There is much evidence that herbivory is a key innovation for the tremendous success of insect. In this thesis I have investigated different aspects of host plant utilization and phenotypic plasticity using the polyphagous comma butterfly, Polygonia c-album. Even though external conditions affect a phenotypic plastic response, the outcome is often influenced by a genetic background which may differ among populations. In Paper I we suspected the genetic background to seasonal polymorphism to be X-linked. However, results from interspecific hybridization between two populations suggested that diapause response is instead inherited in a mainly autosomally additive fashion, with a possible influence of sexual antagonism on males. In Paper II we showed that female oviposition preference is not a plastic response influenced by larval experience, but has a genetic background coupled to host plant suitability. Further, there is a strong individual correlation between larval host plant acceptance and female host plant specificity (Paper III). We believe this to be a larval feed-back genetically linked to female host specificity: offspring to ‘choosy’ specialist mothers benefit by remaining on the original host while offspring to less discriminating generalist mothers should risk inspecting the surroundings, thus compensating for potential poor female choice. In the larval mid-gut, genes are differentially expressed depending on host plant diet (Paper IV). Therefore, we expected to find fitness consequences of host plant switch. However, although growth rate was affected in a few treatments, larvae were generally surprisingly good at adjusting to new diets (Paper V). To conclude, host plant choice in both female and larval life stage is connected to performance. Combined with increased understanding about the plastic response to diet intake and seasonal polymorphism we have gained further insights into the processes of local adaptations and speciation in the Lepidoptera. / At the time of the doctoral defense, the following papers were unpublished and had a status as follows: Paper 3: Submitted Manuscript; Paper 5: Manuscript