identifier	taxonID	type	CVterm	format	language	title	description	additionalInformationURL	UsageTerms	rights	Owner	contributor	creator	bibliographicCitation
871D87BB6D7AFFD9FC207CE0FB41FA30.text	871D87BB6D7AFFD9FC207CE0FB41FA30.taxon	http://purl.org/dc/dcmitype/Text	http://rs.tdwg.org/ontology/voc/SPMInfoItems#GeneralDescription	text/html	en	Francosuchus (Kuhn 1933)	<div><p>‘ FRANCOSUCHUS ’ KUHN, 1933</p> <p>[NOMEN DUBIUM]</p> <p>Type species: Francosuchus broilii Kuhn, 1933.</p> <p>Comments: Because the genotype specimen is lost (see below) and no casts are known, and the original description and figures (see Fig. 2A, B, D) of ‘ Francosuchus broilii ’ do not allow it to be distinguished from Paleorhinus or Parasuchus, we here consider ‘ Francosuchus broilii ’ and ‘ Francosuchus ’ to be nomina dubia (see below).</p> </div>	https://treatment.plazi.org/id/871D87BB6D7AFFD9FC207CE0FB41FA30	Public Domain	No known copyright restrictions apply. See Agosti, D., Egloff, W., 2009. Taxonomic information exchange and copyright: the Plazi approach. BMC Research Notes 2009, 2:53 for further explanation.		Plazi	Butler, Richard J.;Rauhut, Oliver W. M.;Stocker, Michelle R.;Bronowicz, Robert	Butler, Richard J., Rauhut, Oliver W. M., Stocker, Michelle R., Bronowicz, Robert (2014): Redescription of the phytosaurs Paleorhinus (‘ Francosuchus’) angustifrons and Ebrachosuchus neukami from Germany, with implications for Late Triassic biochronology. Zoological Journal of the Linnean Society 170 (1): 155-208, DOI: 10.1111/zoj.12094, URL: http://doi.wiley.com/10.1111/zoj.12094
871D87BB6D7AFFDBFC767E54FE9DFDB4.text	871D87BB6D7AFFDBFC767E54FE9DFDB4.taxon	http://purl.org/dc/dcmitype/Text	http://rs.tdwg.org/ontology/voc/SPMInfoItems#GeneralDescription	text/html	en	Francosuchus broilii O. Kuhn 1933	<div><p>‘ FRANCOSUCHUS BROILII ’ KUHN, 1933</p> <p>[NOMEN DUBIUM]</p> <p>‘ Francosuchus broilii nov. spec. ’; Kuhn, 1933: 97, 127, fig. 6. pl. 5, figs 1, 2</p> <p>‘ Francosuchus broilii ’; Kuhn, 1936: 65</p> <p>‘ Francosuchus broilii Kuhn’; Kuhn, 1938: 318</p> <p>‘ P. broilii (Kuhn, 1933) ’; Gregory, 1962: 670</p> <p>‘ P. (F.) broilii O. Kuhn, 1933 ’; Westphal, 1976: 109</p> <p>‘ Francosuchus broilii Kuhn, 1932 ’; Chatterjee, 1978: 115 ‘ Paleorhinus broilii Kuhn, 1932 ’; Hunt &amp; Lucas, 1991: 489</p> <p>‘ Paleorhinus sp. ’; Long &amp; Murry, 1995: 36</p> <p>Holotype: BSPG lost specimen, skull lacking the lower jaws and damaged at its right posterolateral corner, presumed destroyed in World War II (Kuhn, 1933: fig. 6. pl. 5, figs 1, 2; Fig. 2A, B, D).</p> <p>Locality and horizon: Discovered in 1931 in bed 13 of Kuhn (1933, 1936), Ebrach quarry, Bamberg district, Upper Franconia (Oberfranken) region of northern Bavaria, south-eastern Germany. Blasensandstein of the Sandsteinkeuper, laterally equivalent to the lower Kieselsandstein (lower Hassberge Formation) of the Middle Keuper (Late Triassic: late Carnian, Tuvalian).</p> <p>Comments: The loss of the holotype and the absence of known casts mean that assessing the status of ‘ Francosuchus broilii ’ is difficult. The only available images are two photographs in dorsal and ventral views and one single schematic outline drawing in left lateral view (Kuhn, 1933: fig. 6. pl. 5, figs 1, 2; Fig. 2A, B, D), and a description of the specimen was given by Kuhn (1933: 123–127). These images and Kuhn’s description indicate a skull missing its right posterolateral part, the margin of the left external naris, and much of the palate. Although Kuhn (1933: 123) suggested that the rostrum was complete, Kuhn (1936: 65) noted that the rostrum was collected in two parts that did not fit together directly, and that the skull proportions of this species were therefore uncertain (see also Gregory, 1962; Long &amp; Murry, 1995). Kuhn (1933) also indicated that most sutures could not be observed. Kuhn (1933) did not provide an explicit diagnosis of the species, but his diagnosis of the genus Francosuchus (Kuhn, 1933: 131) does not include any clearly autapomorphic features, and in proportions and general shape the skull appears to have been similar to the holotype of ‘ F.’ angustifrons (see below), as well as to other non-phytosaurid phytosaurs such as Parasuchus and Paleorhinus (Chatterjee, 1978; Stocker, 2010). The skull was clearly different in morphology from the holotype specimen of Ebrachosuchus neukami, particularly with regard to the lack of an anteroposteriorly extended infratemporal fenestra (see below). We are unable to distinguish ‘ Francosuchus broilii ’ with confidence from most other non-phytosaurid phytosaurs, particularly Paleorhinus bransoni (Long &amp; Murry, 1995; Stocker, 2010) and Parasuchus hislopi (Chatterjee, 1978). For these reasons, we consider ‘ Francosuchus broilii ’ and the genus ‘ Francosuchus ’ to be nomina dubia referable to Phytosauria.</p> </div>	https://treatment.plazi.org/id/871D87BB6D7AFFDBFC767E54FE9DFDB4	Public Domain	No known copyright restrictions apply. See Agosti, D., Egloff, W., 2009. Taxonomic information exchange and copyright: the Plazi approach. BMC Research Notes 2009, 2:53 for further explanation.		Plazi	Butler, Richard J.;Rauhut, Oliver W. M.;Stocker, Michelle R.;Bronowicz, Robert	Butler, Richard J., Rauhut, Oliver W. M., Stocker, Michelle R., Bronowicz, Robert (2014): Redescription of the phytosaurs Paleorhinus (‘ Francosuchus’) angustifrons and Ebrachosuchus neukami from Germany, with implications for Late Triassic biochronology. Zoological Journal of the Linnean Society 170 (1): 155-208, DOI: 10.1111/zoj.12094, URL: http://doi.wiley.com/10.1111/zoj.12094
871D87BB6D78FFDBFEE67ACFFCCBF932.text	871D87BB6D78FFDBFEE67ACFFCCBF932.taxon	http://purl.org/dc/dcmitype/Text	http://rs.tdwg.org/ontology/voc/SPMInfoItems#GeneralDescription	text/html	en	Francosuchus latus O. Kuhn 1933	<div><p>‘ FRANCOSUCHUS LATUS ’ KUHN, 1933</p> <p>[NOMEN DUBIUM]</p> <p>‘ Francosuchus latus nov. spec. ’; Kuhn, 1933: 97, 127, fig. 5 ‘ Francosuchus latus ’; Kuhn, 1936: 69 ‘ Francosuchus latus Kuhn’; Kuhn, 1938: 318 ‘ P. broilii (Kuhn, 1933) ’; Gregory, 1962: 670 ‘ P. (F.) latus O. Kuhn, 1933 ’; Westphal, 1976: 109 ‘ Francosuchus latus Kuhn, 1932 ’; Hunt &amp; Lucas, 1991: 489 ‘ Paleorhinus sp. ’; Long &amp; Murry 1995: 36</p> <p>Holotype: BSPG lost specimen, skull missing the rostrum, missing and presumed destroyed in World War II (Kuhn, 1933: fig. 5: Fig. 2C).</p> <p>Locality and horizon: Bed 13 of Kuhn (1933, 1936), Ebrach quarry.</p> <p>Comments: The holotype consisted of a partial skull lacking the rostrum. Unfortunately the only available figure of the skull (Kuhn, 1933: fig. 5; Fig. 2C), showing the specimen in dorsal view, is very indistinct and does not allow a detailed assessment of morphology or the recognition of autapomorphic characters, although it does indicate a phytosaur skull broadly similar in morphology to the holotype of ‘ F. broilii ’, the holotype of ‘ F.’ angustifrons (see below), as well as to other non-phytosaurid phytosaurs such as Parasuchus and Paleorhinus (Chatterjee, 1978; Stocker, 2010). Moreover, the description of the specimen by Kuhn (1933: 127–129) does not clearly distinguish diagnostic features. As a result, we consider ‘ F. latus ’ a nomen dubium referable to Phytosauria.</p> </div>	https://treatment.plazi.org/id/871D87BB6D78FFDBFEE67ACFFCCBF932	Public Domain	No known copyright restrictions apply. See Agosti, D., Egloff, W., 2009. Taxonomic information exchange and copyright: the Plazi approach. BMC Research Notes 2009, 2:53 for further explanation.		Plazi	Butler, Richard J.;Rauhut, Oliver W. M.;Stocker, Michelle R.;Bronowicz, Robert	Butler, Richard J., Rauhut, Oliver W. M., Stocker, Michelle R., Bronowicz, Robert (2014): Redescription of the phytosaurs Paleorhinus (‘ Francosuchus’) angustifrons and Ebrachosuchus neukami from Germany, with implications for Late Triassic biochronology. Zoological Journal of the Linnean Society 170 (1): 155-208, DOI: 10.1111/zoj.12094, URL: http://doi.wiley.com/10.1111/zoj.12094
871D87BB6D78FFDBFEC27F51FC42FD0E.text	871D87BB6D78FFDBFEC27F51FC42FD0E.taxon	http://purl.org/dc/dcmitype/Text	http://rs.tdwg.org/ontology/voc/SPMInfoItems#GeneralDescription	text/html	en	Paleorhinus WILLISTON 1904	<div><p>PALEORHINUS WILLISTON, 1904</p> <p>Type species: Paleorhinus bransoni Williston, 1904. Diagnosis: Characterized by the following autapomorphies: discrete anteroposteriorly extending row of 6–7 slightly raised nodes on the lateral surface of the jugal; low paired ridges on lateral surface of squamosal enclosing a shallow, slit-like fossa.</p> <p>Comments: Paleorhinus is a historically problematic wastebasket taxon that is in need of a full revision that is beyond the scope of the present paper. Here, we merely report characters that appear to be uniquely shared between the genotype species, Paleorhinus bransoni, and ‘ Francosuchus ’ angustifrons and thus support referral of the latter to the genus Paleorhinus (see Phylogenetic Analysis, below). An assessment of whether Paleorhinus is a subjective junior synonym of Parasuchus is beyond the scope of this work (see above).</p> </div>	https://treatment.plazi.org/id/871D87BB6D78FFDBFEC27F51FC42FD0E	Public Domain	No known copyright restrictions apply. See Agosti, D., Egloff, W., 2009. Taxonomic information exchange and copyright: the Plazi approach. BMC Research Notes 2009, 2:53 for further explanation.		Plazi	Butler, Richard J.;Rauhut, Oliver W. M.;Stocker, Michelle R.;Bronowicz, Robert	Butler, Richard J., Rauhut, Oliver W. M., Stocker, Michelle R., Bronowicz, Robert (2014): Redescription of the phytosaurs Paleorhinus (‘ Francosuchus’) angustifrons and Ebrachosuchus neukami from Germany, with implications for Late Triassic biochronology. Zoological Journal of the Linnean Society 170 (1): 155-208, DOI: 10.1111/zoj.12094, URL: http://doi.wiley.com/10.1111/zoj.12094
871D87BB6D78FFC5FC8D7B75FB41FB72.text	871D87BB6D78FFC5FC8D7B75FB41FB72.taxon	http://purl.org/dc/dcmitype/Text	http://rs.tdwg.org/ontology/voc/SPMInfoItems#GeneralDescription	text/html	en	Paleorhinus angustifrons (KUHN 1936)	<div><p>PALEORHINUS ANGUSTIFRONS (KUHN, 1936)</p> <p>‘ Francosuchus angustifrons n. sp. ’; Kuhn, 1936: 69, textfig. 1–3, pl. 9, fig. 2a, b, pl. 10, figs 3, 5 ‘ Francosuchus angustifrons Kuhn’; Kuhn, 1938: 318 ‘ Francosuchus angustifrons ’; Huene, 1939: 142 ‘ Ebrachosuchus angustifrons ’; Huene, 1939: fig. 2 ‘ P. broilii (Kuhn, 1933) ’; Gregory, 1962: 670 ‘ P. (F.) angustifrons O. Kuhn, 1936 ’; Westphal, 1976: 109, fig. 7c ‘ P. neukami Kuhn, 1936 ’; Hunt &amp; Lucas, 1991: 489, 494 ‘ Paleorhinus sp. ’; Long &amp; Murry 1995: 36</p> <p>Holotype: BSPG 1931 X 502, skull missing the anterior portions of the premaxillae (Kuhn, 1936: textfig. 1–3, pl. 9, fig. 2a, b, pl. 10, figs 3, 5; Figs 3–8).</p> <p>Locality and horizon: Bed 9 (Kuhn, 1933, 1936) of the Ebrach quarry, Bamberg district, Upper Franconia region of northern Bavaria, Germany. Blasensandstein of the Sandsteinkeuper, laterally equivalent to the Hassberge Formation of the Middle Keuper (Late Triassic: late Carnian).</p> <p>Diagnosis: Species of Paleorhinus characterized by the following autapomorphic features: (1) stepped lateral rim of external naris that is strongly swollen and rugose at posterior end; (2) paired depressions on the anterior portions of the nasals (immediately posterior to the external nares) and anterior portions of the frontals; (3) foramen in ectopterygoid enlarged and subcircular in outline; (4) suborbital foramen elongate and boomerang shaped; (5) large postparietal foramen at junction between supraoccipital and parietal.</p> <p>Comments: Several of the features identified here as autapomorphic for Paleorhinus angustifrons (characters 1, 2, 5) are also known in phytosaur material from Krasiejów in southern Poland (e.g. ZPAL Ab III 111, 200; Dzik, 2001) that is currently under study by R.B. If the Krasiejów phytosaur material proves to represent a species distinct from P. angustifrons then the status of these character states will probably change, to synapomorphies supporting a sistergroup relationship between P. angustifrons and the Krasiejów phytosaur.</p> <p>DESCRIPTION OF HOLOTYPE OF PALEORHINUS ANGUSTIFRONS</p> <p>General: Although generally well preserved, the surface of the skull (Figs 3–8) has been reconstructed in some places (e.g. left ectopterygoid, pterygoid wing of right quadrate, much of the surface of the skull roof adjacent to the left orbit). Most of the cranial sutures are tightly closed or even fused. The skull as a whole is dorsoventrally shallow (Figs 3D, 4D), as is typical of early phytosaurs generally (e.g. Langston, 1949; Stocker, 2010), although this has undoubtedly been exaggerated by substantial dorsoventral crushing (for example, the occipital condyle appears to be strongly flattened dorsoventrally). As preserved, the skull measures approximately 320 mm from the broken anterior tip of the rostrum to the most posteriorly positioned part of the squamosals, and 184 mm transversely at the lateral extents of the quadrate condyles. The rostrum is broken a short distance anterior to the external nares, and most of the premaxilla has been lost. It is not possible to accurately reconstruct the missing length of the premaxilla, although com- parison with the Krasiejów phytosaur specimen ZPAL Ab III 200 suggests a skull length of approximately 500 mm. Most external skull elements have a weakly sculptured surface, although the rather smooth appearance might at least partially be due to preservation and/or preparation.</p> <p>Fenestral morphology: The dorsally directed external nares are teardrop shaped in dorsal view (Figs 3A, 4A, 5A: ‘en’), being rounded posteriorly and tapering anteriorly (each is approximately 40 mm in total length, and 13 mm wide at the widest point). Their</p> <p>◀</p> <p>posterior margins are set substantially anterior (c. 25 mm) to the anterior corners of the internal antorbital fenestrae (sensu Witmer, 1997a), and to the anterior corners of the antorbital fossae (separated by c. 20 mm). This is similar to the condition in Paleorhinus bransoni (FMNH UC 632; Lees, 1907; Stocker, 2010), Parasuchus hislopi (Chatterjee, 1978), Wannia scurriensis (TTU P-00539; Langston, 1949; Stocker, 2010, 2013), and in phytosaur specimens from Krasiejów (e.g. ZPAL Ab III 200, 1943, Dzik &amp; Sulej, 2007). In all of these taxa the nares are also set distinctly anterior to the fossa. By contrast, in Ebrachosuchus neukami, the anterior margins of the internal antorbital fenestrae are positioned very close to the posterior rims of the external nares (separated by only c. 7 mm; BSPG 1931 X 501). In phytosaurid phytosaurs the posterior rims of the external nares are positioned posterior to the anterior corners of the internal antorbital fenestrae (Stocker, 2010). In lateral view (Figs 3D, 4D), the dorsal rims of the external nares of Paleorhinus angustifrons are positioned below the level of the frontoparietal skull table, although they expand dorsally slightly above the level of the surface of the nasals.</p> <p>The internal antorbital fenestra (Figs 3A, D, 4A, D, 5C: ‘afen’) faces dorsolaterally and has a lenticular or eye-shaped outline (c. 58 mm long, 21 mm deep). In dorsal view (Figs 3A, 4A), the long axis of the fenestra extends subparallel to the skull midline. It is surrounded along nearly its entire margin by an antorbital fossa (Figs 3A, D, 4A, D, 5C: ‘afos’), which is bounded by an external antorbital fenestra (sensu Witmer, 1997a). The antorbital fossa is excavated into the maxilla anterodorsally and anteroventrally, and into the lacrimal posterodorsally and posteroventrally. The fossa has a maximum length of 80 mm, and its posterior expansion onto the lacrimal is more than twice as long as the anterior expansion onto the maxilla (Fig. 5C). Ventrally, the fossa is discontinuous at the anteroventral border of the internal antorbital fenestra (i.e. there is a short section of the border of the internal antorbital fenestra that does not have an adjacent fossa), but dorsally, the anterior and posterior expansions of the fossa are connected by a broad depressed area on the ascending process of the maxilla and the anterior process of the lacrimal (Fig. 5C). The nasal suture forms the dorsal rim of the antorbital fossa. As far as can be determined, the jugal reaches the posteroventral rim of the antorbital fossa, but not the rim of the internal antorbital fenestra, so that the external rim of the internal antorbital fenestra appears to be formed only by the maxilla and lacrimal (Fig. 4D). The antorbital fossa is similarly well developed in Paleorhinus bransoni (TMM 31100- 101; Stocker, 2010), Parasuchus hislopi (Chatterjee, 1978), and Wannia scurriensis (TTU P-00539; Langston, 1949; Stocker, 2010, 2013), but is strongly reduced in Ebrachosuchus neukami, in which it is almost limited to the lacrimal at the posterior margin of the internal antorbital fenestra (BSPG 1931 X 501; see below). The antorbital fossa is reduced or absent in ‘ Paleorhinus ’ sawini and Phytosauridae (Stocker, 2010).</p> <p>The dorsolaterally facing orbit (Figs 3A, D, 4A, D: ‘orb’) is ovoid in outline, being moderately longer than deep (left orbit measured at widest points: anteroposterior length = 43 mm; dorsoventral depth = 33 mm) and with rounded anterior and posterior margins. As in other non-phytosaurid phytosaurs (Ebrachosuchus neukami, BSPG 1931 X 501; Paleorhinus bransoni, TMM 31100-101, Stocker, 2010; Parasuchus hislopi, Chatterjee, 1978; the Krasiejów phytosaur specimens, ZPAL Ab III 200, Dzik &amp; Sulej, 2007) the ventral margin of the orbit is slightly ventral to the dorsal margin of the internal antorbital fenestra (Figs 3D, 4D), whereas the orbit is distinctly elevated in leptosuchomorph phytosaurs such as Pravusuchus (AMNH FR 30646) and Mystriosuchus (GPIT 261/001; Stocker, 2010: fig. 7), probably as a result of the proportionally deeper skulls of these taxa.</p> <p>The dorsolaterally and slightly anteriorly (because of the transverse expansion of the skull across the ventral margin of the quadrates) facing infratemporal fenestra (Figs 3A, D, 4A, D, 6A: ‘itf ’) has a trapezoidal outline, being slightly longer anteroposteriorly at its ventral margin than at its dorsal margin (anteroposterior length along ventral edge = 41 mm; anteroposterior length along dorsal edge = 34 mm; dorsoventral height at midpoint of fenestra, measured at 90° to the jugal/quadratojugal bar = 41 mm). Its margins are formed by the quadratojugal, jugal, postorbital, and squamosal. The dorsal margin of the infratemporal fenestra is positioned approximately at the level of the midpoint of the orbit (i.e. the rear part of the skull is slightly downturned relative to the orbital region); at its anteroventral corner the infratemporal fenestra extends beneath the most posterior part of the orbit. The shape of the infratemporal fenestra is similar to that of Paleorhinus bransoni (TMM 31100-101; Stocker, 2010), ‘ Paleorhinus ’ sawini (TMM 31213-16; Long &amp; Murry, 1995), the Krasiejów phytosaurs (ZPAL Ab III 200, Dzik &amp; Sulej, 2007), and Parasuchus hislopi (Chatterjee, 1978), but it is markedly different from the autapomorphic, highly anteroposteriorly elongated infratemporal fenestra of Ebrachosuchus neukami (BSPG 1931 X 501; see below), and also differs from the dorsoventrally elongated fenestra seen in many phytosaurid phytosaurs (e.g. Hungerbühler, 2002; Stocker, 2010).</p> <p>The dorsally facing supratemporal fenestra has a subtriangular outline that tapers to an apex posterolaterally (Figs 3A, 4A, 5F: ‘stf ’). As is usual in archosaurs, the supratemporal fenestra is bordered anteriorly and anterolaterally by the postorbital, anteromedially, medially and posteriorly by the parietal, and posterolaterally by the squamosal. As in other non-phytosaurid phytosaurs, as well as Angistorhinus (TMM 31098-1) and Brachysuchus megalodon (UMMP 10336), the parietal–squamosal bars are in the same horizontal plane as the skull roof (Stocker, 2010). Moreover, the supratemporal fenestrae are completely open dorsally, unlike the condition in most leptosuchomorph taxa in which they are partially or entirely obscured by expansions of the postorbital and/or squamosal (e.g. Pravusuchus, AMNH FR 30646; Stocker, 2010). However, a supratemporal fossa anterior to the fenestra, as present in, for example, Euparkeria (SAM-PK-5867), is absent (Fig. 5F), with the posterior and medial edges of the postorbital, the anterolateral, lateral and posterolateral edges of the parietal, and the anteromedial edge of the squamosal forming a sharp rim. This sharply defined rim of the supratemporal fenestra is only interrupted in its posterolateral corner, where the supratemporal fenestra opens onto a slight dorsal depression on the dorsal surface of the squamosal.