Phenotypic versus genotypic disparity in the Eurasian orchid genus Gymnadenia: exploring the limits of phylogeny reconstruction. Issue 4 (19th May 2021)

• Record Type: Journal Article
• Authors: Bateman, Richard M.
• Is Part Of: Systematics and biodiversity. Volume 19:Issue 4(2021)
• Page Start: 400
• Page End: 422
• Publication Date: 2021-05-19
• Subjects: evolution -- Gymnadenia -- in situ morphometrics -- internal transcribed spacer -- morphology -- phylogeny reconstruction -- RAD-seq -- speciation -- species circumscription -- transcriptome; Biodiversity -- Periodicals; Biology -- Classification -- Periodicals; Natural history -- Periodicals; Biodiversity; Biology; Classification; Periodicals; 578
• DOI: 10.1080/14772000.2021.1877845
• Languages: English
• ISSNs: 1478-0933

Abstract

Abstract : Recent phylogenetic trees of the phenotypically diverse Eurasian orchid genus Gymnadenia s.l., based on Sanger sequencing of nrITS and next-generation RAD-seq data, largely agree on species boundaries but disagree radically regarding relationships among those species. Moreover, both contrasting topologies receive support from a recent transcriptome study. But these genetic studies are not matched by a taxonomically comprehensive analysis of phenotype. Population-level in situ morphometric analyses of 41 characters for 150 plants spanning seven putative species are used to determine overall phenetic similarities of the species, thereby allowing both phenotypic species circumscription and comparison of estimated phenotypic and genotypic disparities. The three widely recognized short-spurred lineages prove highly morphometrically divergent from each other and from the long-spurred species, which differ from each other only subtly. Phenetic disparity correlates strongly with genetic disparity for RAD-seq data but not with that for nrITS data. The RAD-seq topology is appealing because it implies a fractal evolutionary pattern, but it also requires that later-divergent, recently evolved species are far more geographically widespread than their putative antecedents. Current knowledge of this intensively studied genus allows stronger conclusions on circumscription of taxa than on determining their evolutionary relationships; the clade is unequivocally a single genus that, on present evidence, consists of approximately 12 species. I emphasize the importance of rigorously describing and analysing phenotype, of tree rooting, and of critically appraising reconstructed phylogenies, irrespective of the robustness of the data and the determination with which they are analysed.