About 15,000 years earlier, the Northern half of Europe and North America was buried under a few kilometres of ice. Since then, many organisms have colonized and rapidly adapted to the new, vacant habitats. Some, like the threespine stickleback fish, have done so more successfully than others: from the sea, stickleback have adapted to a multitude of lake and stream habitats with a vast array of complex phenotypes and life histories. Previous studies showed that most of these "ecotypes" differ in multiple divergently selected genes throughout the genome. But how are well-adapted ecotypes of one habitat protected from maladaptive gene flow from ecotypes of another, adjacent habitat? According to a From the Cover meta-analysis in this issue of Molecular Ecology (Samuk et al., 2017), low recombination rate regions in the genome offer such protection. While inversions have often been highlighted as an efficient way to maintain linkage disequilibrium among sets of adaptive variants in the face of gene flow, Samuk et al. (2017) show that variation in recombination rate across the genome may perform a similar role in threespine stickleback. With this study, theoretical predictions for the importance of low recombination regions in adaptation are for the first time tested with a highly replicated population genomic data set. The findings from this study have implications for the adaptability of species, speciation and the evolution of genome architecture.