Coevolution is the reciprocal evolution of the genomes of two species over time. Thompson (2005) proposed that geographic variation in the interaction of species will produce a geographic mosaic of coevolution where space can be used as a proxy for time to study coevolution. Geographic variation in among interacting species is due to a Species 1 Genotype x Species 2 Genotype x Environment (GxGxE) interaction. To measure the G x G forces that produce coevolution it is necessary to account for the environmental effects. Nuismer and Gandon (2008) developed a model where these GxG interactions could be measured using a double reciprocal transplant among interacting species at three sites. We used this model to test for the coevolution of tall goldenrod, Solidago altissima and the gall-inducing fly Eurosta solidaginis. We planted gardens of plants from three sites Duluth, a forest site, Moorhead, a prairie site, and Cedar Creek on the prairie/forest border at each of those sites. We then caged flies on the plants to create all possible combination of flies from the three sites and plants from the three sites at each garden site and measure fly survival, gall growth, and mortality due to parasitism
Results/Conclusions
Fly survival was significantly influenced by the site where the experiment was conducted, the origin of the plant, the origin of the fly, the plant origin x site interaction, fly origin x site interaction, and the fly origin x plant origin. The strong interaction of fly origin and plant origin indicated that there was a strong coevolutionary interaction with flies from both the prairie and forest biomes doing much better on their local plants. Gall size and shape which strongly influences the vulnerability of the fly to natural enemies was also strongly influenced by the interaction of fly origin and plant genotype, but only weakly influenced by environmental effects. We conclude that there is a strong geographic mosaic of coevolution among plants, flies, and natural enemies.