, X + XI and XII [18], so that there are only two exits for CNs X–XII contrary to the derived condition retained in the matrices used in the previous studies [30,4245]. However, it should be noted the combination of a single foramen for CNs IX–XI, which is present on the right side, and two foramina for CN XII is also likely to be, as the latter are usually seen in derived theropods [28] including most allosaurs such as Sinraptor [56], Shaochilong [28], Carcharodontosaurus [23], Giganotosaurus [27] and Murusraptor [31]. In addition, if CN IX has a separate exit, it may open around the columellar recess anterior to the metotic strut as in Carcharodontosaurus [23], Murusraptor [31] and Siamraptor (Fig 4B). Therefore, given the uncertainty in identifying the exact nerve assignments of these foramina, the possibility that the third foramen on the left side of Acrocanthosaurus represents an additional foramen of CN XII cannot be excluded. Accordingly, this taxon was treated as potentially exhibiting both the primitive and derived conditions in the present analysis. As a result, in combination with the presence of the three foramina in Siamraptor, the derived character state is revealed as an unambiguous synapomorphy of Carcharodontosauria convergent with Megalosauroidea, so that the intermediate condition of Acrocanthosaurus might be an autapomorphic feature of this taxon.

The caudal tympanic recess seen in Siamraptor is also known in Acrocanthosaurus and coelurosaurs, but has been reported as absent in a basal allosauroid Allosaurus [14], later-diverging carcharodontosaurids Giganotosaurus and Carcharodontosaurus [21,70], and a later-diverging neovenatorid Murusraptor [31]. Although the true absence of such a recess can be difficult to assess [89], the presence of a distinct caudal tympanic recess remains important in the present context. If the presence of this recess represents a plesiomorphic condition of Allosauroidea, as suggested by Dufeau [70], it appears to have been retained in Carcharodontosauria but secondarily lost in later-diverging taxa of Carcharodontosauridae and Neovenatoridae.

The fusion of the chambers for the caudal diverticulum of the rostral tympanic recess (cRTr) and the medial basioccipital diverticulum (MBd) seen in Siamraptor is also known in ornithomimosaurs, oviraptorosaurs, and therizinosaurs, but is unknown in more basal theropods [70]. Therefore, it can be considered an autapomorphy of Siamraptor among allosauroids, but it is worth considering that it could be a plesiomorphy or synapomorphy of Carcharodontosauria, as it gives some idea to understand the unique condition of Murusraptor. In Murusraptor, the position of cRTr and MBd is occupied by the “basisphenoidal recess”, which has a posteriorly elongated shape as in cRTr and a paired opening at the dorsal apex of the basisphenoid recess as in MBd [31,70]. In the latter scenario, the caudal chamber is originally occupied only by cRTr in Allosauroidea, but occupation by MBd was initiated at the latest in Carcharodontosauria and then completed in Neovenatoridae. For Murusraptor, this scenario is consistent with the presence of only one pair of posteriorly elongated chambers connected only to the basisphenoid recess, but is inconsistent with the lower degree of inflation.

Conclusions

The braincases of Siamraptor recovered in the present study provided unknown aspect of its cranial anatomy, including two diagnostic features, many similarities and several differences with other allosauroids, and many characters enriching the data matrix of the phylogenetic analysis confirming its phylogenetic position as an earliest-branching carcharodontosaur. Detailed comparisons with other allosauroid braincases revealed several synapomorphies of Carcharodontosauria, some of which are inherited by carcharodontosaurids but not by neovenatorids. Some characters observed in the braincases also have functional implications, such as the presence of the specialized thermoregulatory system and adaptations for supporting large head that are shared with carcharodontosaurids. In addition, somewhat upturned head posture was indicated for Siamraptor by the angles of the occiput and the lateral semicircular canal relative to the skull roof. Because the earliest-branching carcharodontosaur Siamraptor is still known from limited elements, future works in Nakhon Ratchasima will provide further information on this taxon and recover unknown aspects of the earliest stage of carcharodontosaurian evolution.

