Nest predation often is reported to be the major cause of nest failure in birds and by extension the predominant influence on annual fecundity. However, other factors such as the ability of some individuals to renest after a failed nest or to lay additional clutches after successful ones, both of which may in turn be influenced by food availability, can also contribute to individual fecundity. We evaluated the relative importance of nest predation, multiple brooding, renesting, and food availability on mean annual fecundity of a migratory songbird population, using AICc from multiple regression models. In a 16-yr study of black-throated blue warblers Dendroica caerulescens, these four variables combined explained 87% of the annual variance in fecundity. Individually, however, food abundance explained more of the variability (35%) than did predation (29%), double-brooding (19%), or renesting (15%). These results suggest that measuring food availability may be essential to understanding annual differences in reproductive output. They also demonstrate that failing to account for the contribution of food availability, renesting, and multiple brooding in studies of avian demography can result in serious underestimates of mean annual fecundity, potentially biasing calculations of population growth rate, source-sink dynamics, habitat quality, or population viability.