18 
THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 123, No. 1, March 2011 
FIG. 1. Parameter estimates and 95% confidence intervals of apparent survival (A) and detection (recapture/ 
reobservation) probability (B) of different age and male or female classes of Chestnut-backed Antbird in northern Costa 
Rica. Parameter estimates were derived using AICc weighted model averaging in Program MARK. 
Estimates of cp were similar for all adults, but 
estimates of (p for juveniles were not informative 
due to large 95% confidence intervals (Fig. 1A). 
Detection probability differed markedly between 
males and females with females being detected 
less frequently once marked (Fig. IB). 
Territory Dynamics. —Seventy-two banded in¬ 
dividuals (51 males, 21 females) were followed 
for at least 1 year within the 300-ha focal plot for 
a total of 160 territory-years (Table 2). False 
disappearance rate of males and females differed 
(2 and 6 instances for males and females, 
respectively), and we adjusted overall turnover 
rates. Turnover was similar between males and 
females, and was dominated by disappearances 
(86%) as opposed to switching (14%; Table 3). 
Turnover was slightly higher for juveniles. We 
found no evidence of differential switching rate 
between males and females ( X 2 = 0.115, df = 1, 
P = 0.73), but our small sample size of switching 
resulted in low statistical power. 
We documented nine cases of territory switch¬ 
ing (4 females, 4 males; one male switched twice). 
The mean (± SE) distance moved was 372 ± 
124 m (range = 145-840), and did not differ 
between males and females (ages pooled, t = 
0.466, df = 7, P = 0.655; x = 334 and 420 m for 
males and females, respectively). Juveniles tended 
to move greater distances (all pooled; t = 2.182, 
df = 7, P = 0.065; x = 590 and 263 m for all 
juveniles and all adults, respectively). Five of the 
nine cases involved switching to adjacent territo¬ 
ries; the greatest distance moved was by a juvenile 
female. Results were similar when analyzed in 
terms of territory-widths (—163 m). Genetic 
parentage analysis revealed that none of the 
juveniles in our analyses was the offspring of 
the social mate (Woltmann 2010); we detected no 
instances ol juveniles inheriting the natal territory. 
Undetected switches involving longer distances 
outside the study area likely occurred, but our 
focal plot size (300 ha) and dimensions (the most 
distant territories were >2 km apart) suggest that 
we could have detected greater movements than 
the mean distance of <400 m. 
Four occupied territories became vacant over 