Supporting information

S1 File. Data matrix used in the phylogenetic analysis in TNT format.

https://doi.org/10.1371/journal.pone.0345155.s001

(TNT)

Acknowledgments

This contribution is a part of the JTDP 2007–2014, which was a collaborative research project between FPDM and NRRU. We thank all members participated in this project, who are Yoshikazu Noda, Kazuo Terada, Toru Sekiya, Hirokazu Yukawa (FPDM), Keiichi Kojima (Monju Elementary School, Fukui), Pratueng Jintasakul, Wickanet Songtham, Jaroon Duangkrayom, Nareerat Boonchai, Wilailuck Naksri, Kongkaew Jenjitpaiboon, Nuntida Yunkratok, Wanida Khonchoho, Praphas Chansom, Mayuree Yosklang (NRRU), Yupa Thasod (Chiang Mai University, Thailand), Rattanaphorn Hanta (Suranaree University of Technology, Thailand), Atsushi Yabe (National Science Museum, Japan; previously FPDM) for collecting, and Prasart Anankasa and his team for their patient work at the site. We also thank Yasuyo Deyama (Fukui Prefectural University) for preparing some photos for the figures.

References

  1. 1. Kubo T, Shibata M, Naksri W, Jintasakul P, Azuma Y. The earliest record of Asian Eusuchia from the Lower Cretaceous Khok Kruat Formation of northeastern Thailand. Cretac Res. 2018;82:21–8.
  2. 2. Shibata M, Jintasakul P, Azuma Y. A New Iguanodontian Dinosaur from the Lower Cretaceous Khok Kruat Formation, Nakhon Ratchasima in Northeastern Thailand. Acta Geologica Sinica (Eng). 2011;85(5):969–76.
  3. 3. Shibata M, Jintasakul P, Azuma Y, You H-L. A new basal hadrosauroid dinosaur from the lower Cretaceous Khok Kruat formation in Nakhon Ratchasima Province, Northeastern Thailand. PLoS One. 2015;10(12):e0145904. pmid:26716981
  4. 4. Chokchaloemwong D, Hattori S, Yukawa H, Shibata M, Naksri W. A new ornithomimosaurian material from the Khok Kruat Formation, Lower Cretaceous of Thailand. Mem Fukui Prefect Dinosaur Museum. 2024;23:1–7.
  5. 5. Chokchaloemwong D, Hattori S, Cuesta E, Jintasakul P, Shibata M, Azuma Y. A new carcharodontosaurian theropod (Dinosauria: Saurischia) from the Lower Cretaceous of Thailand. PLoS One. 2019;14(10):e0222489. pmid:31596853
  6. 6. Malafaia E, Mocho P, Escaso F, Ortega F. A new carcharodontosaurian theropod from the Lusitanian Basin: evidence of allosauroid sympatry in the European Late Jurassic. J Vertebr Paleontol. 2020;40(1).
  7. 7. Meesook A. Cretaceous. In: Ridd MF, Barber AJ, Crow MJ, editors. The Geology of Thailand. London: The Geological Society of London. 2011. pp. 169–84.
  8. 8. Dong Z, Zhou S, Zhang Y. Dinosaurs from the Jurassic of Sichuan. Palaeontol Sin. 1983;162:1–136.
  9. 9. Dong ZM, Zhang YH, Li XM, Zhou ZW. A new carnosaur from Yongchuan County, Sichuan Province. Science Newsletter. 1978;23:302–4.
  10. 10. Currie PJ, Zhao X-J. A new carnosaur (Dinosauria, Theropoda) from the Jurassic of Xinjiang, People’s Republic of China. Can J Earth Sci. 1993;30(10):2037–81.
  11. 11. Yuhui G. Yangchuanosaurus hepingensis - a new species of carnosaur from Zigong, Sichuan. Vertebr Palasiat. 1992;30:313–24.
  12. 12. Gao Y. A complete carnosaur skeleton from Zigong, Sichuan: Yangchuanosaurus hepingensis. Chengdu: Sichuan Science and Technology Press; 1999.
  13. 13. Madsen JH. Allosaurus fragilis: a revised osteology. Utah Geol Surv Bull. 1976;109:1–163.
  14. 14. McClelland BK. Anatomy and kinesis of the Allosaurus skull. Texas Tech University; 1990.
  15. 15. Chure DJ, Loewen MA. Cranial anatomy of Allosaurus jimmadseni, a new species from the lower part of the Morrison Formation (Upper Jurassic) of Western North America. PeerJ. 2020;8:e7803. pmid:32002317
