This study was designed to find evidence for the causal factors responsible
for the length of the brooding cycle (brooding, preening, feeding, explori
ng, dustbathing, and then brooding again) of broody hens with chicks. In pa
rt I, we examined brooding and non-brooding behaviors during the first 8 da
ys after the chicks hatched. Eight groups of birds were monitored continuou
sly during the 14-h daily light period with a time-lapse video recorder. Th
e amount of time spent brooding per cycle declined over days, while the amo
unt of time spent in other activities increased. Nonetheless, the total cyc
le length remained approximately constant at about half an hour between day
s 3 and 7. Correlations between brooding and non-brooding within cycles wer
e small, but positive, as were correlations between successive brooding cyc
les. Frequency distributions of cycle length showed large variability both
within and between hens. Autocorrelation functions provided evidence of sig
nificant periodicity in brooding behavior, but the periods also varied grea
tly both from day to day and between hens. These results all suggested that
an ultradian clock was not responsible for the observed constancy in cycle
length over days. In part II, 18 broody hens with chicks were deprived of
dust from day 6 to day 8. Deprivation of dust per se had only small effects
on the behavior of the hens. Reintroduction of dust on day 8, however, led
to immediate, sustained dustbathing by all hens, which! affected the lengt
h of the brooding cycle and the sequence of behaviors within a cycle. These
results were interpreted as a reflection of competition between behavior s
ystems, although the existence of an ultradian clock could not be ruled out
.