Research Article
Print
Research Article
A large brachyopoid from the Middle Triassic of northern Arizona and the diversity of brachyopoid temnospondyls from the Moenkopi Formation
expand article infoCalvin So‡, Arjan Mann§
‡ George Washington University, Washington, United States of America
§ Field Museum of Natural History, Chicago, United States of America

Corresponding author: Calvin So ( calvincfso@gmail.com )

Academic editor: Florian Witzmann
© 2024 Calvin So, Arjan Mann.
This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Citation: So C, Mann A (2024) A large brachyopoid from the Middle Triassic of northern Arizona and the diversity of brachyopoid temnospondyls from the Moenkopi Formation. Fossil Record 27(1): 233-245. https://doi.org/10.3897/fr.27.117611
ZooBank: urn:lsid:zoobank.org:pub:C084A289-5E44-4FD0-8FBB-AEEB61A83DD1
Open Access

Abstract

Brachyopoids represent a diverse and late surviving temnospondyl group, lasting until the Early Cretaceous. Here, we report on brachyopoid material previously assigned to Hadrokkosaurus bradyi that represents a distinct brachyopoid taxon, characterised by a smaller number of large, robust mandibular teeth, a feature rarely observed in other temnospondyls. We also revisit an angular previously referred to Hadrokkosaurus potentially belonging to other temnospondyl taxa present in the Middle Triassic of North America. In light of the abundance of material of possible taxa distinct from Hadrokkosaurus, we express the need to re-examine previously collected specimens as new information changes the landscape of palaeontology. Parsimony analyses using exclusively mandibular characters recover the new brachyopoid taxon from the locality in a polytomy with Hadrokkosaurus and Vanastega at the base of Brachyopoidea, adding to a diversity of mandibular morphology of temnospondyls in the Middle Triassic of North America.

Key Words

Amphibian, fossil, paleontology, phylogeny, temnospondyl

Introduction

The Triassic Period (ca. 252 to 201 Ma) is a critical stage of vertebrate recovery, evolution and survival for several major clades between two mass extinctions (Dal Corso et al. 2020). This period records the first occurrences of several major temnospondyl amphibian clades, including Brachyopidae, Chigutisauridae and lissamphibians (Piveteau 1936; Rage and Roček 1989; Ruta and Benton 2008). The general consensus considers lissamphibians to be modern day surviving temnospondyls (Bolt 1969; Pardo et al. 2017; Schoch et al. 2020; Kligman et al. 2023; but see Marjanović and Laurin (2007); Ruta and Coates (2007); Anderson et al. (2008) for other hypotheses for lissamphibian origins), which emphasises the importance of temnospondyl research in contextualising the evolution and origin of modern amphibians.

Following the Permo-Triassic extinction, brachyopid and chigutisaurid temnospondyls appeared and diversified across the globe, spreading across both Northern and Southern Hemispheres. Their presence would last until the Early Cretaceous, represented by Koolasuchus cleelandi (Warren et al. 1997) . The brachyopoids of what would become North America in northern Pangea during the Triassic are Hadrokkosaurus bradyi and Vigilius wellesi. Originally, Hadrokkosaurus bradyi was assigned the generic name Taphrognathus (a name occupied by a conodont; Welles (1947); Welles (1957)). Hadrokkosaurus once referred to a large set of specimens from two different quarries (V3922 and V4207) approximately 160 kilometres apart in Arizona from the Holbrook Member of the Moenkopi Formation (Welles and Estes 1969; Warren and Marsicano 2000). V3922 produced the holotype mandible of Hadrokkosaurus (UCMP 36199) and several disarticulated skeletal elements referred to Hadrokkosaurus. In V4207, a brachyopoid skull (UCMP 38165) was excavated and referred to Hadrokkosaurus by Welles (1957) and Welles and Estes (1969), but eventually placed in its own genus, Vigilius wellesi, by Warren and Marsicano (2000). However, in establishing Vigilius, they also assigned several mandibular, cranial and postcranial elements from V3922 to Vigilius (Warren and Marsicano 2000).

Here, we present a detailed description of the material from V3922 and uncover new temnospondyl diversity. Some of the material outside of the holotype cursorily addressed in previous publications suggests the presence of a third brachyopoid taxon in the Holbrook member of the Moenkopi Formation. The previously unknown brachyopoid taxon is recognised from several mandibular elements, including an incomplete right mandible that is preserved from the symphysis to the suture between the first and second coronoid. It is characterised by noticeably wider and rounder teeth that are fewer in number compared to Hadrokkosaurus. Phylogenetic analyses place this new taxon within Brachyopoidea.

Methods

The specimens were studied in person at the University of California Museum of Paleontology and the Field Museum of Natural History. Images of the specimens were photographed using a Canon EOS 7D with a Canon Zoom Lens EF 24-105 mm F/4L IS USM.

Terminology

The definition of Brachyopoidea and its relationship with Plagiosauridae is relevant to this study. Warren and Hutchinson (1983) found a monophyletic Brachyopoidea consisting of Brachyopidae and Chigutisauridae. Yates and Warren (2000) recovered Brachyopoidea as a paraphyletic grade towards Plagiosauridae and Laidleria (Plagiosauroidea). The majority of studies that included temnospondyl systematics in the past decade used the dataset of Schoch (2013) as the base, which also recovered Brachyopoidea forming a grade towards Plagiosauridae and Laidleria. Most recently, Witzmann and Schoch (2024) recovered a monophyletic Brachyopoidea sister to a monophyletic Plagiosauroidea, though they used Plagiosauroidea both including and excluding Laidleria. In this study, we follow the most recent results Witzmann and Schoch (2024) and operate on the definition of Brachyopoidea as the clade formed by Brachyopidae and Chigutisauridae.

Institutional abbreviations

UCMP, University of California Museum of Paleontology, Berkeley, California, USA; MCNAM-PV, Museo de Ciencias Naturales y Antropológicas Juan Cornelio Moyano paleovertebrados collection, Mendoza, Argentina; QM F, Queensland Museum, Brisbane, Queensland, Australia

Anatomical abbreviations

a, angular; af, adductor fossa; aMf, anterior Meckelian fenestra; cd, coronoid dentition; c1, first coronoid; c2, second coronoid; c3, third coronoid; d, dentary; dt, marginal dentition; pMf, posterior Meckelian fenestra; psy, postsymphyseal foramen; pra, pre-articular; pos, postsplenial; prs, presplenial; sa, surangular; sf, symphyseal fang; sym, mandibular symphysis.

