The role of kinship in helping decisions among white-fronted bee-eaters

Summary

White-fronted bee-eaters (Merops bullockoides) are cooperative breeders in which half of all nesting attempts are assisted by non-breeding adults in addition to the breeding pair. These “helpers” assist in all aspects of nesting and their aid significantly increases fledging success. The social unit in these bee-eaters is an extended family clan in which multiple pairs may breed simultaneously (plural breeding). As a result, helpers often have to choose how to allocate their aid among several potential recipients to whom they are unequally related. Using five years of data from a color-marked and genealogically known population of white-fronted bee-eaters in Kenya, we examined the role of kinship in three helping “decisions”: 1) whether or not to become a helper, 2) whom to help, and 3) how much help to provide.

• 1) Kinship between potential donor and recipient was a significant predictor of the likelihood of helping. Fully 44% of the individuals that could have become helpers did not do so. Many of these non-helpers were unrelated mates (“in-laws”) that had paired into their partner's clan and had no genetic link to the available recipients. Others were natal clan members with only distant genealogical ties to the breeders. The conditional probability of helping (see text for definition) decreased significantly with decreasing r _h(coefficient of relatedness) between the potential helper and nestling recipients. The presumed cost of helping was a second significant predictor of whether or not a bird became a helper. We considered the expending of effort in providing aid and the delaying of personal reproduction to be costs of helping. Potential helpers that had engaged in high effort activities in the few weeks prior to the helping opportunity showed a decreased likelihood of becoming helpers. Similarly, paired birds (which by virtue of having a mating partner had a greater potential of breeding) were less likely to become helpers than were single individuals.

• 2) Bee-eater helpers preferentially chose to aid their closest genetic relatives. The average relatedness between helpers and the nestlings receiving their aid was 0.33. This degree of kinship was significantly greater than that expected if helpers had selected recipient nests randomly from among those available within their clans. Further evidence of kin discrimination came from analysis of 115 instances in which a helper had two or more nests of differing relatedness simultaneously available as potential recipients. In 108 cases (94%), the helper aided the most closely related one.

• 3) The magnitude of the helping contribution (as measured by provisioning rate) showed no pattern with relatedness between donor and recipient. These data allow testing of various theoretical models for the allocation of aid among kin. Since bee-eaters helped at only one nest at a time, and selected the nest with the most closely related young available, their behavior conforms to an allor-none case of a diminishing returns model and is inconsistent with alternative models based upon the distribution of aid in proportion to relatedness. We conclude that kinship is a major determinant in the decisions both of whether to become a helper, and of whom to help, in white-fronted bee-eaters.