  16. 16. Lessner EJ, Cranor C, Hunt-Foster R, Holliday CM. Endocranial anatomy of Allosaurus supports neural trends among non-avian theropod dinosaurs. J Vertebr Paleontol. 2023;43:1–9.
  17. 17. Eddy DR, Clarke JA. New information on the cranial anatomy of Acrocanthosaurus atokensis and its implications for the phylogeny of Allosauroidea (Dinosauria: Theropoda). PLoS One. 2011;6(3):e17932. pmid:21445312
  18. 18. Franzosa J, Rowe T. Cranial endocast of the Cretaceous theropod dinosaur Acrocanthosaurus atokensis. J Vertebr Paleontol. 2005;25:859–64.
  19. 19. Kellermann M, Cuesta E, Rauhut OWM. Re-evaluation of the Bahariya Formation carcharodontosaurid (Dinosauria: Theropoda) and its implications for allosauroid phylogeny. PLoS One. 2025;20(1):e0311096. pmid:39808629
  20. 20. Stromer E. Wirbeltiere-Reste der Baharijestufe (unterestes Canoman). Ein Skelett-Rest von Carcharodontosaurus nov. gen. Abhandlungen der Bayerischen Akademie der Wissenschaften, Mathematische Abteilung, Neue Folge. 1931;9:6–16.
  21. 21. Brusatte SL, Sereno PC. A new species of Carcharodontosaurus (Dinosauria: Theropoda) from the Cenomanian of Niger and a revision of the genus. J Vertebr Paleontol. 2007;27:902–16.
  22. 22. Sereno P, Dutheil D, Iarochene M, Larsson H, Lyon G, Magwene P, et al. Predatory dinosaurs from the sahara and late cretaceous faunal differentiation. Science. 1996;272(5264):986–91. pmid:8662584
  23. 23. Larsson HCE. Endocranial anatomy of Carcharodontosaurus saharicus (Theropoda: Allosauroidea) and its implications for theropod brain evolution. In: Tanke DH, Carpenter K, editors. Mesozoic Vertebrate Life. Bloomington: Indiana University Press; 2001.
  24. 24. Canale JI, Apesteguía S, Gallina PA, Mitchell J, Smith ND, Cullen TM, et al. New giant carnivorous dinosaur reveals convergent evolutionary trends in theropod arm reduction. Curr Biol. 2022;32(14):3195–3202.e5. pmid:35803271
  25. 25. Coria RA, Currie PJ. The braincase of Giganotosaurus carolinii (Dinosauria: Theropoda) from the Upper Cretaceous of Argentina. J Vertebr Paleontol. 2002;22:802–11.
  26. 26. Carabajal AP, Canale JI. Cranial endocast of the carcharodontosaurid theropod Giganotosaurus carolinii Coria Salgado, 1995. Neues Jahrb für Geol und Paläontologie - Abhandlungen. 2010;258:249–56.
  27. 27. Paulina-Carabajal A, Nieto MN. Brief Comment on the Brain and Inner Ear of Giganotosaurus carolinii (Dinosauria: Theropoda) Based on CT Scans. Ameghiniana. 2019;57(1):58.
  28. 28. Brusatte SL, Chure DJ, Benson RBJ, Xu X. The osteology of Shaochilong maortuensis, a carcharodontosaurid (Dinosauria: Theropoda) from the Late Cretaceous of Asia. Zootaxa. 2010;2334(1).
  29. 29. Porfiri JD, Novas FE, Calvo JO, Agnolín FL, Ezcurra MD, Cerda IA. Juvenile specimen of Megaraptor (Dinosauria, Theropoda) sheds light about tyrannosauroid radiation. Cretac Res. 2014;51:35–55.
  30. 30. Coria RA, Currie PJ. A new megaraptoran dinosaur (Dinosauria, Theropoda, Megaraptoridae) from the Late Cretaceous of Patagonia. PLoS One. 2016;11(7):e0157973. pmid:27439002
  31. 31. Paulina-Carabajal A, Currie PJ. The braincase of the theropod dinosaur Murusraptor: osteology, neuroanatomy and comments on the paleobiological implications of certain endocranial features. Ameghiniana. 2017;54(5):617.
  32. 32. Aranciaga Rolando AM, Novas FE, Agnolín FL. A reanalysis of Murusraptor barrosaensis Coria & Currie (2016) affords new evidence about the phylogenetical relationships of Megaraptora. Cretac Res. 2019;99:104–27.
