Coexistence mechanisms facilitate the establishment and ultimate persistence of multiple species in communities. As such, coexistence mechanisms provide the foundation for species diversity. While we now know a great deal about the conditions necessary for coexistence mechanisms to promote diversity, we are left with the question of whether such conditions – normally thought to be properties of species – are adaptive. Any mechanism for which the conditions required to promote coexistence are maladaptive cannot be expected to be common in nature.
For the storage effect, the conditions necessary for it to promote coexistence are clear: species must exhibit buffered population growth, the environment must be coupled with competition, and species must respond to the environment differently. Temporal patterns of annual plant germination play a critical role in each condition and have been hypothesized to produce a storage effect. We investigate selection on patterns of germination in a single-species annual plant model where germination and seed production respond to a fluctuating environment. We focus specifically on the role of a fluctuating environment on shaping mean germination responses and germination sensitivity to environmental fluctuations.
Results/Conclusions
Similar to previous studies of bet-hedging, we found a negative association between environmental variation and the average germination rate of adaptive phenotypes. Dormancy is advantageous in variable environments because fitness is buffered in years of low yield compared to non-dormant phenotypes. Thus, the same process that selects for dormancy in a variable environment favors buffered population growth, a necessary condition for the storage effect.
We also found that germination patterns that match environmental fluctuations are favored. The direction and magnitude of selection on germination sensitivity are each proportional to 1) the correlation between germination and seed yield environments and 2) the magnitude of seed yield variation. Thus, variable germination fractions are adaptive when seed yield is predicted by environments that cue germination. Further, when resource capture fluctuates, adaptive germination phenotypes exhibited a negative association between germination rate and resource availability.
Our results suggest that the outcomes of natural selection on individual species considered alone are buffered population growth and coupling of environment and competition, both necessary conditions for the storage effect. Further studies in multispecies communities are necessary to evaluate the potential for species to evolve different germination responses to their common environment.