Modeling the evolutionary and ecological consequences of selection and adaptation in heterogeneous environments

The evolution of adaptations of natural populations in heterogeneous environments poses many challenges. According to the conventional paradigm of adaptive evolution, most traits of existing organisms should be well adapted to the whole wide range of conditions to which they are exposed in their natural environments. However, many traits in most organisms appear to be poorly adapted at many specific conditions, while many other traits appear to be well adapted to conditions that occur only very rarely or not at all in their natural environment. At the same time, many apparently available resources are exploited very little or not at all by the existing species in ecological communities. To help account for these discrepancies and contradictions, I model the effects of the frequency distribution of the contributions to reproductive output of different conditions, habitats, and events, and the major effects of the correlations between the selection forces at different conditions on the selection and adaptation of each of the characteristic traits of species in naturally heterogeneous environments. The force of selection at any one condition i is proportional to the reproductive output fraction F-i of the exposed population. F-i is the product of the population fraction H-i and the individual reproductive output R-i. Selection at any one locus or trait A in a large mixed population establishes an allele or trait A* that maximizes the mean fitness W-Lambda(A) = Sigma(i)[(W-i(A)F-i], weighted over all conditions according to the product of their fitness functions and relative reproductive outputs. Consequently, the adaptation of any trait to any one rare specific condition depends very little on the specific selection coefficient! Specific adaptations depend almost only on the correlations between the fitness of the overall optimal trait A* and the fitness of the specific optimal trait A(i)* at the specific condition. Thus, species are expected to be better adapted and to exploit mostly their more abundant and more positively correlated resources. The consequences for ecological communities are that superior competitive species are not likely to evolve to exclude coexisting rare species that exploit rare negatively correlated habitats. Also, the exploitation by any one species of a range of resources with negatively correlated fitness functions is genetically and ecologically unstable. Such resources may remain unexploited by existing species. Opening a large new habitat may select against adaptation for some previously utilized resources with negatively correlated fitness functions, but may select for adaptation for some previously unutilized rare resources or habitats with positively correlated fitness functions. The opposite effects are predicted for the elimination of some large utilized resource.