The resource concentration hypothesis predicts that herbivorous insect density scales positively with plant density because insects are better able to identify, and remain longer in, patches with denser plant resources. While some studies support this hypothesis, others do not. Different explanations have been proposed for this discrepancy, including variation in insect dispersal ability and diet breadth. We test the resource concentration hypothesis using the Karner blue butterfly (Lycaeides melissa samuelis), a specialist that relies on wild blue lupine (Lupinus perennis) as its sole host plant. We extended this hypothesis to test whether Karner blue density also scales positively with nectar plant resources. Our findings did not support the resource concentration hypothesis and demonstrate that the spatial segregation of nectar and host plant resources relative to each other can influence the location and abundance of Karner blues on the landscape. This is because the location of resources relative to each other influences the energy and time butterflies expend for flight activity, and thereby influences resource acquisition. During early summer when first brood Karner blues emerge, nectar and host plants were spatially segregated, and Karner blue density peaked at intermediate densities of nectar and host plants occurring at ratios approximately equal to 1:1. During late summer, we found no significant relationships between second brood Karner blues and nectar plants or host plants when there was no correlation between nectar and host plants. Conservation practitioners of specialist insects with low vagility can strategically manage the distribution of plant resources to minimize insect time and energy expenditure and promote resource acquisition for all of an insect’s life stages.