Dispersal in a Population of White-Browed Sparrow Weavers

Abstract

Dispersal behavior of a color-marked population of the communal White-browed Sparrow Weaver (Plocepasser mahali) was studied over a threeyear period in Zambia, Africa. Birds invaded communal groups having a higher occurrence of preferred feeding cover more than groups with low occurrence. Successful invasions were negatively correlated with the size of the group, perhaps because of group territorial defense. Movements that resulted in a bird staying in a group for at least six months or that resulted in a breeding attempt during this period were more frequent in smaller groups than larger groups. These movements, combined with female helpers who became breeders at their natal site, were used to approximate birth-to-breeding distances. The mean distance of this sample was 205 m, with a skewed distribution toward short distances. Excluding cases in which birds bred in the natal group, there was not significant difference in distances traveled by males and females. Cohorts from a group periodically invaded a neighboring group, frequently resulting in the disappearance of the invaded group's breeding female. Her replacement was by a member of the invading cohort. Individuals who dispersed relatively far appeared unable to breed as quickly as birds dispersing a short distance. Monitoring resources within the neighborhood while remaining at the natal site may be an adaptive habit for making a sedentary life style compatible with a successful dispersal phase. Studies of communal species suggest that one sex tends to breed in its parental area while the other disperses to neighboring areas (Woolfenden and Fitzpatrick 1978, Ligon and Ligon 1978). The adaptive significance of these traits is unclear, although the importance of territorial acquisition has been proposed (Woolfenden and Fitzpatrick 1978). While the fitness value of inheriting a profitable resource may favor residency at the natal site, it is also adaptive to seek new resources if breeding opportunities at that site are scarce. Acquiring such resources is the ultimate selective advantage of dispersal, and the mode of dispersal is likely to be affected by social behavior in communal species. Dispersal modified by the effects of kinship has been documented (Woolfenden and Fitzpatrick 1978, Ligon and Ligon 1978). How an individual disperses may therefore depend on the size of its communal unit, since social relations and opportunities for social behavior should increase with group size. Another determinant of dispersal may be the relative quality of resources among communal territories, because individuals should be more likely to disperse when their territories have poorer resources than elsewhere. Here I examine how resource characteristics and social behavior influence patterns of dispersal in the communal White-browed Sparrow Weaver (Plocepasser mahali). In particular, I ask: 1) How does resource quality and communal unit size correlate with the frequency of movements among communal groups? 2) What factors influence the frequency distribution of dispersal distances for movements that result in breeding attempts? 3) In what way does social behavior influence an individual's chances of successfully invading and breeding in another group? I also consider the possible significance of these results to population structure in this species. White-browed Sparrow Weavers are ploceid finches, locally resident and often common through East and Central Africa. They inhabit dry bush and acacia country, thorn scrub, and mopane woodland. Easy to watch, they live in groups of 2 to 12, although only one breeding pair occurs in a group. Other group members are typically family members and assist with rearing the young of the breeding pair. Birds forage entirely on the ground for seeds and insects; communal groups defend a territory with a permanent roosting/nesting site in the center. Nests are singly distributed in a tree or several closely spaced trees, with approximately two to three nests per breeding pair. Additional details on the natural history of this species are given by Collias and Collias (1978a, b) and Lewis (in press). STUDY AREA AND METHODS My study area was 1 km2 of mopane woodland (named for its dominant tree, Colophosper-