JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.. The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to International Journal of Plant Sciences.Editor: Susan J. Mazer Premise of research. Genetic variation for ecologically important traits is necessary for populations to adapt to environmental change. Many authors have called for a greater emphasis on directly measuring quantitative genetic variation in rare species, which are expected to have reduced amounts of genetic variation due to genetic drift in small populations. The extent of among-population differentiation for quantitative traits may also help to evaluate the likelihood that genetic rescue/translocation will be a successful conservation strategy. Despite these merits, relatively few studies measure quantitative genetic variation for ecologically important traits as a function of population size.Methodology. Sixteen populations of the endangered plant Hypericum cumulicola were sampled, capitalizing on previous work that has estimated relative effective population sizes and demonstrated minimal migration between populations. This context allows more direct inference about the role of drift in small populations on quantitative genetic variation, the focus of this study. Using controlled pollinations and a greenhouse common garden, quantitative genetic variation within populations and differentiation among populations were estimated for six putatively ecologically important traits.Pivotal results. There were few significant estimates of genetic variation for most traits irrespective of population size. There was a positive correlation between population size and genetic variation for antherstigma distance, a floral trait expected to influence the degree of self-fertilization. There was also limited differentiation among populations for all traits.Conclusions. Limited genetic variation for ecologically important traits in these populations could slow or limit adaptive responses to future environmental change, possibly increasing extinction risk. The smallest populations will be particularly sensitive to environmental/habitat changes that result in reduced pollinator visitation. Limited phenotypic differentiation, combined with previous evidence of strong heterosis in crosses between small populations, suggests little risk of outbreeding depression if genetic rescue efforts become necessary to preserve this species.