Phylogeny and phylogeography of Zieria (Rutaceae)

Abstract

Zieria (Anthophyta: Rutaceae) is a predominately Australian genus of shrubs and small trees, consisting of 59 Australian species and one species endemic to New Caledonia. Although there has been considerable taxonomic revision of this genus in recent decades, largely based on morphology, a comprehensive molecular phylogenetic study that would include as many currently-recognised taxa as possible was clearly needed. Using five chloroplast markers (rpl32–trnL, trnL–F, trnQ–5'rps16, trnS–G and 3'trnV– ndhC) and two nuclear DNA markers (ITS and ETS) this study undertook phylogenetic analyses in order to understand relationships, evolutionary processes and taxonomic issues in Zieria, and to interpret biogeographic patterns. Both Bayesian inference and maximum parsimony methods were employed. Almost all species of Zieria were represented in separate analyses of the cpDNA and nrDNA datasets, with the monotypic sister genus Neobyrnesia used as the outgroup. The results show that relationships within Zieria are complex; widespread incongruence was revealed between cpDNA, nrDNA and currently recognised taxa. A combination of factors is suggested to explain this, including regional cpDNA introgression (chloroplast capture) and incomplete lineage sorting, and the need for taxonomic revision of some taxa. The nrDNA provided greater support for monophyly of species than cpDNA, and potentially a better indication of phylogeny. However, deeply divergent paralogues of nrDNA were detected in some taxa, making the assessment of phylogenetic relationships more challenging, and highlighting some possible pitfalls in using nrDNA for phylogenetic reconstructions.

The phylogenetic relationship of Australian taxa and Z. chevalieri, the single endemic New Caledonian species, is of particular interest in terms of the long-standing debate over the history of New Caledonia and its flora. The placement of Z. chevalieri differed between cpDNA and nrDNA trees. The cpDNA phylogeny placed Z. chevalieri as sister to all other species in the genus, suggesting that the earliest divergence was between lineages of New Caledonia and Australia, and that differentiation of species occurred through vicariance. In contrast, Z. chevalieri was placed higher in the nrDNA tree, which could suggest later dispersal, over water or via past exposed land in the Tasman Sea rather than older vicariance between Australia and New Caledonia. However, the results did not favour one hypothesis over another, and the debate will continue.

A phylogeographic study of one widespread species, Zieria arborescens, was also undertaken, focusing on populations in the wet forests of Victoria and Tasmania. The cpDNA was not highly informative but provided some evidence of Tasmanian–eastern Victorian connections, interpreted as past geographic connectivity during periods of low sea-level now severed by Bass Strait.

Finally, a number of taxonomic issues arising from the results of this project are discussed, and recommendations are made for reappraising and possibly re-circumscribing several taxa. In some cases this might require the description of new species based on variation between populations (e.g. Z. smithii and Z. furfuracea), or recircumscription of taxa to effect monophyly (e.g. Z. arborescens). Some taxa might be considered conspecific based on noteworthy similarities of morphological and genetic characters (as seen in this study) and the geographic proximity of populations (e.g. Z. buxijugum and Z. parrisiae; Z. alata and Z. madida; Z. verrucosa and Z. vagans; Z. caducibracteata and Z. arborescens; and Z. minutiflora and Z. obovata).