Results

Systematic Paleontology

Temnospondyli Zittel, 1888

Stereospondyli Zittel, 1888

Brachyopoidea Lydekker, 1885

Hadrokkosaurus Welles, 1957

Hadrokkosaurus bradyi Welles & Estes, 1969

Holotype

UCMP 31699 (right mandible lacking only the articular).

Horizon and locality

Uppermost channel sandstone of Holbrook Member, Moenkopi Formation; early Anisian, lowermost Middle Triassic. V3922, Geronimo (Holbrook) fossil vertebrate quarry near Holbrook, Coconino County, north-eastern Arizona.

Referred material

UCMP 36200, anterior right dentary; UCMP 36201, partial right dentary; UCMP 36203, partial left dentary; UCMP 36205, partial left pre-articular; UCMP 36836, left pre-articular; UCMP 36837, left pre-articular; UCMP 36838, right surangular.

Revised diagnosis

(modified from Ruta and Bolt (2008)). A brachyopid temnospondyl with the following unique combination of features: total length of angular ventrolateral margin greater than or equal to half of total lateral mandible length; angular posteriormost margin straight in lateral aspect; greatest depth of angular lateral surface less than or equal to greatest depth of dentary lateral surface; ventral margin of posterior Meckelian fenestra formed only by angular; anterior Meckelian fenestra in middle third of postsplenial mesial lamina.

Description

The holotype right mandible possesses several features that maintain its status as a brachyopoid, such as the long postglenoid area and the curvature of the mandible that can be extrapolated to fit a broad and short-snouted temnospondyl (Fig. 1). It maintains the full plesiomorphic complement of ten bones found with most temnospondyls. The postglenoid area is substantially long, as present in other brachyopoid mandibles (Warren and Marsicano 2000). Several partial and fragmentary dentaries besides the holotype have been referred to Hadrokkosaurus.

Figure 1. 

UCMP 36199 holotype right mandible of Hadrokkosaurus bradyi photographed and illustrated in: labial view (A, C); lingual view (B, D); ventral view (E, G); and dorsal view (F, H).

The dentary of Hadrokkosaurus is thin (Fig. 1). The dorsal surface of the dentary (i.e. the dental shelf) is narrow, contributing to the gracile and narrow appearance of the mandible, but towards the symphysis, the shelf broadens to accommodate the larger symphyseal tusk. The dentition of the dentary is represented by small, lingually recurving teeth, with wider than long bases. There is a count of 32 tooth positions. Towards the tips, the dentary teeth are labiolingually compressed and carinated. UCMP 36200, UCMP 36201, UCMP 36203, UCMP 75434 and UCMP 152391 (Suppl. material 1: fig. S1) are partially preserved dentaries; they are likely also Hadrokkosaurus, based on the narrowness of the dental shelf and tooth sockets that would be implanted with small teeth.

The dentary forms a wide parabolic shape that curves towards the symphysis (Fig. 1). The straight linear measurement from the anteriormost tip of the symphysis to the posteriormost extent of the dentary measures approximately 21 cm. From the anteriormost tip of the mandibular symphysis to the anteriormost sutural contact between the first and second coronoid, it measures 9.5 cm. At the mandibular symphysis, the dentary is vertically short, but, as it continues posteriorly, it deepens considerably. The dentary sutures to the presplenial and postsplenial ventrally. The symphysis is formed entirely by the dentary. Lingually, the large postsymphyseal foramen is bounded by the dentary dorsally, the first coronoid posteriorly and the presplenial ventrally. The foramen exits into an open Meckelian canal that opens on the dentary beginning on the lingual surface and ends on the ventral surface as it continues towards the symphysis (Fig. 1B, D, E, G). The size of the postsymphyseal foramen is large and comparable to the condition in Bathignathus poikilops (Damiani and Jeannot 2002). The symphyseal tusk is angled posterodorsally on the symphyseal plate (Fig. 1). The dental row is framed lingually and labially by ridges. In the transverse aspect of the dentary, the ridges are formed by the dental shelf on the lingual side, while the labial ridge is formed by a thin lamina running along the length of the dentary. The dentary teeth are implanted within individual sockets.

While the dental shelf is narrow, the teeth are even narrower, resulting in the dorsal exposure of the dental shelf along the length of the dentary (Fig. 1F, H). The labial lamina of the dentary is shortest at the symphysis, but becomes taller towards the posterior. The lingual side of the dental shelf possesses a lamina that projects ventrally, contributing to the lingual surface of the mandible. The lingual lamina primarily projects ventrally for most of the dentary, but, towards the posterior extent of the element, the lamina possesses a 90-degree torsion before suturing to the third coronoid. It overlies and sutures to the first and second coronoids dorsally. The lamina ends where the dentary sutures to the third coronoid posteriorly, leaving the dental shelf occupied by the last few teeth on the dentary without a lingual lamina. The surface of the dentary lingual lamina is smooth until the portion just before the suture to the third coronoid, where the texture of the surface changes drastically. It is marked with anteroposteriorly orientated striations. This roughness may have been the attachment site for musculature. The labial side of the dentary is deep and forms the majority of the anterior labial surface of the mandible (Fig. 1A, C).

A deep trench runs along the labial side of the dentary (Fig. 1A, C). Ruta and Bolt (2008) discuss this trench; they note that this groove is present and homologous in the dentaries from V3922 and in other brachyopoids (Damiani and Kitching 2003), but they do not consider it to be the oral sulcus. They suggest that an external mandibular artery may have been set within the lateral groove (Morales and Shishkin 2002) and question its identity as an oral sulcus. Lydekkerinids have been described to have an oral sulcus that extends from the posterior mandible on to the dentary and towards the mandibular symphysis (Jupp and Warren 1986; Jeannot et al. 2006), which is the case in Hadrokkosaurus. The mandible of Brachyops allos (Warren 1981) possesses a “groove” in the same topological position, but it is identified as an oral sulcus. Similar grooves can be observed across other trematosaurians (e.g. Sulej (2007); Schoch (2019)) and capitosaurians (e.g. Morales and Shishkin (2002); Eltink et al. (2016)), suggesting the feature to be broadly distributed across stereospondyls. The groove on the labial surface of the dentary has been widely discussed amongst other descriptions of stereospondyls. Given the aquatic nature of stereospondyls, it is likely that the groove is a lateral line sulcus.