  33. 33. Aranciaga Rolando AM, Motta MJ, Agnolín FL, Manabe M, Tsuihiji T, Novas FE. A large Megaraptoridae (Theropoda: Coelurosauria) from Upper Cretaceous (Maastrichtian) of Patagonia, Argentina. Sci Rep. 2022;12(1):6318. pmid:35474310
  34. 34. Rauhut OWM, Pol D. Probable basal allosauroid from the early Middle Jurassic Cañadón Asfalto Formation of Argentina highlights phylogenetic uncertainty in tetanuran theropod dinosaurs. Sci Rep. 2019;9(1):18826. pmid:31827108
  35. 35. Racey A. Mesozoic red bed sequences from SE Asia and the significance of the Khorat Group of NE Thailand. Geol Soc London, Spec Publ. 2009;315(1):41–67.
  36. 36. Yukawa H, Noda Y, Azuma Y, Suzuki S, Jintasakul P. Sedimentary facies and depositional environment of the Lower Cretaceous Khok Kruat Formation, Nakhon Ratchasima in NE Thailand. In: Abstracts of International Symposium on Asian Dinosaurs in Fukui. 2014. pp. 109–10.
  37. 37. Amiot R, Wang X, Zhou Z, Wang X, Lécuyer C, Buffetaut E, et al. Environment and ecology of East Asian dinosaurs during the Early Cretaceous inferred from stable oxygen and carbon isotopes in apatite. J Asian Earth Sci. 2015;98:358–70.
  38. 38. Cappetta H, Buffetaut E, Suteethorn V. A new hybodont shark from the Lower Cretaceous of Thailand. Neues Jahrb für Geol und Paläontologie - Monatshefte. 1990;1990(11):659–66.
  39. 39. Buffetaut E, Suteethorn V. A new species of the ornithischian dinosaur Psittacosaurus from the Early Cretaceous of Thailand. Palaeontology. 1992;35:801–12.
  40. 40. Buffetaut E, Suteethorn V, Le Loeuff J, Khansubha S, Tong H, Wongko K. The dinosaur fauna from the Khok Kruat Formation (Early Cretaceous) of Thailand. In: Int Conf Geol Geotechnol Miner Resour Indochina (GEOINDO 2005). 2005. pp. 28–30.
  41. 41. Racey A, Goodall JGS. Palynology and stratigraphy of the Mesozoic Khorat Group red bed sequences from Thailand. Geol Soc London, Spec Publ. 2009;315(1):69–83.
  42. 42. Coria RA, Currie PJ, Ortega F, Baiano MA. An early Cretaceous, medium-sized carcharodontosaurid theropod (Dinosauria, Saurischia) from the Mulichinco Formation (upper Valanginian), Neuquén Province, Patagonia, Argentina. Cretac Res. 2019;104:319.
  43. 43. Carrano MT, Benson RBJ, Sampson SD. The phylogeny of Tetanurae (Dinosauria: Theropoda). J Syst Palaeontol. 2012;10:211–300.
  44. 44. Zanno LE, Makovicky PJ. Neovenatorid theropods are apex predators in the Late Cretaceous of North America. Nat Commun. 2013;4:2827. pmid:24264527
  45. 45. Apesteguía S, Smith ND, Juárez Valieri R, Makovicky PJ. An Unusual New Theropod with a Didactyl Manus from the Upper Cretaceous of Patagonia, Argentina. PLoS One. 2016;11(7):e0157793. pmid:27410683
  46. 46. Maddison WP, Maddison DR. Mesquite: a modular system for evolutionary analysis. 2019. Available from: http://www.mesquiteproject.org
  47. 47. Goloboff PA, Catalano SA. TNT version 1.5, including a full implementation of phylogenetic morphometrics. Cladistics. 2016;32(3):221–38. pmid:34727670
  48. 48. Goloboff PA, Farris JS, Nixon KC. TNT, a free program for phylogenetic analysis. Cladistics. 2008;24(5):774–86.
  49. 49. Owen R. Report on British fossil reptiles, part II. Reports Br Assoc Adv Sci. 1842;11:60–204.
  50. 50. Marsh OC. A new order of Jurassic reptiles (Coeluria). Am J Sci. 1881;21:339–40.
  51. 51. Gauthier JA. Saurischian monophyly and the origin of birds. Mem Calif Acad Sci. 1986;8:1–55.
  52. 52. Marsh OC. Notice of new dinosaurian reptiles. Am J Sci Arts. 1878;14:14–516.
  53. 53. Benson RBJ, Carrano MT, Brusatte SL. A new clade of archaic large-bodied predatory dinosaurs (Theropoda: Allosauroidea) that survived to the latest Mesozoic. Naturwissenschaften. 2010;97(1):71–8. pmid:19826771
