The Temnospondyls were an ancient group of Tetrapods, which are the sister group to the modern Lisamphibians (Frogs, Salamanders, and Caecilians), although they were in many ways more Reptile-like, with many apparently able to live completely away from water for much of their lives, while others colonised marine environments. The first Temnospondyls appeared in the Carboniferous, rapidly expanding to become one of the most diverse and abundant groups of terrestrial Vertebrates. The Temnospondyls remained a major group throughout the Permian and Triassic, but were badly affected by the End Triassic extinction from which they never recovered. A few Temnospondyl lineages persisted through the Jurassic and into the Early Cretaceous, when they finally disappeared.
The Trematosaurids were a group of Temnospondyls which migrated into marine environments in the Early Triassic, evolving to occupy a large marine predator role which had become vacant during the End Permian Extinction, and spreading throughout much of the world's environments. Trematosaurids have been described from Madagascar, Greenland, Australia, Pakistan, Spitsbergen, European Russia, the Russian Far East, Germany, and Jordan.
To date, only a single species of Marine Trematosaurid (several others are known from clearly non-marine environments) has been described from Australia, Erythrobatrachus noonkanbahensis, which described in 1972 by John Cosgriff and Neil Garbutt on the basis of a composite of fragmentary cranial remains from the Early Triassic Blina Shale of the central-southern Kimberley region of far northwestern Western Australia, collected during a series of expeditions to the area in the 1960s.
The Blina Shale records a transgressive delta setting (setting where the land is extending into the sea over a delta system), which would have faced onto the East Gondwana interior rift seaway. These deposits record a mixture of saline, brackish, and freshwater environments, and have produced non-Trematosaurid Temnospondyls such as the Rhytidosteid, Deltasaurus kimberleyensis, the Capitosauroid, Warrenisuchus aliciae, and the Brachyopids, Batrachosuchus henwoodi, and Banksiops townrowi. A variety of other fossils have also been found within the Blida Shale, including the ubiquitous Triassic Saurichthyid Actinopterygian, Saurichthys, a variety of Ceolacanths, the Ptychoceratodontid Lungfish, Ptychoceratodus philippsi, the Sagenodontid Lungfish, Aphelodus anapes, the Ceratodontid Lungfish, Asiatoceratodus tiguidensis, several possible Sharks, Conodonts, Insects, Spinicaudatans and Lingulid Brachiopods, possible Pseudomonotid Bivalves, indeterminate Ammonoids, some possible Mollusc egg cases, burrow traces, palynomorphs (fossil pollen and spores), Achritarchs, Horsetails, and possible Lycopods.
Temnospondyls from the Blina Shale are typically found in a disarticulated state, either as individual bones or accumulations, and show signs of having been transported before their final deposition, including weathering and sorting by size. This is consistent with deposition in beds with preserved ripple marks and thin cross-lamination, which is suggestive of a shallow, tidal environment.
In a paper published in the Journal of Vertebrate Palaeontology on 22 February 2026, Benjamin Kear of the Department of Palaeobiology at the Swedish Museum of Natural History, Nicolás Campione of the Palaeoscience Research Centre at the University of New England, Mikael Siversson of the Western Australian Museum, and the School of Molecular and Life Sciences at Curtin University, Mohamad Bazzi of the Department of Earth and Planetary Sciences at Stanford University, and Lachlan Hart of the School of Education and Earth and Sustainability Science Research Centre at the University of New South Wales, as well as the Australian Museum Research Institute, reassess that material assigned to Erythrobatrachus noonkanbahensis from the Blinda Shale deposits, and draw new conclusions about the presents of Trematosaurid Temnospondyls in the Early Triassic of Western Australia.
The original material assigned to Erythrobatrachus noonkanbahensis included the holotype, WAM 62.1.46, two topotype specimens, WAM 71.6.22 and WAM 62.1.50, and a high-definition plaster caste of the holotype, WAM 62.1.59 (in taxonomy, a holotype is the specimen from which a species is described, any other specimen considered to belong to the same species as the holotype therefore belongs to that described species, but if the holotype is found to belong to the same species as the previously described holotype of another species, then the newer species is considered invalid; a topotype is a specimen asigned to a species which comes from the same location as the holotype). All of these specimens were placed in the collection of the Western Australian Museum, but when Kear et al. came to look for them, only WAM 62.1.50 and WAM 62.1.59 could be found, WAM 62.1.46 and WAM 71.6.22 having apparently been loaned to John Cosgriff in 1984, at which time he was working at Wayne State University in Detroit, Michigan.
