Current climate models predict a shift to warmer, drier conditions in the southwestern US. While major shifts in plant distribution are expected to follow these climate changes, interactions among species and intraspecific genetic variation rarely have been incorporated into models of future plant distributions. We examined the drought-related mortality of pinyon pine (Pinus edulis) in northern Arizona focusing on trees that showed genetically-based resistance or susceptibility to a nonlethal herbivore, the shoot-boring moth, Dioryctria albovittella. Because moth resistant trees have outperformed susceptible trees during 20 years of study, and herbivory has been shown to increase drought related mortality, we expected higher mortality rates in susceptible trees. However, our field observations and greenhouse experiments showed several unexpected patterns relevant to understanding the consequences of climate change: (1) The mortality of adult P. edulis resistant to the moth was three times higher than the mortality of trees susceptible to the moth. (2) Over a few years, differential mortality caused a shift in stand structure from resistant dominated to equality (3 : 1 resistant : susceptible to 1 : 1). (3) Adult moth resistant trees suffered significantly greater water stress than adult moth susceptible trees, suggesting that variation among the two groups in drought tolerance may be a mechanism for differential mortality. (4) When grown under drought conditions in the greenhouse, seedlings from resistant mothers died sooner than seedlings from susceptible mothers. These data support the hypothesis that drought can act as an agent of balancing selection and that drought resistance is a heritable trait. Taken together, our findings suggest that genetic variation in a population can be an important factor in determining its response to future climate change, and argue for the inclusion of genetics into models developed to understand the consequences of climate change.