One major challenge in biology is to understand how microevolutionary processes generate macroevolutionary patterns. My research group studies plants and plant-feeding insects, as phylogenetic studies indicate that speciation in these groups is driven by divergent specialization in traits of importance for the species interaction. Here, we focus on the evolution of floral scent, a complex trait of crucial importance for plant-insect interactions. After several years of eco-evolutionary studies in a plant model system, Arabis alpina, we are appling a combination of field studies, crossing experiments, chemical ecology and genomics to investigate how floral scent is diversified in natural populations. In particular, we test how a striking floral scent variation among populations relates to geographic variation in their interaction with insects. Initially, the available A. alpina reference genome allows us to couple ecological surveys, field transplant experiments and genomic studies aimed at reconstructing the phylogeographic background to floral scent variation and identifying candidate genes for scent compounds through re-sequencing of natural populations and lab-generated F2-crosses. We have started this work, and have recently received funding from VR to continue these studies and to add also gene expression analyses of genes involved in the floral scent production. Collectively, these studies address several unresolved questions about how complex traits like floral scent are molded and diversified through selection from each local ecological network of interacting insects.