</p> <p>The posttemporal fenestra (Figs 3C, 4C, 6C: ‘ptf’) is present on the occipital surface approximately halfway between the foramen magnum and the lateral tip of the paroccipital process, being placed entirely dorsal to the level of the foramen magnum. It is elongate and slit like, with margins formed by the otooccipital ventrally, and the squamosal and supraoccipital dorsally, with the parietal being excluded from its margin by a small contact between the latter bones along its dorsal rim.</p> <p>A small postparietal foramen (Figs 3C, 4C, 6C: ‘ppfor’) is present at the dorsal extremity of the midline of the occipital surface, and is bordered by the parietals (dorsally and laterally) and the supraoccipital (ventrally). Whereas the ventral margin of this fenestra is straight, its dorsal margin is strongly arched. The foramen connects to what appears to be a sinus positioned above the posterior part of the endocranial cavity (visible in CT data). A similar foramen is also present in one of the Krasiejów phytosaur specimens (ZPAL Ab III 200), but has not been reported to our knowledge for any other phytosaur, and it is considered here as an autapomorphy of Paleorhinus angustifrons. Harris (2006) and Balanoff, Bever &amp; Ikejiri (2010) discussed a postparietal foramen in a similar position in several sauropods that is proposed to be associated with the dural venous sinus (Balanoff et al., 2010). Openings in a similar position have also been described in the small ornithopod Dysalotosaurus; in this species, a single, rather large foramen in juveniles becomes subdivided into two smaller foramina by a posterior process of the parietals during ontogeny (Pompeckj, 1920; Janensch, 1955). The same variation might also be present in phytosaurs, because a second skull from Krasiejów (ZPAL Ab III 111) appears to show a subdivided foramen.</p> <p>The foramen magnum (Figs 3C, 4C, 6C: ‘fm’) is a large, circular opening that is mostly bordered by the otooccipital, with only a small portion of the dorsal margin made up by the supraoccipital, and a small part of the floor possibly being contributed by the basioccipital.</p> <p>The subtemporal fenestrae (Fig. 3B, 4B: ‘subtf ’) are large and posterolaterally placed. They are elongate and ovoid in outline, being slightly transversely wider anteriorly than posteriorly. The margin of each subtemporal fenestra is formed by the ectopterygoid anteriorly, the pterygoid anteromedially and medially, the quadrate posteromedially and posteriorly, and the quadratojugal and jugal laterally. The lateral process of the pterygoid and the main body of the ectopterygoid ventrally underlie the anterior end of the fenestra.</p> <p>The suborbital fenestra (= postpalatine fenestra of Sereno, 1991; = suborbital foramen of Stocker, 2010) is elongate (Figs 3B, 4B, 7C, D: ‘sub’) and is similar to that of the Krasiejów phytosaur specimens (ZPAL Ab III 200, Dzik &amp; Sulej, 2007), but it is curved more strongly along its length in BSPG 1931 X 502, with a boomerang shape. This boomerang shape appears to be autapomorphic for Paleorhinus angistifrons. Its margins are formed by the ectopterygoid and maxilla laterally, and the palatine medially. The choanae (Figs 3B, 4B, 7C: ‘cho’) are larger but also slit-like, with rounded anterior and tapering posterior ends (although the posterior margins are not completely preserved), and they diverge slightly from one another anteriorly. Their anterolateral margins are formed by the maxillae, their lateral margins by the palatines, and their medial and posteriormost margins by the posteriorly expanding vomers. The choanae are positioned medial to the antorbital fenestra and posterior to the external nares.</p> <p>Premaxilla: Only the most posterior parts of the premaxillae are preserved (Figs 3, 4: ‘pm’); on the right side there appear to be no parts of the premaxillary tooth row preserved, whereas on the left the most anterior three of the preserved alveoli appear to be in the premaxilla. On the dorsal surface of the skull the premaxilla contacts the maxilla, the ‘septomaxilla’, and the nasal. Externally, the premaxillae form the midline of the rostrum anterior to the external nares, and each premaxilla possesses a short posteromedial process that projects between the anterior parts of the ‘septomaxillae’ (Fig. 5A: ‘pmp’), but is separated by the latter and an anterolateral process of the nasal from the margins of the external nares. Laterally, a short posterolateral process of the premaxilla appears to project between the nasal and the maxilla, ventral to the anterior part of the external naris, as also occurs in Paleorhinus bransoni (TMM 31100-101; Stocker, 2010) and Parasuchus hislopi (Chatterjee, 1978). The premaxillae are dorsoventrally compressed, although they have probably been slightly crushed, and their dorsal surface is gently convex transversely. The external surfaces of the premaxillae are only weakly ornamented, with faint longitudinal striations.</p> <p>On the ventral surface (Figs 3B, 4B, 7B, C), the transversely flat-to-gently-concave premaxillary palate is formed by medially extending shelves (one contributed by each premaxilla) that form a broad interpremaxillary fossa. At their lateral margins, adjacent to the alveoli, these shelves are thickened into distinct alveolar or palatal ridges (approximately 6 mm in width at the most anterior preserved point) that continue posteriorly onto the maxillae (Figs 3B, 4B, 7B: ‘alvr’). These alveolar ridges are visible in lateral view, projecting ventrally just below the external margin of the snout (Fig. 3D, 4D). The premaxillary palate tapers in transverse width towards its posterior termination, which is placed slightly posterior to the level of the posterior margin of the external nares. Posteriorly, the premaxillae contact the vomers (Fig. 7C).</p> <p>The few premaxillary alveoli preserved are round in outline, much smaller than the mid-maxillary alveoli and closely spaced. There is no notable change in size, spacing or shape of the alveoli at the premaxilla– maxilla boundary (Fig. 7B: ‘pm-max’).</p> <p>CT data show that, anterior to the external nares, an anteroposteriorly extending, centrally positioned, sediment-infilled premaxillary cavity (pneumatic paranasal sinus) with a sub-quadrate outline is present above the premaxillary and maxillary palates (Fig. 8A: ‘mcv’). It is likely that this cavity housed diverticula of the antorbital air sinuses, possibly as an adaptation to resisting torsion (Witmer, 1997a). Lateral to this median sinus and adjacent to the alveoli is an anteroposteriorly extending alveolar neurovascular canal (Fig. 8A–C: ‘nvc’) that would have contained the maxillary branch of the trigeminal nerve, and associated arteries and veins (Witmer, 1997a). The canals connect via foramina with each of the thecodont alveoli, and also have connections with the median sinus. More posteriorly, the median sinus is continuous with the airway, external nares, and antorbital cavity (Fig. 8B, C).</p> <p>Maxilla: The maxilla (Figs 3, 4: ‘max’) is an anteroposteriorly elongate and dorsoventrally shallow bone. Seventeen (empty) alveoli are present in the left maxilla (assuming that the most anterior three alveoli belong to the premaxilla), and 16 are present in the right maxilla. In lateral view, the left maxilla has a gently convex ventral margin. On the right maxilla the ventral margin appears to be nearly straight along its entire length, although this is probably due to deformation. The ventral margin of the maxilla is strongly convex in ‘ Paleorhinus ’ sawini (TMM 31213-16; Long &amp; Murry, 1995).</p> <p>Externally, the maxilla of Paleorhinus angustifrons contacts the premaxilla and nasal dorsally, the lacrimal posteriorly (both dorsal and ventral to the internal antorbital fenestra), and the jugal posteriorly. The bone has a slender main body with a posterodorsally extending ascending process (Fig. 5C: ‘asc’) that is placed at about mid-length of the bone. The part of the main body placed anterior to the ascending process is subequal in height to that part placed posterior to it. The ascending process is strongly inclined posteriorly, forming an angle of more than 45° from the vertical. It is very slightly curved and meets the anterior process of the lacrimal above the antorbital fenestra at about two-thirds of the length of the fenestra.</p> <p>The major feature of the lateral surface of the maxilla is the large internal antorbital fenestra and surrounding antorbital fossa. The antorbital fossa extends some 9 mm anterior to the antorbital fenestra onto the base of the ascending process (Fig. 5C: afos), where it is shallow and demarcated by low ridges that define the external antorbital fenestra. At the anterior end of the fossa there is a small foramen that connects anteriorly to the alveolar canal, as shown by CT data. On the left side, this foramen also appears to have a connection to the central airway. Dorsally, the antorbital fossa covers the entire height of the majority of the ascending process of the maxilla, and is bordered at its dorsal extremity by a distinct step formed by the nasal (which represents the margin of the external antorbital fenestra). Ventrally, the maxillary antorbital fossa rapidly narrows in dorsoventral height posteriorly and disappears at about one-third of the length of the antorbital fenestra. At approximately mid-length of the ventral margin of the internal antorbital fenestra a posteriorly expanding antorbital fossa appears again, but this is probably entirely placed on the lacrimal. The external surface of the maxilla is strongly convex below the posterior end of the antorbital fossa (due to the anterior extension of a longitudinally orientated ridge that is present on the jugal; see below), and becomes flat to gently concave anteriorly.</p> <p>On the ventral surface (Figs 3B, 4B, 7B, C), the alveoli generally have a subcircular outline. The preserved alveoli are small anteriorly (the fourth most anterior alveolus preserved on the left side, probably representing the most anterior alveolus of the maxilla, has an anteroposterior diameter of 4 mm) and increase in size posteriorly, reaching a maximum diameter in the region lateral to the choanae (the 13th alveolus preserved on the left side, which is probably the 10th maxillary alveolus, has an anteroposterior diameter of 10 mm). Alveolus size decreases again slightly at the posterior end of the tooth row, which is placed just anterior to the anterior rim of the orbit. The alveoli are closely spaced in the anterior and posterior ends of the maxilla, but become more widely spaced towards its mid-length, as evidenced by the mesiodistally expanded interalveolar septa. Medial to the tooth row, the alveolar/palatal ridges are continuous with those of the premaxillae, but become narrower and less pronounced posteriorly although the ridges are present as a distinct thickening along the entire tooth row. The maxilla ends just posterior to the tooth row and lacks an elongate posterior process.</p> <p>The maxilla is contacted by the ectopterygoid immediately posterior and medial to the last alveolus (Fig. 7D). The maxilla forms the anterolateral margin of the slit-like suborbital fenestra; anterior to this fenestra the maxilla has a medially extending shelf that forms the anterolateral and anterior margins of the choana (Figs 3B, 4B, 7C), and which prevents contact between the premaxilla and the palatine. Contact between the premaxilla and the palatine previously was cited as a synapomorphy of Phytosauria (Sereno, 1991) but is absent in many taxa including Wannia scurriensis (Stocker, 2010, 2013), the Krasiejów phytosaur specimens (e.g. ZPAL Ab III 200), and Nicrosaurus kapffi and N. meyeri (Hungerbühler, 1998).</p> <p>‘Septomaxilla’: Kuhn (1936) reconstructed the ‘septomaxillae’ (for discussion of the homology of this element see Sereno, 1991; Hungerbühler, 2002; Stocker, 2010; Nesbitt, 2011) as forming the entire internarial bar in Paleorhinus angustifrons; such a situation would be unusual, because the internarial bar is typically formed by both the ‘septomaxilla’ and the nasal or predominantly by the nasal in phytosaurs (e.g. Stocker, 2010). However, a clear ‘septomaxilla’–nasal contact cannot be identified. Given that a ‘septomaxilla’ is generally present in phytosaurs, we assume that this element is present (Figs 3A, 4A, 5A: ‘sm’), but that the ‘septomaxillae’ and nasals are indistinguishably fused. In comparison with other phytosaurs, it seems most likely that the contact occurs at approximately the midpoint of the internarial bar, where the bar narrows in transverse width and the dorsal margin is slightly concave. In lateral view, the internarial bar is raised along the anterior half of its length to a position more dorsal than the lateral borders of the external nares (Figs 3D, 4D, 5B).</p> <p>The ‘septomaxillae’ articulate with one another along the midline posteriorly and slightly flare mediolaterally anteriorly, where they terminate a short distance anterior to the anterior corner of the external nares (Figs 3A, 4A), as indicated by Kuhn (1936: fig. 1). However, differing from the reconstruction of Kuhn, the anterior processes of the ‘septomaxillae’ are separated from one another by posteromedial processes of the premaxillae (Fig. 5A), as in the Krasiejów phytosaur specimens (e.g. ZPAL Ab III 200; Dzik &amp; Sulej, 2007) and Paleorhinus bransoni (TMM 31100-101; Stocker, 2010). Anterolaterally, the ‘septomaxillae’ are separated from the nasals by a groove (Fig. 5A: ‘gr’) that extends anteriorly from the anterior corner of the external naris. CT data show that this groove continues anteriorly as a foramen for a short distance, apparently within the premaxilla. On the right side, a small connection of this foramen with the median paranasal sinus is visible in CT data. The ‘septomaxillae’ differ from those of Wannia scurriensis (TTU P-00539), which do not articulate with one another on the midline or form any part of the internarial septum (Stocker, 2010, 2013).</p> <p>Nasal: The nasals (Figs 3A, D, 4A, D: ‘na’) are anteroposteriorly elongate elements that form the posterior and lateral margins of the external nares, and probably the posterior half of the internarial septum (although the ‘septomaxilla’–nasal contact is unclear: see above). They contact the premaxillae and maxillae laterally, the prefrontals posterolaterally, and the frontals posteriorly. The nasals are flat to slightly convex transversely, and are strongly rugose along their lateral margins, dorsal to the antorbital fossa and along the raised lateral rim of the external naris. The part of the nasal forming the rim of the external naris is distinctly raised and thickened, most prominently along the posterolateral and posterior margins, where it also flares laterally (Fig. 5A: ‘fl’), contributing to the teardrop-shape of the nares. In lateral view, the rim of the external naris is therefore stepped, being low and concave anteriorly, but raised into a prominent knob-like convexity posteriorly, with a much smaller knob anteriorly (Fig. 5B: ‘convx’). A similar condition is absent in Parasuchus hislopi (Chatterjee, 1978), Wannia scurriensis (Stocker, 2013) and Paleorhinus bransoni (TMM 31100-101; Stocker, 2010), but does appear to be present in the Krasiejów phytosaur specimens (ZPAL Ab III 200). The anterior processes of the nasals forming the lateral margins of the nares continue anteriorly some 13 mm anterior to the external nares. Here, the processes are slightly expanded mediolaterally and form the lateral borders of the grooves anterior to the nares mentioned above, and thus together with the ‘septomaxillae’ exclude the premaxillae from the external nares. There is a small depression on the nasal, immediately posterior to the rim of the external naris (Figs 3A, 4A, 5A: ‘ndp’); a similar depression is present in the Krasiejów phytosaur specimens (ZPAL Ab III 200, Dzik &amp; Sulej, 2007: fig. 10).</p> <p>Laterally the nasals form the dorsal margin of the external antorbital fenestrae, thus bordering the antorbital fossae, and they have clearly defined sutures with the prefrontals. Kuhn (1936: fig. 1) identified the nasal–frontal suture in a posterior position, level with the anterior corner of the orbit. The iden- tification of the nasal–frontal suture in this area was probably based upon paired depressions on the skull roof between the prefrontals (Figs 3A, 4A, 5E: ‘fdp’), which are also seen in some other other basal phytosaurs (M. R. Stocker, pers. obs. of TMM 31025- 172, Paleorhinus bransoni; the Krasiejów phytosaurs, ZPAL Ab III 200). However, the actual nasal–frontal suture is placed more anteriorly (Figs 3A, 4A, 5E: ‘na-fr’), nearly level with the posterior margin of the antorbital fenestra, as also occurs in the Krasiejów phytosaurs (ZPAL Ab III 200, Dzik &amp; Sulej, 2007), Parasuchus hislopi (Chatterjee, 1978), and Paleorhinus bransoni (Stocker, 2010). Thus, the paired posterior depressions are positioned on the anterior part of the frontals, rather than the nasals, and each nasal forms a sharply pointed posterior process separated from the other element by the anterior processes of the frontals. The posteriormost extension of the nasal reaches approximately level with the mid-point between the antorbital fenestra and the orbit.</p> <p>Prefrontal: The prefrontal forms the anteromedial and most of the anterior margin of the orbit (Figs 3A, 4A: ‘prf ’), and is a substantially larger element than that identified by Kuhn (1936). It is approximately triangular in dorsal view, tapering to points anteriorly, posteriorly, and laterally. Its dorsal and lateral surfaces are strongly rugose and pitted (it is one of the most strongly ornamented of the skull roof elements), and it is thickened at its orbital margin. This thickening is offset from the dorsal surface of the prefrontal by a marked dorsal step. A shallow, crescentic fossa is present on the dorsal surface just anterior to, and curving parallel to, this thickening of the orbital margin (Fig. 5E: ‘prod’; Fig. 8F). A similar feature was described by Hungerbühler (2002) in Mystriosuchus as the ‘pre-orbital depression’, and its strong development in Mystriosuchus was considered by him to be autapomorphic for that genus. Laterally, the exact position of the prefrontal–lacrimal suture is difficult to determine, but it probably occurs close to the sharp junction between the nearly horizontal skull roof and the nearly vertical lateral surface of the skull anteroventral to the orbit. The prefrontal–nasal suture is clearly defined (Fig. 5E: ‘na-prf’), whereas that with the frontal is clearly defined anteriorly, but becomes difficult to trace more posteriorly. The prefrontal clearly forms the anteromedial and anterior parts of the orbital rim.</p> <p>Frontal: The surface of the left frontal is heavily reconstructed (Fig. 8F, G) and the definition of the margins of the frontals generally is poor: their sutures with the nasals and the prefrontals are clear anteriorly, but posteriorly the sutures with the postfrontals and parietals are difficult to recognize. The dorsal surfaces of the frontals (Figs 3A, 4A: ‘fr’) are covered with rugose and pitted ornamentation similar to that of the prefrontal and other skull roof bones. The frontals are gently concave transversely (judging from the surface of the right frontal), and towards their anterior ends they each bear a small depression on their dorsal surfaces (see above). These depressions (Figs 3A, 4A, 5E: ‘fdp’) are mainly marked by raised posterior and posterolateral rims and are continuous anteriorly with the rather smooth dorsal surface of the nasals, whereas the surfaces of the frontals posterior to and the prefrontals lateral to the depressions are notably rugose. The frontals contribute to the medial rims of the orbits, and they are slightly thickened at their lateral margins (seen clearly on the right side and in CT images: Fig. 8G) as also occurs in many other phytosaurs (Stocker, 2010). On the right side the ventral surface of the frontal is partially exposed. The contribution of the frontal to the medial margin of the orbit curves medioventrally towards the contact with the laterosphenoid (at the anteriormost extent of the latter bone). CT data reveal a marked transverse concavity on the medial parts of the ventral surfaces of the frontals for the reception of the olfactory bulbs (Fig. 8G). This concavity is narrow between the orbits but widens anteriorly (Fig. 8F, G).