The physiological cost of reproduction was investigated in a perennial plant species, Sidalcea oregana ssp. spicata. Two different classes of estimates of current reproductive investment were compared, with the goal of identifying the most powerful predictor of future reproductive effort. ''Static'' estimates, those typically used to measure current reproductive investment (i.e., biomass or nutrients allocated to reproduction), neglect the physiological processes that influence actual reproductive expenditure. Conversely, ''dynamic'' estimates that incorporate reproductive photosynthesis or respiration, nectar production, or reproductive nutrient resorption offer a more accurate assessment of current reproductive investment. I assessed the potential for nutrient resorption from senescing reproductive structures to mitigate the cost of reproduction. Significant proportions of nitrogen and phosphorus were resorbed from floral structures, but structures differed in resorption potential and efficiency. Moreover, dynamic estimates of nutrient investment were better predictors of future reproductive effort than were static estimates. Dynamic estimates appear to be closer indicators of realized physiological cost of reproduction. By taking a dynamic perspective of reproductive investment, we may be better equipped to address the potential for evolutionary change in life history and sex allocation in plants.