Julien Benoit

The tapinocephalid dinocephalian Moschops is one of the most iconic taxa of the middle Permian, yet its paleobiology remains one of the least well‐known. Here, we address some aspects of paleoneurology and paleobiology of Moschops using CT scanning of four well preserved skulls from the upper Abrahamskraal Formation of South Africa. Two of the specimens preserve articulated scleral ossicle rings, the dimensions of which are more consistent with a diel activity pattern, although intermediate between diurnal, nocturnal, and cathemeral. The bony labyrinth for the inner ear (preserved in one specimen) has comparatively short semicircular canals, and the agility score aligns with those of modern large herbivores that are able to swim. This is consistent with previous hypotheses inferring that some tapinocephalid dinocephalians may have been semi‐aquatic. The four individuals form a partial ontogenetic series, and we show that the endocast grows markedly with body size in Moschops , as does the encephalization quotient. The latter is an unusual pattern for vertebrates. We also report the first possible evidence for the preservation of soft brain tissues in a non‐mammalian synapsid, which sparks exciting perspectives for future studies of non‐mammalian synapsid paleoneurology.

The maxillary canal system in Cynognathus and Diademodon is described. The hypothesis that expansion of the maxillary sinus represents a synapomorphy of the clade Cynognathia is supported. A trend towards the reduction of the inferior palpebral ramus is proposed.

Brain size and encephalization are correlated with extinction risks, making them central to understanding evolutionary outcomes in the face of the climatic and ecological challenges posed by mass extinction events. Yet, the end-Permian mass extinction—the most significant extinction event in the history of Life—has received little attention from a paleoneurological perspective. Here we use Synchrotron Radiation and CT scanning to study the evolution of brain endocasts of mammalian forerunners, the Permo-Triassic Synapsida, across this extinction event. During mass extinction events, it is expected that the brain would adapt either by growing larger to enhance behavioral response to the challenges, or shrinking to save energy. To test these predictions, we conducted a broad phylogenetic survey of synapsids across the end of the Permian and, unexpectedly, found stasis of the encephalization quotient. Furthermore, analysis of evolutionary rates indicates that stabilizing selection does not account for the observed stasis. We hypothesize that a lack of resources may have impeded neuroplasticity and prolonged a period of stasis that began in the late Permian. Brain size resumed varying after the crisis, marking the first steps toward the evolution of the mammalian brain.

Oviparity was likely the plesiomorphic reproductive condition for non-mammalian Synapsida, the stem-mammal group. Yet, despite nearly two centuries of research, no definitive fossil eggs of late Palaeozoic or early Mesozoic synapsids have been discovered. Here, three perinate specimens of the dicynodont genus Lystrosaurus from the Early Triassic of the South African Karoo Basin are examined using high-resolution CT and synchrotron scanning. One specimen, NMQR 3636, displays a tightly curled posture suggestive of an in ovo position and completely lacks tusks. Crucially, the lower jaw symphysis remains unfused—a developmental trait found only in pre-hatching embryos of modern birds and turtles. No calcified eggshell is preserved, so the egg might have been soft and leathery. The large size of the reconstructed egg suggests a precocial, non-milk-feeding developmental strategy. As a non-cynodont synapsid, Lystrosaurus offers a rare and valuable glimpse into reproductive biology far removed from the mammalian crown group. Unlike the more derived, mammal-like cynodont Kayentatherium , whose egg size aligns with lactation, Lystrosaurus anchors the plesiomorphic condition deep within Synapsida. Its reproductive strategy may have played a crucial role in its resilience and ecological dominance following the end-Permian mass extinction.

L'étude des caractères basicrâniens et endocraniens chez les vertébrés fossiles et actuels connait un regain d'intérêt depuis quelques années. Ces caractères bénéficient en effet d'une bonne réputation en tant que marqueurs phylogénétiques, ainsi que d'un biais taphonomique favorable à leur fossilisation. C'est pourquoi dans ce volume, la région auditive et l'oreille interne (Partie I) et l'endocrâne (Partie II) de mammifères fossiles appartenant au clade des Afrotheria ont été scannés par (micro)tomographie à rayon-X, décrits, étudiés et comparés à ceux des espèces actuelles. Il s'agit plus précisément des Afrotheria fossiles que sont les Bibymalagasia (Chapitre 1), le Macroscelididae Chambius (Chapitre 2), de nombreux Hyracoidea du Paléogène (Chapitre 3) ainsi que des Tethytheria basaux tels que le proboscidien Numidotherium, l'embrithopode Arsinoitherium et le sirénien Prorastomus (Chapitre 4). Cette étude met en lumière l'intéressant signal phylogénétique fournit par les caractères de la région auditive et de l'endocrâne, capable de soutenir certains clades moléculaires (e.g. Afroinsectivora, ‘Panelephantulus'). Elle montre aussi comment, sous certaines pressions de sélection, ces caractères peuvent converger de façon remarquable (e.g. chez les téthythères). Cette étude apporte aussi des éléments qui appuient l'hypothèse d'une origine Africaine des Afrotheria, ainsi que celle faisant de l'ancêtre commun des afrothères un mammifère plutôt ‘ongulé' que ‘insectivore'. Ces inférences fondées sur l'observation des représentants fossiles les plus anciens de ce groupe sont importants pour notre compréhension de l'origine et de la diversification des afrothères et des mammifères placentaires en général.