Three coronoid bones are present as in other temnospondyls (Fig. 1B, D, F, H). The first coronoid is a long, splint-like element on the lingual surface of the mandible, wedged between the dentary dorsally and the presplenial ventrally. Posteriorly, the first coronoid is sutured to the second coronoid. The first coronoid frames the posteriormost tip of the Meckelian canal forming the postsymphyseal foramen. The second coronoid forms an interdigitating suture with the first coronoid anteriorly. It is foreshortened as it compensates for a lengthened first coronoid. Ventrally, the second coronoid is sutured to the postsplenial. Posteriorly, it is sutured to the third coronoid.

The third coronoid is tooth-bearing (Fig. 1B, D, F, H). It is positioned more dorsally compared to the other coronoids, almost reaching the tips of the crowns of the marginal dentition. The third coronoid sutures to the second coronoid anteriorly. Ventrally, it is sutured to the pre-articular. The body of the third coronoid is lingually expanded to form the anterior margin of the adductor chamber (Fig. 1F, H). A process of the third coronoid extends posteriorly, lingual to the posterior process of the dentary to form the anterior half of the labial margin of the adductor chamber. The posterior process is well exposed in labial view and forms an interdigitating suture with the surangular. The third coronoid also possesses a lamina that descends from its body and contributes to the lingual surface of the mandible. The third coronoid teeth are smaller than the marginal teeth, but they are similar in shape. There are eight tooth positions forming a row on the third coronoid. The coronoid process is formed by the third coronoid without contribution by the dentary.

The presplenial is short and trough-shaped, positioned near the symphysis on the ventral surface of the mandible (Fig. 1B, D, E, G). The presplenial forms the ventral margin of the canal into which the postsymphyseal foramen exits. It sutures to the dentary dorsally within the canal and to the postsplenial posteriorly. The suture between the presplenial and postsplenial is interdigitating. The presplenial also forms a suture with the first coronoid lingually towards its posterior. The suture between the presplenial and the postsplenial is interdigitating and visible on the lingual surface of the mandible. The suture continues around the ventral mandibular mandible, where it is obscured by plaster. In labial view, the presplenial is barely visible as a narrow splint, where it also sutures to the labial component of the dentary. It does not participate in the mandibular symphysis.

The postsplenial is longer than the presplenial (Fig. 1B, D, E, G). Anteriorly, it is similarly trough-shaped, but towards the posterior, it twists and becomes flat and primarily exposed lingually. It forms interdigitating sutures with the presplenial anteriorly, with the angular posteroventrally and with the pre-articular posteriorly. On the lingual surface, the postsplenial sutures to the first, second and third coronoid dorsally. On the labial surface, the postsplenial sutures to the dentary dorsally.

The angular is poorly ornamented and forms the majority of the floor of the adductor chamber (Fig. 1). It is trough-shaped, contributing to the ventral labial and lingual surfaces of the mandible. The angular has a low exposure on the labial surface of the mandible, reaching only the mid-point of the height of the dentary (Fig. 1A, C). As the angular curves lingually around the ventral mandible to form the adductor chamber floor, it contributes to a narrow ventral portion of the lingual surface. The angular extends posteriorly to contribute to the ventral surface of the postglenoid area, along the length of which it sutures to the surangular dorsally on the labial surface. The anterior angular on the labial surface sutures to the dentary dorsally. It forms a straight suture with the pre-articular on the lingual surface. Anteriorly, the angular forms an interdigitating suture to the postsplenial.

The surangular of Hadrokkosaurus is a large element on the labial surface of the mandible (Fig. 1A, C). It forms the posterior labial margin of the adductor chamber (Fig. 1F, H), where it forms an interdigitating suture with the third coronoid anteriorly and stepped suture to the angular ventrally on the labial side. The surangular forms a straight simple suture with the angular on the ventral postglenoid area. The surangular would underlie the articular, which is not preserved. It forms the labial half of the postglenoid area, where it forms a simple straight suture with the retro-articular process of the pre-articular on the dorsal surface of the postglenoid area. The surangular forms a low preglenoid process, only slightly taller than the prearticular wall of the adductor chamber (Fig. 1A, C).

The pre-articular is tall and forms the majority of the posterior lingual surface of the mandible (Fig. 1B, D). It forms the lingual wall of the adductor chamber (Fig. 1F, H). UCMP 36836, UCMP 36837 and UCMP 36838 are referred to as partial pre-articulars that share with the holotype a dorsal process that curls lingually (Suppl. material 1: fig. S1E, F). UCMP 36839 could be the postglenoid process of the pre-articular or the surangular, but there is not enough information preserved to discern its identity (Suppl. material 1: fig. S1H). The pre-articular forms a simple suture with the third coronoid anterodorsally and to the angular ventrally. It shares with the angular an interdigitating suture with the postsplenial. The pre-articular forms the lingual half of the postglenoid area, where it forms a simple suture with the surangular labially on the dorsal postglenoid area.

The articular is not preserved in the mandible. It may have been poorly ossified or it could have been disarticulated during the preservation of the mandible. However, the surangular and pre-articular preserve the facet upon which the articular would sit.

Temnospondyli Zittel, 1888

Stereospondyli Zittel, 1888

Brachyopoidea Lydekker, 1885

Brachyopoidea indet.

Horizon and locality. Uppermost channel sandstone of Holbrook Member, Moenkopi Formation; early Anisian, lowermost Middle Triassic. V3922, Geronimo (Holbrook) fossil vertebrate quarry near Holbrook, Coconino County, north-eastern Arizona.

Referred material. UCMP 36202, partial posterior left dentary; UCMP 36833, partial anterior right mandible; UCMP 36834, near complete right dentary; UCMP 36385, partial right dentary; UCMP 152390, right dentary fragment.

Description. UCMP 36833 is a well-preserved anterior right mandible that demonstrates different morphology from Hadrokkosaurus (Figs 2, 3). Welles (1947) previously noted that some of the dentaries referred to Hadrokkosaurus had different tooth morphology than the type. These additional specimens share the same features as UCMP 36833. While these specimens provide significant morphological detail, they exhibit the same preservation quality as Hadrokkosaurus. UCMP 36202 is a disarticulated posterior left dentary, with partial dentition preserved (Suppl. material 1: fig. S2). It is noticeably laterally compressed in preservation. The teeth are large and robust, as in UCMP 36833 and unlike in Hadrokkosaurus. UCMP 36834 is a well-preserved right dentary, retaining most if not all of the morphology (Fig. 3). It preserves the same tooth morphology as in UCMP 36833. UCMP 152390 is a mid-section fragment of a right dentary; it also exhibits the same tooth morphology as in UCMP 36833.

Figure 2. 

UCMP 36833, an incompletely preserved right mandible of the novel brachyopoid photographed and illustrated in: ventral view (A, C), dorsal view (B, D), lingual view (E, G) and labial view (F, H).

