ISOLATION, POPULATION-SIZE AND EXTINCTION - THE CLASSICAL AND METAPOPULATION APPROACHES APPLIED TO VASCULAR PLANTS ALONG THE DUTCH RHINE-SYSTEM

The recent rise of conservation biology has renewed interest in the effects of population size and isolation on population extinction. Empirical evidence for the role of isolation is however scarce, especially in plants. When applying the classical approach, associated with the MacArthur-Wilson model, to patterns of plant species distribution there is seldomly a large contribution of isolation, if any. A more promising way would be to analyse extinction and colonization processes in time of separate species, here called the metapopulation approach. In this paper a data set of 143 sites along the Dutch Rhine system is compared between two years (1956 and 1988). The multiple regression analysis of the 1956 distribution of total species number (classical approach) revealed significant contributions of site area and environmental variance within sites (26% and 5.7% of total variation, respectively). Isolation explained an additional 4.2% of total variation and is relatively unimportant. Logistic regression analysis of population extinction between 1956 and 1988 (the metapopulation approach) demonstrated that for some species large populations with close neighbours had a higher chance of persistence. For other species only population size was important for extinction and yet other species were only influenced by isolation. For some species (Eryngium campestre, Medicago falcata and possibly Plantago media) both extinction and colonization were influenced by isolation suggesting that populations of these species were integrated in a metapopulation structure. Further analysis revealed that this structure must be mostly of the equal sized source-recipient type. These results are discussed in relation to population biology characteristics of the different reaction types and the existence of metapopulations in higher plants compared to animals. The results demonstrate that the metapopulation approach, i.e. process analysis of individual species, is a more powerful tool when analysing isolation effects than the MacArthur-Wilson approach, i.e. pattern analysis of total species number. It is argued that both the metapopulation concept and metapopulation method deserve more attention in plant population biology.