Technical Abstract: Quantifying an invasive species’ negative impacts across its introduced range will be quite challenging if the impacts vary unpredictably from site to site or population to population. Little emphasis, however, has been placed on quantifying such inter-population variation in the impacts of individual invasive species. We studied the response of a native grass (Festuca rubra) to competition from four geographically dispersed invasive plant (Melilotus alba) populations in order to determine if some populations of this invader have greater competitive impacts than others. Despite our greenhouse study’s relatively large number of experimental units, we did not see evidence that per-gram-of-biomass competitive effects varied by invader population. Therefore, in some cases it should be possible to estimate invasive weed impacts with simple competition models that ignore some forms of phenotypic variation, as long as the models control for invader biomass per unit area (i.e. invader yield). Invader yield must be controlled for, because it varies dramatically from population to population and invasive weed impacts intensify with increasing yield.