Pair and Group Reproductive Success, Polyandry, and Cooperative Breeding in Harris' Hawks

Abstract

-The average size of Harris' Hawk (Parabuteo unicinctus) social units in New Mexico was 2.7 individuals, and pairs were most common (49%; n = 61). Groups of more than two hawks included both adult(74%) and immature-plumaged (26%) members (n = 76). Immatures rarely provided food to nestlings, but adult supernumeraries did. Electrophoretic analyses of two groups containing two adult-plumaged males did not suggest polyandry. At least in southeastern New Mexico, Harris' Hawk groups consist primarily of a monogamous pair with helpers. Pairs and groups showed no differences in clutch size, number of young produced per successful nest, or number of offspring fledged per year. Pair nests failed less often (16%) than group nests (46%) during the incubation period. Groups reared larger offspring and tended to initiate second nests more frequently than pairs. The overall lack of correlation between reproductive output and group size suggests that kin selection has not been a major influence in the evolution of the Harris' Hawk breeding system. Received 6 June 1986, accepted 22 December 1986. MADER (1975a, b) described a breeding population of Harris' Hawks (Parabuteo unicinctus) in Arizona that consisted both of pairs and of putatively polyandrous trios. Since publication of Mader's work, the Harris' Hawk often has been cited as one of only a few species known to exhibit polyandry or a cooperatively polyandrous breeding system (e.g. Woolfenden 1976; Brown 1978, 1983; Faaborg and Patterson 1981; Koenig and Pitelka 1981; Ligon 1983; Woolfenden and Fitzpatrick 1984; Oring 1986). In contrast, 18 of 19 (95%) Harris' Hawk nests in Texas were tended only by pairs, suggesting that polyandry and cooperative breeding are rare in at least one Texas population (Griffin 1976). Cooperative breeding should entail some fitness advantage over available alternative strategies. Benefits resulting from cooperation may be shared mutually by all group members or biased toward those individuals of the group that have greater behavioral leverage (Emlen and Vehrencamp 1983). These advantages may be expressed in terms of increased reproductive success, improved quality of young (e.g. production of healthier or heavier fledglings), decreased stress on breeders (Morton and Parry 1974, Ricklefs 1980, Emlen 1984), or increased probability of survival (Woolfenden and Fitzpatrick 1984, Faaborg 1986). Because survival probabilities and stress are difficult to measure, most fieldworkers have concentrated on obtaining estimates of reproductive success. Data on reproductive success show that, in general, large groups produce more fledglings than smaller groups of the same species (Brown 1978; Emlen 1978, 1984). There are exceptions, however, and in many cases this positive relationship is not significant. Koenig (1981) compared productivity measures for 15 species and found that per-capita reproductive success tended to decrease with increased group