séminaire du pôle évolution du vivant - vendredi 4 avril

Biogeographic patterns in the Mediterranean realm: two case studies (Araceae and Boraginaceae)

Guilhem Mansion, Institute for Systematic Botany, université de Zürich, Suisse

vendredi 4 avril 2008 à 11h, amphi Monge, bâtiment Gabriel 

Despite the remarkable species-richness of the Mediterranean flora and its well-known geological history, few studies have investigated its temporal and spatial origins. Most importantly, the relative contribution of geological processes and long distance dispersal to the composition of contemporary Mediterranean biotas remains largely unknown. Our main goals in this study were to elucidate the time frame and biogeographic events associated with the diversification of two plant families in the Mediterranean basin. More specifically, we examined the alternative paleo- and neoendemic hypotheses for the origin of the endemics. The presence of numerous reliable fossils and the occurrence of many species endemic to either fragment or oceanic islands make the Araceae and Boraginaceae particularly suitable for integrative studies including phylogenetic analyses of DNA sequences, Bayesian dating methods, and ancestral area reconstructions, in combination with paleogeographic, paleoclimatic, and ecological evidence. The present study revealed an Anatolian origin for most of the investigated genera and underlines the importance of the Eastern Mediterranean region as a possible reservoir for plant evolution in the Mediterranean Basin. Furthermore, a recurrent theme emerging from our analyses was that land connections and interruptions, caused by repeated cycles of marine transgressions-regressions between the Tethys and Paratethys, favored geodispersalist expansion of biotic ranges from western Asia into the western Mediterranean basin and subsequent allopatric speciation at different points in time from the Late Eocene to the Late Oligocene. Finally, only one case of paleoendemism was detected, often contradicting earlier intuitive hypotheses.