We examined three hypotheses concerning hatching asynchrony in the Common Tern (Sterna hirundo). Survival of third-hatching "C-chicks" was significantly lower than that of "A-" and "B-chicks" in broods of three. In 2 yr when conditions did not appear favorable, survival was significantly higher in manipulated broods in which chicks hatched synchronously (73%) than in nonmanipulated broods (56%). Chicks in synchronous broods grew significantly faster than C-chicks and at a rate similar to A- and B-chicks. These results were inconsistent with the brood-reduction hypothesis, which predicts that hatching asynchrony will maximize brood success under conditions of food limitation. Chicks hatching from C-eggs grew significantly faster and survived at nonsignificantly higher rates when an older sibling was removed experimentally. These results were consistent with the hypothesis that C-chicks serve as insurance against loss of an older sibling. All three siblings, however, survived in 26% of nonmanipulated broods, which indicates that the sole function of the C-chick was not insurance. Predation on tern eggs was common and was correlated with numbers of migrating Ruddy Turnstones (Arenaria interpres). The percentage of time adults incubated was lower when only one egg had been laid than when two or three eggs were laid, and egg predation was most frequent during this initial stage. Although hatching asynchrony did not maximize chick survival, incubation before laying is completed may maximize overall nest success by protecting eggs from predators.