252 
THE WILSON JOURNAL OF ORNITHOLOGY • Vol 123, No. 2. June 2011 
1,690 m asl), in the central eastern Andes south of 
Apolo, La Paz Department, Bolivia. The area 
contains semi-humid forest fragments, montane 
savannas, and farm pastures. The forest fragments 
are drier than wet forests and have fewer species 
(Foster and Gentry 1991). Local farmers frequent¬ 
ly bum the montane savannas and farm pastures 
(Miranda 2005). Consequently, the vegetation 
consists of a thin cover of grass and other forbs 
mixed with gnarled shrubs and small tress 
including Atchomea spp. (Euphorbiaceae) (Foster 
and Gentry 1991) with dominant Schefflera 
morototoni trees (Araliaceae) scattered throughout 
the study area. The wet season normally occurs 
from October to March, and it is dry between 
April and August. Average annual temperatures 
range from 16 to 20 Q C; October is the hottest 
month with temperatures reaching a maximum of 
28° C (SENAMHI 2006). 
Nest Monitoring. —I found active nests by 
intensively searching (following Martin and 
Geupel 1993) in montane savanna habitat from 
late October 2005 to early March 2006. I 
monitored nests every 2—4 days to ascertain 
length ot nesting periods and nesting success. 
The incubation period was calculated as the 
number ot days between laying of the last egg 
and hatching of that egg (Gill 1995). The nestling 
period was considered the number of days 
between hatching of the last egg and fledging of 
the last nestling. Nests found in the building 
process were followed until completion. 
I observed parental care of birds at nests at 1 to 
4 day intervals, either from 0630-1200 or 1400- 
1830 hrs from a distance of 20 m using 10 X 45 
binoculars. Adults did not appear to be bothered 
by my presence at this distance. 
I collected incubation period data at five nests, 
distinguishing three types of parental behavior: 
incubation, perching in trees in the nesting area, 
and absence from the nesting area. I collected data 
from an additional five nests during the nestling 
period distinguishing the following types of 
parental behavior: ( 1 ) Lotal attendance that 
included parental care at the nest and parental 
care at the nest from a distant tree perch 
(vigilance), (2) parental provisioning of nestlings, 
and (3) foraging trips. Parental care at the nest 
consisted of brooding nestlings or perching near 
the nest (~5-50 cm). Parental care from a distant 
tree perch consisted of perching in trees (includ¬ 
ing the nest tree) within a 10-m radius of the nests, 
which occurred each time a parent arrived with a 
food item or provisioned nestlings. I recorded the 
number of visits and the rate at which parents 
provisioned nestlings (food item/hr/parentI. 1 
recorded each "'foraging trip" as the length of 
lime a parent was absent from the nesting area. 
Observation of nests allowed me to distinguish 
nestling’s traits by their age. 
I estimated nesting success for the incubation 
and the nestling periods separately and combined. 
Nest success was defined as at least one egg 
hatching, and nestling success as at least one 
young fledging (Gill 1995). I observed parental 
activities, checked the contents of nests, and 
searched for fledglings in the nesting area to 
verify nesting activity (Green 2004). I considered 
nests to be (I) active, if parents displayed nearby 
activity, or if individuals were seen within the 
nest; (2) predated, if there was sign of predation 
such as eggshell fragments, nest material scattered 
around the area, or nestlings had disappeared prior 
to the time of potential fledging; (3) deserted, if 
the parents did not attend or stopped attending the 
nest; and (4) other, if climatic conditions ot 
location of nest influenced nesting success. 
Statistical Analyses .—Average time (min/hn 
that each parent spent incubating, providing 
parental care at the nest, providing parental care 
from a tree perch, and provisioning rates as well 
as foraging trips was compared using a paired /• 
test. Repeated measure ANOVA tests were used 
to analyze six nestling-age categories based on 
extent of feather development and nestling size: 
I = 2-5 days, 2 = 6-9 days. 3=10-13 days. 4 = 
14-17 days, 5 = 18-22 days, and 6 = 23-27 days. 
Post-hoc tests were conducted for repeated 
measures with Bonferroni corrections if differ¬ 
ences were significant. 
I checked normal distribution of the data using 
Kolmogorov-Smirnov tests, and data homogene¬ 
ity using the F-max test. The compound symmetry 
assumption was verified with Huynh-Feldt epsilon 
approaches when repeated measures ANOVAs 
were needed. A Wilcoxon test was used for 
comparing differences between males and fe¬ 
males, and a Friedman lest was used tor 
comparing age categories if data violated these 
assumptions. Statistical analyses were conducted 
using SYSTAT 11.0 (2004) at the significance 
level of ot = 0.05. 
The daily probability of survival was calculated 
following Mayfield (1961. 1975). This estimate 
results from the total number of days when nests 
are active (exposure days) divided by the number 
