Adaptive significance of avian helping behaviour

Abstract

Mumme and Koenig’ have provided a useful summary of the debate among ornithologists over the adaptive significance of helping behaviour in birds. I would like to suggest that there are, in the debate, two confusions whose removal would help to clarify it further. Both my points were originally made by Dawkin+. The first confusion arises because in this debate parental care is regarded uncontroversially as ‘shaped by the process of natural selection’, whereas kin selection (inclusive fitness theory) is considered separately as a possible explanation for ‘aid toward nondescendant relatives”. But selection for parental altruism towards young and selection for altruism to relatives other than offspring are of the same kind, namely kin selectior+. Kin selection for altruism is the natural selection of genes for altruism via the sharing of these genes with relatives3e4. A gene for altruism is one whose bearer acts so that there is a decrease in its own survival or offspring number (cost) and an increase in the survival or offspring number of another individual (benefit). In kin-selection theory, such a gene spreads if Hamilton’s condition, (relatedness) x (benefit) - (cost) > 0 is met3. So, since offspring are relatives, the theory applies as much to the evolution of care of offspring as to, say, sib altruism. Therefore, the removal of phenomena such as sib altruism into a theoretical domain separate from parental care is unjustified*, and could be why some participants in the ‘helping debate’ are suspicious of the kinselection explanation of helping. The second confusion is the idea that a kin-selected gene for helping must alone be responsible for the care behaviour observed. Say it is true, as seems likely, that natural (kin) selection for parental care has favoured birds that ‘respond positively to begging young regardless of the context”. Now imagine that there is genetic variation in the timing of departure of young from the natal territory. Late-departing young might then meet sib nestlings and start feeding them through a misdirection of parental, care. Assuming that care is costly, the gene for departing late in this situation would, in the fullest sense, be a gene for altruism, because as in the earlier definition its bearers would care for others at a lossto themselves. So, if Hamilton’s condition were satisfied, the gene would spread and the result would be a population of cooperative breeders with latedispersing sib altruists. But the gene for sib altruism would not itself have been responsible for the neural machinery of care: neurally, the behaviour would still be parental care*. Further, the sib altruists’ care behaviour would have been kinselected without necessarily having been modified compared to parental care’. Therefore, the sides in the helping debate could be closer than they think: helping behaviour in birds could logically be both misdirected, unmodified parental care and prevalent in populations as the adaptive result of kin selection for care to nondescendant relatives.