Colonization and persistence ability explain the extent to which plant species fill their potential range

Aim: How species traits and environmental conditions affect biogeographical dynamics is poorly understood. Here we test whether estimates of a species'evolutionary age, colonization and persistence ability can explain its current 'range filling'(the ratio between realized and potential range size). Location: Fynbos biome (Cape Floristic Region, South Africa). Methods: For 37 species of woody plants (Proteaceae), we estimate range filling using atlas data and distribution models, evolutionary age using molecular phylogenies, and persistence ability using estimates of individual longevity (which determines the probability of extinction of local populations). Colonization ability is estimated from validated process-based seed dispersal models, the arrangement of potential habitat, and data on local abundance. To relate interspecific variation in range filling to evolutionary age, colonization and persistence ability, we use two complementary model types: phenomenological linear models and the process-based metapopulation model of Levins. Results: Linear model analyses show that range filling increases with a species' colonization and persistence ability but is not affected by species age. Moreover, colonization ability is a better predictor of range filling than its component variables (local abundance and dispersal ability). The phylogenetically independent interaction between colonization and persistence ability is significant (P < 0.05) for 97% of 180 alternative phylogenies. While the selected linear model explains 42% of the variance in arcsine transformed range filling, the Levins model performs more poorly. It overestimates range filling for realistic parameter values and produces unrealistic parameter estimates when fitted statistically. Main conclusions: Colonization and local extinction seem to shape Proteaceae range dynamics on ecological rather than macroevolutionary time-scales. Our results suggest that the positive abundance-range size relationship in this group is due primarily to the effect of abundance on colonization. In summary, this study contributes to a process-based understanding of range dynamics and highlights the importance of colonization for the future survival of Fynbos Proteaceae. © 2007 The Authors Journal compilation © 2007 Blackwell Publishing Ltd.