  54. 54. Sereno PC, Brusatte SL. Basal abelisaurid and carcharodontosaurid theropods from the lower Cretaceous Elrhaz Formation of Niger. Acta Palaeontologica Polonica. 2008;53(1):15–46.
  55. 55. Cau A, Dalla Vecchia FM, Fabbri M. A thick-skulled theropod (Dinosauria, Saurischia) from the Upper Cretaceous of Morocco with implications for carcharodontosaurid cranial evolution. Cretac Res. 2013;40:251–60.
  56. 56. Paulina-Carabajal A, Currie PJ. New information on the braincase of Sinraptor dongi (Theropoda: Allosauroidea): ethmoidal region, endocranial anatomy, and pneumaticity. Vertebr Palasiat. 2012;50:85–101.
  57. 57. Stovall JW, Langston W. Acrocanthosaurus atokensis, a new genus and species of Lower Cretaceous Theropoda from Oklahoma. Am Midl Nat. 1950;43:696.
  58. 58. Sampson SD, Witmer LM. Craniofacial anatomy of Majungasaurus crenatissimus (Theropoda: Abelisauridae) from the Late Cretaceous of Madagascar. J Vertebr Paleontol. 2007;27:32–102.
  59. 59. Currie PJ. Braincase Anatomy. In: Currie PJ, Padian K, editors. Encyclopedia of Dinosaurs. New York: Academic Press; 1997. pp. 81–4.
  60. 60. Benson RBJ. A description of Megalosaurus bucklandii (Dinosauria: Theropoda) from the Bathonian of the UK and the relationships of Middle Jurassic theropods. Zool J Linn Soc. 2010;158:882–935.
  61. 61. Barker CT, Hone DWE, Naish D, Cau A, Lockwood JAF, Foster B, et al. New spinosaurids from the Wessex Formation (Early Cretaceous, UK) and the European origins of Spinosauridae. Sci Rep. 2021;11(1):19340. pmid:34588472
  62. 62. Welles SP. Dilophosaurus wetherilli (Dinosauria, Theropoda): osteology and comparisons. Palaeontogr Abteilung A. 1984;185:85–180.
  63. 63. Chure DJ. A new species of Allosaurus from the Morrison Formation of Dinosaur National Monument (UT-CO) and a revision of the theropod family Allosauridae. Columbia University; 2000.
  64. 64. Paulina-Carabajal A. El neurocráneo de los dinosaurios Theropoda de la Argentina. Osteología y sus implicancias filogenéticas. Universidad Nacional de La Plata; 2009.
  65. 65. Franzosa JW. Evolution of the brain in Theropoda (Dinosauria). The University of Texas at Austin; 2004.
  66. 66. Hopson JA. Paleoneurology. In: Gans C, Glenn R, Ulinski P, editors. Biology of the Reptilia Volume 9 Neurology A. London, New York, and San Francisco: Academic Press; 1979. pp. 39–146.
  67. 67. Witmer LM, Ridgely RC. New insights into the brain, braincase, and ear region of tyrannosaurs (Dinosauria, Theropoda), with implications for sensory organization and behavior. Anat Rec (Hoboken). 2009;292(9):1266–96. pmid:19711459
  68. 68. Currie PJ. Cranial anatomy of Stenonychosaurus inequalis (Saurischia, Theropoda) and its bearing on the origin of birds. Can J Earth Sci. 1985;22(11):1643–58.
  69. 69. McKeown M, Brusatte SL, Williamson TE, Schwab JA, Carr TD, Butler IB, et al. Neurosensory and Sinus Evolution as Tyrannosauroid Dinosaurs Developed Giant Size: Insight from the Endocranial Anatomy of Bistahieversor sealeyi. Anat Rec (Hoboken). 2020;303(4):1043–59. pmid:31967416
  70. 70. Dufeau DL. The evolution of cranial pneumaticity in Archosauria: patterns of paratympanic sinus. Ohio University; 2011.
  71. 71. Holliday CM, Porter WR, Vliet KA, Witmer LM. The frontoparietal fossa and dorsotemporal fenestra of archosaurs and their significance for interpretations of vascular and muscular anatomy in dinosaurs. Anat Rec (Hoboken). 2020;303(4):1060–74. pmid:31260177
