Covariation and composition of arthropod species across plant genotypes of evening primrose, Oenothera biennis

Abstract

Genetic variation in plants has broad implications for both the ecology and evolution of species interactions. We addressed how a diverse community of arthropod species covary in abundance among plant genotypes of a native herbaceous plant (Oenothera biennis), and if these effects scale‐up to shape the composition, diversity, and total abundance of arthropods over the entire lifetime of plants (two years). In a field experiment, we replicated 14 plant genotypes of O. biennis across five field habitats and studied the arthropod communities that naturally colonized plants. Genetic variation in O. biennis affected the abundance of 45% of the eleven common species in 2002, and 75% of sixteen common species in 2003. We examined the strength of correlations in mean abundance of arthropod species among plant genotypes and found that species responded independently to variation among genotypes in the first year of the study, whereas species formed positively covarying clusters of taxa in the second year (rmean=0.35). The strength of these correlations did not consistently correspond to either taxonomy or functional attributes of the different species. The effects of plant genetic variation on the abundance and covariation of multiple arthropod species was associated with cascading effects on higher levels of community organization, as plant genotype and habitat interacted to affect the species composition, diversity, and total abundance of arthropods in both 2002 and 2003, though the specific effects varied across years. Our results suggest that plants may employ generalized resistance strategies effective against multiple herbivores, but such strategies are unlikely to be effective against entire functional groups of species. Moreover, we show that genotypic variation in plants is an important ecological factor that affects multiple levels of community organization, but the effects of plant genotype vary in both space and time.