Figure 3. 

UCMP 36834, a complete right dentary of the novel brachyopoid photographed in: lingual view (A), ventral view (B) and dorsal view (C).

UCMP 36833 preserves the dentary, the first coronoid, the presplenial, the anterior second coronoid and the anterior postsplenial (Fig. 2). The mandibular symphysis of UCMP 36833 is partially reconstructed in plaster. Based on UCMP 36834, consisting of a nearly complete dentary, the curvature of the mandible suggests it would have accompanied a very wide and parabolic skull (Figs 2, 3). The ornamentation that is typically present on the temnospondyl mandible is significantly eroded, though there are hints of its distribution present. It appears the ornamentation may have been more polygonal on UCMP 36833 in the symphyseal area and more represented by elongate grooves and ridges towards the posterior. Otherwise, the other specimens belonging to this new brachyopoid taxon do not preserve ornamentation. The dentition consists of tooth bases that are anteroposteriorly compressed ovals in the cross section of the base. The teeth are slightly lingually recurved. They also possess a slight labiolingual narrowing at the crown, but are far less labiolingually compressed than the teeth in Hadrokkosaurus. Although slightly eroded, the consistent shape across all teeth in all specimens of the unidentified brachyopoid shows that the lack of carinae is not a result of taphonomic processes. Generally, the tooth morphology can be broadly extrapolated to be larger at the base, rounder overall and fewer in number to accommodate the limited space of the dentary. Amongst brachyopoids, this tooth morphology is found only in Koolasuchus cleelandi from the Early Cretaceous of Australia (Warren et al. 1997) and an incomplete mandible from the Late Triassic of Argentina (Marsicano 2005). The straight linear measurement from the anteriormost tip of the mandibular symphysis to the anteriormost sutural contact between the first and second coronoid of UCMP 36833 measures 10.7 cm. Measured from the anteriormost tip of the symphysis to the posteriormost extent of the mandible, UCMP 36834 measures approximately 14 cm. In total, there are approximately 21 tooth positions present on UCMP 36834, far fewer than the count of 32 on the Hadrokkosaurus mandible and the count of 40 on the Koolasuchus mandible.

The dentary is more robust compared to the dentary of Hadrokkosaurus (Figs 2, 3). The dental shelf is notably wider. The dentary forms most of the labial surface of the anterior mandible, similar to the condition of the anterior mandible of Koolasuchus (Warren et al. 1997). It is low anteriorly towards the symphysis and deepens to become a tall and robust element towards the posterior. The width of the dentition is wide enough to span the width of dorsal facing surface of the dentary on which the dentition sits. This differs from Hadrokkosaurus, in which the teeth are smaller, leaving a partially exposed dorsal-facing surface of the dentary. The entire width of the dental shelf is occupied entirely by the width of the dentition, resulting in a dorsally unexposed dental shelf unlike in Hadrokkosaurus. The dentary is markedly exposed on the lingual surface of the mandible, extending ventrally from the dentary shelf. The lingual lamina of the dentary curves horizontally and then posteriorly from the shelf. It forms straight sutures ventrally with the first and second coronoid. It sutures to the presplenial and postsplenial labially and ventrally. On the labial surface, there is a shallow groove that is likely homologous with the “horizontal groove” noted by Welles (1947) in Hadrokkosaurus and by Damiani and Kitching (2003) in Vanastega (Fig. 2F, H). As mentioned before, this feature is found broadly across stereospondyls.

The mandibular symphysis is poorly preserved in UCMP 36833 and reconstructed in plaster; however, based on the anterior extent of the presplenial, the symphysis is formed by the dentary alone (Fig. 2B, D, E, G). Although the symphysis is not preserved in its entirety in UCMP 36833, it is well-preserved in UCMP 36834, where the reduction in the height of the dentary can be seen at the mandibular symphysis. The symphysis widens posteriorly and accommodates a pair of symphyseal fangs in both UCMP 36833 and 36834. The first coronoid is a long element that begins at the posterior extent of the symphyseal shelf. The first coronoid composes the lingual wall and margin of the postsymphyseal foramen. It extends posteriorly, where it forms a double scarf suture with the second coronoid. The first coronoid forms a straight suture to the dentary dorsally and to the presplenial ventrally. It is a relatively shorter element compared to the first coronoid in Hadrokkosaurus.

A postsymphyseal foramen is present in UCMP 36833 (Fig. 2A, C, E, G). The foramen exits into a distally widening Meckelian canal positioned on the ventral aspect of the mandible. The foramen and canal are entirely exposed on the mandible ventrally, compared to UCMP 36199, in which the postsymphyseal foramen and canal initially appear on the lingual surface of the mandible before the canal curves ventrally. When viewed at the symphyseal surface, the canal forms a ventrally opening concavity. The presplenial contributes to the ventral margin and the first coronoid forms the lingual margin. Other temnospondyl taxa possess a postsymphyseal groove that lies lingually or sometimes ventrally on the mandible and participate in the mandibular symphysis (Damiani 2001; Jeannot et al. 2006).

Temnospondyli Jaekel, 1909

Temnospondyli indet.

Horizon and locality. Uppermost channel sandstone of Holbrook Member, Moenkopi Formation; early Anisian, lowermost Middle Triassic. V3922, Geronimo (Holbrook) fossil vertebrate quarry near Holbrook, Coconino County, north-eastern Arizona.

Referred material. UCMP 36210, partial ventral angular.

Description. UCMP 36210 is a partial right angular that would have floored the adductor chamber of the right mandible (Fig. 4). It was noted by Ruta and Bolt (2008) to belong to a different temnospondyl than Hadrokkosaurus due to a “boss-like process” upon the floor that is absent in Hadrokkosaurus (Fig. 1F, H; Fig. 4). The pit-and-ridge ornamentation is slightly worn, but noticeable on the ventral side, unlike in Hadrokkosaurus. In light of the presence of another temnospondyl taxon in this locality, the process may have belonged to a temnospondyl with a strong adductor muscle inserted to produce a stronger bite or hold the prey of the animal. A similar process has been reported in the contemporaries Plagiosternum and Gerrothorax (Schoch and Witzmann 2011), but noted by Ruta and Bolt (2008) to also be present in Aphaneramma (Nilsson 1943), Dvinosaurus (Shishkin 1973), Archegosaurus (Gubin 1997), Acroplous and Trimerorhachis.

Figure 4. 

UCMP 36210, a right angular in dorsal view (A) and oblique view (B). The arrow points to the boss-like process.

