540 
THE WILSON JOURNAL OF ORNITHOLOGY • Vol 124. No. 3. September 2012 
brownish-black plumage, and dark brown to 
grayish-yellow irides. Adults were classified as 
males or females in the field from behavioral 
differences and the larger size of males (135 vs. 
97 g in body mass; AMC, unpubl. data); (~30 vs. 
27 cm total length) (Hilty and Brown 1986). 
However, gender of first-year birds could not be 
identified in the field. We marked captured birds 
with two plastic color bands to identify individ¬ 
uals, one colored band to recognize breeding 
group and the other, individuals in the group. All 
fledglings front five nests were banded to examine 
if first-year birds act as helpers in the next 
breeding season and if they remain in natal 
groups. We used a general linear mixed model 
(GLMM) to examine the variation in clutch size 
as a function of group size where we specified 
group identity as a random-effect variable. We 
had 18 observations from six groups. We used the 
glmmPQL function implemented in the R package 
MASS to lit the GLMM (Venables and Ripley 
2002). This function used pseudo- and penalized 
quasi-likelihood to estimate the parameters of the 
model (Bolker et al. 2009), Values are presented 
as means ± SD. 
RESULTS 
Most breeding events occurred from March to 
June (it = 35). All seven breeding groups located 
every year exhibited a cooperative breeding 
system. Groups varied in size from three to seven 
individuals in each breeding event (5 ± 1.3, // = 
27), not including nestlings, and also varied by 
gender and age class composition (e.g.. 1-2 adult 
males. 1-3 adult females, and 1—3 first-year birds 
or subadults). We observed two adult males (by 
size) in one breeding event; one was much more 
active vocalizing from a high perch and delivering 
vocalizations that have solely been observed for 
dominant males, suggesting primary and second¬ 
ary males in that group. Individual colored bands 
placed on 13 nestlings in live nests indicated 
breeding groups raise the helpers of the following 
years, and that groups have high fidelity to 
breeding territories across years. Nine of 13 
marked fledglings remained in their group and 
spatial territory for at least 2 consecutive years. 
We did not re-encounter the remaining four 
fledglings and they may have been depredated 
or dispersed to another area. 
Nests. We found 38 nests attended by seven 
groups. Nests averaged 4.6 ± 2.1 m above ground 
(it - 34). but four were at a height of 15 m. Most 
were on bifurcations of lateral branches of a 
variety of trees or tall shrubs of 11 species in 
Clusiaceae, Melastomataceae. Monimiaccac, Ru- 
taeeae. Solanaceae, and Tiliaceae. Introduced 
trees including Eucalyptus camaldulesis (Mynu- 
ceae). Finns pa tula (Pinaceae). and Cupresm 
lusitanica (Cuprcssaceae) were also used for 
nesting. Nests were in trees and tall shrubs at 
the edge of forest fragments (;i = 16), or in 
isolated trees in pastures at a mean distance of 
12.5 m (range = 5-81 m, n = 22) from the nearest 
forest edge. The mean distance of nests to the 
river edge was 13.8 m (range = 0-66 m. n = 3U 
Nests of Red-bellied Gracklcs were ovoid 
open-cup structures of two layers of sticks and 
roots that had different diameters (internal layer - 
0.5 t 0.2 mm. n - 85 items; external layer: 1.0 1 
0.3 mm. n = 75 items; n = 3 nests), Nests went 
lined with dry leaves. There appeared to be little 
variation in materials used, although one group 
added an extra external layer of an epiphyte 
(Tillaiidsia usneoides , Brotneliaceae). Mean mea¬ 
surements were based on a subset of eight nests: 
inside cup depth = 77 ± 14 mm, outside cup 
depth = 155 ± 22 mm. inside opening = 103 ± 1 
X 112 ± 8 mm, outside opening = 152 ± 15 8 
182 ± 24 mm. We observed construction of only 
one nest, and this nest was completed between 4 
and 8 days; we only observed one adult female 
building it. Nest materials were collected in the 
surrounding area (<100 nr) by the female and 
other members of the group, which also supplied 
her with food. We did not quantify the rate of 
delivery of nest materials or food items to the 
female. The male did not participate actively in 
construction of this nest. 
Eggs ami Clutch Size. —Eggs were elliptical in 
shape and light blue in color with purplish brown 
spots and stripes that w'ere denser towards the 
w ider end (Fig. 2). Fresh egg mass averaged 7J) 
± 0.3 g (n = 12, from 4 nests), and their external 
measurements averaged 28.8 ± 1.0 mm X 21.0 - 
0.4 mm (/? = 35. from 11 nests ). Females laid one 
egg per day. and clutch size averaged 3.1 - O ' 
(range - 2-4. from 21 nests). Breeding group size 
and clutch size were positively correlated (slope 
= 0.25 ± 0.09 (SE). P = 0.03). 
Incubation Period. —Incubation periods lasted 
16 ± 1 days (15. 15, 17 days, respectively; n = 31- 
Eggs lost up to 11 .9% of their weight during this 
period at a rate of 0.07 ± 0.01 g per day (R : 
0.92. n = 4; Fig. 3). Nest attentiveness averaged 
69 ± 18% during incubation of the total 
