The seasonal decline of avian clutch size may result from the conflict between the advantage of early breeding (greater offspring value) and the advantage of a delay in lay date (improved body condition and hence clutch size). We tested predictions of a condition-dependent individual optimization model based on this trade-off (Rowe et al. 1994) in a long-distance migrant, the greater snow goose (Chen caerulescens atlantica), using data on condition, migration, and reproductive decisions of individuals. We closely tracked radio-marked females at their main spring staging area and on their breeding grounds. Our results were consistent with predictions of the model. Early-arriving females had a longer prelaying period and initiated their nests earlier than late arrivals. After controlling statistically for arrival date, we determined that females with high premigration condition had an earlier lay date than those in low condition. After controlling for the seasonal decline (i.e., lay date), we observed that clutch size was not related to premigration condition. Moreover, we took advantage of an unplanned manipulation of the prebreeding condition that occurred during our long-term study. We found that a reduction in condition caused a delay in lay date. However, after controlling for the seasonal decline, it did not affect clutch size. Our study indicates that geese simultaneously adjust their lay date and clutch size according to their premigration condition and migratory behavior as predicted by the condition-dependent optimization model.