Facultative Helping Does Not Influence Reproductive Success or Survival in Cooperatively Breeding White-Browed Scrubwrens

Abstract

1. The most common form of cooperative breeding in birds involves a pair and their adult offspring ('helpers') provisioning young. Many studies show that pairs with helpers have higher reproductive success than pairs alone, but the differences could be due to confounding variables, like parental or territory quality, rather than to helping behaviour. 2. One method of testing whether helping increases reproductive success is to compare the success of intact groups with those from which helpers have been removed. The disadvantage is that variables other than provisioning by helpers (e.g. group size) are affected, which themselves could affect reproductive success. Alternatively, one can attempt to control statistically for confounding variables, but this is difficult in territorial species because it may be impossible to ensure that all confounding variables are measured. 3. We took advantage of natural variability in provisioning behaviour among subordinate white-browed scrubwrens, Sericornis frontalis, to assess the effect of helping. Scrubwrens are a small passerine in the family Pardalotidae in which male offspring may remain on the natal territory as adults. We compared the reproductive performance of pairs, groups with nonhelping subordinates and groups with helping subordinates, using data from 4-years' study of an individually colour-banded population resident in the Australian National Botanic Gardens, Canberra. 4. The total rate of feeds to nestlings was greater when helpers were present, but we could detect no increase in nestling weight. Furthermore, we found no effect of helping behaviour on reproductive performance or reproductive success. Our measures included: (i) duration of the nesting cycle and interval between nesting attempts; (ii) success of individual nesting attempts, measured in over 20 ways; and (iii) reproductive success over the whole breeding season, measured as the total number of fledglings and independent young. 5. There was also no effect of helping on survival of the female or dominant male. 6. We conclude that helping behaviour does not have a substantial, consistent effect on reproductive success, but cannot eliminate the possibility that it might increase reproductive success by a small amount or in some territories or years. Alternatively, helping behaviour might be the nonadaptive consequence of gene flow from populations in which it is adaptive. Nonetheless, we reject Jamieson's hypothesis that helping behaviour is a nonselected consequence of strong selection on parental behaviour, because, unlike many other species, helping is not an invariant consequence of natal philopatry.