In avian species where only one parent incubates, that parent must divide i
ts time between the mutually exclusive activities of incubation and foragin
g in such a way as to maintain both body condition and clutch temperature w
ithin certain limits. In a uniparental incubator, the starling, we experime
ntally reduced the rate at which unattended clutches of eggs cooled down an
d monitored the resulting changes in the parent's incubation strategy. Oppo
site to the predictions of standard models of time allocation during incuba
tion, parents spent a much greater percentage of each 24-h period incubatin
g when the rate of clutch cooling was reduced. Incubation bouts lasted sign
ificantly longer on experimental nests than on control nests, both during t
he daytime and overnight. Mean foraging bout duration did not differ betwee
n the two groups of nests. These results are consistent with the hypotheses
that parental foraging success cues the end of a foraging bout, and that p
arental energy level cues the end of an incubation bout. However, most prev
ious studies suggest that parents spend less time incubating when the rate
of clutch cooling is slow. If parental energy level cues departure, these r
esults can be explained only if the amount of time available for incubation
is constrained in these cases by the time a parent must spend foraging in
order to maintain body condition. Such parents should take more time away f
rom incubation when the unattended clutch cools slowly, as this is when the
cost of being absent is minimized. (C) 1999 The Association for the Study
of Animal Behaviour.