</p> <p>Lacrimal: The lacrimal forms the anteroventral rim of the orbit and the posterior rim of the antorbital fenestra (Figs 3A, D, 4A, D: ‘lc’). It contacts the maxilla anteriorly, both dorsal and ventral to the internal antorbital fenestra (although the exact position of the lacrimal–maxilla suture ventral to the internal antorbital fenestra is uncertain), the jugal ventrally, and the prefrontal and nasal dorsally. Most of the anterior portion of the lacrimal is emarginated laterally by the antorbital fossa (Fig. 5C), which is continuous from the anterodorsal process to the anterior expansion of the ventral end of the main body. At its posteriormost point, the antorbital fossa is separated from the orbit by a raised surface only 7.7 mm wide. Posterodorsally, there is a marked step on the lateral surface of the element along the orbital margin; the prefrontal–lacrimal suture appears to be situated just medial to this step. Whereas the orbital margin and skull roof dorsal to this step are strongly rugose, the orbital margin and surface of the lacrimal ventral to the step are smooth. The nasolacrimal foramen is visible in CT data on the internal surface of the orbital margin, at about the mid-height of the orbit, and is probably positioned on the boundary between the lacrimal and prefrontal as in other phytosaurs (Witmer, 1997a), although the suture is not visible. Anteriorly, the nasolacrimal canal extends dorsomedial to the antorbital fenestra, but becomes very narrow and difficult to trace in CT data. In contrast to the situation in Machaeroprosopus pristinus, where the canal divides into two branches, one leading anteriorly and one ventrally (Camp, 1930), only a single canal is evident in Paleorhinus angustifrons.</p> <p>Jugal: The jugal is a triradiate element whose contacts with surrounding elements are relatively clearly visible on the left side of the skull (Figs 3A, D, 4A, D: ‘jg’). The dorsal process of the jugal articulates with the postorbital along an almost horizontal suture level with the ventral border of the orbit, and forms part of the ventral rim of the orbit and the ventral part of the anterior rim of the infratemporal fenestra. The jugal–postorbital suture appears to be overlapping, with the dorsal process of the jugal extending dorsally along the medial surface of the posterior margin of the jugal–postorbital bar. There is a small and very shallow fossa on the dorsal process of the jugal (Figs 3A, 4A, 6A: ‘fo’), adjacent to the anteroventral corner of the infratemporal fenestra (most clearly visible on the left side), the anterior margin of which is defined by a low break-in-slope. A similar fossa is present in some other phytosaurs (e.g. Mystriosuchus planirostris, Hungerbühler, 2002: fig. 12C; Pravusuchus hortus, Stocker, 2010: fig. 6; Krasiejów phytosaur specimen ZPAL Ab III 200).</p> <p>The transversely compressed and dorsoventrally shallow posterior process of the jugal forms the anterior part of the ventral border of the infratemporal fenestra. This posterior process tapers in dorsoventral height towards its posterior termination, and it is extensively overlapped along its dorsal edge by the anterior corner of the quadratojugal. The posterior process of the jugal extends close to the quadrate on the lateral surface, but the clearly visible suture with the quadratojugal on the medial surface of the left jugal demonstrates that it does not contact the quadrate.</p> <p>The dorsoventrally deep anterior process of the jugal contacts the maxilla laterally along an anterodorsallyto-posteroventrally trending suture, which begins immediately ventral to the posterior corner of the internal antorbital fenestra and ends at the posterior end of the maxillary tooth row. The anterior process also forms an elongate suture with the lacrimal that coincides with the ventral margin of the antorbital fossa. The jugal apparently thus forms part of the posteroventral rim of the antorbital fossa (margin of the external antorbital fenestra), but is excluded from entering the rim of the internal antorbital fenestra by the contact between the maxilla and the lacrimal (contra Kuhn, 1936). Medially, the jugal is contacted by the ectopterygoid adjacent to the jugal–maxilla contact (Fig. 7D).</p> <p>In lateral view, the jugal is proportionally shallower dorsoventrally and longer anteroposteriorly than the deep and short element shown in Parasuchus hislopi (Chatterjee, 1978), and the very dorsoventrally deep jugal of ‘ Paleorhinus ’ sawini (Long &amp; Murry, 1995: fig. 24A), but similar to that of some of the Krasiejów phytosaur specimens (e.g. ZPAL Ab III 200, Dzik &amp; Sulej, 2007) and Paleorhinus bransoni (Long &amp; Murry, 1995: fig. 24D). The ventral margin of the jugal is only gently concave in lateral view in Paleorhinus angustifrons, rather than strongly concave as occurs in Wannia scurriensis (TTU P-00539), and Parasuchus hislopi (Chatterjee, 1978). As a result, only a small part of the ectopterygoid is visible in lateral view in Paleorhinus angustifrons. However, it is possible that these features of the jugal may have been affected by dorsoventral crushing of the skull.</p> <p>Approximately 10 mm ventral to the ventral rim of the orbit (at approximately two-thirds of the dorsoventral height of the jugal), the ornamentation of the lateral surface of the jugal takes the form of a linear row of six to seven slightly raised nodes, extending anteroposteriorly from the posteroventral rim of the antorbital fossa to the anterior margin of the small fossa adjacent to the infratemporal fenestra (Figs 3A, D, 4A, D, 5D, 6A: ‘jgb’). A similar nodular row is present in the Krasiejów phytosaur specimens (ZPAL Ab III 200, Dzik &amp; Sulej, 2007) and referred specimens of Paleorhinus bransoni (TMM 31100-101; M. R. Stocker, pers. obs), but this row is absent in other phytosaur taxa. The lateral surface of the jugal immediately dorsal and ventral to this row of nodes is smooth. At approximately 10 mm (one-third of the dorsoventral height of the element) from the ventral margin of the jugal there is a marked ridge that extends anteriorly from the base of the posterior process of the jugal onto the maxilla, fading out beneath the antorbital fenestra (Fig. 5D: ‘jmri’). This ridge separates the row of nodes and associated smooth surface texture (above) and a more typical rugose surface texture (below, and on the posterior process).</p> <p>On the medial surface of the jugal there is a large, concave area ventral to the orbit, which is separated from the medial surface of the posterior process by a pronounced internal ridge that extends posterodorsally from the contact with the ectopterygoid onto the dorsal process of the jugal. The medial surface of the dorsal process of the jugal is slightly bevelled along the margin of the infratemporal fenestra.</p> <p>In CT cross sections (Fig. 8F) an anteroposteriorly extending canal is visible within the anterior part of the jugal. Anteriorly, this canal enters the bone dorsomedially approximately along the ectopterygoid– jugal contact, posterior to the maxillary tooth row and lateral to the suborbital fenestra (at approximately 60% of the anteroposterior length of the fenestra). It extends posterolaterally and slightly dorsally into the jugal for approximately 24 mm, and exits the jugal on its medial surface immediately posterior to the termination of its contact with the ectopterygoid. Posteriorly the course of this canal continues as a narrow trough visible for around 35 mm on the medial surface of the base of the posterior process of the jugal. Anteriorly, a much smaller canal appears to branch off from the main canal within the jugal and opens dorsomedially through a foramen on the dorsal surface of the ectopterygoid lateral to the suborbital fenestra. A foramen piercing anteriorly into the jugal body on its medial surface is also present in Pravusuchus (AMNH FR 30646), ‘ Machaeroprosopus ’ zunii (UCMP 27159), and ‘ Phytosaurus ’ doughtyi (ANMH FR 4919) (Stocker, 2010). Stocker (2010) suggested that this canal may have housed the maxillary branch of the trigeminal nerve.</p> <p>Postorbital: The postorbital is a triradiate element (Figs 3A, D, 4A, D: ‘po’), the entire external surface of which is weakly rugose. The central body and ventral process of the postorbital form most of the posterior rim of the orbit and the dorsal half of the anterior rim of the infratemporal fenestra. The posterior process of the postorbital forms the anterior half of the dorsal rim of the infratemporal fenestra, whereas the posterior and medial processes of the postorbital together form the anterolateral rim of the supratemporal fenestra. The exact positions of the sutures between the postorbital and the parietal and postfrontal are not clear. The posterior process of the postorbital terminates dorsal to the midpoint of the infratemporal fenestra (contra Kuhn, 1936, who reconstructed this as a substantially longer process), and forms the anterior half of the transversely narrow postorbital–squamosal bar. As in Paleorhinus bransoni (Stocker, 2010) and the Krasiejów phytosaur specimens (ZPAL Ab III 200, Dzik &amp; Sulej, 2007), the posterior process of the postorbital tapers to a point in dorsal view in Paleorhinus angustifrons, and this point fits into a slot on the anterior process of the squamosal (Figs 3A, 4A). Stocker (2010: character 22) identified this as the plesiomorphic condition for Phytosauria, and described derived conditions present in many phytosaurid phytosaurs. The postorbital– squamosal bar of Paleorhinus angustifrons is strongly compressed dorsoventrally.</p> <p>On the medial surface of the postorbital, the ridge extending from the dorsal process of the jugal continues dorsally, ending at approximately the postorbital–postfrontal contact, and marks a sharp boundary between the orbital facet (facing anteromedially) and a broad medially and slightly posteriorly facing surface. Based on CT data, there does not seem to be a contact between the postorbital and the laterosphenoid.</p> <p>Postfrontal: The margins of the postfrontal are not clear, but the bone presumably forms the posterodorsal corner of the orbital rim, and contacts the frontal anteriorly and medially, the parietal posteromedially, and the postorbital laterally and posterolaterally (Figs 3A, 4A, 5F: ‘pof’). The dorsal surface of the postfrontal is weakly rugose, with an ornamentation similar to that of the other skull roof elements. The postfrontal is slightly thickened at its orbital rim. Its posteromedial margin may contact the laterosphenoid ventrally. On the right side, a pronounced, oval groove is present in the ventral surface of the postfrontal (and possibly also on the adjacent portions of the frontal and parietal) just lateral to the laterosphenoid contact.</p> <p>Parietal: The contacts of the parietals (Figs 3A, 4A, 5F: ‘pa’) with the frontals, postfrontals, and postorbitals are poorly demarcated as a result of fusion of the skull roof bones, combined with the presence of strongly ornamented bone surfaces and some reconstruction of the bone surface. The dorsal surfaces of the parietals bear a slight midline convexity at the interparietal suture that is most prominent at approximately midlength but that continues posteriorly as far as the posterior termination of the main bodies of the parietals: a similar feature was described for the holotype of Pravusuchus (AMNH FR 30646; Stocker, 2010). There are shallow concavities within the parietals on either side of this convexity. The main bodies of the parietals expand laterally anteriorly to form the anterolateral corners of the supratemporal fenestrae. The transversely orientated frontoparietal suture seems to be slightly posterior to the posterior margin of the orbits. Posteriorly, the posterolateral corners of the parietals are drawn out posterolaterally (Figs 5F, 6C: ‘lpw’) to form the parietal component of the anterolaterally-toposteromedially compressed parietal–squamosal bar. This bar, and the parietal as a whole, is not depressed ventrally relative to the infraorbital portion of the skull roof (see above). The almost vertical lateral parietal wing forms essentially the entire dorsal part of the posterior margin of the supratemporal fenestra in dorsal view, dorsally overlapping a medial process of the squamosal and tapering to a point posterolaterally. In posterior view, the posterolateral parietal wings form a substantial component of the dorsal margin of the occiput (Figs 3C, 4C, 6C: ‘lpw’), extending laterally to slightly beyond the lateral margins of the posttemporal fenestrae. The posterolateral wing dorsally overlaps a medial process of the squamosal on the occiput, with the parietal being excluded from the dorsal margin of the posttemporal fenestra by a small contact between the supraoccipital and the squamosal. Medially, the parietals form the entire dorsal and lateral margins of the small postparietal foramen.</p> <p>Squamosal: Four processes extend from the main body of the squamosal (Figs 3, 4: ‘sq’): an anteriorly (and slightly medially) directed postorbital process forming the posterior half of the postorbital– squamosal bar; a medially (and slightly anteriorly) directed parietal process forming the lateralmost and ventral parts of the parietal–squamosal bar; an anteroventral process extending along the lateral surface of the quadrate and contacting the quadratojugal (Fig. 6: ‘avpsq’); and a short ventrally directed opisthotic process (Fig. 6B, D: ‘opsq’). As in non-phytosaurid phytosaurs and Angistorhinus (Stocker, 2010: character 24), the squamosal of Paleorhinus angustifrons lacks a distinct posterior process that extends posteriorly beyond the paroccipital process of the opisthotic.</p> <p>The postorbital–squamosal bar is strongly compressed dorsoventrally. As in Paleorhinus bransoni (TMM 31100-101; Stocker, 2010) and the Krasiejów phytosaur specimens (ZPAL Ab III 200, Dzik &amp; Sulej, 2007), the dorsal surface of the squamosal has an anteroposteriorly extending concavity or furrow that begins at the most posterior corner of the squamosal and extends onto the postorbital–squamosal bar (Figs 5F, 6D: ‘sqd’). This furrow becomes shallower anteriorly, fading out on the dorsal surface of the postorbital. The lateral edge of the postorbital– squamosal bar is a sharp ridge that extends along the lateral margins of both the postorbital and the squamosal to reach the most posterolateral corner of the squamosal. This ridge overhangs the dorsal margin of the laterotemporal fenestra and the base of the anteroventral process of the squamosal, and thus separates the strongly ornamented dorsal surface from the smooth lateral surface of the anteroventral process. Above the posterior edge of the quadrate this ridge bifurcates, sending off a short posteroventrally extending ridge along the lateral surface of the squamosal (Figs 3D, 4D, 6A, B: ‘sqr’). These two ridges enclose a small, elongate, teardrop-shaped concavity. An identical arrangement of paired ridges enclosing a small slit-like concavity is present in referred specimens of Paleorhinus bransoni (TMM 31025-172; Stocker, 2010: fig. 9) and the Krasiejów phytosaur specimens (ZPAL Ab III 200, Dzik &amp; Sulej, 2007) but is absent in other basal phytosaurs. It is here identified as autapomorphic for the genus Paleorhinus.</p> <p>The main body of the squamosal is dorsoventrally shallow, as in other basal phytosaurs (Stocker, 2010: character 25). In dorsal view, the squamosal tapers posteriorly to a posterolaterally projecting point, similar to the condition in the Krasiejów phytosaur specimens (ZPAL Ab III 200, Dzik &amp; Sulej, 2007) and Paleorhinus bransoni (Stocker, 2010: fig. 9). In lateral view, the squamosal extends beyond the quadrate head posteriorly, but only a short distance posterior to the quadrate condyles. The squamosal is drawn out posteroventrally into a short opisthotic process (Fig. 6B, D: ‘opsq’), the posterior surface of which articulates with the paroccipital process.</p> <p>The anteroventral process of the squamosal (Fig. 6: ‘avpsq’) is anteroposteriorly broad dorsally but tapers along its length and forms slightly more than 60% of the posterior margin of the infratemporal fenestra, articulating with the quadratojugal ventrally and the quadrate posteriorly. The anteroventral process is transversely compressed and smooth and unornamented on its lateral surface.</p> <p>Medially, the parietal process of the squamosal forms the most posterolateral part of the parietal– squamosal bar and the lateral part of the dorsal margin of the slit-like posttemporal fenestra (Figs 3A, C, 4A, C). The process tapers medially, being overlapped dorsally by the lateral process of the parietal, and reaches almost to the medial margin of the supratemporal fenestra. On the occiput, the parietal process articulates medially with a thin lateral process of the supraoccipital above the posttemporal fenestra and ventrally with the dorsal margin of the paroccipital process.</p> <p>Quadratojugal: The exact contacts of the subtriangular quadratojugal with surrounding elements are poorly defined, but it forms the posteroventral rim of the infratemporal fenestra (Figs 3, 4, 6A: ‘qj’). The quadratojugal has a long, anteriorly tapering anterior process that articulates with the posterior process of the jugal along an elongate anterodorsalto-posteroventrally inclined suture beneath the infratemporal fenestra. The process thus forms approximately three-quarters of the ventral margin of the infratemporal fenestra. Posteriorly the quadratojugal overlaps the lateral edge of the quadrate laterally, and dorsally it articulates with the anteroventral process of the squamosal and a laterally extending tongue of the quadrate that forms the roof of the posteriorly facing quadrate foramen (Fig. 6A). The quadratojugal forms the lateral margin of this foramen. CT data suggest that a narrow and anteroposteriorly short longitudinal canal is present in the thickened main body of the quadratojugal, parallel to the quadrate foramen, but not connected to the latter, although the resolution of the data is not sufficiently high to resolve the details of this canal. This canal appears to open posteriorly a short distance lateroventral to the quadrate foramen and anterolaterally through a small opening onto the lateral side of the quadratojugal.</p> <p>Quadrate: In posterior view (Figs 3C, 4C, 6D: ‘qd’), the main body of the quadrate is transversely expanded at its ventral margin to form the condyles for articulation with the mandible (Fig. 6D: ‘qdc’); dorsally it is tightly sutured to the squamosal and the paroccipital process of the opisthotic. The width of the skull across the quadrate condyles at their ventralmost part (c. 180 mm) is greater than that across the squamosals (134 mm), giving the skull a somewhat trapezoidal outline in posterior view, although this outline has probably been modified by post-mortem dorsoventral compression. In ventral view the quadrate condyles are anteroposteriorly narrow and the medial condyle is set slightly further posteriorly than the lateral condyle (Figs 3B, 4B). The articular surface of the condyles is flat-to-gently convex laterally and gently concave medially, although the morphology appears to have been distorted by post-mortem compression.</p> <p>The quadrate articulates laterally with the quadratojugal, and the quadrate forms the ventral, medial, and dorsal margins of the large quadrate foramen (Figs 3, 4, 6D: ‘qf’). The dorsal margin of the foramen is formed by a lateral flange of the quadrate (Fig. 6B, D: ‘lfqd’), which articulates with the quadratojugal and the anteroventral process of the squamosal. Medial to the quadrate foramen, the main body of the quadrate is slightly raised into a low vertical ridge. Anteriorly, the quadrate is thickened in the area of the quadrate foramen, so that the latter has the appearance of a short channel between the quadrate and the quadratojugal. The foramen opens into a canal that runs almost directly anteriorly, exiting along the anterior surface of the quadrate/ quadratojugal (Figs 3B, 4B).</p> <p>The pterygoid wing of the quadrate extends anteromedially (Figs 3, 4, 6D, 7A: ‘ptqd’). It is robust, curves slightly laterally towards its anterior end, and meets the posterior wing of the pterygoid in an extensive suture whereby the anterior end of the pterygoid wing of the quadrate is medially overlapped by the posterior wing of the pteryoid. The ventral margin of the pterygoid wing is placed slightly above the quadrate condyles and is flexed medially, to form an expanded medial shelf that extends along the full length of the pterygoid–quadrate plate. Dorsal to the shelf, the pterygoid wing is strongly concave dorsoventrally. The lateral surface of the pterygoid wing is dorsoventrally convex and forms the posteromedial border of the subtemporal fenestra.