  72. 72. Tsuihiji T, Watabe M, Tsogtbaatar K, Tsubamoto T, Barsbold R, Suzuki S, et al. Cranial osteology of a juvenile specimen of Tarbosaurus bataar (Theropoda, Tyrannosauridae) from the Nemegt Formation (Upper Cretaceous) of Bugin Tsav, Mongolia. Journal of Vertebrate Paleontology. 2011;31(3):497–517.
  73. 73. Sanders RK, Smith DK. The endocranium of the theropod dinosaur Ceratosaurus studied with computed tomography. Acta Palaeontol Pol. 2005;50:601–16.
  74. 74. Nieto MN, Paulina-Carabajal A. 3D models related to the publication: Brief comment on the brain and inner ear of Giganotosaurus carolinii (Dinosauria: Theropoda) based on CT scans. M3. 2020;6(2):e108.
  75. 75. Benoit J, Legendre LJ, Farke AA, Neenan JM, Mennecart B, Costeur L, et al. A test of the lateral semicircular canal correlation to head posture, diet and other biological traits in “ungulate” mammals. Sci Rep. 2020;10(1):19602. pmid:33177568
  76. 76. Cerroni MA, Canale JI, Novas FE. The skull of Carnotaurus sastrei Bonaparte 1985 revisited: insights from craniofacial bones, palate and lower jaw. Hist Biol. 2021;33:2444–85.
  77. 77. Bonaparte JF, Novas FE, Coria RA. Carnotaurus sastrei Bonaparte, the horned, lightly built carnosaur from the Middle Cretaceous of Patagonia. Contrib Sci. 1990;1–42.
  78. 78. Tykoski RS, Rowe T. Ceratosauria. In: Weishampel DB, Dodson P, Osmólska H, editors. The Dinosauria. 2nd ed. Berkeley: University of California Press; 2004. pp. 47–70.
  79. 79. Snively E, Russell AP. Functional variation of neck muscles and their relation to feeding style in Tyrannosauridae and other large theropod dinosaurs. Anat Rec (Hoboken). 2007;290(8):934–57. pmid:17654673
  80. 80. Tsuihiji T. Reconstructions of the axial muscle insertions in the occipital region of dinosaurs: evaluations of past hypotheses on marginocephalia and tyrannosauridae using the extant phylogenetic bracket approach. Anat Rec (Hoboken). 2010;293(8):1360–86. pmid:20665814
  81. 81. Taquet P, Welles SP. Redescription du crâne de dinosaure théropode de Dives (Normandie). Annales de Paléontologie. 1977;63:191–206.
  82. 82. Carrano MT, Loewen MA, Sertich JJW. New materials of Masiakasaurus knopfleri Sampson, Carrano, and Forster, 2001, and implications for the morphology of the Noasauridae (Theropoda:Ceratosauria). Smithson Contrib to Paleobiol. 2011;95:1–53.
  83. 83. Charig AJ, Milner AC. Baryonyx walkeri, a fish-eating dinosaur from the Wealden of Surrey. Bull Nat Hist Museum London. 1996;53:11–70.
  84. 84. Sues HD, Frey E, Martill DM, Scott DM. Irritator challengeri, a spinosaurid (Dinosauria: Theropoda) from the Lower Cretaceous of Brazil. J Vertebr Paleontol. 2002;22:535–47.
  85. 85. Marsh AD, Rowe TB. A comprehensive anatomical and phylogenetic evaluation of Dilophosaurus wetherilli (Dinosauria, Theropoda) with descriptions of new specimens from the Kayenta Formation of northern Arizona. J Paleontol. 2020;94:1–103.
  86. 86. Rauhut OW. Braincase structure of the Middle Jurassic theropod dinosaur Piatnitzkysaurus. Can J Earth Sci. 2004;41(9):1109–22.
  87. 87. Sadleir R, Barrett PM, Powell HP. The anatomy and systematics of Eustreptospondylus oxoniensis, a theropod dinosaur from the middle Jurassic of Oxfordshire, England. Monogr Palaeontogr Soc. 2008;160:1–82.
  88. 88. Allain R. Discovery of megalosaur (Dinosauria, Theropoda) in the middle Bathonian of Normandy (France) and its implications for the phylogeny of basal Tetanurae. J Vertebr Paleontol. 2002;22:548–63.
  89. 89. Bronzati M, Langer MC, Rauhut OWM. Braincase anatomy of the early sauropodomorph Saturnalia tupiniquim (Late Triassic, Brazil). J Vertebr Paleontol. 2018;38:e1559173.