Phylogenetic analysis

The matrix is derived from the dataset of Ruta and Bolt (2008) (Suppl. material 1: data S3). We made additions to the taxon sampling in the dataset. Our additions include UCMP 36833 as an operational taxonomic unit (OTU), Keratobrachyops australis (Warren, 1981b) and Plagiosuchus pustuliferus (Damiani et al. 2009). The matrix is available for download under project 5265 on Morphobank.org (http://morphobank.org/permalink/?P5265). Three additional characters were added regarding the postsymphyseal foramen, as follows:

126. Postsymphyseal foramen: absent (0) or present (1).

127. Postsymphyseal foramen position: the foramen is on the lingual surface of the mandible (0) or the foramen is on the ventral surface of the mandible (1).

128. Postsymphyseal foramen and the Meckelian canal: the foramen opens to a flat surface of the mandible (0) or the foramen opens into an exposed Meckelian canal (1).

We ran the analyses using TNT 1.6 (Goloboff and Morales 2023) under the New Technology search method. Parallel searches were run under equal weights (EW) and implied weights (IW; Goloboff (1993)). The modified dataset consists of 59 taxa and 128 unordered characters. The EW analysis was conducted for 3,000 additional sequences. The subsequent equally most-parsimonious trees (MPTs) were subjected to an additional round of tree bisection reconnection (TBR). We recovered 328 MPTs with a length of 757 steps (CI = 0.210; RI = 0.571). We explored different values of the concavity constant (k) for IW to better explore the data. We specifically used a concavity constant of k = 3 as a generally accepted default value (Goloboff 1993). We also employed k = 12, which has been shown to be more effective at identifying topologies (Goloboff et al. 2018). We conducted an IW analysis with k = 3 for 3,000 additional sequences. The resultant MPTs were subjected to an additional round of TBR. This analysis recovered three MPTs of fit 64.29 (CI = 0.204; RI = 0.555). The IW analysis with k = 12 was conducted and subjected to an additional round of TBR. It recovered one MPT of fit 31.82 (CI = 0.209; RI = 0.567).

The strict consensus topology from the EW analysis resulted in poor resolution (Fig. 5A). A large polytomy of temnospondyls, tetrapodomorphs and other early tetrapods was recovered. Higher-nested stereospondyls grouped together in a monophyly, but the internal relationships were not reconcilable with previously-published topologies. Nominal brachyopoids are found to group together, but their monophyly includes Mastodonsaurus, Kupferzellia and Plagiosuchus. The IW analysis under k = 3 produced a more resolved topology (Fig. 5B). Nominal Temnospondyli was recovered, but it includes several amniote-line tetrapods. The analysis also recovered a monophyletic Brachyopoidea (Brachyopidae + Chigutisauridae) to the exclusion of Keratobrachyops. The IW analysis under k = 12 produced a clade that included all nominal temnospondyls, except for Edops, which diverges before a clade including nominal “lepospondyls” and “reptiliomorphs (Fig. 5C). The nominal brachyopoid relationships recovered in this analysis mirror the results of the EW analysis, but both include Mastodonsaurus and Kupferzellia and resolves Plagiosuchus as a highly-nested brachyopoid.

Figure 5. 

Results of the phylogenetic analyses. Strict consensus and Bremer supports for EW analysis (A); strict consensus and relative fit difference (RFD) for IW (k = 3) analysis (B); strict consensus for IW (k = 12) analysis (C). The green box highlights nominal brachyopoids (Brachyopidae + Chigutisauridae). RFD shows the ratio of the amount of favourable evidence relative to the amount of contradictory evidence (Goloboff and Farris 2001). In the case of an RFD of 0.12, the amount of contradictory evidence is 88% of the amount of favourable evidence, equivalent to a conflict of 25 characters versus 22 characters.

Discussion

Taxonomic identifications

UCMP 36833 and UCMP 36834 both exhibit a strong curvature from the symphysis to the rest of the mandible, suggesting that a complete set of left and right mandibles would correspond to a widely parabolic skull of a temnospondyl. During the Middle Triassic of North America, temnospondyls were broadly represented by the capitosauroids, trematosauroids, plagiosaurids and brachyopoids. Given that most capitosauroids and trematosauroids, except for the capitosauroid Sclerothorax (Schoch et al. 2007), possessed longirostrine skulls, the unidentified temnospondyl is unlikely to belong to these taxa. Brachyopoids are stereospondyls that possess brevirostrine skull morphology that would match the widely curving contour of UCMP 36833 and UCMP 36834. Plagiosaurids also possess similar brevirostrine skulls, but in North America, they are limited to Greenland. Additionally, the diagnostic feature of pustular dermal ornamentation of plagiosaurine plagiosaurids is not present on any of the specimens (Damiani et al. 2009; Schoch and Witzmann 2012), although the surfaces are too eroded to confidently exclude pustular ornamentation. Sclerothorax was an early-diverging Early Triassic capitosauroid with a short and broad skull morphology (Schoch 2007) and could be considered here as a possible candidate; however, the preserved mandibles are sharply curved, aligning the posterior half of the mandible parallel to the mid-line. Coronoid teeth are not present on at least the first and second coronoid in UCMP 36833, unlike the continuous row of coronoid teeth of Gerrothorax (Schoch and Witzmann 2012). In plagiosaurids, symphyseal fangs are either absent or present as small, rudimentary fangs (Warren and Davey 1992), which is in stark contrast to the large symphyseal fangs observed in UCMP 36834 (Fig. 3). Based on these observations, we provisionally assign the unidentified temnospondyl to Brachyopoidea, which is supported by our phylogenetic results.

Angulars with a boss-like process as in UCMP 36210 are observed in several other taxa as previously noted. In this context, plagiosaurids and trematosauroids are relevant as temnospondyls present in the Middle Triassic with the process. Pustular ornamentation is diagnostic for plagiosaurine plagiosaurids, which are not present on UCMP 36210. Instead, the ventral surface of UCMP 36210 exhibits typical temnospondyl pit and ridge ornamentation. UCMP 36210 has no overlap of anatomy with the unidentified brachyopoid, which does not preserve an angular. Furthermore, with the process present on non-stereospondyl temnospondyls as well, the process may be the result of ecology rather than phylogeny. At this stage, we are unable to establish any further identification of UCMP 36210 to a more specific level than Temnospondyli.

