More than a half century ago, the British ornithologist David Lack suggeste
d that parent birds may use brood reduction to track uncertain food, a proc
ess facilitated by the asynchronous hatching of their young. Lack sketched
the logic of asymmetric sibling rivalry: the phenotypic handicap imposed up
on last-hatched marginal offspring renders their growth and survival condit
ional upon uncertain ecological conditions while buffering first-hatched co
re offspring from the inimical effects of overcrowding during periods of st
ringency. Though subjected to numerous indirect tests in short-term studies
, the central prediction of Lack's hypothesis - that parents use marginal o
ffspring to track unpredictable brood-rearing conditions and thus achieve a
secondary adjustment of clutch size - has never been tested directly. Here
we present the results of a 7-year study of marsh-nesting red-winged black
birds (Agelaius phoeniceus) showing that (1) brood size tracks interannual
variability in growth and survival of nestlings, (2) the growth and mortali
ty of marginal but not core offspring is contingent upon stochastic environ
mental conditions (mean air temperature) during brood rearing, (3) the mort
ality of marginal but not core offspring is strongly affected by developmen
tal uncertainty in the form of both experimental and natural alterations of
brood size, (4) the phenotypic handicap of hatching asynchrony buffers cor
e offspring from poor growth conditions, but (5) its effects upon marginal
nestlings are reversible when growth conditions are favourable and especial
ly when brood size is reduced either experimentally or via hatching failure
. The presence of marginal offspring ensures that blackbird parents are not
left with a too small brood when brood-rearing conditions are favourable.
Parents create two castes of progeny: marginal offspring that are strongly
affected by both ecological and developmental stochasticity, and core offsp
ring that are not.