Understanding the selective pressures shaping the number of offspring per breeding event is a key area in the study of life–history strategies. However, in species with parental care, costs incurred in offspring production, rather than rearing, have been largely ignored in both theoretical and empirical studies until relatively recently. Furthermore, the few experimental studies that have manipulated production costs have not yet teased apart effects that operate via the parental phenotype from effects on the quality of the resulting young. To examine whether increased egg production influences parental brood rearing capacity independently of effects operating via egg quality, we experimentally increased egg production in gulls and then examined their capacity to rear a control clutch. We found that the capacity of parents to rear the control brood was substantially reduced solely as a consequence of having themselves produced one extra egg. The paradox that, in many species, parents apparently aim for fewer young per breeding event than the experimentally and theoretically demonstrated optimum, has partly arisen from the failure to take into account the constraints imposed by production costs.