544 
THE WILSON JOURNAL OF ORNITHOLOGY • Voi 124. No. 3. September 2012 
descriptions by Sclater and Salvin (1879) and 
Ochoa and Cuervo (1998), although we found one 
group that consistently built its nests with an outer 
layer of an epiphyte. The closest relatives of the 
Red-bellied Grackle are the Oriole Blackbird 
(Gymnomystax mexicanus) and the Velvet-fronted 
Grackle (Lamp rops a r ranagrinus) (Cadena 
et al. 2004, Eaton 2006). The bulky cup nest of 
the Red-bellied Grackle is more similar to the 
nest of the Oriole Blackbird (Jaramillo and Burke 
1999) but not to the hanging basket nest of the 
Velvet-fronted Grackle (Maillard and Herrera 
2007). Egg coloration (but not size) is similar 
among (he three species, all having pale blue or 
greenish eggs with dark brown or purplish spots 
(Skutch 1967. Maillard and Herrera 2007). 
Cooperative breeding likely evolved indepen¬ 
dently in the Icteridae. but is most pervasive in 
South American species of gcackles (Orians el al. 
1977, Jaramillo and Burke 1999. Fraga 2008). The 
Oriole Blackbird, among Red-bellied Grackle 
relatives, is a solitary breeder (Jaramillo and 
Burke 1999) and the Velvet-fronted Grackle has 
apparently been observed breeding cooperatively 
(J. A. Tobias cited by Fraga 2008), but published 
information is inconclusive (Jaramillo and Burke 
1999, Maillard and Herrera 2007). The scarcity of 
breeding ecology data and missing taxa in the 
phylogenv of the Icteridae make unclear if 
occurrence of cooperative breeding in the Red- 
bellied Grackle is phylogenetieally or ecologically 
constrained. Observation of solitary breeding 
individuals of Red-bellied Grackles (Hilly and 
Brown 1986) reported prior to our study, if true, 
would imply either plasticity or temporal or 
geographic variation in breeding strategies. We 
suspect those reports resulted from incomplete 
observations. Recently, helpers attending a nest of 
Red-bellied Grackles have been observed in Huila 
in the southern part of its range (J. J. Leon and S. 
Vargas Troncoso. pers. comm.). 
Nest attentiveness across the incubation period 
was variable, similar to that reported for other 
neotropical montane birds (Martin et al. 2007). 
The weight loss of eggs during incubation 
(11.9%) is surprisingly small in comparison to 
the —18% loss documented for 475 species (Rahn 
and Ar 1974) or the known estimates for other 
neotropical birds <e,g„ Greeney 2006). Hatching 
in our Study was roughly synchronous (i.e., all 
eggs hatched within 24 hrs irrespective of their 
laying time) in two cases, contrary to the 
asynchronous hatch reported by Ochoa and 
Cuervo (1998). who inferred the presence of 
nestlings from observations of adults carrying 
food to the nest. However, feeding the incubating 
individual in the nest is rather common and could 
be mistaken for provisioning of young. 
We do not know if eggs of a single clutch were 
laid by one or more females, or the extent of 
extra-group matings. Nestling development (i.e.. 
during incubation) ranged from 16 to 18 days, 
similar to other neotropical icterids of similar size 
(Skutch 1996. Jaramillo and Burke 1999). Nest¬ 
ling growth rate was slightly lower than in other 
neotropical birds of similar nestling period length 
(—0.33; Martin el al. 2011). This comparatively 
lower growth rate for Red-bellied Grackles may 
be due to the overall smaller body size of species 
included in that analysis as slower growth is 
expected for larger species of comparable nestling 
duration (Case 1978, Starck and Ricklefs 1998). 
Juveniles in at least two groups stayed up to 
3 years in their natal territory and delayed 
dispersal to help their group in provisioning 
nestlings. There are a number of ecological and 
fitness benefits of delayed dispersal in juveniles 
We recorded the first breeding attempt of a 5 year- 
old female, banded as a nestling in 2006 during 
more recent observations (2010-2011. unpubl. 
data). It is possible that sexual maturation of Red- 
bellied Grackles is achieved after several years. 
We have not found any individual banded as 
young that have reached the size of dominant 
adult males. All fledglings captured during the 
first years of the study exhibited a size similar 
to that of adult females. Thus, it is difficult to 
ascertain whether there is a pattern in failure to 
disperse w ith respect to gender of fledglings. 
It remains to be confirmed if breeding i^ 
concentrated along the relatively short stretch ot 
habitat along the river watershed, where a 
maximum of seven groups was breeding at any 
given time. Group territories were maintained 
year after year (unpubl. data); juveniles, by 
remaining with their group, could eventually 
occupy their natal territory as their reproductive 
territory in following generations. 
ACKNOWLEDGMENTS 
This project was largely funded by our families ant >iut 
personal funds. We thank Idea Wild, Optics for the Tropics, 
and G. J. Colorado for providing field equipment. Plant 
material was identified by F. J. Roldan (Herbarium 
Universidad de Aniioquiu. Medellin. Colombia). " c 
appreciate the effort of all the people that helped us a* 
