Evolutionary stability of plant-pollinator networks: efficient communities, hysteresis, and a pollination dilemma

Mutualistic interactions between species such as pollination and plant-mycorrhiza interactions are ubiquitous in nature and essential for ecosystem functioning. Often dozens or even hundreds of species with different degree of specialization form complex networks of interdependence. How the complexity evolved and is maintained are fundamental questions in ecology. Here, we present a new game theoretical approach to model complex mutualistic interactions, which we apply to pollination networks. The theoretical analysis revealed multiple evolutionary stable network structures that form a gradient from generalism toward specialism with increasing availability of pollination service. In particular, we found that efficient communities evolve only under pollination oversupply, but that pollination shortage selects for inefficient network structures due to a pollination dilemma. These results suggest that availability of pollination services is a key factor structuring pollination networks and offer a new explanation for the geographical differences in pollination faunas that have long been recognized by ecologists. The study bridges the gap between network studies, game theory, and the natural history of pollination, which have hitherto been studied largely independently.