Phylogenetic results

The results of the TNT phylogenetic analysis using only mandibular characters differ from the topology recovered in Ruta and Bolt (2008) in several areas of the tree. We recover relationships far from currently accepted temnospondyl phylogenies, though there is recovery of nominal brachyopoid relationships (Fig. 5). UCMP 36833 is consistently recovered as a brachyopoid. In analyses under EW and IW when k = 12, UCMP 36833 is recovered as part of a grade of “brachyopids” leading to Chigutisauridae. However, under IW when k = 3, UCMP 36833 is recovered as sister to Hadrokkosaurus, highly nested within chigutisaurids.

A consistent pattern emerges from the analyses, in which Keratobrachyops does not fall within a traditional brachyopoid monophyly. Mastodonsaurus and Kupferzellia are found to be more closely related to brachyopoids than Keratobrachyops in the analyses under EW and IW when k = 12 (Fig. 5A, C), while in the IW analysis when k = 3, Keratobrachyops is recovered as a trematosauroid (sensu Schoch (2013)). In the context of these analyses, it is possible that the proposed brachyopoid features of Keratobrachyops may be homoplasies, but the exclusion of skull roof characters should cause some scepticism regarding these results. The results of the phylogenetic analyses highlight a need to revisit and refine mandibular characters to better test relationships and homology.

Palaeoecological and palaeogeographic interpretations

The Moenkopi Formation spreads widely across the southwest United States and, given its coverage, we expect to capture a broad sampling of Middle Triassic temnospondyls. The Moenkopi Formation exhibits a co-occurrence of Brachyopoidea, Capitosauria and Trematosauria, but lacks representation of plagiosaurids, despite their rich record in Greenland and western Europe. Brachyopoids would eventually become the latest surviving stereospondyl clade (Warren et al. 1997) and one of two surviving temnospondyl clades past the end-Triassic mass extinction, the other being trematosauroids (Maisch et al. 2004), making brachyopoids an important taxon for understanding temnospondyl faunal turnover.

The unidentified brachyopoid has a novel dental ecomorphotype observed in the temnospondyls in this ecosystem. The tooth morphology of tetrapods is widely considered to correlate with diet (Hotton 1995; Evans et al. 2007), including amphibians (Gregory et al. 2016). Following this, the robust dentition in the unidentified brachyopoid may have enabled a different diet and lifestyle than Hadrokkosaurus, Vigilius and other temnospondyls of the Middle Triassic Moenkopi Formation (Fig. 6). Few temnospondyl fossils preserve teeth like the unidentified brachyopoid. A likely brachyopoid, MCNACM-PV-3195, was also described to possess relatively large teeth from the Late Triassic of Brazil (Marsicano 2005). Marsicano (2005) also noted that the only other brachyopoid taxon possessing proportionately large teeth is Koolasuchus. The size of the dentition in the unidentified brachyopoid is comparable to the dentition in Koolasuchus (Warren et al. 1997), a chigutisaurid with few, but large teeth. However, this tooth condition does not appear to be a synapomorphy of chigutisaurids as Siderops and Compsocerops do not have as large dentition of the dentaries (Warren and Hutchinson 1983; Sengupta 1988) and our phylogeny recovers the unidentified brachyopoid on the grade of brachyopoids outside of chigutisaurids. Additionally, the tooth count of the unidentified brachyopoid is far fewer compared to coeval Moenkopi Formation stereospondyls, such as Eocyclotosaurus (98–100 teeth; Rinehart et al. (2015)), further supporting ecological differentiation of the unidentified brachyopoid. The variation seen between Hadrokkosaurus and the unidentified Moenkopi brachyopoid may reflect differentiation into different ecological niches. This would be in line with the novel dental ecomorphotype of the unidentified brachyopoid. Differentiation into a more diverse feeding regime spanning broader ecological niches could have contributed to the post-Triassic success of the brachyopoids, enabling brachyopids in the Northern Hemisphere to co-exist with phytosaurs and sphenosuchian crocodiles (Warren et al. 1997).

Figure 6. 

UCMP 31865, the holotype skull of Vigilius wellesi (A). The maxilla of the holotype does not preserve any dentition, but the marginal tooth sockets are small (B), especially when compared to the sockets and dentition of the dentary teeth in UCMP 36834 (C). Hadrokkosaurus dentition and sockets also appear to be larger than the sockets of Vigilius (D).

The presence of multiple large-bodied stereospondyls at this locality strongly suggests that they diversified into niches occupied by other local aquatic tetrapods. This is contrary to the expectation that diversity decreases progressively towards higher trophic levels (Evans et al. 2005), assuming shared resources in lower trophic levels. In this case, it appears that the Moenkopi Formation brachyopids were able to exploit resources that did not overlap with other large aquatic tetrapods and between the brachyopids. The horizon of V3922 is a channel sandstone deposit, which would have been established by a meandering river in a floodplain. Seasonal floods may have provided the necessary nutrients to this locality, supporting multiple temnospondyl taxa. The expansion into novel niches may be what allowed brachyopoids to survive past the End Triassic extinction, as niches previously occupied by other large-bodied stereospondyls disappeared.

At the formation level, the ecosystem supported brachyopids, trematosauroids and capitosauroids. However, if this locality was able to support diverse stereospondyl taxa without necessarily competing for resources, then it begs the question as to why some Triassic stereospondyl clades are not represented, such as the plagiosaurids. Plagiosaurids are a diverse and common component at the higher latitudes of northern Pangean assemblages (e.g. Schoch et al. (2014); Damiani et al. (2009); Witzmann and Schoch (2024)). They are also observed in southern Pangea at higher latitudes (Dias-Da-Silva and Milner 2010; Gee and Sidor 2022), which leaves a conspicuous geographic gap near the palaeoequator. It has been demonstrated that plagiosaurids have the capacity to endure habitats with salinity fluctuations and low level of nutrients (Witzmann and Soler‐Gijón 2010; Sanchez and Schoch 2013), which would have been advantageous in the severe seasonality at the palaeoequator, allowing for possible co-occurrence with other stereospondyl clades. Indeed, plagiosaurids are also present alongside other large stereospondyls in northern Pangea (Milner et al. 1996; Nonsrirach et al. 2021), where there have been less environmental fluctuations in salinity. Overlapping brevirostrine morphology between brachyopoids and plagiosaurids may be a factor in limiting their co-occurrence, but there is evidence of their co-occurrence in some localities (Warren 1985; Nonsrirach et al. 2021), The absence of plagiosaurids in some communities have previously been explored, pointing to a possible preference for marginal habitats which are potentially present in these systems, but not preserved (Gee and Sidor 2021).

