Reply from J. Komdeur and B.J. Hatchwell

Abstract

We welcome Barnard’s comments1xBarnard, C. Trends Ecol. Evol. 1999; 14: 448Abstract | Full Text | Full Text PDF | PubMedSee all References1 regarding our article2xKomdeur, J. and Hatchwell, B.J. Trends Ecol. Evol. 1999; 14: 237–241Abstract | Full Text | Full Text PDF | PubMed | Scopus (130)See all References2 on kin recognition in avian societies. Indeed, we agree that certain arguments about the function and mechanism of kin-biased behaviour would be more forceful without referring to the long-questioned but nevertheless, resistant tenets of the kinship literature. However, we feel that this is necessary for a proper evaluation of the experimental evidence for kin discrimination, and to discuss the mechanism and adaptive value of kin recognition in cooperative and noncooperative species. For a nonbreeding helper to maximize fitness by providing aid to their closest genetic relatives, it is crucial to discriminate between kin and non-kin, and even between kin differing in their degree of relatedness. If using a strict definition of ‘kin recognition’ (kin discrimination through recognition of relatedness), we agree with Barnard that ‘kin-recognition’ is not the sole mechanism to achieve kin discrimination. In our review, we have defined the term kin recognition as the ability to discriminate between kin and non-kin either through strict ‘kin recognition’ or through other mechanisms (e.g. associative learning). Whatever definition or terminology is used, the messages of our review remain the same: that associative learning is the most likely mechanism enabling helpers to discriminate kin from non-kin; and, that there is a lack of empirical studies in this area.Helpers of several cooperatively breeding bird species preferentially allocate their aid to closest kin3xEmlen, S.T. : 228–253See all References3, suggesting that kinship effects are not incidental (see Box 1; Ref. 2xKomdeur, J. and Hatchwell, B.J. Trends Ecol. Evol. 1999; 14: 237–241Abstract | Full Text | Full Text PDF | PubMed | Scopus (130)See all ReferencesRef. 2). The four categories of recognition mechanisms that we considered in our article are helpful in trying to understand these examples of kin preference, whether or not kin discrimination is demonstrated. Although we realize that kin recognition through phenotype matching might involve recognition alleles, we feel that kin discrimination either through phenotype matching and/or recognition alleles, associative learning and spatially based recognition are heuristically useful and, importantly, from an empirical perspective, they make exclusive predictions about where helpers should direct their care. We sought to emphasize that the nature and precision of cues for discrimination available to helpers will vary with the ecology and life history of particular species, and are unlikely to be universal.