A fossilOsmundafrom the Jurassic of Sweden—reconciling molecular and fossil evidence in the phylogeny of Osmundaceae

Abstract

1The systematic classification of Osmundaceae has long remained controversial. Recent 2 molecular data indicate that Osmunda is paraphyletic, and needs to be separated into 3Osmundastrum and Osmunda s. str. Here we describe an exquisitely preserved Jurassic 4Osmunda rhizome (O. pulchella sp. nov.) that combines diagnostic features of Osmundastrum 5and Osmunda, calling molecular evidence for paraphyly into question. We assembled a new 6 morphological matrix based on rhizome anatomy, and used network analyses to establish 7 phylogenetic relationships between fossil and extant members of modern Osmundaceae. We 8 re-analysed the original molecular data to evaluate root-placement support. Finally, we 9 integrated morphological and molecular data-sets using the evolutionary placement algorithm. 10Osmunda pulchella and five additional, newly identified Jurassic Osmunda species show 11 anatomical character suites intermediate between Osmundastrum and Osmunda. Molecular 12 evidence for paraphyly is ambiguous: a previously unrecognized signal from spacer sequences 13 favours an alternative root placement that would resolve Osmunda s.l. as monophyletic. Our 14 evolutionary placement analysis identifies fossil species as ancestral members of modern 15 genera and subgenera. Altogether, the seemingly conflicting evidence from morphological, 16 anatomical, molecular, and palaeontological data can be elegantly reconciled under the 17 assumption that Osmunda is indeed monophyletic; the recently proposed root-placement in 18Osmundaceae-based solely on molecular data-likely results from un-or misinformative 19 out-group signals. 20

Using more than the oldest fossils: Dating Osmundaceae by three Bayesian clock approaches

“…In their original study, Ronquist et al included even poorly preserved fossils (14 of their 45 fossils could only be coded for 4-10% of the 343 morphological characters), yet recovered topologies did not change and the median node ages were similar to those obtained with subsets excluding these poorly preserved fossils. In our study, the 19 rhizome fossils used in the TE analyses, coded for up to 33 morphological traits (Bomfleur et al 2014b), were placed as expected based on the analyses by Bomfleur et al We have no morphological matrix for the frond fossils, so they could not be used in TE dating. The strengths of TE dating, the parallel optimisation of node heights and phylogenetic placement of fossils are also its major weaknesses: The method forces workers to carefully analyse and score morphological traits, but where comprehensive morphological data are unavailable, it cannot be used.…”

“…A few regions were excluded due to ambiguities in the alignment (29 bp in rbcL-accD, 129 bp in rbcL-atpB, 51 bp in rps4-trnS, 39 bp in trnG-trnR, and 146 bp in trnL-trnF) and missing data (rps4), resulting in a matrix of 13 species and 8616 nucleotide positions. We excluded the outgroup, Gleicheniales, because of long-branch attraction (Bomfleur et al 2014b) and follow these authors in rooting Osmundaceae between Todea/Leptopteris and Osmunda/Osmundastrum instead of between Osmundastrum and the remaining three genera (Yatabe et al 2005, Metzgar et al 2008. Sampling covers the seven species of Osmunda from the three subgenera Claytosmunda, Osmunda, and Plenasium; the single species of Osmundastrum, three of the ten species of Leptopteris, and both species of Todea (the second species lacked morphological data and was therefore excluded from some analyses).…”

“…The first two occur mostly in the Northern Hemisphere extending into the humid tropics, the latter two in South Africa and Australasia (Kubitzki 1990). Royal ferns have an exceptional fossil record (Miller 1971, Tidwell and Ash 1994, Bomfleur et al 2014b, Wang et al 2014, with more than 150 species, 17 genera, and at least three extinct families ranging from the Permian to Neogene. The fossils are mostly foliar remains, but many are anatomically preserved (permineralised) axes.…”

“…Within-Osmundaceae divergence events have not been the focus of a previous molecular clock study. The discovery of an unusually well preserved 182-190 million year-old Osmunda fossil from Sweden (Bomfleur et al 2014a) recently prompted a critical re-evaluation of the rhizome fossils of modern Osmundaceae (Bomfleur et al 2014b). This reappraisal (involving morphological data matrices) provided the opportunity to estimate the timing of major evolutionary events within the family, using both TE and FBD molecular clock dating.…”

“…Our phylogenies are based on ~8.6 kb of plastid DNA data for 13 extant species, 33 morphological characters scored for 19 permineralised rhizomes (data from Bomfleur et al 2014b), and 17 frond compression/ impression fossils, assessed on the basis of key autapomorphic features for the present study. Incorporating fossils into tree models as undertaken here in principle allows one to infer speciation and extinction rates with reasonable confidence as shown with simulated data by Heath et al (2014) and is probably an improvement over inferring these rates only from neontological data.…”