Incorporating phylogenetic uncertainty on phylogeny‐based palaeontological dating and the timing of turtle diversification

Abstract

Methods improving the performance of molecular dating of divergence time of clades have improved dramatically in recent years. The calibration of molecular dating using the first appearance of a clade in the fossil record is a crucial step towards inferring the minimal diversification time of various groups and the choice of extinct taxa can strongly influence the molecular dates. Here, we evaluate the uncertainty on the phylogenetic position of extinct taxa through non-parametric bootstrapping. The recognition of phylogenetic uncertainty resulted in the definition of the Bootstrap Uncertainty Range (BUR) for the age of first appearance of a given clade. The BUR is calculated as the interval of geological time in which the diversification of a given clade can be inferred to have occurred, based on the temporal information of the fossil record and the topologies of the bootstrap trees. Divergence times based on BUR analyses were calculated for three clades of turtles: Testudines, Pleurodira and Cryptodira. This resulted in extensive uncertainty ranges of topology-dependent minimal divergence dates for these clades.

Allaeochelys libyca, a New Carettochelyine Turtle from the Middle Miocene (Langhian) of Libya

“…The strict consensus tree of these 10,000 equally parsimonious solutions, however, retrieves a monophyletic PanCarettochelys, Trionychia and Pan-Trionychia (i.e., Adocusia and Trionychia) and thereby reveals that the great number of trees found in the first analysis is not the result of homoplasy, but rather of many rogue taxa with shifting positions. A monophyletic Pan-Trionychia is consistent with most of the recent phylogenetic analyses (e.g., Meylan and Gaffney 1989;Joyce 2007;Danilov and Parham 2008;Tong et al 2009;Joyce et al 2011;Anquetin 2012;Sterli et al 2013).…”

“…Newer versions of this matrix were not used (e.g., Joyce et al 2011;Anquetin 2012;Sterli et al 2013) because they universally added characters and taxa not relevant to trionychian systematics and because rampant homoplasy with the existing analysis already proved difficult to handle computationally. The newly added taxa are Kizylkumemys khoratensis Tong et al, 2005 (as described by Tong et al 2005); K. schultzi Nessov, 1977(as described by Nessov 1977; Anosteira pulchra (Clark, 1932) (Young and Chow, 1962) (as described by Young and Chow 1962); Allaeochelys crassesculpta (Harrassowitz, 1922) (personal observation of material by W.G.…”

A review of the fossil record of turtles of the clade Thalassochelydia

“…In a few studies, thalassochelydian turtles have been found to be closely related to Cretaceous turtles, in particular the Early Cretaceous protostegid Santanachelys gaffneyi Hirayama, 1998 and the Early Cretaceous sandownid Sandownia harrisi Meylan et al, 2000(Hirayama 1998Meylan et al 2000;Joyce 2007;Mateus et al 2009;Sterli et al 2013;Anquetin et al 2015). If these connections are corroborated by future work, Thalassochelydia may become significantly more speciose than presented herein by including species from the Cretaceous and Paleogene.…”

“…Joyce (2007) included an expanded sample by scoring Plesiochelys etalloni, Portlandemys mcdowelli, Jurassichelon oleronensis (his "Thalassemys" moseri), and Solnhofia parsonsi as terminal taxa, but they were found in a paraphyletic arrangement. More recent global phylogenetic analyses of turtles continued to include these species as separate terminal taxa Parham 2006, 2008;Sterli 2010;Anquetin 2012;Rabi et al 2013;Sterli et al 2013;Zhou et al 2014;Zhou and Rabi 2015), but none found them to form a monophyletic group.…”

A Review of the Fossil Record of Turtles of the CladeBaenidae

“…The only exception is Naomichelys speciosa , which was only known by a single, large entoplastron and has since been shown to be a representative of the clade Solemydidae (Hirayama et al 2000; Joyce et al 2014). Williams (1950) recognized a similar set of turtles from the Late Cretaceous to Eocene of North America, his Baenoidea, but also included Meiolaniidae on the basis of the presence of a biconvex cervical vertebra IV, but this characteristic is now interpreted as a homoplasy and meiolaniids are no longer thought to be closely related to baenids (e.g., Gaffney 1996;Joyce 2007;Anquetin 2012;Sterli et al 2013). The influential classification of was centered on the same set of Late Cretaceous to Eocene turtles, similar to previous authors, but Baenidae was expanded to include Trinitichelys hiatti from the Early Cretaceous (Aptian-Albian) of Texas and the idiosyncratic Compsemys victa Leidy, 1856 from the Late Cretaceous to Paleocene of the American West .…”