A search of the palaeontological collection of Wayne State University could not locate these specimens, although WAM 62.1.46 was subsequently found in a search for potentially related specimens in the collection of the University of California Museum of Paleontology, where it had been identified as cf. Tertrema sp., and given the identifier UCMP 65858. The collection of the University of California Museum of Paleontology was also found to contain a second high-definition plaster caste of this specimen, listed as UCMP 65850. The University of California Museum of Paleontology has subsequenty returned WAM 62.1.46 to the Western Australian Museum.
Also found within the collection of the University of California Museum of Paleontology was a box labelled WAM 62.1.50, however, this was found to be empty, having been 'withdrawn for study' by John Cosgriff in August 1968.
Specimen WAM 62.1.46, the holotype of Erythrobatrachus noonkanbahensis is a steinkern internal cast from the naso-frontal region of the skull and vomero-palatine section of the palate. This, along with the plaster casts WAM 62.1.59 and UCMP 65850, show Erythrobatrachus noonkanbahensis to have had an elongate skull with a basally constricted rostrum, dorsolaterally facing orbits that are positioned close to the lateral jaw margin, broad nasals that contact the lacrimals posterolaterally, and possibly the septo-maxilla near the external bony nasal opening, anteriorly narrow interpterygoid vacuities that are bordered by transversely broad palatines, ctopterygoids apparently lacking large palatal tusks, at least at the ectopterygoid-palatine suture, and a narrow cultriform process of the parasphenoid that divides the interpterygoid vacuities along the palatal midline, and extends to a point level with the anterior edges of the choanae.
Specimen WAM 62.1.50 is an external impression of the vomerine palate showing multiple dental rows and anterior margins of the choanae. This is recorded as a paratype of Erythrobatrachus noonkanbahensis on its Western Australia Museum label (a paratype is a specimen other than the holotype of a species which is used in the formal description of that species), but as 'cf. Aphaneramma' (refer to Aphaneramma) on the label of the empty box at the University of California Museum of Paleontology, a label which Kear et al. assume reflects Cosgriff's original thoughts on the classification of the specimen. Aphaneramma is a cosmopolitan Trematosaurid Temnospondyl also known from the Early Triassic of Pakistan, Madagascar, Russia, and Svarlbard.
WAM 62.1.50 appears to be similar in proportions to the skull of Aphaneramma gavialimimus, a large (skull-lenght about 400 mm) species of Aphaneramma described from Madagascar in 2017. It also has fine longitudinal bone ridges, which have previously been observed in members of the genera Aphaneramma, Wantzosaurus, and Cosgriffius. The choanae of WAM 62.1.50 are longitudinally offset, such that the left opening would have been displaced anteriorly relative to the right, something which has also been recorded in other specimens of Aphaneramma. In their 1972 description of Erythrobatrachus noonkanbahensis, Cosgriff and Garbutt identify this as being the result of displacement of the right choanae, which they believe was 'compressed and pushed forward from its original position', but which Kear et al. consider may be a diagnostic feature of the genus. Notably, WAM 62.1.50 shows several rows of vomerine teeth (teeth on the roof of the mouth), which are absent in WAM 62.1.46, suggesting the two do not belong to the same species. Vonerine teeth are found in Aphaneramma, as well as some other genera of Trematosaurid Temnospondyls, although the size and arrangement of those of WAM 62.1.50 do not appear to exactly match any previously described taxa. For this reason, Kear et al. return WAM 62.1.50 to the designation cf. Aphaneramma.
Cosgriff and Garbutt apparently viewed the additional specimens assigned to Erythrobatrachus noonkanbahensis, WAM 71.6.22 and WAM 62.1.50, as developmental stages of the species, noting that they were smaller than the holotype, WAM 62.1.46. As WAM 71.6.22 could not be located, this assessment could not be evaluated for this specimen, but Kear et al.'s study clearly shows that the smaller size of WAM 62.1.50 only relates to its fragmentary nature, and that it was clearly derived from quite a large animal. Furthermore, it differs significantly in morphology to WAM 62.1.46, and cannot be assigned to the same species.
This expands the diversity of Temnospondyls known from the Blina Shale, and expands our knowledge of how that assemblage relates to wider Temnospondyl faunal distributions in the Early Triassic. This includes widespread Australian species such as Deltasaurus kimberleyensis, Warrenisuchus aliciae, and Banksiops townrowi, taxa also known from South Africa, such as the genus Batrachosuchus, species not found anywhere else, such as Erythrobatrachus noonkanbahensis, and now an example of the globally distributed genus Aphaneramma. This also increases the distribution of these marine Temnospondyls, raising the possibility that their distribution was not just due to expansion along the continuous coastal Tethyan periphery of the Pangean supercontinent, but may also have involved longer distance, ocean-crossing dispersals between Laurasia and Gondwana across the Tethys Ocean.
See also...
