Behav Ecol Sociobiol (1984) 14:137-146 Behavioral Ecology and Sociobiology 9 Springer-Verlag 1984 Co-operative breeding by the Australian Bell Miner Manorina melanophrys Latham: A test of kin selection theory Michael F. Clarke Department of Zoology, University of Melbourne, Parkville, Victoria 3052, Australia Received March 29, 1983 / Accepted July 20, 1983 Summary. A population of Bell Miners Manorina melanophrys consisted of two social units (" clans"), each consisting of one or more monoga- mous pairs and numerous helpers. Individuals of a breeding pair occupied one home range, while non-breeding individuals occupied partially over- lapping home ranges. Attendants (i.e. parents and helpers) showed clear preferences as to whom they aided (fed) and how much aid they provided. A significant correlation was found between the amount of aid given by an attendant and the genet- ic relationship between the attendant and the recip- ient. This provides strong quantitative evidence for the theory of kin selection. Only 14% of eggs laid produced juveniles which lived longer than 3 months. It is suggested that the extremely high ju- venile mortality favours individuals which help rel- atives, even while such individuals have dependent offspring of their own. Introduction Co-operative breeding in birds is characterised by the participation of "helpers", i.e. individuals other than the breeding pair, in the reproductive activites of that pair. Helping behaviour may affect the fitness of the recipient breeding pair and/or the inclusive fitness of the helper (Brown 1978). Inclusive fitness has two components, direct fitness and indirect fitness (Brown 1980; Brown and Brown 1981). Direct fitness is mediated through the genes in an individual's own offspring while indirect fitness is mediated through copies of the genes in the offspring of related individuals. Direct enhancement of the fitness of a recipient pair by the presence of helpers may occur if the pair fledges more young, shows increased survival as a result of decreased energy expenditure on breeding, or is able to control a better territory or achieve better protection against predators (Brown 1978; Kinnaird and Grant 1982). A help- er's fitness may increase directly through the expe- rience it gains in feeding young, through decreased dispersal leading to lower mortality, through terri- tory inheritance and through possible reciproca- tion of helping by the aided young (Rowley 1965; Brown 1978 ; Ligon and Ligon 1978 ; Woolfenden and Fitzpatrick 1978; Reyer 1980). However, there may be also indirect benefits to the helper if it is a close relative of the breeders whose reproduc- tive success it is increasing, through the process of kin selection (Hamilton 1964; Maynard Smith 1964; Brown 1978; Ricklefs 1975; Reyer 1980). A basic prediction of kin selection theory is that acts of aid-giving should occur more often among closely related individuals than among dis- tantly related ones. Brown (1975) suggested that "a positive correlation between altruistic aid-giv- ing and genetic relationship probably exists, but it remains to be confirmed generally by quantita- tive studies". Massey (1977) provided evidence that in primates a correlation exists between the amount of agonistic aid given (in the form of de- fense of relatives from aggressors) and the degree of genetic relationship. Brown and Brown (1980) concluded that in the Mexican Jay Aphelocoma ul- tramarina "the feeding of fledgling Mexican Jays by parent-helpers is not correlated with relatedness in any obvious rigorous way" An ideal system for testing such a correlation exists in the Australian Bell Miner (Meliphagidae) (Swainson 1970; Smith and Robertson 1978) be- cause an individual may help at several different nests belonging to different breeding pairs, and aid-giving can easily be quantified as the number