We conducted a manipulative experiment to investigate how raising chicks af
fects the body condition (body mass scaled by body size) and body compositi
on (percent fat vs. lean mass) of black-legged kittiwakes (Rissa tridactyla
). For 4 consecutive years (1991-1994) we removed eggs from randomly select
ed nests and then compared adults raising chicks with adults that had their
eggs removed. At the end of the chick-rearing period, adults raising chick
s were significantly lighter for their size than adults that had their eggs
removed. Adults raising chicks also had a significantly lower percent body
fat (by 28%) than adults from manipulated nests. The difference in percent
body fat between the two groups was apparent at all levels of condition, s
uggesting that adults that are raising chicks apportion their reserves diff
erently than adults that are working only to meet their own metabolic needs
. End-of-season body condition of adults from manipulated and unmanipulated
nests varied signi ficantly among 5 years of study, and appeared to reflec
t differences in local foraging conditions. In all years, females were in w
orse condition than males at the end of the breeding season. This sex-speci
fic condition difference did not, however, appear to indicate a greater sho
rt-term reproductive cost among females. Females were lighter for their siz
e than males in both the manipulated and unmanipulated groups. Our results
suggest that adult kittiwakes compromise their body condition and body comp
osition during chick rearing to increase the likelihood of successfully fle
dging young, even though such adjustments may decrease their own post-repro
ductive survival probabilities. Prior to estimating the body composition of
the experimental birds, we evaluated the usefulness of several noninvasive
techniques for predicting fat mass in kittiwakes. We used cross-validation
techniques to compare multiple regression models that included total body
electrical conductivity (TOBEC), total body water (TBW), and morphometric m
easurements as independent variables. The most parsimonious model for predi
cting fat mass was based on TOBEC and mass measurements. TBW and morphometr
ics were of little utility in predicting fat mass in kittiwakes. Previous s
tudies that have evaluated the usefulness of TOBEC as a predictor of fat ma
ss have shown mixed results. We suggest that the size of the experimental s
ubject relative to the size of the TOBEC measurement chamber may affect the
accuracy of this technique.