</p> <p>Pterygoid and epipterygoid: The pterygoid is a complex element consisting of several major processes (Figs 3B, C, 4B, C: ‘pt’). The posterior wing or quadrate process curves posterolaterally away from the midline (Fig. 7A: ‘qdpt’), and its lateral margin overlaps the pterygoid wing of the quadrate to form the pterygoid– quadrate plate. The posterior wing is strongly concave dorsoventrally in medial view and flexes medially at its ventral margin to form a ventral medial shelf continuous with that on the pterygoid wing of the quadrate. Posteriorly the posterior wing is forked, with the dorsal part of the wing reaching slightly further posteriorly (almost level with the posterior end of the occipital condyle) than the ventral process (Figs 3C, 4C). At the proximal end of the posterior wing of the pterygoid, the ventral surface of the pterygoid is drawn out medially into a small medially extending plate of bone (Fig. 7A: ‘mppt’). This plate has a broadly subrectangular outline, but its posteromedial corner is drawn out posteriorly such that the posterior margin of the plate is concave in ventral view. The posterior part of this medial plate forms the articular surface for the basipterygoid process of the basisphenoid. The left and right articular surfaces for the basipterygoid processes are broadly separated from one another, so that an interpterygoid vacuity is present at this point, through which the cultriform process of the parabasisphenoid can be seen (Fig. 7A: ‘clt’).</p> <p>There is a low thickening of the medial margin of the base of the posterior process of the pterygoid, immediately lateral to the basipterygoid articulation. Dorsal to this thickening, CT data show that the pterygoid begins to be drawn out dorsally to form the dorsoventrally expanded posterior wing (Fig. 8H). CT data indicate the presence on the right side of a dorsoventrally extending epipterygoid bone at this point (Fig. 8H). Basally, the epipterygoid is strongly expanded anteroposteriorly and slightly transversely expanded, forming a sheet that laterally overlies the dorsal expansion of the pterygoid. The epipterygoid tapers dorsally, articulating with a cup-like surface on the ventrolateral margin of the laterosphenoid (Fig. 8H).</p> <p>Anterior to the basipterygoid articulation, a lateral flange of the pterygoid extends anterolaterally and ventrally to contact the ectopterygoid and palatine (Fig. 7A, D: ‘lfpt’). The lateral flange is rather short and the articulation with the ectopterygoid appears to be sutural in contact, with only a slight dorsal overlap of the pterygoid onto the ectopterygoid, whereas the palatine has a posteriorly orientated process that appears to articulate with a small cup-shaped fossa on the anterior margin of the lateral flange of the pterygoid (Fig. 7D).</p> <p>Medial and anterior to the lateral flange, sheet-like anterior flanges of the pterygoid extend dorsomedially, and approach one another very closely along the ventral midline (Fig. 7A, C, D: ‘afpt’). These anterior flanges are badly damaged, possibly from the original preparation, and much of the bone surface is missing. Posteriorly, they are offset dorsally from the basipterygoid articulation and, to a lesser extent, from the lateral flange by a distinct step. They articulate with the palatines at their lateral margins and with the vomers anteriorly. At their medial margins they are markedly thickened into sharp ridges that arise from the medial edges of the medial rectangular sheets that form the basicranial articulation, and the ridges are near continuous anteriorly with the ridges of the vomers (Fig. 7C, D). CT cross sections show that the anterior flanges are drawn out at their medial margins into vertical sheets, between which extends the cultriform process of the parabasisphenoid (Fig. 8F). These flanges continue anteriorly dorsal to the vomers (Fig. 8E), which underlie them ventrolaterally, terminating a short distance anterior to the posterior margins of the choanae.</p> <p>Ectopterygoid: The left ectopteryoid is damaged along its posterior margin and has been partially reconstructed with plaster, but the right ectopterygoid is relatively well preserved (Figs 3, 4; ‘ect’). It contacts the pterygoid posteromedially and the palatine anteromedially. The ectopterygoid extends anterolaterally as a broad bar of bone from the contact with the pterygoid and the palatine to its contact with the posterior end of the maxilla and the adjacent anterior part of the jugal, where the bone is expanded posteriorly (Fig. 7D). It forms the posterolateral margin of the suborbital fenestra. A large elliptical foramen is present on the ectopterygoid adjacent to the medial contact with the pterygoid (Fig. 7D: ‘ectf ’). A foramen is also present in the ectopterygoid in other phytosaurs (often associated with a groove), but the very rounded nature of the opening and the lack of an associated groove appear to be unique to Paleorhinus angustifrons. In the Krasiejów phytosaurs this foramen is elongated and slit-like rather than oval (ZPAL Ab III 200, Dzik &amp; Sulej, 2007: fig. 10; R. Bronowicz, pers. obs.). CT data show that the foramen opens into an anterolaterally extending cavity that terminates within the bone (Fig. 8G). Doyle &amp; Sues (1995) suggested that this opening was related to the antorbital sinus system, and Witmer (1997a) discussed the possibility that somewhat similar ‘ectopterygoid recesses’ in theropods represented cranial pneumaticity.</p> <p>Palatine: The palatine is an elongate, horizontally orientated element (Figs 3B, 4B: ‘pal’), the main body of which comprises a horizontal shelf that forms most of the lateral region of the posterior portion of the palate. At its posterior end the palatine articulates with the anteromedial corner of the ectopterygoid; medially it articulates with the pterygoid along much of its length, although large parts of both bones, which obviously were very thin in this region, are missing. The palatine forms the medial margin of the slit-like and curved suborbital fenestra, although on both sides the palatine has an unfinished edge along the margin of this fenestra (Fig. 7C, D). Anterior to the suborbital fenestra the horizontal main body of the palatine tapers in mediolateral width as it articulates with a medially extending shelf of the maxilla (Figs 3B, 4B, 7C, D); this medially extending shelf excludes contact between the palatine and premaxilla (Fig. 7C: ‘atpal’). Medial to the main body of the palatine, there is a dorsomedially extending and vaulted flange (Fig. 7C, D: ‘dmfpal’) that articulates with the pterygoid posteriorly and with the vomer anteriorly. These two portions of the palatine are separated by a marked step, which is continuous posteriorly with the step separating the anterior pterygoid flanges from the basipterygoid articulations (as described above).</p> <p>Vomer: The vomers form the entirety of the septum separating the choanae (Figs 3B, 4B, 7C: ‘v’); they are separated slightly from one another, and are transversely compressed with their ventral margins forming sharp ridges (Fig. 5D). At their anterior end the vomers maintain their sharp median ridges, but also become dorsoventrally flattened and expand transversely (visible in CT data) to form a plate that appears to underlie the posteriormost parts of the premaxillae and laterally contact the maxillae, although the exact position of the contacts between the vomers and the premaxillae are unclear. At the posterior end of the choanae, the vomers are also expanded transversely (and form the posteriormost corners of the margins of the choanae), and underlie the anterior extensions of the pterygoids (Fig. 7C: ‘pexv’; Fig. 8E). The sharp median ridges on the vomers are near continuous posteriorly with those of the pterygoids but disappear anteriorly at about the contact with the premaxillae and do not continue anteriorly onto the ventral surface of the latter bones.</p> <p>Supraoccipital: The external surface of the posterodorsally facing supraoccipital (Figs 3B, C, 4B, C, 6C: ‘so’) is exposed broadly in dorsal view and slightly raised at its midline, although without forming a sharp median crest. The surface of the supraoccipital is more dorsally than posteriorly directed, although this might be slightly exaggerated by dorsoventral compression. The supraoccipital forms only a very small part of the median dorsal margin of the foramen magnum. Laterally, the supraoccipital con- tacts the otooccipital ventrally and forms the medial margin of the posttemporal fenestra (Fig. 6C). Dorsally, it sends a slender lateral process above the posttemporal fenestra, forming the dorsomedial margin of the fenestra, contacting the squamosal laterally, and excluding the parietal from the margin of this fenestra. CT data show that a small foramen enters the supraoccipital from the endocranial cavity and extends posterolaterally and slightly ventrally through this bone, until it exits at about the supraoccipital–parietal suture; this foramen most probably marks the course of the mid-cerebral vein towards its posterior exit (Sedlmayr, 2002).</p> <p>Otooccipital: The opisthotic and exoccipital appear to be indistinguishably fused to one another; therefore, the two elements are described together here as the otooccipital (Figs 3C, 4C, 6C: ‘oto’). The exoccipital portion forms the lateral and dorsolateral margin, as well as part of the ventrolateral margin, of the foramen magnum. A small, oval facet for the contact with the proatlas is present on the otooccipital along the dorsolateral margin of the foramen magnum, adjacent to the suture with the supraoccipital. The contacts of the otooccipital with the basioccipital are unclear, so it is uncertain how much of the occipital condyle it forms, although it seems that the basioccipital forms a small portion of the ventral margin of the foramen magnum. On the lateral surface of the opisthotic portion of the left otooccipital, the metotic foramen and fenestra ovalis are visible as two deep recesses separated by the descending ramus of the opisthotic (= crista interfenestralis). The metotic foramen is slightly more ventrally placed than the more anteriorly positioned fenestra ovalis, so that the crista interfenestralis runs slightly obliquely posterodorsally. The anteroventral base of the crista is expanded, but the structure rapidly narrows towards its posterodorsal end. Only a very small and shallow stapedial groove (Fig. 7A: ‘stgr’) extends from the fenestra ovalis posterolaterally onto the base of the paroccipital process and disappears at about the posterolateral edge of the crista interfenestralis.</p> <p>The paroccipital processes of the opisthotic portions of the otooccipitals extend posterolaterally, and their external surfaces face posterodorsally (Figs 3C, 4C, 6C, D: ‘par’). The processes are anteroposteriorly expanded and dorsoventrally shallow at their bases, such that they form a flattened, elongate, triangular platform in ventral view. They become more compressed anteroposteriorly and gently expanded dorsoventrally towards their distal ends, although the distal tips of the processes appear slightly incomplete on both sides. Nevertheless, it can be established that the paroccipital processes did not reach the posterolateral edge of the squamosal. The paroccipital process forms the ventral margin of the elongate, slit-like posttemporal fenestra. Laterally, the paroccipital process shares a tight and immobile suture with the quadrate along the anterior margin of the process. Distally, the paroccipital process is buttressed by the opisthotic process of the squamosal (Fig. 6D).</p> <p>Basioccipital and parabasisphenoid: The basioccipital (Figs 3B, 4B: ‘bo’) forms most of the apparently heavily crushed occipital condyle (Figs 3B, C, 4B, C: ‘occ’), which is strongly compressed dorsoventrally and has a transversely convex ventral surface. There is no distinct transverse constriction or neck separating the occipital condyle from the body of the basioccipital. The surface of the basioccipital is more ventrally than posteriorly directed, although this is probably exaggerated by compression. In ventral view the basioccipital is flared laterally towards its anterior end, forming the posterior parts of the basitubera (Figs 3B, C, 4B, C, 7A: ‘bt’). The basitubera are separated from one another by a narrow, concave, anteroposteriorly extending slot (Fig. 7A: ‘slt’). This slot extends posteriorly as a shallow depression to the rim of the basioccipital condyle, and is separated from small longitudinal depressions on either side of the basioccipital (positioned close to the suture with the exoccipital) by low, flattened ridges. Lateral to the narrow groove separating the basitubera, a small posteroventrally directed depression (Fig. 7A: ‘dep’) is present on each side on the posteromedial surface of the tubera, adjacent to the suture with the basisphenoid.</p> <p>The basitubera extend further laterally than the lateral edges of the occipital condyle in ventral view, and they extend a short distance ventral to the occipital condyle in posterior view. Their ventral surfaces are flattened, although this might at least partially be due to dorsoventral compression.</p> <p>The basisphenoid is expanded transversely at its anterior and posterior ends. Posteriorly the basisphenoid forms the anterior three-quarters of the basitubera, whereas anteriorly it forms the basipterygoid processes (Fig. 7A: ‘bpt’) that extend laterally and ventrally to articulate with the pterygoids. The transverse width across the basitubera is slightly greater than the transverse width across the basipterygoid processes. There is a deep, cone-shaped concavity that covers the ventral midline of the basisphenoid just anterior to the contact with the basioccipital (Fig. 7A: ‘conc’); this concavity has a subcircular outline and is delimited posteriorly by a low transversely extending ridge that separates it from the narrow concave slot on the basioccipital described above. Anteriorly, there is no clearly defined ridge between the basipterygoid processes, such that the concavity opens anteriorly.</p> <p>Anterior to the main body of the basisphenoid, the cultriform process (Fig. 7A: ‘clt’) is poorly exposed but extends anteriorly as a mediolaterally narrow process, the anterior termination of which is hidden by the pterygoids. CT data show that the cultriform process has a ‘V’-shaped cross section (Fig. 8G: ‘clt’) that leads posteriorly into the sella turcica, and extends anteriorly as a slender process terminating approximately level with the posterior ends of the maxillary tooth rows. There is no subsellar recess at the base of the cultriform process, unlike the condition in many dinosaurs (Witmer, 1997b).</p> <p>Laterosphenoid: The right laterosphenoid is visible through the orbit and temporal fenestrae. It is an anteroposteriorly elongate bone that is triangular in outline in lateral view. The laterosphenoid has a long, pointed anterior process that articulates with the frontal dorsally and reaches almost to half the length of the orbit. The lateral process of the laterosphenoid is short and anteroposteriorly broad and appears to contact the postfrontal but not the postorbital. On the lateral surface of the bone, the anterior and lateral processes are separated by a rounded break-in-slope that extends dorsally and slightly anteriorly from the ventrolateral margin of the laterosphenoid. Posterior to this break-in-slope, the lateral surface of the bone is dorsoventrally concave, whereas anterior to it the lateral surface is slightly concave anteroposteriorly and slightly convex dorsoventrally. The posterior suture with the parietal cannot be clearly distinguished, and the posteroventral contact with the prootic is covered by matrix.</p></div> 	https://treatment.plazi.org/id/871D87BB6D78FFC5FC8D7B75FB41FB72	Public Domain	No known copyright restrictions apply. See Agosti, D., Egloff, W., 2009. Taxonomic information exchange and copyright: the Plazi approach. BMC Research Notes 2009, 2:53 for further explanation.		Plazi	Butler, Richard J.;Rauhut, Oliver W. M.;Stocker, Michelle R.;Bronowicz, Robert	Butler, Richard J., Rauhut, Oliver W. M., Stocker, Michelle R., Bronowicz, Robert (2014): Redescription of the phytosaurs Paleorhinus (‘ Francosuchus’) angustifrons and Ebrachosuchus neukami from Germany, with implications for Late Triassic biochronology. Zoological Journal of the Linnean Society 170 (1): 155-208, DOI: 10.1111/zoj.12094, URL: http://doi.wiley.com/10.1111/zoj.12094
871D87BB6D66FFF6FC647D69FAA1FB5C.text	871D87BB6D66FFF6FC647D69FAA1FB5C.taxon	http://purl.org/dc/dcmitype/Text	http://rs.tdwg.org/ontology/voc/SPMInfoItems#GeneralDescription	text/html	en	Ebrachosuchus neukami O. Kuhn 1936	<div><p>EBRACHOSUCHUS NEUKAMI KUHN, 1936</p> <p>‘ Ebrachosuchus neukami nov. gen. nov. spec.’; Kuhn, 1936: 77, figs 4, 5, pl. 8, 10</p> <p>‘ Ebrachosuchus neukami Kuhn’; Kuhn, 1938: 318</p> <p>‘ P. broilii (Kuhn, 1933) ’; Gregory, 1962: 670, fig. 2B</p> <p>‘ P. (F.) neukami O. Kuhn 1936 ’; Westphal, 1976: 109, fig. 7D</p> <p>‘ Francosuchus broilii Kuhn, 1932 ’; Chatterjee, 1978: fig. 16B</p> <p>‘ P. neukami Kuhn, 1936 ’; Hunt &amp; Lucas, 1991: 489</p> <p>‘ Paleorhinus neukami Kuhn, 1936 ’; Long &amp; Murry, 1995: 36</p> <p>Holotype: BSPG 1931 X 501, complete skull missing the lower jaws (Kuhn, 1936: textfigs 4, 5, pl. 8, fig. 1a–e, pl. 10, fig. 4; Figs 9–14).</p> <p>Locality and horizon: Bed 9 (Kuhn, 1933, 1936) of the Ebrach quarry, Bamberg district, Upper Franconia region of northern Bavaria, Germany. Blasensandstein of the Sandsteinkeuper, laterally equivalent to the Hassberge Formation of the Middle Keuper (Late Triassic: late Carnian).</p> <p>Previous diagnosis: Occiput strongly compressed dorsoventrally; orbit directed dorsally; interpterygoid vacuity large, elongate; rostral elongation extreme (prenarial length twice that of postnarial length); upper dentition with 58 teeth [i.e. in combined premaxilla and maxilla] (Long &amp; Murry, 1995: 36).</p> <p>Emended diagnosis: Characterized by the following autapomorphic features: (1) preorbital length more than 3.8 times that of the orbit + postorbital length (convergently present in the pseudopalatine phytosaur Mystriosuchus planirostris); (2) more than 50 teeth in the premaxilla and maxilla combined; (3) pronounced, sharp flange extending along the lateral surface of the dorsal (postorbital) process of the jugal and the ventral (jugal) process of the postorbital that is continuous posteriorly with the lateral margin of the postorbital–squamosal bar; (4) infratemporal fenestra is substantially longer anteroposteriorly than deep dorsoventrally, terminates anteriorly beneath the midpoint of the orbit; (5) quadrate foramen very large, approximately two-thirds of width of foramen magnum; (6) alveolar ridges absent from the anterior maxilla and only poorly developed on the premaxilla.</p> <p>DESCRIPTION OF HOLOTYPE OF EBRACHOSUCHUS NEUKAMI</p> <p>General: The skull (Figs 9–14) is essentially complete (missing lower jaws) although the teeth are mostly missing and there is minor damage to some areas (e.g. lateral rims of the external nares, ectopterygoid). As in Paleorhinus angustifrons, most of the cranial sutures are tightly closed or fused. The skull has been crushed dorsoventrally, exaggerating its apparently low profile (Figs 9C, D, 10D). The rostrum is very long and slender, with the preorbital length being nearly 3.9 times as long as the orbit + postorbital length (skull length ∼ 635 mm; preorbital length ∼ 505 mm; orbit + postorbital length ∼ 130 mm; transverse width across squamosals = 144 mm). Among phytosaurs, a similarly elongate and slender rostrum is only known in Mystriosuchus planirostris (e.g. Hungerbühler, 1998, 2002), in which the ratio of preorbital length to orbit + postorbital length may approach 3.9–4.0 (e.g. SMNS 9134). The skull roof of Ebrachosuchus is strongly ornamented, particularly on the lateral portions of the nasals, the prefrontals, dorsal parts of the lacrimals, frontals, parietals, postfrontals and dorsal parts of the postorbitals.</p> <p>Fenestral morphology: The external nares (Figs 9, 10A, D, 11C, D: ‘en’) have a sub-triangular outline, being broad and rounded posteriorly and tapering anteriorly. Their lateral rims (formed by the nasals) are broken on both sides, although the lateral rim is more complete on the left side. The nares face dorsally and slightly laterally. By comparison with Paleorhinus angustifrons, the nares are relatively short and broad (maximum width of the external naris is ∼60% of the narial length in Ebrachosuchus neukami and ∼33% of the narial length in Paleorhinus angustifrons) with a much broader internarial septum. The external naris of Paleorhinus bransoni appears more similar in its proportions to Paleorhinus angustifrons than to Ebrachosuchus (Lees, 1907; Stocker, 2010). The external nares of Ebrachosuchus are positioned notably proportionately more posteriorly than those of Paleorhinus angustifrons, Paleorhinus bransoni (Lees, 1907; Stocker, 2010), and ‘ Paleorhinus ’ sawini (Long &amp; Murry, 1995): the posterior rim of the external naris in Ebrachosuchus is still positioned anterior to the anterior corner of the internal antorbital fenestra, but is only separated from it by a very short distance (c. 7 mm; see above). The posterior rims of the external nares of Ebrachosuchus are not so strongly rugose and raised as in Paleorhinus angustifrons, and, in contrast to the latter species, are not raised above the surface of the nasals.