Conclusions

The Moenkopi Formation has been shown to be an ecosystem with a diverse assemblage of stereospondyls. The novel identification of another brachyopoid in the Moenkopi Formation highlights the need to critically re-examine closely-collected material; unobserved diversity may be hiding amongst them. Further exploring the historical collections and localities from the Holbrook Member of the Moenkopi Formation allows us to contribute to a bigger picture of ancient local systems in the Middle Triassic of south-western North America. The diversity of brachyopoid mandibles may be a clue to their specialised morphology enabling exploration of different roles in the ecosystem, which may have allowed the clade to survive the end-Triassic extinction.

Acknowledgements

The authors would like to thank James Clark for his patience and advice. The authors are significantly grateful to Pat Holroyd for allowing a collection visit and aid in tracking down fossil specimens. The authors would like to thank Ken Angielczyk, Bill Simpson and Adrienne Stroup for organising and accommodating rapid collections visit to the Field Museum of Natural History. Further, the authors would like to thank Jason Pardo for his guidance and support of this research. Finally, the authors would like to thank the Museum für Naturkunde Berlin, Bryan Gee and reviewer 2 for their publication support.

References

  • Brown B (1933) A new genus of Stegocephalia from the Triassic of Arizona. American Museum Novitates 640: 1–4.
  • Cabrera Á (1944) Sobre un estegocéfalo de la Provincia de Mendoza. Notas Mus. La Plata 9: 421–429.
  • Dal Corso J, Bernardi M, Sun Y, Song H, Seyfullah LJ, Preto N, Gianolla P, Ruffell A, Kustatcher E, Roghi G, Merico A, Hohn S, Schmidt AR, Marzoli A, Newton RJ, Wignall PB, Benton MJ (2020) Extinction and dawn of the modern world in the Carnian (Late Triassic). Science Advances 6(38): eaba0099. https://doi.org/10.1126/sciadv.aba0099
  • Damiani RJ (2001) Parotosuchus (Amphibia, Temnospondyli) from the Cynognathus Assemblage Zone (Early Triassic) of South Africa: cranial morphology and relationships. Alcheringa 25(4): 351–379. https://doi.org/10.1080/03115510108619226
  • Damiani RJ, Jeannot AM (2002) A brachyopid temnospondyl from the lower Cynognathus Assemblage Zone in the northern Karoo Basin, South Africa. Palaeontologia Africana 38: 57–69.
  • Damiani RJ, Warren A (1996) A new look at members of the Superfamily Brachyopoidea (Amphibia, Temnospondyli) from the Early Triassic of Queensland and a preliminary analysis of brachyopoid relationships. Alcheringa 20(4): 277–300. https://doi.org/10.1080/03115519608619472
  • Damiani RJ, Schoch RR, Hellrung H, Werneburg R, Gastou S (2009) The plagiosaurid temnospondyl Plagiosuchus pustuliferus (Amphibia: Temnospondyli) from the Middle Triassic of Germany: anatomy and functional morphology of the skull. Zoological Journal of the Linnean Society 155(2): 348–373. https://doi.org/10.1111/j.1096-3642.2008.00444.x
  • Dias-Da-Silva S, Milner AR (2010) The pustulated temnospondyl revisited—a plagiosternine plagiosaurid from the Lower Triassic of Brazil. Acta Palaeontologica Polonica 55(3): 561–563. https://doi.org/10.4202/app.2009.0131
  • Dong ZM (1985) A Middle Jurassic labyrinthidont (Sinobrachyops placenticephalus gen et sp. nov) from Dashanpu, Zigong, Sichuan Province. Vertebrata Palasiatica 23(4): 301.
  • Eltink E, Da-Rosa ÁAS, Dias-da-Silva S (2017) A capitosauroid from the Lower Triassic of South America (Sanga do Cabral Supersequence: Paraná Basin), its phylogenetic relationships and biostratigraphic implications. Historical Biology 29(7): 863–874. https://doi.org/10.1080/08912963.2016.1255736
  • Evans AR, Wilson GP, Fortelius M, Jernvall J (2007) High-level similarity of dentitions in carnivorans and rodents. Nature 445(7123): 78–81. https://doi.org/10.1038/nature05433
  • Fortuny J, Gastou S, Escuillié F, Ranivoharimanana L, Steyer JS (2018) A new extreme longirostrine temnospondyl from the Triassic of Madagascar: phylogenetic and palaeobiogeographical implications for trematosaurids. Journal of Systematic Palaeontology 16(8): 675–688. https://doi.org/10.1080/14772019.2017.1335805
  • Goloboff PA, Morales ME (2023) TNT version 1.6, with a graphical interface for MacOS and Linux, including new routines in parallel. Cladistics 39: 144–153. https://doi.org/10.1111/cla.12524
  • Goloboff PA, Torres A, Arias JS (2018) Weighted parsimony outperforms other methods of phylogenetic inference under models appropriate for morphology. Cladistics 34(4): 407–437. https://doi.org/10.1111/cla.12205
  • Gregory AL, Sears BR, Wooten JA, Camp CD, Falk A, O’Quin K, Pauley TK (2016) Evolution of dentition in salamanders: relative roles of phylogeny and diet. Biological Journal of the Linnean Society 119(4): 960–973. https://doi.org/10.1111/bij.12831
  • Hellrung H (2003) Gerrothorax pustuloglomeratus, ein Temnospondyle (Amphibia) mit knöcherner Branchialkammer aus dem Unteren Keuper von Kupferzell (Süddeutschland): Gerrothorax pustuloglomeratus, a temnospondyle (Amphibia) with a bony branchial chamber from the Lower Keuper of Kupferzell (South Germany). Staatliches Museum für Naturkunde.
  • Hotton III N (1955) A survey of adaptive relationships of dentition to diet in the North American Iguanidae. American Midland Naturalist 88–114. https://doi.org/10.2307/2422301
  • Howie A (1972) A brachyopid labyrinthodont from the Lower Trias of Queensland. Proceedings of the Linnean Society of New South Wales 96: 268–277.
  • Jeannot AM, Damiani R, Rubidge BS (2006) Cranial anatomy of the Early Triassic stereospondyl Lydekkerina huxleyi (Tetrapoda: Temnospondyli) and the taxonomy of South African lydekkerinids. Journal of Vertebrate Paleontology 26(4): 822–838. https://doi.org/10.1671/0272-4634(2006)26[822:CAOTET]2.0.CO;2
