Seasonal patterns of singing in the willow warbler: evidence against the fertility announcement hypothesis

Abstract

Males of many bird species use a variety of behaviour patterns that reduce their chances of being cuckolded. The ‘fertility announcement’ hypothesis (Møller 1991, American Naturalist, 138, 994–1014) proposes that song might be one such paternity guard. According to this hypothesis, paired males would announce their female's fertile status by singing. This has been interpreted as an honest signalling, evolutionarily stable strategy. Contrary to the predictions of this hypothesis, male willow warblers, Phylloscopus trochilus, sang very little when females were fertile. Intrusions by other males in the fertile period were not less common when males sang at higher rates. Mate guarding and singing are best interpreted as two conflicting behaviours during this period, the former being directed to the fertile female and the latter to attracting a second female, or an extrapair female. A survey of recent studies suggests that, in most passerine species studied, males do not sing during the fertile period of their females. The different conclusions of Møller (1991) are probably due to his use of population-wide estimates of the timing of singing behaviour and egg laying. Breeding asynchronies within populations would be responsible for the apparent matching between the peaks of singing activity and fertility.