</p> <p>The internal antorbital fenestra (Figs 9, 10A, D, 11D–F: ‘afen’) of Ebrachosuchus faces laterally and has a lenticular or eye-shaped outline (c. 43 mm long; 16 mm deep). The fenestra is absolutely and proportionally smaller than in Paleorhinus angustifrons (the anteroposterior lengths of the internal antorbital fenestra and the orbit are nearly identical in Ebrachosuchus, whereas the anteroposterior length of the internal antorbital fenestra is 140% or more of the length of the orbit in Paleorhinus angustifrons, Paleorhinus bransoni, the Krasiejów phytosaur specimens, and Parasuchus hislopi: Chatterjee, 1978; Dzik &amp; Sulej, 2007; Stocker, 2010), despite the fact that the skulls of Ebrachosuchus and Paleorhinus angustifrons are closely similar in size (Ebrachosuchus is very slightly larger, based on width across the squamosals and the orbital size). In Ebrachosuchus the antorbital fossa is greatly reduced: it is limited to an area on the lacrimal at the posteroventral corner of the internal antorbital fenestra. On the maxilla, there is only a slight transverse expansion of the anteroventral margin of the border of the internal antorbital fenestra, which might represent a remnant of the maxillary antorbital fossa. Similar reduction of the antorbital fossa occurs in ‘ Paleorhinus ’ sawini and most phytosaurids (Stocker, 2010).</p> <p>◀</p> <p>The orbit of Ebrachosuchus (Figs 9, 10A, D, 11F, G: ‘orb’) faces dorsolaterally and is oval in outline (left orbit: anteroposterior length = 44 mm; dorsoventral depth = 33 mm). As in Paleorhinus angustifrons and other basal phytosaurs (see above) the ventral margin of the orbit of Ebrachosuchus is positioned slightly ventral to the dorsal margin of the internal antorbital fenestra.</p> <p>The laterally and slightly dorsally facing infratemporal fenestra of Ebrachosuchus (Figs 9, 10A, D, 12A: ‘itf ’) is highly distinctive and differs greatly from that of Paleorhinus angustifrons and other basal phytosaur taxa. Rather than having a subquadrate outline it is strongly extended anteriorly (anteroposterior length = 68 mm; maximum dorsoventral height = 37 mm), tapering in length towards its anteroventral corner, and terminates anterior to the midpoint of the orbit. The infratemporal fenestra is therefore much larger than the orbit, contrasting strongly with the typically trapezoidal infratemporal fenestra of Paleorhinus angustifrons, Paleorhinus bransoni, the Krasiejów phytosaur specimens, ‘ Paleorhinus ’ sawini, and Parasuchus hislopi, which in all cases is subequal in size to the orbit and terminates anteriorly beneath the posterior margin of the orbit (Chatterjee, 1978; Long &amp; Murry, 1995; Dzik &amp; Sulej, 2007; Stocker, 2010). By contrast, the infratemporal fenestra is greatly enlarged and extends beneath the orbit in all phytosaurids, including Angistorhinus (Mehl, 1913, 1915), although in these taxa the infratemporal fenestra is also proportionately much deeper dorsoventrally. The most dorsal part of the infratemporal fenestra of Ebrachosuchus is positioned somewhat below the mid-height of the orbit.</p> <p>The dorsally facing supratemporal fenestra of Ebrachosuchus (Figs 9A, 10A, 11H: ‘stf’) has a teardrop-shaped outline that tapers to an apex posterolaterally. As in Angistorhinus, Brachysuchus megalodon, and non-phytosaurid phytosaurs such as Paleorhinus angustifrons (see above), the supratemporal fenestrae are in the same horizontal plane as the skull roof and completely open dorsally. Although a true supratemporal fossa is absent, the posterior temporal bars (the parietal–squamosal bars) are strongly inclined posterodorsally, so that their anterodorsal surfaces form somewhat depressed surfaces that are continuous with the posterior rims of the supratemporal fenestrae. The posterodorsally orientated parietal–squamosal bars contrast with the condition in Paleorhinus angustifrons, in which the bars are essentially vertically orientated.</p> <p>The posttemporal fenestra (Figs 9E, 10C, 12B, C: ‘ptf ’) is elongate and slit like, with margins formed by the otooccipital ventrally, and probably the squamosal and supraoccipital dorsally, although sutures are not clearly identifiable. As in Paleorhinus angusifrons, the posttemporal fenestra is placed dorsal to the foramen magnum and approximately halfway between the foramen magnum and the lateral tip of the paroccipital process. A notable groove extends medially and slightly ventrally along the supraoccipital–otooccipital suture from the posttemporal fenestra towards the dorsal margin of the foramen magnum, fading out shortly before the margin of the foramen magnum (Fig. 12B). A groove extending laterally from the lateral margin of the posttemporal fenestra is bordered by the overhanging squamosal shelf and the posterodorsally facing paroccipital process. Similar grooves are present in Paleorhinus angustifrons, but are less strongly defined.</p> <p>No single large post-parietal foramen is present, although much smaller paired openings are present (Fig. 12B: ‘for’) in an equivalent position to the large post-parietal foramen of Paleorhinus angustifrons (see above).</p> <p>The foramen magnum (Figs 9E, 10C, 12B: ‘fm’) is broad and oval in outline, with a straight dorsal margin, although the very low and broad shape has probably been exaggerated by dorsoventral compression of the skull. As in Paleorhinus angustifrons, the foramen magnum is bordered by the otooccipitals dorsolaterally, laterally, and ventrolaterally, with the supraoccipital forming only a small median part of the dorsal rim. Whether the basioccipital participates in the ventral margin of the foramen cannot be established with certainty, although it seems likely that any basioccipital contribution was very small.</p> <p>The subtemporal fenestrae (Figs 9B, 10B) are also very similar to those of Paleorhinus angustifrons, in both their shape and their relative size. The suborbital fenestra (Figs 9B, 10B, 13B, C: ‘sub’) is elongate and slit-like, with its long axis extending anterolaterally-to-posteromedially. It is proportion- ately narrower and lacks the curvature seen in Paleorhinus angustifrons. The margins of the suborbital fenestra are formed predominantly by the ectopterygoid posterolaterally and the palatine anteromedially, with the maxilla forming its lateral corner. The choanae (Figs 9B, 10B, 13B: ‘cho’) are slit-like with rounded anterior margins, but their posterior margins appear to have broken away. The choanae are positioned medial to the antorbital fenestra and posterior to the external nares.</p> <p>Premaxilla: The premaxilla is strongly compressed dorsoventrally (Figs 9, 10: ‘pm’), with a dorsal surface that is gently convex transversely. The external surface is generally smooth and unornamented, with the exception of a number of longitudinally extending grooves and lineations, which are more common and clearly defined along the posterior half of the element. The interpremaxillary suture is open and the premaxillae have slightly separated from one another post-mortem such that a clear line of sediment, 3 mm thick, divides them on the dorsal surface. The anterior termination of the premaxilla is slightly downturned in lateral view, and is rounded and expanded transversely to form the terminal rosette or bulb, as occurs in many other phytosaurs. The terminal rosette (Fig. 11A) is as wide as long, as in Paleorhinus bransoni (Long &amp; Murry, 1995), the Krasiejów phytosaur specimens (Dzik &amp; Sulej, 2007), Parasuchus hislopi (Chatterjee, 1978), and Mystriosuchus westphali (Hungerbühler, 2002), but differing from the rosettes of Angistorhinus (Mehl, 1915) and Pravusuchus (AMNH FR 30646; Stocker, 2010), which are wider than long. Immediately posterior to the rosette, the premaxillae of Ebrachosuchus are slightly contracted transversely in dorsal view to a minimum width of the snout of 22 mm. From this point on, the articulated premaxillae gradually expand posteriorly to a width of some 40 mm at the anterior end of the suture with the maxilla. The dorsal and ventral margins of the premaxilla are straight in lateral view along most of its length, but the dorsal margin becomes slightly concave and the ventral margin slightly convex posteriorly as the premaxilla arches gently upwards towards the external nares. There is no ‘rostral crest’, unlike the condition in most pseudopalatine phytosaurs (Stocker, 2010). In dorsal view, the lateral margins of the premaxillae are straight.</p> <p>The exact number of premaxillary teeth (none of which is preserved) is uncertain, because the exact position of the premaxilla–maxilla contact cannot be determined [there is no constriction at this contact, unlike the condition in some other phytosaurs (e.g. Hungerbühler, 2000) and no identifiable change in alveolar diameter that might correspond to the transition], and the premaxillary alveoli and palate are poorly preserved. However, there appear to be at least 30 premaxillary alveoli posterior to the terminal rosette. The terminal rosette clearly contains two large (7–8 mm diameter) alveoli that open ventrally and slightly posteriorly on the anterior margin (Figs 11A, 14A: ‘1’, ‘2’). Posterior to these enlarged alveoli, two additional ventrolaterally opening alveoli are clearly visible in CT data (Fig. 14A: ‘3’, ‘4’) but are poorly preserved on the external surface, and are followed by a short edentulous region (where the premaxillae are constricted transversely; Figs 11A, 14A: ‘edt’). Hungerbühler (2000) identified Ebrachosuchus as possessing a ‘bipartite’ upper dentition, in which the dentition is divided into a ‘tip-of-snout set’ and a large ‘post-tip set’, encompassing the entire post-rosette dentition. Thus, the complete premaxillary tooth count is estimated at 34–35. This count is higher than that reported for most other phytosaurs (see Gregory, 1962: table 1). The premaxillary alveoli are subcircular and gradually increase in size posteriorly: the first postrosette alveolus in the left premaxilla has a diameter of 3 mm, whereas a posterior alveolus (estimated as the 25th post-rosette alveolus) has a diameter of 6 mm. The alveoli face ventrally and slightly laterally. The teeth were well separated, with the interalveolar septa generally being at least half the anteroposterior width of the alveoli or even more; there is some irregularity in the spacing of the teeth. There also seems to be slight left–right asymmetry both in the spacing of the teeth and in the size of the alveoli, although the latter might be influenced by preservation.</p> <p>On the ventral surface, the premaxillary palate is formed by medially extending shelves that have been damaged and become displaced and deformed along much of their length (Figs 9, 10, 11B, 14B). At their lateral margins, adjacent to the alveoli, these shelves are slightly thickened into weakly developed alveolar or palatal ridges along some parts of the premaxilla, but in other areas alveolar ridges appear to be absent. Alveolar ridges are also only weakly developed on the better-preserved palatal surfaces of the maxillae. The relatively weak development of the alveolar ridges distinguishes Ebrachosuchus from other phytosaurs, in which alveolar ridges are generally well developed, and is therefore considered autapomorphic. The premaxillary palate tapers in transverse width towards its posterior termination where it contacts the vomers, although sutures are difficult to trace in this area. Nevertheless, the preserved portions of the suture with the maxilla on the palate show this suture to be straight.</p> <p>Dorsally, the premaxillae possess posteromedial processes (Fig. 11C: ‘pmp’) that project between the ‘septomaxillae’ but do not appear to form any more than the anterior base of the internarial septum. This is similar to the condition in Paleorhinus angustifrons (see above). The exact nature of the contacts in the region where the premaxilla, maxilla, and nasals converge is unclear in Ebrachosuchus.</p> <p>CT data show that a median premaxillary cavity similar to that of Paleorhinus angustifrons, presumably housing a pneumatic diverticulum of the antorbital air sinuses (Witmer, 1997a), extends for the complete length of the premaxilla (Fig. 14B: ‘mvc’). Along much of the length of the premaxilla, the bases of the alveoli are continuous with, or incomplete separated from, this cavity, and a discrete alveolar neurovascular canal (as seen along at least part of the posterior premaxilla of Paleorhinus angustifrons) does not appear to be present, suggesting that the neurovascular bundle may have been carried within the median cavity. More posteriorly, the median sinus is continuous with the airway, external nares, and antorbital fossa. CT data indicate that each premaxilla is slightly dorsoventrally thickened adjacent to the midline interpremaxillary suture on the dorsal surface (Fig. 14B).</p> <p>Maxilla: As in Paleorhinus angustifrons, the maxilla of Ebrachosuchus is an anteroposteriorly elongate and dorsoventrally shallow bone (Figs 9, 10A, D, 11E: ‘max’), although the exact positions of its sutures with the premaxilla and nasals anteriorly are unclear on the external surface. Eighteen alveoli on the left side and 19 on the right side appear to belong to the maxillae. As a result, the complete tooth count for the upper jaw of Ebrachosuchus exceeds 50; this extremely high tooth count seems to be autapomorphic for this species (Gregory, 1962). Two complete maxillary crowns are preserved in situ in the left maxilla, and one is preserved in the right maxilla. In lateral view, the ventral margin of the maxilla is very subtly concave anteriorly, and then is nearly straight along its posterior half.</p> <p>Externally, the maxilla contacts the premaxilla anteriorly, the nasal dorsally, the lacrimal posteriorly (ventrally and dorsally to the internal antorbital fenestra), and the jugal posteriorly. The maxilla forms the anteroventral and anterodorsal rims of the antorbital fenestra (Fig. 11E) – in this area of the skull the internal and external antorbital fenestrae are essentially synonymous because no maxillary antorbital fossa is present. The maxilla is slightly thickened into a ridge where it forms the ventral rim of the antorbital fenestra. Below this ridge, the lateral surface of the maxilla is dorsoventrally concave, with the alveolar margin being drawn out laterally as a lateral ridge (best seen on the right side; the left side is distorted). The ascending process of the maxilla is slender and strongly posterodorsally directed. It slightly tapers posteriorly and meets the anterior process of the lacrimal dorsal to and at about the mid-length of the antorbital fenestra. The lateral surface of the ascending process is strongly sculptured, as is the case with the lateral parts of the nasals, prefrontals, and the dorsal part of the lacrimal. The ventral margin of the posterior half of the internal antorbital fenestra is formed by the medially inset anteroventral expansion of the lacrimal, with the suture between the lacrimal and the maxilla running along the lateral margin of the antorbital fossa.</p> <p>On the ventral surface, the alveoli generally have a subcircular outline (Fig. 13B). All of the alveoli, with the exception of those at the posterior end of the tooth row, are similar in size, with a diameter of ∼ 7 mm. The smallest alveoli at the posterior end of the tooth row have a diameter of 4–5 mm. The consistent alveolar size along the maxillary tooth row differs from the condition in Paleorhinus angustifrons, in which there is greater heterodonty in tooth size along the maxilla (inferred from variation in alveolar size). Furthermore, the largest alveoli in Paleorhinus angustifrons are considerably larger in absolute size than those of Ebrachosuchus (9 mm vs. 7 mm in diameter), although both skulls are of closely comparable overall size. Medial to the tooth row, the alveolar ridge is almost absent, and is only indicated by a slight thickening of the medial alveolar margins in a few places. As in the premaxilla, the alveoli are well spaced, with a notably wider spacing occurring between several teeth that are placed just in front of the level of the external nares.</p> <p>The maxilla is contacted by the ectopterygoid immediately posterior and medial to the last alveolus (Fig. 13C). The maxilla forms the anterolateral margin of the slit-like suborbital fenestra (Fig. 13B, C); this fenestra is more broadly separated from the maxillary alveoli than in Paleorhinus angustifrons. Anterior to the suborbital fenestra, the medial palatal margin of the maxilla is drawn out medially into a palatal flange that is elongate and triangular in outline in ventral view (Fig. 13B). At its broadest extent, this medial flange forms the anterolateral margin of the choana, thus separating the palatine from the anterior expansion of the vomer and the premaxilla.</p> <p>‘Septomaxilla’: The ‘septomaxillae’ appear to form the anterior half of the internarial septum (Figs 9, 10A, D, 11C: ‘sm’). Their posterior extensions terminate anterior to the point indicated by Kuhn (1936), although the exact position of the suture with the nasals is difficult to determine. The ‘septomaxillae’ articulate with one another along the midline posteriorly, and their anterior processes terminate a short distance anterior to the external nares. They are separated from one another on the midline anteriorly by posteromedial processes of the premaxillae (Fig. 11C), as in Paleorhinus angustifrons, the Krasiejów phytosaur specimens (ZPAL Ab III 200, 1943, Dzik &amp; Sulej, 2007), and Paleorhinus bransoni (TMM 31100-101; Stocker, 2010). Anterolaterally, the ‘septomaxillae’ are separated from the nasals by a groove that extends anteriorly from the anterior corner of the external naris (Fig. 11C: ‘gr’). This groove is longer than that in Paleorhinus angustifrons.</p> <p>Nasals: The nasals are elongate and transversely expanded elements that form the very broad, nearly flat area (only slightly convex transversely) of the skull roof between the orbits and the external nares (Figs 9, 10A, D, 11C–G: ‘na’). Their maximal transverse width (c. 48.5 mm) is more than 1.7 times the transverse width of the frontals between the orbits (c. 28 mm), rather than being subequal in width as in Paleorhinus angustifrons and Paleorhinus bransoni (Stocker, 2010).</p> <p>Anteriorly, the nasals apparently form the posterior half of the broad internarial septum and the posterior and lateral margins of the external nares, and appear to terminate at a point approximately level with the anterior termination of the ‘septomaxillae’. The internarial septum is broad and has a flattened dorsal surface. Unlike the condition in Paleorhinus angustifrons, the posterior rims of the external nares are not raised into rugose ridges and are not raised above the level of the main bodies of the nasals (Fig. 11C). The depression on the anterior portion of the nasal immediately posterior to the naris, which is present in Paleorhinus angustifrons, is absent in Ebrachosuchus (Fig. 11C). The nasals are broken along the lateral rims of the external nares on both sides; however, the left rim is sufficiently well preserved to indicate that this rim was raised and rugose along nearly its entire length, rather than just posteriorly as in Paleorhinus angustifrons. The posterolateral parts of the narial rims flare abruptly laterally in the posteriormost quarter of the nares, and the lateral narial rims slightly overhang the lateral surfaces of the anterior ends of the maxillae along their entire length (Fig. 11C). Laterally, the nasal contacts the maxilla, lacrimal, and prefrontal, and is excluded from the rim of the antorbital fenestra by a maxilla–lacrimal contact. The nasals are gently convex transversely, and their external surfaces are rugosely ornamented laterally but relatively smooth at the midline internasal suture (Fig. 11C). The exact position of the suture with the frontals is not clear, but appears to be positioned between the prefrontals some 7 mm anterior to the anterior margin of the orbit – there is some evidence for an interdigitating transverse suture at this point.