Cambaytherium, Nakusia, and Kalitherium are closely related early Eocene mammals from the Indo-Pakistan region that have been assigned to Perissodactyla (Laurasiatheria)or Anthracobunidae. The latter have been variously considered artiodactyls or perissodactyls, but more recently are usually placed at the base of the order Proboscidea or of the more inclusive Tethytheria (Afrotheria). We present new evidence from the dentition, skull, and postcranial skeleton of Cambaytherium, from Gujarat, India (ca. 54.5 Ma), that cambaytheres occupy a pivotal position as the sister taxon of Perissodactyla. Cambaytherium was more robust than basal perissodactyls such as ″Hyracotherium″ and Homogalax, and had a body mass of ~25-27 kg based on humeral, radial, and dental regressions. Perissodactyl synapomorphies include a transverse nasal-frontal suture, twinned molar metaconids, and an astragalus with deeply grooved trochlea and a saddleshaped navicular facet. Like perissodactyls, cambaytheres are mesaxonic and have hooflike unguals and a cursorially-adapted skeleton. Plesiomorphic traits compared to basal perissodactyls include bunodont molars with large conules and almost no hint of bilophodonty, unmolarized premolars, sacrum with four vertebrae, humerus with distally extensive pectoral crest and distal articulation lacking a capitular tail, distal radius without discrete scaphoid and lunate fossae, femur with low greater trochanter, calcaneus robust and wide with rounded ectal facet, astragalus wide with moderately long neck and vestigial astragalar foramen, navicular and cuboid short and wide, metapodials short and robust, and Mc I and Mt V present. In most or all of these traits cambaytheres are intermediate between phenacodontid condylarths and perissodactyls but closer to the latter. Our phylogenetic analyses place cambaytheres just outside perissodactyls, and place anthracobunids among primitive perissodactyls. However, similarities between cambaytheres and anthracobunids suggest that they are closely related, and future discovery of skeletal material of anthracobunids will provide a test of this hypothesis. Our results indicate that Anthracobunidae are not Proboscidea or tethytheres, and suggest that the origin of Perissodactyla may have taken place on the drifting Indian plate. How the progenitors of perissodactyls reached India is more problematic but might have involved land connections with Afro-Arabia during the Paleocene. Field work and research supported by the National Geographic Society.
Located in
Library
/
RBINS Staff Publications
Pipimorpha and its crown-group Pipidae possess one of the most extensive fossil records among anurans, which extends into the Early Cretaceous in both Laurasia and Gondwana. This is probably linked to the highly aquatic lifestyle of pipids, which is probably also characteristic of early pipimorphs. In South America, pipids are currently represented only by Pipa, but the fossil record documents an evolutionary radiation of Shelaniinae (a taxon endemic to South America) in the Cretaceous; shelaniines seem to have become extinct in the Eocene. Fewer pipimorph fossils are known from Africa. Our recent redescription of the mid-Late Cretaceous (Coniacian–Santonian) taxon Pachycentrata taqueti from In Becetèn (Niger) partly fills this gap. Our new phylogenetic analysis of Cretaceous and Paleogene pipimorphs shows that this taxon diversified in a West Gondwanan block until about the mid-Cretaceous, but after that, pipimorphs show two distinct evolutionary radiations, one in South America (Pipinae), and the other (Xenopodinae) in Africa. This pattern appears to reflect the breakup of West Gondwana simultaneously with the opening of the South Atlantic during the Cretaceous. This probable vicariant pattern yields slightly different ages for the South Atlantic opening depending on the accepted topology. The tree constrained to reflect the topology of extant taxa supported by molecular data shows a last dispersal between both continents before the Cenomanian (more than 100 Ma), whereas the unconstrained topology that reflects only morphological data is compatible with a more recent last faunal dispersal among pipids. Under this unconstrained topology, the fossil record is too poor to give a reliable minimal age for this last dispersal, but molecular dating analyses suggest that this event harks back to the Mesozoic.
Located in
Library
/
RBINS Staff Publications 2024 OA
