The genes underlying adaptations are becoming known, yet the causes of selection on genes—a key step in the study of the genetics of adaptation—remains uncertain. We address this issue experimentally in a threespine stickleback species pair showing exaggerated divergence in bony defensive armor in association with competition‐driven character displacement. We used semi‐natural ponds to test the role of a native predator in causing divergent evolution of armor and two known underlying genes. Predator presence/absence altered selection on dorsal spines and allele frequencies at the Msx2a gene across a generation. Evolutionary trajectories of alleles at a second gene, Pitx1, and the pelvic spine trait it controls, were more variable. Our experiment demonstrates how manipulation of putative selective agents helps to identify causes of evolutionary divergence at key genes, rule out phenotypic plasticity as a sole determinant of phenotypic differences, and eliminate reliance on fitness surrogates. Divergence of predation regimes in sympatric stickleback is associated with coevolution in response to resource competition, implying a cascade of biotic interactions driving species divergence. We suggest that as divergence proceeds, an increasing number of biotic interactions generate divergent selection, causing more evolution in turn. In this way, biotic adaptation perpetuates species divergence through time during adaptive radiation in an expanding number of traits and genes.
The genes underlying the evolution of differences between species are quickly being identified in many species, but the causes of natural selection on these genes are largely unknown. We manipulated the presence of a native predator to test the effect of contrasting predation regimes on the evolution of defensive armor and at two key genes underlying armor variation between two coexisting stickleback species. The predator altered the pattern of natural selection on armor and on two underlying loci, leading to divergent evolutionary trajectories in the next generation. The study shows how direct manipulation can yield insights into the mechanisms of evolution, in this case the role of a biotic interaction. Beyond illuminating the relationships between natural selection on phenotype and genotype this experiment also demonstrates how evolution in habitat use, driven by competition, can lead to changes in the strength of other species interactions that ultimately drive further divergence. This is an empirical example of how trophic complexity can facilitate diversification and suggests that diverse and evolving biotic interactions could be a core component that sustains species divergence and speciation in adaptive radiations.
Evolutionary response of armor (A) and allele frequencies at two underlying genes (B). Dots above the line indicate more armor (longer spines or higher frequency of the limnetic alleles linked to longer spines) in the treatment ponds relative to control ponds. Black dots indicate overall mean with standard error. Individual colored dots represent pond pairs (F1 families).