</p> <p>In CT data, a subtle longitudinal concavity is present on the ventral surface of each nasal (Fig. 14E: ‘naconc’), immediately adjacent to the internasal suture, and is visible for approximately 40 mm of nasal length, beginning shortly posterior to the external nares and reaching close to the posterior margin of the internal antorbital fenestra. This concavity covers approximately half of the width of the ventral surface of each nasal; in this area, the bone is dorsoventrally thick, whereas it is notably thinner more laterally. Furthermore, the lateral margin of the concavity seems to coincide roughly with the distinction between the rugose lateral part and the more smooth medial part of the external surface of the nasal. It is not clear what soft tissue structure this concavity represents, although Walker (1961) and Witmer (1997a) identified a median concavity on the ventral surface of the nasals of the aetosaur Stagonolepis as the impression of the nasal gland. The CT data for Paleorhinus angustifrons show only a single, very subtle median concavity spanning both nasals, which is mainly marked in its posterior part and is missing the distinction between a thickened medial and thinner lateral part.</p> <p>Prefrontal: The prefrontal forms the anteromedial and most of the anterior margin of the orbit (Figs 9, 10A, D, 11E–G: ‘prf’). It has convex lateral and medial margins in dorsal view, giving it a broadly oval outline. Its dorsal and lateral surfaces are strongly rugose and deeply pitted; as in Paleorhinus angustifrons, the prefrontal of Ebrachosuchus is one of the most strongly ornamented of the skull elements and is strongly thickened at its orbital margin. Also as in Paleorhinus angustifrons, the thickened orbital margin is distinctly offset from the rest of the surface of the bone, and a crescentic ‘pre-orbital depression’ (Hungerbühler, 2002) is present anterior to the orbit (Fig. 11E–G: ‘prod’). In contrast to Paleorhinus angustifrons, several irregularily arranged ridges subdivide this depression. Laterally, the prefrontal is broadly separated from the margin of the antorbital fenestra by the lacrimal. There is a small foramen in the prefrontal–lacrimal suture close to the anterior end of the prefrontal. This foramen pierces the skull roof and opens ventrally into the antorbital cavity (visible in CT data). The suture with the frontal and nasal is curved along its length.</p> <p>Frontal: The anterior contact of the frontals (Figs 9, 10A, 11G, H: ‘fr’) with the nasals is poorly defined but is placed between the prefrontals (see above). The paired depressions present on the anterior frontals of Paleorhinus angustifrons are absent in Ebrachosuchus. Laterally the frontal forms clear sutures with the pre- and postfrontal, and contributes to the dorsomedial rim of the orbit. Posteriorly the frontals form a transversely extending interdigitated suture with the parietals. The dorsal surfaces of the frontals are covered with rugose and pitted ornamentation similar to that of the prefrontal and other skull roof bones. Transversely, the frontals are slightly raised at the midline interfrontal suture, concave lateral to this suture, and then strongly raised and thickened at their orbital margins, much more so than in Paleorhinus angustifrons. The width of the frontals between the orbits is relatively narrow: it is approximately equal to the transverse width of the parietals. By contrast, in Paleorhinus angustifrons and the Krasiejów phytosaur specimens (ZPAL Ab III 200, Dzik &amp; Sulej, 2007) the frontals are substantially broader than the transversely narrow parietal. In CT data a marked transverse concavity for the olfactory bulbs is visible on the ventral surface of the frontals (Fig. 14H).</p> <p>Lacrimal: The lacrimal forms the anteroventral margin of the orbit and the posteroventral, posterior, and posterodorsal rims of the internal antorbital fenestra, contacting the maxilla both dorsal and ventral to the fenestra (Figs 9, 10A, D, 11E–G). The lacrimal additionally contacts the prefrontal and nasal dorsally, and the jugal ventrally. The lacrimal is extensively excavated on its lateral surface by the antorbital fossa, which is deepest anteriorly and shallows posteriorly. The posterior margin of the lacrimal antorbital fossa is not defined by a sharp rim, but rather merges smoothly with the lateral surface of the lacrimal adjacent to the orbit. The lateral surface of the lacrimal adjacent to the orbit has a smooth surface texture, but more dorsally and anteriorly the lacrimal is rugose and forms a rim that overhangs the antorbital fossa. The contact of the lacrimal with the maxilla appears to run anteroposteriorly, and is positioned medial to the ridge on the maxilla that forms the ventral margin of the antorbital fossa and fenestra. The nasolacrimal foramen is visible in CT data on the internal surface of the orbit just ventral to the midheight of the orbit, and is placed on the suture between the lacrimal laterally and the prefrontal medially. The narrow nasolacrimal canal extends anteriorly within the lacrimal (Fig. 14F: ‘lccn’), exiting the lacrimal dorsomedially, just posterior to the posterior rim of the internal antorbital fenestra. Two smaller canals branch off the main nasolacrimal canal (Fig. 14F), one exiting laterally just ventral to the dorsal rim of the lacrimal antorbital fossa, and one exiting ventrally. CT data also demonstrate that the medial margin of the lacrimal has a strongly developed ventral flange (Fig. 14F: ‘lcfl’), medioventral to the nasolacrimal canal. This flange defines a clearly defined and deep ventral fossa on the lacrimal, into which the second of the small canals extending from the nasolacrimal duct opens. Anteriorly, this flange converges upon and merges with the main body of the lacrimal medial to the lacrimal antorbital fossa. Posteriorly, this flange is continuous with a low ridge on the ventral surface of the prefrontal. A similar medioventral flange of the lacrimal is not present in Paleorhinus angustifrons.</p> <p>Jugal: The triradiate jugal has a distinctive morphology that reflects the anterior extension of the infratemporal fenestra beneath the orbit (Figs 9, 10A, D, 11F, 12A). The dorsal process of the jugal begins anterior to the midpoint of the orbit and extends primarily posteriorly and slightly dorsally. Rather than facing laterally, the dorsal process of the jugal is twisted along its length to face primarily dorsally and is expanded transversely. Its lateral margin forms a sharp flange that overhangs the infratemporal fenestra and that is continuous with a similar flange on the lateral margin of the postorbital (Fig. 11F: ‘por’). The dorsal process forms the ventral and posteroventral margins of the orbit and the anterodorsal corner of the rim of the infratemporal fenestra. The dorsal process is overlapped dorsally by the ventral process of the postorbital. As in Paleorhinus angustifrons and some other phytosaurs (see above), the infratemporal fenestra extends as a shallow fossa onto the jugal at the point at which the dorsal process of the jugal connects to the posterior process of the jugal (Fig. 11F: ‘fo’). In Ebrachosuchus, the margin of this fossa is defined by a sharp ridge (preserved on the left jugal; the right jugal is reconstructed in this region) that is continuous with the lateral flange of the dorsal process, unlike the subtle break-in-slope present in Paleorhinus angustifrons. A shallow, triangular depression extends anterior to this ridge on the lateral side of the jugal below the lacrimal antorbital fossa (Fig. 14G: ‘jgfo’), being separated from the lacrimal antorbital fossa by a broad, swollen area at the suture between the lacrimal and jugal.</p> <p>The elongate posterior process of the jugal is shallow dorsoventrally (Fig. 12A), much shallower relative to its length than in other basal phytosaurs (e.g. Dzik &amp; Sulej, 2007; Stocker, 2010) including Paleorhinus angustifrons. The posterior process forms the anterior half of the ventral margin of the infratemporal fenestra, and tapers posteriorly. It forms an elongate, posteroventrally sloping suture with the quadratojugal and terminates anterior to the quadrate. Its lateral surface is reconstructed on the left side.</p> <p>The short and dorsoventrally deep anterior process of the jugal contacts the maxilla along a deep anterodorsally-to-posteroventrally extending suture, which begins at the posteroventral corner of the antorbital fenestra and reaches to the posterior end of the maxillary tooth row. The anterior process also forms an elongate posterodorsally-to-anteroventrally extending suture with the lacrimal, although the exact position of this suture is unclear; it probably coincides with the ventral margin of the lacrimal antorbital fossa. The lateral surface of the anterior process is rugose, and possesses a strong anteroposteriorly extending ridge (Fig. 14G: ‘jgri’) ventral to the concavity described above, which is continuous with the posterior process of the jugal; this ridge is similar to that present in Paleorhinus angustifrons. However, Ebrachosuchus lacks the discrete row of nodes on the lateral surface of the jugal that is present in Paleorhinus angustifrons.</p> <p>Medially, the jugal is contacted by the ectopterygoid adjacent to the jugal–maxilla contact as in other phytosaurs (Fig. 13C). The medial surface of the jugal is not visible due to sediment infilling the subtemporal fenestra.</p> <p>In CT data, an anteroposteriorly extending canal is visible within the jugal (Fig. 14G: ‘jgc’), as in Paleorhinus angustifrons (see above). Anteriorly, this canal enters the bone from the dorsomedial side adjacent to the ectopterygoid–jugal contact, a short distance posterior to the last maxillary alveolus and lateral to the midpoint of the suborbital fenestra. It extends posterolaterally into the jugal for less than 10 mm, and exits the bone via a ventral and posteriorly facing opening situated on the suture between the posterolateral process of the ectopterygoid (see below) and the jugal (Fig. 13C: ‘for’). The canal differs from that preserved within the jugal of Paleorhinus angustifrons in being relatively shorter, and in not extending as a groove along the medial surface of the posterior process of the jugal. In addition, there do not appear to be additional narrower canals branching off the main canal in Ebrachosuchus, unlike the condition in Paleorhinus angustifrons (see above).</p> <p>Postorbital: The postorbital is triradiate with long anterior and posterior processes and a short and broad medial process (Figs 9, 10A, D, 11F, H, 12A: ‘po’). It is ornamented on its dorsal surface with small, irregular pits. This ornamentation becomes less marked towards the sutures with the jugal, squamosal, and parietal. The postorbital forms the majority of the posterior rim of the orbit (where it is not strongly thickened relative to the rest of the bone), the anterodorsal rim of the infratemporal fenestra, and the anterolateral rim of the supratemporal fenestra. The ventral (descending) process of the postorbital is directed more anteriorly than ventrally, and is folded outwards at its lateral margin so that it forms a partly dorsally facing and nearly horizontal surface. The lateral margin of this horizontal surface of the ventral process forms a sharply developed flange that overhangs the anterodorsal part of the infratemporal fenestra (Figs 11F, 12A: ‘por’) and is continuous with a flange on the dorsal process of the jugal (see above) and the anterior process of the squamosal. The ventral process becomes thickened dorsoventrally towards the orbital margin. CT data show that the ventral process of the postorbital is very massive at its base.</p> <p>The interdigitating contact between the postorbital and the postfrontal is best seen on the right side, with the suture curving posteromedially away from the orbital rim. The postfrontal suture forms approximately two-thirds of the medial sutural contacts of the postorbital. With the parietal, the medial process of the postorbital forms a short, anteroposteriorly extending suture anterior to the most anterior point of the supratemporal fenestra. The posterior process of the postorbital terminates approximately at the level of the posterior margin of the infratemporal fenestra. The postorbital–squamosal bar is slightly more expanded transversely than in Paleorhinus angustifrons, with its lateral margin being developed into a distinct flange, continuous with the lateral flange of the ventral process, which overhangs the infratemporal fenestra (Fig. 14A). The posterior process of the postorbital tapers to a rounded tip in dorsal view in Ebrachosuchus, and this point appears to fit into a slot in the anterior process of the squamosal, as in other basal phytosaurs (see above). The postorbital–squamosal bar is strongly compressed dorsoventrally.</p> <p>The medial surface of the postorbital is not exposed. CT data suggest that contact between the laterosphenoid and postorbital is absent.</p> <p>Postfrontal: The postfrontal is a nearly circular element that forms the posteromedial rim of the orbit, contacting the postorbital laterally and posterolaterally, the parietal posteromedially, and the frontal anteromedially (Figs 9, 10A, 11H: ‘pof ’). Its external surface is strongly rugose on the medial side of the bone, but smooth towards the postorbital contact. It is thickened and slightly raised at its orbital margin. CT data suggest that at least a point contact between the laterosphenoid and the postfrontal is present ventrally.</p> <p>Parietal: The parietals contact the postorbitals laterally, the postfrontals anterolaterally, and the frontals anteriorly, and form the medial, anteromedial, and much of the posterior rims of the supratemporal fenestrae (Figs 9, 10A, 11H). Posteriorly, they contact the supraoccipital on the occiput and have long posterolateral processes for the contact with the squamosal. The dorsal surfaces of the parietals are strongly ornamented, with a shallow, broad fossa present on each parietal anteriorly, and a deeper arcuate pit present on each parietal posteriorly (Fig. 11H: ‘dpa’). The parietals are slightly thickened at their medial margins where they articulate with one another, although there is no development of a distinct sagittal crest. A parietal foramen is absent. The main bodies of the parietals expand laterally anteriorly, forming the anteromedial rims of the supratemporal fenestrae. The dorsal surface of the parietal forms a sharp, overhanging edge above the lateral braincase wall on the medial margin of the supratemporal fenestra. This morphology is mainly caused by the presence of a marked depression on the parietal–prootic suture on the lateral braincase wall. Posteriorly, the posterolateral corner of each parietal is drawn out to form the parietal components of the parietal–squamosal bars, which, as described above, are strongly inclined posterodorsally, so that their anterodorsal surfaces form somewhat depressed surfaces that are continuous with the posterior rims of the supratemporal fenestrae. The posterolateral parietal wing dorsally overlaps a medial process of the squamosal, and tapers to a point posterolaterally. It extends laterally up to the level of the lateralmost point of the supratemporal fenestra.</p> <p>In posterior view, the posterolateral wings of the parietals form the dorsal margin of the occiput, which they slightly overhang posteriorly. They contact the supraoccipital along most of their length, but laterally each wing overlaps a medial process of the squamosal. Whether the posterolateral wings contribute to the dorsal margins of the posttemporal fenestra is unclear. Medially, the parietals appear to form the dorsal margins of the paired small postparietal fenestrae.</p> <p>Squamosal: As in Paleorhinus angustifrons, the squamosal possesses four processes (postorbital, parietal, anteroventral, opisthotic), but lacks a posterior process extending posteriorly beyond the paroccipital process of the opisthotic (Figs 9, 10 A, C, D, 11H, 12A, C: ‘sq’).</p> <p>The postorbital process of the squamosal forms the posterior part of the postorbital–squamosal bar, the posterodorsal margin of the infratemporal fenestra, and the posterolateral rim of the supratemporal fenestra. The postorbital–squamosal bar is strongly compressed dorsoventrally and expanded transversely. In dorsal view, its anterior end is forked, with the lateral processes almost reaching the level of the anterior end of the supratemporal fenestra, whereas the medial process ends just posterior to the midlength of this opening. There is a shallow concavity on the dorsal surface of the squamosal, lateral to the contact between its parietal and postorbital processes, but Ebrachosuchus lacks the well-developed, anteroposteriorly extending furrow that is present in Paleorhinus angustifrons, Paleorhinus bransoni (TMM 31025-172; Stocker, 2010), and the Krasiejów phytosaur specimens (ZPAL Ab III 200, Dzik &amp; Sulej, 2007). Most of the dorsal ornamentation is placed within this concavity, whereas the rest of the surface of the squamosal is rather smooth. The lateral edge of the postorbital–squamosal bar is a sharp flange that overhangs the infratemporal fenestra and extends along the lateral margins of the postorbital and squamosal to reach the most posterolateral corner of the squamosal. This ridge does not bifurcate posteriorly, unlike the condition in Paleorhinus angustifrons, referred specimens of Paleorhinus bransoni (Stocker, 2010: fig. 9), and the Krasiejów phytosaur specimens (ZPAL Ab III 200, Dzik &amp; Sulej, 2007), but it is somewhat thickened in this region.</p> <p>The main body of the squamosal is dorsoventrally shallow in lateral view, as in other basal phytosaurs (e.g. Paleorhinus bransoni, TMM 31025-172; the Krasiejów phytosaur specimens, ZPAL Ab III 200). In lateral view, the squamosal extends beyond the quadrate head posteriorly, but only slightly posterior to the quadrate condyles. In dorsal view, the squamosal tapers posterolaterally but is rounded, rather than pointed at its posterolateral corner as occurs in Paleorhinus angustifrons. The squamosal is drawn out posteroventrally into a short opisthotic process (Fig. 12A, C: ‘opsq’), the posterior surface of which articulates with the paroccipital process and extends slightly further laterally than the latter.</p> <p>The anteroventral process of the squamosal tapers along its length and forms nearly the entire posterior margin of the infratemporal fenestra, articulating with the quadratojugal ventrally and the quadrate posteriorly (Fig. 12A: ‘avpsq’). The anteroventral process is transversely compressed, smooth, and unornamented on its lateral surface. Dorsally, the process is separated from the dorsal surface of the squamosal by the lateral ridge extending posteriorly from the dorsal rim of the infratemporal fenestra, which laterally overhangs the base of the anteroventral process.</p> <p>Medially, the parietal process forms the most posterolateral part of the parietal–squamosal bar and at least the lateral part of the dorsal margin of the slit-like posttemporal fenestra. The process tapers medially, and is dorsally overlapped by the posterolateral wing of the parietal. The parietal process reaches almost to the medial margin of the supratemporal fenestra. The contacts of the squamosal with the supraoccipital and parietal above the posttemporal fenestra on the occiput are unclear.</p> <p>Quadratojugal: The quadratojugal is a hook-shaped element (Figs 9, 10A, D, 12A: ‘qj’) that has an elongate, dorsoventrally compressed, tapering anterior process (broken off on the left side and reconstructed, but complete on the right side), which forms the posterior half of the ventral rim of the infratemporal fenestra. The process articulates with the posterior process of the jugal along an elongate anterodorsalto-posteroventrally inclined suture. The posterior process of the jugal overlaps the ventral part of the quadratojugal laterally and ends at about the level of the ventral process of the squamosal, a short way anterior to the quadrate condyles. The lateral surface of the anterior process of the quadratojugal is gently concave dorsoventrally, and it is thickened slightly at its dorsal margin. Posteriorly, the quadratojugal contacts the quadrate, separating it from the jugal. Dorsally the quadratojugal articulates with the anteroventral process of the squamosal, and forms a small ventral part of the posterior margin of the infratemporal fenestra. Posterodorsally it articulates with a laterally extending tongue of the quadrate that forms the roof of the quadrate foramen (Fig. 12C: ‘ltqd’). The quadratojugal forms the lateral margin of the large, posteriorly facing, quadrate foramen (Fig. 12A, C: ‘qf ’), which is considerably larger than that of Paleorhinus angustifrons or other phytosaurs. In Ebrachosuchus the quadrate foramen is ∼66% of the width of the foramen magnum, whereas it is 50% or less in other phytosaurs including Paleorhinus angustifrons, Mystriosuchus westphali (Hungerbühler, 2002, GPIT 261/001), Angistorhinus grandis (Mehl, 1915, FMNH UC 631), and Parasuchus hislopi (Chatterjee, 1978). The enlargement of the quadrate foramen appears therefore to be autapomorphic for Ebrachosuchus. On the right side, a short, narrow canal is present within the quadratojugal anterolateral to the quadrate foramen, and opens medially into the subtemporal fenestra (visible in CT data).