Mimetic song in superb lyrebirds: species mimicked and mimetic accuracy in different populations and age classes

Mimetic songs of superb lyrebirds, Menura novaehollandiae, were investigated to evaluate the effects of population and age class on model selection and mimetic accuracy. Two geographically isolated populations with similar habitat and sympatric avian species mimicked the same vocalizations from almost identical species of avian models and failed to mimic the same suite of potential models. Mimetic song had a similar structure at the two sites: diversity of song items and performance rate were not significantly different. However, the populations differed in the frequency some models were repeated in a bout of song. We compared the acoustic structure of lyrebird copies to the physically challenging song of the eastern whipbird, Psophodes olivaceus, and the complex contact call of the yellow-tailed black cockatoo, Calyptorhynchus funereus. Lyrebirds could not match the duration of the sustained monotone note of the whipbird nor the demanding frequency bandwidth of the explosive whipcrack. There was a significant positive correlation between whipcrack bandwidth and diversity of mimetic items which suggests consistent individual differences in mimetic performance. Contact calls of cockatoos and lyrebird copies were also distinguishable but less so than with the whipcrack. Subadults made much poorer copies of these models than did the adult lyrebirds and had fewer identifiable heterospecific song items suggesting that costs are associated with the development of mimetic song. Together, these findings are consistent with a female choice hypothesis where females prefer males that mimic more accurately and have a greater diversity of mimetic items, but experimental verification is needed.