Egg-laying organisms should lay, in a reproductive bout, the number of eggs that maximizes fitness. Lack argued 50 years ago that clutch size for most birds is limited by the amount of food parents can pro,ide for their offspring(1). Clutch sizes, however, are of much debate(2-4). Here I propose and test a new explanation for this pattern that is based on evidence that conspecific brood parasitism is widespread in birds(5-7), specifically when females with their own nests also parasitize conspecific birds(8-13). A graphical model of clutch size shows that the trade-offs a brood parasite faces when allocating eggs to her own nest or to nests of other conspecific females can favour a reduction in the parasite＇s own clutch size. This prediction is supported by a field study of American coots (Fulica americana). Moreover, the cost of receiving parasitic eggs also favours a reduction in clutch size for hosts, introducing a ＇game＇(14) element to clutch size when parasitic females are themselves parasitized. These results indicate that conspecific brood parasitism should no longer be ignored as a force in clutch-size evolution.