Explaining the disjunct distributions of austral plants: the roles of Antarctic and direct dispersal routes

AimDispersal explains the disjunct distributions of many austral plant lineages. However, the role of Antarctica is largely uncertain and the routes of dispersal have remained speculative. Based on niche conservatism we can make predictions about the timing of disjunction establishment, as well as the availability of direct transoceanic, Antarctic stepping-stone, and out-of-Antarctica dispersal routes over time. We evaluate these predictions using molecular divergence time estimates for the establishment of disjunct distributions across multiple plant lineages. LocationSouthern Hemisphere. MethodsWe estimated the timing of disjunction establishment and determined habitat affinities for 72 austral plant groups. We used Wilcoxon rank sum tests to compare the timing of disjunction establishment between cold and temperate climate lineages for the full data set, as well as within several subsets. We compared our results with those from a literature survey. ResultsAs niche conservatism predicts, the timing of disjunction establishment in cold and temperate climate austral lineages is consistent with the availability of the corresponding habitats over time. Our results also suggest that disjunction establishment has involved a combination of Antarctic and direct dispersal routes. For cold climate lineages, both out-of-Antarctica and direct dispersal routes are required to explain the observed estimates, while stepping stone routes cannot be ruled out. It appears that for these lineages the importance of the three dispersal routes differs with environmental, geographical and temporal context. Main conclusionsBoth direct and Antarctic dispersal routes are necessary to explain the establishment of contemporary austral distributions. Evidence that some taxa were, until recently, restricted to Antarctica changes how we view the evolutionary histories of austral floras and the lineages they contain. Moreover, that we detect differences in the importance of alternative dispersal routes suggests that long-distance plant dispersal processes can be explicitly incorporated into models of climate change response.