I examined Lack's (1954, 1968) hypothesis that asynchronous hatching is an adaptive response to food shortage during the breeding season by comparing growth and survival of nestlings in asynchronous and artificially created synchronous broods of American Goldfinches (Carduelis tristis). I also examined the effects of seasonal and environmental factors on nestling growth and survival and on hatching asynchrony. Nestlings in asynchronous broods were more likely to diverge in mass and did so at a faster rate than those in synchronous broods. The lightest nestlings of asynchronous broods grew more slowly than their heavier nestmates and than all nestlings in synchronous broods. Brood reduction occurred more frequently in asynchronous broods. Survival and growth rates decreased throughout the breeding season and during inclement weather. Hatching intervals increased during the season. When brood reduction or differential growth among nestmates occurred in asynchronous broods, suggesting that the energy available for growth was limiting, heavy nestlings in asynchronous broods grew nonsignificantly faster than heavy nestlings in concurrent synchronous broods. This trend implies that when insufficient food is delivered to nestlings, asynchronous hatching may provide a slight advantage for older nestlings. Growth rates of all nestlings, however, were greatest in highly synchronous broods. Explanations other than growth of young must be invoked to explain why extreme synchrony is not more common in goldfinches. I discuss constraints on the normal pattern of hatching asynchrony characteristic of this species. The proximate mechanism for differential feeding and brood reduction is discussed.