</p> <p>Quadrate: The main body of the quadrate is triangular in posterior view (Figs 9E, 10C, 12C: ‘qd’), being transversely expanded at the ventral margin to form the articular condyles for the mandible and becoming narrower dorsally where it is tightly sutured with the squamosal and the paroccipital process. The dorsal end of the bone is remarkably broad and CT data show that it contacts the ventral surface of the squamosal in a saddle-shaped suture, with a broader medial condyle being separated from a lateral condyle by a marked concavity. At the mandibular articulation, the medial condyle extends slightly further ventrally in posterior view than does the lateral condyle (Fig. 12C: ‘qdc’). The width of the skull across the quadrate condyles (185 mm) is greater than that across the squamosals (145 mm), giving the skull a somewhat trapezoidal outline in posterior view. In ventral view the quadrate condyles are anteroposteriorly narrow, and their articular surface appears somewhat saddle-shaped, with shallow medial and lateral concavities separated by a median convexity, although this morphology has probably been distorted by crushing. As in Paleorhinus angustifrons, the medial condyle is set slightly more posteriorly than the lateral condyle.</p> <p>The quadrate articulates laterally with the quadratojugal and forms the ventral, medial, and dorsal margins of the quadrate foramen. The dorsal margin of the foramen is formed by a lateral flange of the quadrate that articulates with the quadratojugal and the anteroventral process of the squamosal (Fig. 12C: ‘ltqd’). Medial to the foramen the main body of the quadrate is thickened. The quadrate foramen passes directly anteriorly through the skull, opening into the subtemporal fenestra (visible in CT data).</p> <p>The pterygoid wing of the quadrate extends anteromedially as a robust flange (Fig. 12C: ‘ptqd’). It forms an extensive suture with the posterior wing of the pterygoid, with the pterygoid medially overlapping the quadrate. The ventral margin of the pterygoid wing is placed dorsal to the quadrate condyles and is flexed medially, to form an expanded medial shelf that extends along the full length of the pterygoid–quadrate plate. In ventral view, the articulation of the quadrate with the pterygoid along the pterygoid–quadrate plate is marked by a small swelling. Dorsal to the medial shelf, the medial surface of the pterygoid wing is strongly concave dorsoventrally.</p> <p>Pterygoid and epipterygoid: The posterior wing or quadrate process of the pterygoid (Figs 9, 10B, 13A, C: ‘pt’) curves posterolaterally from the midline, and its lateral surface overlaps the pterygoid wing of the quadrate. The posterior wing is strongly concave dorsoventrally in medial or posterior view (Fig. 12C), and its ventral margin is drawn out posteromedially as a horizontal medial shelf that is continuous with the medial shelf of the ventral margin of the pterygoid wing of the quadrate. Unlike the situation in Paleorhinus angustifrons, the wing does not appear to be strongly forked at its posterior margin. At the proximal end of the posterior wing of the pterygoid, the pterygoid is drawn out medially to form a subquadrangular, medially extending plate (Fig. 13A: ‘mppt’), the posteromedial corner of which is slightly drawn out posteriorly (best preserved on the right side) as in Paleorhinus angustifrons. The posterior part of this medial plate forms a cup-like articulation on its dorsal surface for the anteroposteriorly broad basipterygoid process of the basisphenoid. As in Paleorhinus angustifrons, an interpterygoid vacuity is present, through which the cultriform process of the parabasisphenoid can be seen, although this vacuity is wider and extends further anteriorly in Ebrachosuchus neukami (see below) and so proportionally more of the cultriform process is exposed.</p> <p>Anterior to the basipterygoid articulation, a short lateral flange of the pterygoid extends anterolaterally and ventrally to contact the ectopterygoid and palatine (Fig. 13A: ‘lfpt’); the posterolateral extremity of this flange is missing on both sides. Medial and anterior to this, flanges of the pterygoid extend dorsomedially to form the roof of the palate, and articulate with one another along the midline (Fig. 12A, C: ‘afpt’). These anterior flanges have been damaged badly and mostly lost, so their articular contacts with the palatines are unknown. Posteriorly, they are offset sharply dorsally from the basipterygoid articulation by a curved ridge extending from this articulation laterally onto the lateral flange of the pterygoid. At their medial margins the anterior flanges are markedly thickened into sharp ridges that arise from the medial edges of the medial sheets that form the basicranial articulation, as in Paleorhinus angustifrons. These ridges are continuous anteriorly with the ridges of the vomers. The anterior flanges are broadly separated from one another for at least 50% of their length, rather than tightly appressed to each other as in Paleorhinus angustifrons. CT data show that these thickened medial parts of the pterygoid anterior flanges are also drawn out dorsally into short curved sheets that extend either side of the cultriform process (Fig. 14G, H). The nature of the contact between the vomers and the anterior flanges of the pterygoids is not clear.</p> <p>CT data indicate that the epipterygoid is at least partially preserved on the right side. It articulates dorsally with a cup-like depression on the ventral margin of the laterosphenoid, and ventrally with the pterygoid, immediately anterior to the basipterygoid articulation. The element is a thin, triangular sheet of bone that seems to be somewhat inclined anteriorly.</p> <p>Ectopterygoid: The ectopterygoids (Figs 9, 10B–D, 13C: ‘ect’) are damaged along their posterior margins and adjacent to their contacts with the pterygoids. The ectopterygoid extends anterolaterally as a broad bar of bone from its medial contact with the pterygoid and palatine to its lateral contact with the posterior end of the maxilla and the adjacent anterior part of the jugal. The ectopterygoid forms the entire posterior margin of the suborbital fenestra and the anterior margin of the subtemporal fenestra (Fig. 13C). On the right side, a narrow slit-like depression, smaller than that of Paleorhinus angustifrons, is present adjacent to the pterygoid–ectopterygoid contact, and forms a shallow, blind pit that does not extend extensively into the ectopterygoid (Fig. 13C: ‘?ectf ’). At its distal end, the ectopterygoid is considerably expanded anteroposteriorly, and as a result has a more extensive contact with the medial surface of the jugal than in Paleorhinus angustifrons. Whereas the anterior expansion is robust, the posterior expansion is thin and rod-like (Fig. 13C: ‘ppect’), and forms the medial border of the foramen in the jugal–ectopterygoid suture described above.</p> <p>Palatine: The palatine (Figs 9B, 10B, 13B, C: ‘pal’) is elongate with a main body that is orientated horizontally, forming most of the lateral region of the posterior portion of the palate. This horizontal main body forms a medially convex shelf with a sharply defined medial margin in ventral view that narrows the palatal vault and more strongly approaches the opposing palatine than in Paleorhinus angustifrons. This medial margin forms a projecting ridge that separates the lateral part of the palatine from the dorsomedial wing. Posteriorly, the palatine articulates with the pterygoid and ectopterygoid adjacent to the pterygoid/ectopterygoid suture and forms the entire anteromedial margin of the suborbital fenestra. Medially, the dorsomedial wings of the palatines appear to have broken away, and thus no contacts with the pterygoids are visible. Anterior to the suborbital fenestra the palatine tapers in transverse width as it articulates with a medially extending shelf of the maxilla (Fig. 13B: ‘atpal’). This maxillary shelf excludes contact between the premaxilla and palatine.</p> <p>Vomer: The vomers form the septum separating the choanae (Figs 9B, 10B, 13B: ‘v’); anteriorly they are expanded to form a plate that contacts the maxilla and premaxilla, although the contacts in this area are unclear. The positions of the posterior contacts with the pterygoid are not clear.</p> <p>Supraoccipital: The external surface of the supraoccipital faces posterodorsally (Figs 9E, 10C, 12B: ‘so’), and is broadly exposed dorsally posterior to the parietal–squamosal bars. There is a low median ridge in the dorsalmost third of the bone that becomes less pronounced posteroventrally; either side of this ridge the surface of the supraoccipital is gently concave. Dorsally, the supraoccipital articulates with the main body of the parietal and the parietal component of the parietal–squamosal bar. Unlike the condition in Paleorhinus angustifrons, there is no large foramen at the junction between the supraoccipital and the parietals, although notable depressions are present in this area; however, CT data suggest that very narrow slit-like foramina exit the bone immediately ventral to the parietal–supraoccipital contact, on either side of the median ridge within the depressions and open anteriorly into a sinus dorsal to the endocast. Ventrally, the supraoccipital is thickened into a low, transversely extending ridge along the extensive articulation with the opisthotic. The supraoccipital forms only a small median part of the dorsal margin of the foramen magnum. At its most lateral corners, the supraoccipital forms the medial rim of the slit-like posttemporal fenestra. There is no sign of separate interparietal or tabular ossifications.</p> <p>Otooccipital: The exoccipital portions of the otooccipitals (see discussion above) form the lateral and dorsolateral margins of the oval foramen magnum as well as most of the ventrolateral margin (Fig. 12B: ‘oto’), although the contacts with the basioccipital are unclear. The parts lateral to the foramen magnum are offset from the dorsal portions by a posterior ridge that is continuous with the ventral margin of the paroccipital processes. On the lateral surface of the opisthotic portion, the metotic fissure and fenestra ovalis are visible as deep recesses separated by the ventral process of the opisthotic. The fenestra ovalis is placed more anteriorly and is roofed over dorsolaterally by a strongly expanded otosphenoidal crest (crista prootica), which continues anteroventrally towards the basisphenoid. A short and narrow stapedial groove is present lateral to the fenestra ovalis on the broad ventral surface of the paroccipital process. A large depression is present below the fenestra ovalis in the ventral part of the otosphenoidal crest; the internal carotid might have entered the braincase through a foramen (which is not visible) within this depression. The metotic fissure seems to be at least partially subdivided by a posterior projection of the opisthotic into a smaller dorsal and a larger ventral opening, the former corresponding to the foramen lacerum posterius for cranial nerves IX–XI, whereas the latter represents the jugular foramen. Posterior to the metotic fissure and separated from it by a low lateral ridge is a triangular depression on the lateral side of the neck of the occipital condyle, in which the foramen for cranial nerve XII is probably placed.</p> <p>The paroccipital processes of the opisthotics extend posterolaterally (Fig. 10C, 12C: ‘par’), with their external surfaces facing posterodorsally and broadly exposed in dorsal view, substantially more so than in Paleorhinus angustifrons. At their bases, the processes are anteroposteriorly expanded and dorsoventrally shallow, but distally they become more compressed anteroposteriorly and gently expanded dorsoventrally, with their tips being gently rounded in posterior view. The paroccipital process forms the ventral margin of the elongate, slit-like posttemporal fenestra. Laterally, the paroccipital process forms a suture with the quadrate along the anterior margin of the process. Distally, the paroccipital process is buttressed by the opisthotic process of the squamosal.</p> <p>Basioccipital and parabasisphenoid: The basioccipital (Figs 9B, E, 10B, C, 13A: ‘bo’) forms most of the dorsoventrally crushed occipital condyle, the ventral surface of which is transversely convex. The neck of the basioccipital separates the condyle from the basitubera, but it is only a few millimetres narrower than the widest point of the condyle. Anteroventrally, the basioccipital flares laterally to form the basioccipital tubera, which form the most posterior parts of the basitubera (Fig. 13A: ‘bt’). The basitubera are separated posteriorly by a broad incision (Fig. 13A: ‘slt’), but connected to one another on the ventral midline by a low transverse ridge of the basisphenoid.</p> <p>The basitubera are expanded broadly, being mediolaterally wider than the occipital condyle, and extend slightly ventral to the condyle. They are proportionally larger, more robust, and broader than in Paleorhinus angustifrons: in Ebrachosuchus the mediolateral width of the basitubera is greater than the mediolateral width of the articular condyles of the left quadrate, whereas the basitubera are substantially narrower than the articular condyles of the quadrate in Paleorhinus angustifrons. The basitubera are flattened along their posteroventral surfaces, presumably for muscle attachment, although this might be a result of crushing.</p> <p>The basisphenoid is an anteroposteriorly compressed element that is expanded transversely at its anterior and posterior ends; posteriorly it forms the anterior part of the basitubera, whereas anteriorly the basipterygoid processes extend laterally, ventrally, and slightly posteriorly to articulate with the pterygoid. The processes thus face anterolaterally and are much broader anteroposteriorly than in Paleorhinus angustifrons. The transverse width across the basitubera is slightly greater than the transverse width across the basipterygoid processes. There is a shallow concavity that covers the ventral midline of the basisphenoid adjacent to the contact with the basioccipital (Fig. 13A: ‘conc’); this concavity is broader than long, rather than being subcircular as in Paleorhinus angustifrons, reflecting the anteroposterior compression of the basisphenoid. The depression is connected to the ventral surface of the cultriform process of the parasphenoid anteriorly by a broad ventral groove.</p> <p>Anterior to the main body of the basisphenoid, the elongate cultriform process extends anterodorsally as a narrow process of the parabasisphenoid. Its anterior termination is hidden by the pterygoids, but CT data show that, as in Paleorhinus angustifrons, it has a V-shaped cross section and terminates a short distance posterior to the end of the maxillary tooth rows.</p> <p>Dentition: Three maxillary teeth are preserved in situ, two in the left maxilla (one from the anterior part of the maxilla and one from the middle part) and one in the posterior half of the right maxilla, although none is well preserved. The maxillary teeth are similar in size to one another (total crown heights of approximately 10–12 mm), despite their differing positions along the tooth row, and are apicobasally short and robust. The apices of the crowns are curved both posteriorly (recurved) and medially. The posterior crown in the right maxilla possesses carinae that cover nearly the entire crown height on the mesiolingual and distolabial surfaces, but preservation is not sufficient to recognize the presence or absence of serrations.</p></div> 	https://treatment.plazi.org/id/871D87BB6D66FFF6FC647D69FAA1FB5C	Public Domain	No known copyright restrictions apply. See Agosti, D., Egloff, W., 2009. Taxonomic information exchange and copyright: the Plazi approach. BMC Research Notes 2009, 2:53 for further explanation.		Plazi	Butler, Richard J.;Rauhut, Oliver W. M.;Stocker, Michelle R.;Bronowicz, Robert	Butler, Richard J., Rauhut, Oliver W. M., Stocker, Michelle R., Bronowicz, Robert (2014): Redescription of the phytosaurs Paleorhinus (‘ Francosuchus’) angustifrons and Ebrachosuchus neukami from Germany, with implications for Late Triassic biochronology. Zoological Journal of the Linnean Society 170 (1): 155-208, DOI: 10.1111/zoj.12094, URL: http://doi.wiley.com/10.1111/zoj.12094
871D87BB6D55FFF2FC157CA5FEB8F9CE.text	871D87BB6D55FFF2FC157CA5FEB8F9CE.taxon	http://purl.org/dc/dcmitype/Text	http://rs.tdwg.org/ontology/voc/SPMInfoItems#GeneralDescription	text/html	en	Paleorhinus undetermined	<div><p>PALEORHINUS SP.</p> <p>‘ Francosuchus sp. ’; Kuhn, 1936: 65, pl. 11, fig. 4</p> <p>Material: BSPG 1931 X 503, partial skull described as Francosuchus sp. (Kuhn, 1936: 65–67, pl. 11, fig. 4; Figs 15, 16).</p> <p>Locality and horizon: Bed 5 of Kuhn (1933, 1936), Ebrach quarry.</p> <p>DESCRIPTION OF BSPG 1931 X 503</p> <p>General: BSPG 1931 X 503 consists of a partial rostrum (Figs 15, 16) preserved largely in articulation or close association, including the complete premaxillae, most of both maxillae, probable fragments of the nasals and ‘septomaxillae’, parts of the right lacrimal and jugal, and parts of the palate including vomers, palatines, ectopterygoids, and probable fragments of the pterygoid. The specimen has undergone some dorsoventral crushing, and the dorsal part of the rostrum (above the ventral margin of the antorbital fenestra) is lost. The rostrum belongs to a moderately larger individual than the holotypes of Paleorhinus angustifrons (BSPG 1931 X 502) and Ebrachosuchus (BSPG 1931 X 501), based upon rostrum width at the level of the anterior margins of the choanae (95 mm in BSPG 1931 X 503, although this has probably been exaggerated by dorsoventral compression; 75 mm in Paleorhinus angustifrons; 80 mm in Ebrachosuchus) and the size of the alveoli. Despite this, the rostrum is shorter in absolute length than that of Ebrachosuchus (length of the rostrum anterior to the external nares is approximately 325 mm in BSPG 1931 X 503, and approximately 390 mm in Ebrachosuchus). The number of alveoli (approximately 44 on each side; see below) is also fewer than in Ebrachosuchus.</p> <p>The position of the posterior margin of the external nares relative to the antorbital fossa can only be determined approximately. The anterior margins of the external nares are placed above the anterior end of the eleventh maxillary tooth position as counted from the back of the tooth row, as opposed to the twelfth position in Paleorhinus angustifrons. The anterior margin of the antorbital fenestra can be established on the basis of the remnants of the ascending process of the maxilla, which is preserved on both sides. Thus, the anterior end of this fenestra was placed above the anterior end of the eighth maxillary alveolus as counted from the back of the tooth row. If the skull of Paleorhinus angustifrons is aligned with this snout so that the anterior ends of the external nares are level, the anterior end of the antorbital fenestra of BSPG 1931 X 503 is placed slightly more anterior, despite the fact that this specimen represents a larger individual. This suggests external nares that were placed proportionately slightly more posteriorly in BSPG 1931 X 503 when compared with Paleorhinus angustifrons.</p> <p>Premaxilla: The premaxilla is strongly compressed and crushed, with a dorsal surface that is gently convex transversely. The external surface generally is smooth and only very weakly ornamented. As in Ebrachosuchus, the dorsal and ventral margins of the premaxilla are straight in lateral view along most of its length, but the dorsal margin becomes slightly concave and the ventral margin slightly convex posteriorly as the premaxilla arches gently upwards towards the external nares. The anterior end of the premaxilla arches first slightly upwards from the premaxillary–maxillary tooth row and the tip is then slightly downturned in lateral view. However, unlike the downturned premaxillary tip in Ebrachosuchus, the anteriormost tip of the snout is placed at approximately the same level as the pos- terior premaxillary tooth row. In dorsal and ventral views, the lateral margins of the premaxillae are straight, but diverge gently posteriorly as the snout becomes broader towards the external nares. The anterior termination of the premaxilla is rounded in dorsal and ventral views and expanded transversely to form the terminal rosette, which is as wide as long, as in Ebrachosuchus. However, in comparison with the latter taxon, the rosette is less strongly expanded relative to the width of the premaxillae posterior to it. Immediately posterior to the rosette, the premaxillae of BSPG 1931 X 503 are slightly contracted transversely.