  • Jenkins FA (1994) Late Triassic continental vertebrates and depositional environments of the Fleming Fjord Formation, Jameson Land, east Greenland. Vol. 32. Commission for Scientific Research in Greenland. https://doi.org/10.7146/moggeosci.v32i.140904
  • Kligman BT, Gee BM, Marsh AD, Nesbitt SJ, Smith ME, Parker WG, Stocker MR (2023) Triassic stem caecilian supports dissorophoid origin of living amphibians. Nature 614(7946): 102–107. https://doi.org/10.1038/s41586-022-05646-5
  • Maisch MW, Matzke AT (2004) Temnospondyl amphibians from the Jurassic of the southern Junggar Basin (NW China). Paläontologische Zeitschrift 79: 285–301. https://doi.org/10.1007/BF02990189
  • Milner AR, Duffin C, Delsate D (1996) Plagiosaurid and capitosaurid amphibian material from the Late Triassic of Medernach, Grand-Duchy of Luxembourg: preliminary note. Bulletin de la Sociètè belge de Gèologie 104: 42–53.
  • Nonsrirach T, Manitkoon S, Lauprasert K (2021) First occurrence of brachyopid temnospondyls in Southeast Asia and review of the Mesozoic amphibians from Thailand. Fossil Record 24: 33–47. https://doi.org/10.5194/fr-24-33-2021
  • Nilsson T (1943) Über einige postkraniale Skelettreste der triassichen Stegocephalen Spitzbergens. Bulletin of the Geological Institute, University of Uppsala 30: 227–272.
  • Pardo JD, Small BJ, Huttenlocker AK (2017) Stem caecilian from the Triassic of Colorado sheds light on the origins of Lissamphibia. Proceedings of the National Academy of Sciences 114(27): E5389–E5395. https://doi.org/10.1073/pnas.1706752114
  • Piveteau J (1936) Une forme ancestrale des Amphibiens Anoures dans le Trias inferieur de Madagascar. Comptes Rendus de l’Académie des Sciences 102: 1607–1608.
  • Rage JC, Rocek Z (1989) Redescription of Triadobatrachus massinoti (Piveteau, 1936) an anuran amphibian from the early Triassic. Palaeontographica A 206(1–3): 1–16.
  • Rinehart LF, Lucas SG, Schoch RR (2015) Eocyclotosaurus appetolatus, a new cyclotosaurid amphibian from the Middle Triassic (Perovkan) Moenkopi Formation of New Mexico, USA. Journal of Vertebrate Paleontology 35(3): e929140. https://doi.org/10.1080/02724634.2014.929140
  • Ruta M, Bolt JR (2008) The brachyopoid Hadrokkosaurus bradyi from the early Middle Triassic of Arizona, and a phylogenetic analysis of lower jaw characters in temnospondyl amphibians. Acta Palaeontologica Polonica 53(4): 579–592. https://doi.org/10.4202/app.2008.0403
  • Sanchez S, Schoch RR (2013) Bone histology reveals a high environmental and metabolic plasticity as a successful evolutionary strategy in a long-lived homeostatic Triassic temnospondyl. Evolutionary Biology 40: 627–647. https://doi.org/10.1007/s11692-013-9238-3
  • Schoch RR (2019) Osteology of the temnospondyl Trematosaurus brauni Burmeister, 1849 from the Middle Buntsandstein of Bernburg, Germany. Palaeodiversity 12(1): 41–63. https://doi.org/10.18476/pale.v12.a4
  • Schoch RR, Fastnacht M, Fichter J, Keller T (2007) Anatomy and relationships of the Triassic temnospondyl Sclerothorax. Acta Palaeontologica Polonica 52(1): 117–136.
  • Schoch RR, Milner AR, Witzmann F (2014) Skull morphology and phylogenetic relationships of a new Middle Triassic plagiosaurid temnospondyl from Germany, and the evolution of plagiosaurid eyes. Palaeontology 57(5): 1045–1058. https://doi.org/10.1111/pala.12101
  • Schoch RR, Werneburg R, Voigt S (2020) A Triassic stem-salamander from Kyrgyzstan and the origin of salamanders. Proceedings of the National Academy of Sciences 117(21): 11584–11588. https://doi.org/10.1073/pnas.2001424117
  • Sengupta DP (1988) Chigutisaurid temnospondyls from the Late Triassic of India and review of the family Chigutisauridae. Palaeontology 38: 313–339.
  • Shishkin MA (1973) The morphology of the early Amphibia and some problems of lower tetrapod evolution. Trudy Paleontologiceskogo Instituta Akademia Nauk SSr 137: 1–257.
  • Sulej T (2007) Osteology, variability, and evolution of Metoposaurus, a temnospondyl from the Late Triassic of Poland. Palaeontologia Polonica 64: 29–139.
  • Warren AA (1981a) The lower jaw of the labyrinthodont family Brachyopidae. Memoirs of the Queensland Museum 20(2): 285–289.
  • Warren AA (1985) Triassic Australian plagiosauroid. Journal of Paleontology 59: 236–241.
  • Warren AA, Hutchinson MN (1983) The last labyrinthodont? A new brachyopoid (Amphibia, Temnospondyli) from the early Jurassic Evergreen Formation of Queensland, Australia. Philosophical Transactions of the Royal Society of London, B, Biological Sciences 303(1113): 1–62. https://doi.org/10.1098/rstb.1983.0080
  • Welles SP (1947) Vertebrates from the upper Moenkopi Formation of northern Arizona. University of California Publications, Bulletin of the Department of Geological Sciences 27(7): 241–294.
  • Welles SP (1993) A review of lonchorhynchine trematosaurs (Labryrinthodontia), and a description of a new genus and species from the lower Moenkopi Formation of Arizona. PaleoBios 14: 1–24.
  • Welles SP, Estes R (1969) Hadrokkosaurus bradyi from the Upper Moenkopi Formation of Arizona: with a review of the brachyopid labyrinthodonts. University of California Publications in Geological Sciences 84: 1–61.
  • Witzmann F, Schoch RR (2024) Osteology and phylogenetic position of Plagiosaurus depressus (Temnospondyli: Plagiosauridae) from the Late Triassic of Germany and the repeated loss of dermal bones in plagiosaurids. Zoological Journal of the Linnean Society: zlae014. https://doi.org/10.1093/zoolinnean/zlae014

Supplementary material

Supplementary material 1 

Supporting information

Calvin So, Arjan Mann

Data type: docx

Explanation note: fig. S1: Referred Hadrokkosaurus specimens; fig. S2: Specimens of the novel brachyopoid taxon; S3: Character changes.

This dataset is made available under the Open Database License (http://opendatacommons.org/licenses/odbl/1.0). The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited.
Download file (6.50 MB)
login to comment