Cynodontia is an important subclade of Therapsida that first occurred in the late Permian. It includes extinct subclades which are the non‐mammaliaform cynodonts and Mammaliaformes, with the latter ultimately giving rise to crown mammals. The systematics of non‐mammaliaform cynodonts has been extensively studied and is relatively well‐resolved, however, there are still many problematic taxa that are difficult to identify and place confidently into the cynodont phylogenetic tree. Cistecynodon parvus is one such taxon, known only from a single specimen, the holotype skull which was found in the Middle Triassic Burgersdorp Formation of South Africa, in strata assigned to the Trirachodon–Kannemeyeria Subzone of the Cynognathus Assemblage Zone. Over the past century Cistecynodon has been variously referred to Eucynodontia (which includes the two major subclades Probainognathia and Cynognathia) and to non‐eucynodont cynodonts, without any emerging consensus. Here the holotype of Cistecynodon parvus (BP/1/2520) is re‐described based on computed tomography (CT) digital reconstruction of the specimen. This new data is applied to determine a phylogenetic matrix which supports that C. parvus is a basal (non‐eucynodont) cynodont. This basal position is reflected in the anatomy of its secondary palate, which is not closed despite the specimen being a subadult. The inner ear, trigeminal canal, parietal foramen and carotid foramina are uniquely derived, likely as adaptations to an obligate fossorial lifestyle, confirming the validity of the taxon and its inferred subterranean habitat.

The middle Permian represents a critical interval in therapsid evolution, when gorgonopsians emerged as some of the first specialized apex predators within terrestrial ecosystems. Despite their significance, the early diversification of Gorgonopsia in Gondwana remains poorly understood due to scarcity and fragmentary material. Here, we describe a nearly complete skull (BP/1/8260) with an occluded lower jaw from the lower Abrahamskraal Formation ( Tapinocephalus Assemblage Zone). The specimen exhibits a distinctive combination of cranial features, including a transversely narrow snout, small orbital and temporal openings, V‐shaped palatine bosses, and a vertically oriented occiput that distinguish it from all known gorgonopsians. Based on its unique morphology, a new taxon, Jirahgorgon ceto sp. nov., is established for this specimen. In our phylogenetic analysis, the new taxon forms a clade with Phorcys dubei , which is here named Phorcyidae fam. nov. Members of Phorcyidae are unique among basal African gorgonopsians in combining a vertical occiput and rubidgeine‐like cranial proportions, indicating that large‐bodied gorgonopsians were present in the Wordian and overturning notions that they were exclusively small carnivores until the Wuchiapingian. Basal skull length analyses indicate that body size evolution in Gorgonopsia, while largely random through time (Brownian motion evolutionary model), is nonetheless structured by shared ancestry. The discovery of Jirahgorgon illustrates the complexity of gorgonopsian evolution, showing an early appearance of large‐bodied, robust morphotypes and highlighting the lower Abrahamskraal Formation as a key resource for understanding the initial radiation of theriodonts.

Paul Ellenberger’s archives provide insight into the Basotho and San peoples palaeontological knowledge in southern Africa, specifically from Lesotho. For instance, it preserves many published and unpublished accounts of Basotho people discovering dinosaur footprints and bones in Lesotho. Moreover, the Basotho folk tale (a “litsomo” in Sesotho) featuring the voracious monster Kholumolumo is here demonstrably linked to dinosaur trace and bone fossils in the main Karoo Basin of southern Africa, suggesting that this well-known local fable functions as a geomyth. For the San, Ellenberger hypothesised that their understanding of fossil footprints reflected a grasp of geoscientific concepts, potentially including the principles of superposition, faunal succession, locomotor biology, and plate tectonics. He connected the San myths of the Agama (!Khau) and All-Devourer (//Khwai-hemm) to fossil footprints, arguing in an unpublished manuscript that the San narrative indicated a historical awareness of not only the fossil track record but also of geoscientific concepts. Despite Ellenberger’s intriguing interpretations, the evidence for this San geomyth is tenuous. It is possible that this geomyth may partly result from Ellenberger’s enthusiasm for linking indigenous knowledge with his interests in fossil footprints. The evidence for San understanding of geological concepts remains speculative, and more robust data is needed to substantiate these claims. Nevertheless, this review paper highlights the scientific contributions of indigenous populations as an early form of citizen science and underscore the complex interactions between indigenous knowledge and scientific inquiry.

A subadult Moschognathus whaitsi from the Eastern Cape Province, South Africa, was scanned using synchrotron radiation X‐ray computed tomography (SRXCT). Its subadult state allowed the cranial bones and teeth to be identified and individually reconstructed in 3D. A complete description of every preserved cranial bone is here produced, with special attention given to the braincase. An evaluation of a frontal abscess has provided insight into the potential behavior of Moschognathus , regarding the long‐standing hypothesis of dinocephalian headbutting. The first 3D description of tapinocephalian dentition shows the unique talon‐and‐heel tooth morphology and expands our understanding of the replacement patterns in the dentition of basal therapsids. We report evidence of at least three successive replacement generations developing in the incisiform dentition of M. whaitsi simultaneously. The arrangement of these successive generations of replacement incisiform teeth is reminiscent of the complex dental batteries described in sauropod dinosaurs. Furthermore, an alternating pattern of replacement is apparent and suggests that the upper and lower rows of functional teeth were comprised of at least two tooth generations.