</p> <p>Although the position of the premaxilla–maxilla contact is difficult to identify with certainty, 27 alveoli appear to present in the right premaxilla with 26 alveoli in the left premaxilla. Thus, the premaxillary tooth count is lower than the 34–35 estimated for Ebrachosuchus. The terminal rosette contains three large (8–10 mm diameter) ventrally and slightly posteriorly opening alveoli on its anterior and lateral margins, and one smaller alveolus (diameter 6 mm) immediately posterior to these. In contrast to Ebrachosuchus, in which the anteriormost two teeth are almost aligned transversely, the teeth in the rosette of this specimen are arranged in a gentle arch. The crown in the fourth alveolus is directed laterally in the left premaxilla, but ventrally in the right premaxilla. Posterior to this ‘tip-of-snout set’, there is a short diastema (length of one alveolus) at the level of the transverse contraction noted above, followed by the remaining 22–23 premaxillary alveoli. The premaxillary alveoli are subcircular and face ventrally and slightly laterally. There is some variation in alveolar diameter along the tooth row: the first alveolus in the right premaxilla is around 5 mm in diameter, whereas the largest alveolus (number 17) in the right premaxilla has a diameter of approximately 9.5 mm. There is a short edentulous section (probably pathological) of the left premaxilla, approximately 23 mm in length, between alveoli 14 and 15, although a shorter gap is also present on the right side between the same alveoli.</p> <p>The premaxillary palate is formed by medially extending shelves that are thickened at their lateral margins, adjacent to the alveoli, into alveolar ridges. These alveolar ridges begin immediately posterior to the ‘tip-of-snout set’ and continue along the entire length of the premaxillary palate. These alveolar ridges are better developed than in Ebrachosuchus (see above). The premaxillary palate tapers in transverse width towards its posterior termination where it contacts the vomers.</p> <p>Dorsally, the contacts of the premaxillae with the maxillae and with fragments of the nasals and ‘septomaxillae’ cannot be clearly identified.</p> <p>Maxilla: Only the ventral parts of the maxillae are preserved. As noted above, the position of the premaxilla–maxilla contact is not certain, but approximately 16 alveoli appear to be present on the right side and at least 17 are present on the left (so that complete tooth counts for the premaxilla and maxilla combined appear to be 43 on both sides). In lateral view, the maxilla has a straight ventral margin along its anterior third and is gently convex along its posterior two-thirds.</p> <p>The contacts of the maxilla with surrounding elements (premaxilla, nasal, lacrimal) cannot be determined externally. The anterior, ventral, and posterior margins of the antorbital fenestra are preserved on the right side: the fenestra is elongate (∼ 58 mm in length) and similar in size to that of Paleorhinus angustifrons, but substantially larger than that of Ebrachosuchus. An antorbital fossa is excavated shallowly into the maxilla at its anterior end but excavated deeply posteriorly. Below the antorbital fenestra, the external surface of the maxilla is flat to gently concave. The ascending process of the left maxilla is well preserved, but probably incomplete. It is a small, triangular dorsal projection positioned at about the mid-length of the maxillary body.</p> <p>The maxillary alveoli have subcircular outlines. As in Paleorhinus angustifrons, the alveoli are relatively small at the anterior end of the maxilla (presumed first maxillary alveolus on the right side has a diameter of 6.5 mm) and increase in size posteriorly, reaching a maximum in the region lateral to the choanae (the presumed 10th right maxillary alveolus has an anteroposterior diameter of 12 mm). Alveolus size decreases again slightly at the posterior end of the tooth row. This morphology differs from Ebrachosuchus, in which there is less variation in alveolar size along the maxillary tooth row. Medial to the tooth row, the alveolar ridge is continuous with that of the premaxilla, but becomes narrower and less pronounced posteriorly although it is present as a distinct thickening along the entire tooth row.</p> <p>The maxilla was contacted by the ectopterygoid immediately medial to the last alveolus, although the contact is disarticulated on the right side (and so the outline of the suborbital fenestra is uncertain) and the ectopterygoid is missing on the left. Lateral to the choanae the maxilla has a medially extending shelf that articulates with the palatine and excludes contact between the premaxilla and the palatine. The medial maxillary shelf lateral to the palatine is relatively narrower than in Paleorhinus angustifrons.</p> <p>Only fragments of the right lacrimal, ‘septomaxillae’, and nasals are present externally, and provide little anatomical information. A flattened fragment of what is probably part of the right jugal is present, and has several low raised bumps (Fig. 16B) that may correspond to the nodular row present in Paleorhinus angustifrons (BSPG 1931 X 502). The jugal body ventral to the nodes is dorsoventrally high, comparable to the Polish phytosaur specimen ZPal Ab III 111, but in contrast to Paleorhinus angustifrons, although this might at least partially be due to dorsoventral crushing of the skull of the latter.</p> <p>Palatine: Both palatines are preserved, the right in articulation with the maxilla and the left element disarticulated and slightly displaced posteriorly. The palatine is elongate and the main body is orientated horizontally. Medially, it possesses a sharp ridge that defines the lateral rim of the palatal vault. A dorsomedially extending flange of the palatine is present, probably overlapping the unpreserved pterygoid within the palatal vault. A dorsomedially directed flange marks the posterior end of the internal choana, which is entirely placed posterior to the external naris and is 38 mm in length. Anterior to the suborbital fenestra the palatine tapers in transverse width as it articulates with a medially extending shelf of the maxilla. In contrast to both Paleorhinus angustifrons and Ebrachosuchus, the palatine margin of the suborbital fenestra is concave, rather than convex. Consequently, the suborbital fenestra is relatively larger than in these taxa and was apparently elongate and teardrop-shaped in outline.</p> <p>Vomer: Part of the right vomer is present, forming the septum separating the choanae and contacting the premaxilla anteriorly. The sutural contacts of the element cannot be established and the posterior end is probably missing.</p> <p>Ectopterygoid: The right ectopterygoid is slightly disarticulated and damaged along its posterior and medial margins. A foramen cannot be identified medially, but this part of the element has broken away. Laterally the ectopterygoid is expanded broadly in an anteroposterior direction, with its posterior part drawn out into a narrow process that would have contacted the medial surface of the jugal. The anterior process for the articulation with the maxilla is long and tapering anteriorly, with the articular surface for the maxilla being set off from the main ectopterygoid body by a small step. The main transverse body of the ectopterygoid is twisted, so that the lateral part is almost horizontal, whereas the medial surface faces anteroventrally. The posterior margin, which formed the anterior rim of the subtemporal fenestra, is strongly concave.</p> <p>A fragmentary bone preserved behind the end of the left maxilla might represent a remnant of the pterygoid, but cannot be identified with certainty at present.</p> <p>Dentition: Parts of nine crowns, five from the anterior portion of the premaxillae, and four from the mid-toposterior portions of the maxillae are preserved, but most are damaged. Those of the anterior premaxilla particularly are preserved poorly, but appear to be relatively small, conical, not strongly recurved, and lack carinae and serrations. Those of the mid-toposterior portions of the maxilla are strongly recurved posteriorly and slightly medially towards their apices, and possess carinae along mesiolingual and distolabial surfaces. Although preservation is poor, serrations are visible along the anterior carina of the most posterior preserved crown in the right maxilla.</p> <p>Taxonomic affinities: BSPG 1931 X 503 differs from Ebrachosuchus in possessing a proportionally shorter snout with lower premaxillary and maxillary tooth counts, increased heterogeneity in alveolar size, a large antorbital fenestra with an anterior antorbital fossa, well-defined alveolar ridges, and a relatively larger suborbital fenestra with a concave anterior border. It is more similar in many features to Paleorhinus angustifrons (with the exception of rostrum length which is uncertain in Paleorhinus angustifrons), but differs in probably possessing an antorbital fenestra that is proportionally slightly smaller in size (the absolute sizes of the fenestrae are nearly identical in Paleorhinus angustifrons and BSPG 1931 X 503, but BSPG 1931 X 503 represents a moderately larger skull) and relatively closer to the external nares and a considerably larger, elongate, and teardrop-shaped suborbital fenestra. Although the bone is not well preserved, BSPG 1931 X 503 appears to possess the nodular row on the jugal that we consider diagnostic of Paleorhinus. For these reasons, we refer BSPG 1931 X 503 to Paleorhinus sp., although the differences noted above may indicate that it represents a species different from Paleorhinus angustifrons. Since these two specimens of Paleorhinus come from different horizons within the quarry, there is a possibility that they might represent a single anagenetic lineage.</p> </div>	https://treatment.plazi.org/id/871D87BB6D55FFF2FC157CA5FEB8F9CE	Public Domain	No known copyright restrictions apply. See Agosti, D., Egloff, W., 2009. Taxonomic information exchange and copyright: the Plazi approach. BMC Research Notes 2009, 2:53 for further explanation.		Plazi	Butler, Richard J.;Rauhut, Oliver W. M.;Stocker, Michelle R.;Bronowicz, Robert	Butler, Richard J., Rauhut, Oliver W. M., Stocker, Michelle R., Bronowicz, Robert (2014): Redescription of the phytosaurs Paleorhinus (‘ Francosuchus’) angustifrons and Ebrachosuchus neukami from Germany, with implications for Late Triassic biochronology. Zoological Journal of the Linnean Society 170 (1): 155-208, DOI: 10.1111/zoj.12094, URL: http://doi.wiley.com/10.1111/zoj.12094
871D87BB6D51FFF2FEAE7E3EFB49FC83.text	871D87BB6D51FFF2FEAE7E3EFB49FC83.taxon	http://purl.org/dc/dcmitype/Text	http://rs.tdwg.org/ontology/voc/SPMInfoItems#GeneralDescription	text/html	en	Phytosauria INDET.	<div><p>PHYTOSAURIA INDET.</p> <p>‘ Mystriosuchus cf. plieningeri ’; Kuhn, 1933: 97, 122, fig. 4</p> <p>‘Unterkiefer von Francosuchus (?)’; Kuhn, 1936: 67, pl. 10, fig. 2</p> <p>‘Unterkiefer von Ebrachosuchus (?)’; Kuhn, 1936: 68, pl. 10, fig. 1</p> <p>‘ Mystriosuchus cf. plieningeri ’; Kuhn, 1936: 91–92, pl. 13, fig. 1</p> <p>Material: BSPG lost specimens, impression of the left posterior part of a skull and partial right lower jaw described as Mystriosuchus cf. plieningeri (Kuhn, 1933: 121–122, fig. 4), and a partial right lower jaw described as Mystriosuchus sp. (Kuhn, 1936: 91–92, pl. 13, fig. 1). Two partial left lower jaws described as? Francosuchus (Kuhn, 1936: 67–68, pl. 10, fig. 2) and? Ebrachosuchus (Kuhn, 1936: 68, pl. 10, fig. 2). All specimens are presumed destroyed in World War II.</p> <p>Locality and horizon: All specimens collected from Bed 9 of Kuhn (1933, 1936), Ebrach quarry.</p> <p>Comments: Gregory (1962: 670–671) suggested that the lower jaw referred by Kuhn (1936) to Mystriosuchus sp. probably represented either Ebachosaurus or Francosuchus (both of which he considered junior synonyms of Paleorhinus), and referred this specimen to Paleorhinus broilii. All of this material is missing, and none of it can be identified beyond Phytosauria on the basis of the available photographs and descriptions.</p> </div>	https://treatment.plazi.org/id/871D87BB6D51FFF2FEAE7E3EFB49FC83	Public Domain	No known copyright restrictions apply. See Agosti, D., Egloff, W., 2009. Taxonomic information exchange and copyright: the Plazi approach. BMC Research Notes 2009, 2:53 for further explanation.		Plazi	Butler, Richard J.;Rauhut, Oliver W. M.;Stocker, Michelle R.;Bronowicz, Robert	Butler, Richard J., Rauhut, Oliver W. M., Stocker, Michelle R., Bronowicz, Robert (2014): Redescription of the phytosaurs Paleorhinus (‘ Francosuchus’) angustifrons and Ebrachosuchus neukami from Germany, with implications for Late Triassic biochronology. Zoological Journal of the Linnean Society 170 (1): 155-208, DOI: 10.1111/zoj.12094, URL: http://doi.wiley.com/10.1111/zoj.12094
871D87BB6D50FFEDFF277DB4FA95FCDF.text	871D87BB6D50FFEDFF277DB4FA95FCDF.taxon	http://purl.org/dc/dcmitype/Text	http://rs.tdwg.org/ontology/voc/SPMInfoItems#GeneralDescription	text/html	en	Ebrachosuchus neukami O. Kuhn 1936	<div><p>ANGUSTIFRONS AND EBRACHOSUCHUS NEUKAMI</p> <p>Our re-examination of the Ebrach phytosaur specimens has revealed a large number of substantial anatomical features that differ between Paleorhinus angustifrons and Ebrachosuchus neukami, including:</p> <p>1. The external nares of Ebrachosuchus are proportionally shorter and broader than those of Paleorhinus angustifrons, with a relatively broader internarial septum.</p> <p>2. The posterior rims of the external nares of Ebrachosuchus are not raised into thickened, rugose ridges, unlike the condition in Paleorhinus angustifrons, and the lateral rims of the external nares lack the distinct profile in lateral view that is present in Paleorhinus angustifrons (see above).</p> <p>3. The external nares of Ebrachosuchus are relatively more posteriorly positioned than those of Paleorhinus angustifrons, being separated by only a very short distance from the anterior corner of the antorbital fenestra.</p> <p>4. The internal antorbital fenestra is proportionately smaller in Ebrachosuchus (anteroposterior length equal to that of the orbit) relative to the condition in Paleorhinus angustifrons (anteroposterior length exceeds that of the orbit).</p> <p>5. The antorbital fossa is reduced in Ebrachosuchus, being limited to a small lacrimal antorbital fossa at the posteroventral corner of the internal antorbital fenestra. By contrast, Paleorhinus angustifrons retains an extensive maxillary antorbital fossa and a proportionately larger lacrimal antorbital fossa.</p> <p>6. The infratemporal fenestra of Ebrachosuchus is strongly expanded anteroposteriorly relative to that of Paleorhinus angustifrons, terminating ventral to the midpoint of the orbit.</p> <p>7. Alveolar ridges are only weakly developed on the premaxilla and maxilla of Ebrachosuchus, but are well developed in Paleorhinus angustifrons and other phytosaurs.</p> <p>8. Weaker heterodonty is present in the maxilla of Ebrachosuchus than in Paleorhinus angustifrons (inferred from variation in alveolar diameter).</p> <p>9. The postorbital–jugal bar is twisted along its length to face dorsally and is expanded transversely and overhangs the infratemporal fenestra in Ebrachosuchus, whereas this morphology is absent in Paleorhinus angustifrons.</p> <p>10. The posterior process of the jugal is shallower dorsoventrally relative to its length in Ebrachosuchus than in Paleorhinus angustifrons (probably correlated with difference 6).</p> <p>11. A strong anteroposteriorly extending ridge and nodular row is present on the lateral surface of the anterior process of the jugal of Paleorhinus angustifrons but is absent in Ebrachosuchus.</p> <p>12. The nasals of Ebrachosuchus are transversely expanded, forming a broad platform between the orbits and external nares, whereas the nasals of Paleorhinus angustifrons are much narrower.</p> <p>13. The combined transverse width of the frontals of Paleorhinus angustifrons is substantially greater than the combined width of the parietals, whereas the parietals are proportionately broader in Ebrachosuchus.</p> <p>14. The concavity on the dorsal surface of the squamosal is much more weakly developed in Ebrachosuchus than in Paleorhinus angustifrons.</p> <p>15. The ridge forming the lateral margin of the postorbital–squamosal bar bifurcates posteriorly in Paleorhinus angustifrons but not in Ebrachosuchus.</p> <p>16. The posterolateral corner of the squamosal is rounded in dorsal view in Ebrachosuchus but pointed in Paleorhinus angustifrons.</p> <p>17. The base of the dorsal process of the quadratojugal is relatively much broader anteroposteriorly in Paleorhinus angustifrons than in Ebrachosuchus.</p> <p>18. The lateral surface of the anterior process of the quadratojugal is dorsoventrally concave in Ebrachosuchus but is flat in Paleorhinus angustifrons.</p> <p>19. The quadrate foramen of Ebrachosuchus is greatly enlarged in size relative to that of Paleorhinus angustifrons, and is approximately two-thirds of the width of the foramen magnum.</p> <p>20. There is a large foramen at the junction between the parietal and supraoccipital in Paleorhinus angustifrons that is absent in Ebrachosuchus.</p> <p>21. The basisphenoid is anteroposteriorly compressed and the basitubera enlarged and anteroposteriorly and transversely expanded in Ebrachosuchus relative to the morphology of these elements in Paleorhinus angustifrons.</p> <p>22. The suborbital foramen is boomerang shaped in Paleorhinus angustifrons and slit-like in Ebrachosuchus.</p> <p>23. A large, subcircular ectopterygoid foramen is present in Paleorhinus angustifrons, as opposed to a slit-like foramen in Ebrachosuchus.</p> <p>24. A pronounced medial ridge on the palatine separates the horizontal main body of the bone from the dorsomedial wing and narrows the palatal vault medially in Ebrachosuchus.</p> <p>The skulls essentially are identical in size and are unlikely to represent individuals of radically different ontogenetic stages, so these substantial observed differences are unlikely to be attributable to ontogeny. The skulls were recovered within 10 cm of one another, within the same bedding plane (Kuhn, 1936), and it is worth considering the possibility of sexual dimorphism. Sexual dimorphism has been proposed for the phytosaur Machaeroprosopus on the basis of morphological differences in the robustness of the premaxilla and the development of a narial crest (Zeigler, Lucas &amp; Heckert, 2003). Major differences between the proposed female and male individuals of Machaeroprosopus are, however, limited to the preorbital region of the skull (Zeigler et al., 2003). By contrast, substantial differences between Paleorhinus angustifrons and Ebrachosuchus are not obvious in the premaxilla (at least, those parts preserved), but numerous and substantial differences do occur in the remaining parts of the preserved skulls. A number of these differences, particularly the reduction of the antorbital fossa and the anterior extension of the infratemporal fenestra (as well as potentially the more posterior position of the external nares) in Ebrachosuchus may be phylogenetically significant. Moreover, while Paleorhinus angustifrons closely resembles Paleorhinus bransoni and the Krasiejów phytosaur specimens, no Ebrachosuchus -like phytosaur has yet been described from North America or Krasiejów. Although sexual dimorphism cannot be excluded, we believe it is considerably more likely that Paleorhinus angustifrons and Ebrachosuchus neukami represent generically distinct taxa. These species may have occupied distinct ecological niches, based upon the substantial elongation of the rostrum of Ebrachosuchus. Other examples of coexisting but morphologically distinct genera of phytosaurs are known, such as the co-occurrence of Nicrosaurus and Mystriosuchus in deposits of the Löwenstein Formation of south-west Germany (see above), so nichepartitioning by different genera of phytosaurs may have occurred widely during the Late Triassic.</p> </div>	https://treatment.plazi.org/id/871D87BB6D50FFEDFF277DB4FA95FCDF	Public Domain	No known copyright restrictions apply. See Agosti, D., Egloff, W., 2009. Taxonomic information exchange and copyright: the Plazi approach. BMC Research Notes 2009, 2:53 for further explanation.		Plazi	Butler, Richard J.;Rauhut, Oliver W. M.;Stocker, Michelle R.;Bronowicz, Robert	Butler, Richard J., Rauhut, Oliver W. M., Stocker, Michelle R., Bronowicz, Robert (2014): Redescription of the phytosaurs Paleorhinus (‘ Francosuchus’) angustifrons and Ebrachosuchus neukami from Germany, with implications for Late Triassic biochronology. Zoological Journal of the Linnean Society 170 (1): 155-208, DOI: 10.1111/zoj.12094, URL: http://doi.wiley.com/10.1111/zoj.12094
