Independently assessing seed-plant molecular phylogenies: does the cordaite fossil record support close relationships between Gnetales and conifers

The study of DNA in living organisms has revolutionised how we conceptualise and analyse the evolutionary history of biological groups, a scientific discipline known as phylogeny reconstruction. Earlier methods relied on the analysis of morphological and anatomical features of organisms in what are termed morphological cladistic analyses, but more recently these have been used only to supplement analyses based on DNA sequences, which are termed molecular cladistic analyses. In some cases the results gained from morphological and molecular analyses contrast strongly with each other, generating damaging uncertainties regarding which set of evidence more correctly infers the evolutionary history of the organisms studied, and undermining the confidence not only of formal classifications but also of any following evolutionary interpretations. One such critical group, the seed plants, originated ca. 363 million years ago in the late Devonian period; they have an extensive fossil record and dominate modern terrestrial ecosystems. Living seed plants represent five major taxonomic groups: cycads, Ginkgo, conifers, Gnetales and angiosperms (flowering plants). Several other major seed-plant groups evident in the fossil record are long extinct, and so cannot yield molecular information. They include a complex of plants that combine fern-like foliage and seeds, termed the pteridosperms, that are implicated in the evolutionary origins of all five extant groups of seed-plants. Such fossils are the only means of bridging the vast morphological gaps that separate the five extant seed-plants groups, which cannot be addressed directly using molecular systematic methods. Of the five extant groups, the Gnetales includes three distinct extant lineages (represented by the genera Welwitschia, Gnetum and Ephedra) that show remarkable combinations of morphological features. Most morphological cladistic studies place Gnetales next to angiosperms, immediately above fossil groups such as the advanced pteridosperm Pentoxylon and the Bennettitales (the Anthophyte hypothesis of seed-plant relationships). However, data-rich (if taxon-poor) molecular phylogenetic studies increasingly place Gnetales either at the base of the conifers (the Gnetifers hypothesis) or within conifers immediately adjacent to pines (the Gnepines hypothesis). This project seeks to (a) independently assess current molecular results with new fossil data from exceptionally well-preserved plants, and (b) integrate the results with morphological data from current investigations of development and the key genes that underpin development, together with molecular phylogenies. Thus, we aim to employ a whole-evidence approach to reconstruct the evolutionary history of seed plants. The Gnepine, Gnetifer and Anthophyte hypotheses of relationships will be tested by carefully targeted analysis of the fossil record, focusing on detailed morphological features that may, or may not, unite Gnetales with conifers. Recent morphological reanalysis of living Gnetales shows that phylogeneticists have previously been misinterpreting their male organs, which are similar to some extinct conifers and to a closely allied extinct group called the cordaitean coniferophytes. This will be the first detailed study designed to determine whether the cordaites have fundamental similarities with Gnetales. Any fundamental similarities will support the Gnepine and especially the Gnetifer hypotheses. Any fundamental dissimilarities may return the focus of gnetalean relationships to the angiosperms inherent in the Anthophyte hypothesis, but may also require substantial revision of inferred relationships among major groups within the coniferophytes. We are therefore confident that this project will help to elucidate the controversial history of seed plants.