Demography, Pollination, and Baker's Law

Baker's Law states that selfing should commonly be selected during dispersal because bottlenecks during colonization limit the availability of mates. Although this truism has broad intuitive appeal, a recent body of theory (Cheptou and Massol 2009; Massol and Cheptou 2011) casts doubt on whether adaptation favors both selfing and dispersal when both parameters are free to evolve. In these models, the joint evolution of dispersal and the selfing rate are considered in a metapopulation, with a spatially and temporally variable pollination environment. Under these conditions, adaptation favors one of two strategies: the "dispersal/outcrosser" syndrome and the "no dispersal/selfing" syndrome. These results appear to contradict the prediction of Baker's Law. These models clarify how variation in the pollination environment per se cannot generate an association between selfing and dispersal. That being said, demographic factors during dispersal episodes are likely to be important in generating patterns consistent with Baker's law. Determining whether Baker's law maintains its predictive utility requires determining whether seed banks, the perennial habit, multiple introductions, or the simultaneous arrival of many founders weaken selection for selfing during the bottleneck associated with a dispersal event. These issues highlight the many assumptions that are necessary for Baker's law to hold.