Largest known Mesozoic multituberculate from Eurasia and implications for multituberculate evolution and biology

Abstract

A new multituberculate, Yubaartar zhongyuanensis gen. and sp. nov., is reported from the Upper Cretaceous of Luanchuan County, Henan Province, China. The holotype of the new taxon is a partial skeleton with nearly complete cranium and associated lower jaws with in situ dentitions. The new species is the southern-most record of a Late Cretaceous multituberculate from outside of the Mongolian Plateau in Asia and represents the largest known Mesozoic multituberculate from Eurasia. The new specimen displays some intriguing features previously unknown in multituberculates, such as the first evidence of replacement of the ultimate upper premolar and a unique paleopathological case in Mesozoic mammals in which the animal with a severely broken right tibia could heal and survive in natural condition. The phylogenetic analysis based on craniodental characters places Yubaartar as the immediate outgroup of Taeniolabidoidea, a group consisting of a North American clade and an Asian clade. This relationship indicates at least a faunal interchange of multituberculates before the K-Pg transition. The new evidence further supports the hypothesis that disparity in dental complexity, which relates to animal diets, increased with generic richness and disparity in body size, and that an adaptive shift towards increased herbivory across the K-Pg transitional interval.

A new, diminutive species ofCatopsalis(Mammalia, Multituberculata, Taeniolabidoidea) from the early Paleocene of southwestern Alberta, Canada

“…We follow Williamson et al (2015) in recognizing a monophyletic Taeniolabidoidea that minimally includes the Taeniolabididae, a North American family that contains Taeniolabis and Kimbetopsalis , and the genera Valenopsalis and Catopsalis , both of which are classified as Taeniolabidoidea incertae sedis. We provisionally concur with Xu et al (2015), Williamson et al (2015), and Mao et al (2016) in regarding Sphenopsalis and Lambdopsalis , from the Mongolian Plateau of East Asia, as constituting a monophyletic family, the Lambdopsalidae, the sister group to Taeniolabididae. The enigmatic Late Cretaceous taxon Bubodens magnus Wilson, 1987, to date known only from an isolated m1 from the Fox Hills Formation of South Dakota, has been previously considered a taeniolabidoid (e.g., Kielan-Jaworowska et al, 2004; Weil and Krause, 2008; Clemens, 2010).…”

“…The discovery of a diminutive species of taeniolabidoid multituberculate in rocks of late early Paleocene age is unexpected, least of all because the general overall trend in taeniolabidoid evolution appears to have been one of increasing body size, but also increasing size in concert with larger molar size, greater cusp number, and increased dental complexity and morphological disparity (Middleton, 1982; Wilson et al, 2012; Williamson et al, 2015). Several recent phylogenetic analyses of multituberculates have included taeniolabidoids (Xu et al, 2015; Williamson et al, 2015; Mao et al, 2016), with the phylogeny of Williamson et al (2015) being the most comprehensive with respect to Taeniolabidoidea. The results of Williamson et al (2015) suggest that Valenopsalis and Catopsalis alexanderi , previously the two smallest North American taeniolabidoids, are the most basal, with larger body size generally being more characteristic of geologically younger, derived species, including C .…”

“…The taxonomic composition of Taeniolabidoidea has varied, but results of recent studies suggest that the superfamily minimally contains four North American genera ( Valenopsalis Williamson et al, 2015, Catopsalis Cope, 1882, Kimbetopsalis Williamson et al, 2015, and Taeniolabis Cope, 1882), all from the early Paleocene to earliest late Paleocene, as well as two genera from the late Paleocene of eastern Asia ( Sphenopsalis Matthew, Granger, and Simpson, 1928 and Lambdopsalis Chow and Qi, 1978). A third eastern Asian genus, Prionessus Matthew and Granger, 1925, has traditionally been considered a taeniolabidoid, but we consider this referral doubtful (contra, e.g., Xu et al, 2015).…”

“…As such, ‘Taeniolabidoidea,’ in any of its previous incarnations, were then considered nonmonophyletic, and the phylogenetic relationships of the various ‘taeniolabidoid’ taxa (e.g., Eucosmodontidae, Microcosmodontidae, Djadochtatherioidea), both to one another and to other multituberculates, remained uncertain. The evolutionary relationships of the various taeniolabidid species have been explored more recently, resulting in Taeniolabidoidea once again being resurrected, although the constituent taeniolabidoid taxa differ among the studies (compare, e.g., Xu et al, 2015, Williamson et al, 2015, and Mao et al, 2016). We follow Williamson et al (2015) in recognizing a monophyletic Taeniolabidoidea that minimally includes the Taeniolabididae, a North American family that contains Taeniolabis and Kimbetopsalis , and the genera Valenopsalis and Catopsalis , both of which are classified as Taeniolabidoidea incertae sedis.…”

Dome-headed, small-brained island mammal from the Late Cretaceous of Romania

“…Our phylogenetic analysis utilizes a dataset that broadly samples multituberculates, which is built on a data matrix that one of us (J.M.) has been iteratively building with colleagues (35,36). This dataset, in turn, has sampled taxa and characters from several previous studies (see refs.…”