Quantitative-genetic approaches have offered significant insights into phenotypic evolution. However, quantitative-genetic analyses fail to provide information about the evolutionary relevance of specific loci. One complex and ecologically relevant trait for plants is their resistance to herbivory because natural enemies can impose significant damage. To illustrate the insights of combined molecular and ecological research, we present the results of a field study mapping quantitative trait loci (QTL) for resistance and tolerance to natural rabbit herbivory in the genetic model, Arabidopsis thaliana. Replicates of the Ler × Col recombinant inbred lines were planted into field sites simulating natural autumn and spring seasonal germination cohorts. Shortly after flowering, herbivores removed the main flowering inflorescence (apical meristem). We found several main-effect QTL for resistance within each seasonal cohort and significant QTL–season interactions, demonstrating that the loci underlying resistance to a single herbivore differ across seasonal environments. The presence of QTL × environment also shows that variation at specific loci is only available to selection in some environments. Despite significant among-line variance components, no QTL for tolerance were detected. The combined results of the quantitative-genetic and QTL analyses demonstrate that many loci of small effect underlie tolerance to damage by rabbits, and counter the hypothesis of locus-specific tradeoffs between resistance and tolerance. The results also provide insights as to the locus-specific nature of evolutionary constraints, i.e. some loci influence flowering time and resistance in both seasonal cohorts. Our results show how linking molecular-genetic tools with field studies in ecologically relevant settings can clarify the role of specific loci in the evolution of quantitative traits.