Introduction

In the late 1980s and early 1990s, behavioral ecologists were first able to estimate, with good reliability, the paternity of broods in socially monogamous birds. Within a very few years, rates of paternity by social fathers had been estimated within a large number of populations. An amazing pattern emerged from these studies, one that changed fundamentally the way that behavioral ecologists viewed mating in birds. The extra-pair paternity rate – the proportion of offspring not sired by social fathers – was, on average, 10–15% (e.g. Birkhead & Møller, 1995; Petrie & Kempenaers, 1998). Though occasionally in the 1–3% range, estimated extra-pair paternity rates of at least 20% were not uncommon, and they reached 50% in a few populations. A substantial proportion of socially monogamous bird species are clearly not sexually monogamous.

From the perspective of some observers, these surprising empirical findings were the leading edge of a revolutionary “paradigm shift” in behavioral ecology currently taking place, with “the traditional concepts of the choosy, monogamous female and the coadapted gene complex increasingly giving way to the realization that sexual reproduction engenders conflicts [and] promotes polyandry …” (Zeh & Zeh, 2001). Though Trivers' (1972) parental investment theory did not explicitly claim that females in socially monogamous species should be sexually monogamous, it did emphasize the potential reproductive benefits of male multiple mating and not the reasons why females (the sex typically investing more heavily in offspring) might multiply mate. Its arguments were based on reasoning about limitations on offspring number.