Evolution with both exploitative and interference competition

When interference is one of the components of a competitive interaction, the addi­tion (or elimination) of a competitor may have a wide variety of effects, depending (inter alia) on the magnitudes of inter- and intraspecific components of interference and what other parameters are affected by the interference.

In the simplest case, interference is entirely intraspecific, and it only affects per capita mortality rate. Under this scenario, adding a second consumer species (also with intraspecific interference) to a system with a single one usually reduces the abundance of the original consumer. The lower abundance reduces the amount of interference because that interference is purely intraspecific. This lowers each consumer’s mortality and therefore reduces the resource intake required for zero pop­ulation growth. In most cases, those resources consumed at the highest rates will be most affected by sympatry of the consumers, and both consumers will evolve some degree of reduction in their relative consumption of these resources. Assuming an identical trade-off in the relative consumption rates of different resource types, the outcome in a 2-consumer-2-resource model is parallel shifts in the C_ij values of both consumers (Abrams 1986a). This qualitative outcome is preserved when the trade-off relationships are not identical, but are sufficiently similar. In this case, the ‘preferred’ resource of both consumers decreases in relative abundance when the two are in sym- patry, so both consumers shift to have somewhat lower consumption of this resource and higher consumption of the resource initially characterized by a lower Cj (Abrams 1986a).

If the interference affecting mortality is purely interspecific, a variety of out­comes are possible. One of these is the outcome of alternative exclusion, originally described by Lotka and Volterra, which would preclude coevolution. If coexistence is still possible, the resource abundances at equilibrium are normally increased in sympatry relative to allopatry. This is again likely to cause parallel shifts in relative consumption rates in the 2-consumer-2-resource case, but in the opposite direction to those produced with intraspecific interference. Interspecific interference is seldom symmetrical (Schoener 1983), so it is possible for very large effects on the abundance of the species that suffers greater interference when the two consumers are sympatric.

In natural systems, interference often has its largest effects on capture rate param­eters; this occurs when individuals steal food items or prevent foraging by other individuals. This type of interference may have different effects on the capture rates of different resources. Probably because of this range of mechanisms, capture rate interference has seldom been included in models of competition generally, and I am unaware of its inclusion in any models of competitive coevolution.

Most empirical work has found that both inter- and intraspecific competition is usually asymmetric (Schoener 1983). If there is asymmetric interference between species, it is tempting to assume that the species that is dominant in interactions between individuals is more likely to exclude the subordinate species. However, greater ability in contests usually comes at a cost to some other component of fitness. It is quite possible for a species with low investment in a trait determining ability in competitive contests to replace a species with a high investment (Matsuda and Abrams 1994). The evolution of contest competition has mostly been studied using models of within-species interactions (Abrams and Matsuda 1994; Kisdi 1999). High investment in contest traits is likely to cause a reduction in population size relative to similar scenarios in which such competition is absent. It is also possible to have temporal cycles in the mean competitive ability (Abrams and Matsuda 1994).

11.7