vers un nouveau fonctionnement des méta-populations – DIAME

One of the most important paradigm in population biology is the metapopulation concept. Dispersal together with extinction are the central parameters. Exchanges among populations or colonization of new habitats are most of the time thought to be random processes where the source population (if we except density) et the arrival population have no influence on the exchanges nor the characteristics of the dispersing individual. Recent theoretical and empirical studies nevertheless suggest that this is not the real situation. So, we can propose that the cause of dispersal, the type of arrival habitat and the dispersal individual phenotype all interact to determine the rate of dispersal, the selected habitat and the settlement success. To demonstrate this, one need to do experiments. We have at our disposal an important to grant (state and region) to build experimental metapopulations (three connected populations, 16 replicates). Using these facilities, we therefore propose, with the help of the ANR grant, to test the effect of dispersal cause (kin competition, competition among congeners) on the disperser phenotype et its settlement success in the arrival patch (either empty or already occupied). In a second experiment, we will frustrate the disperser and examine the consequence on the individual success and metapopulation dynamics. Two model organisms will jointly participate to the experiments, the Large White and the common lizard. Butterflies and lizards are two model systems which has been reccurently used in the study of dispersal evolution and metapopulation dynamics. The results from these experiments should hopefully have important impacts on the understanding of invasion processes, colonization, and conservation of species living in fragmented habitats. In more fundamental terms, we also hope that these results will bring some new knowledge on dispersal evolution, phenotypic plasticity (physiological and behavioural) and local adaptation.