The evolution of avian growth rates in variable environments : a thesis presented in partial fulfillment of the requirements for the degree of Ph.D. in Evolutionary Ecology at Massey University, Palmerston North, New Zealand

Abstract

Tubenoses and swifts develop slowly, and often have a variable food supply. Lack (1968) attributed this to convergent evolution, arguing that slow growth is an adaptation allowing survival in environments with a variable food supply. In this thesis, I test whether there is a general relationship among bird species between slow growth rate and variability in food supply. I analysed data on nestling period, growth rates for mass and wing length, and variability in food supply in birds using phylogenetically independent contrasts. Variability in food supply may be correlated with feeding frequency, and growth rate is correlated with predation risk. I included these potential confounds in my analysis. Variability in food supply was correlated with nestling period, and negatively correlated with mass and wing growth rate, taking average feeding frequency and predation risk into account. I show that nest site preference is incompletely coadapted with growth rate. The correlations between growth rate and variability in food supply could also be explained by the proximate effect of environmental variability on growth. I therefore tested predictions of Lack's hypothesis, in comparison to those of growth models assuming facultative growth adjustments in response to variability in food supply. This further supported Lack's hypothesis. While Lack proposed that slow growth is an adaptation to variability in food supply, he did not explain the underlying mechanism. I examined three possible mechanisms, along with two alternative explanations where slow growth is not an adaptation to a variable food supply, and tested them with comparative data. I developed two of these models using computer simulations which predicted that survival is increased by reducing maximum lean tissue growth rates and increasing maximum fat deposition rates when food supply is variable. I tested predictions from these models using experiments on the Welcome Swallow, corroborating a model that predicts that lean tissue growth is prioritised over fat deposition but that fat deposition is facilitated by reduced lean tissue growth rates. I also tested whether swifts and tubenoses are adapted to an unpredictably, or predictably, variable food supply, and discuss the degree to which chicks of swifts and tubenoses are well designed for survival in environments with a variable food supply.