Estimation of individual fitness from life-history data

Fitness is a vague, poorly developed concept in ecology and evolutionary biology. Although it is generally recognized that fitness is determined by the complete survival and reproductive schedules of individual organisms, experimental studies have rarely attempted to integrate these into a single measure of individual fitness. Instead, particularly in studies of natural selection, components of fitness are measured as surrogates for fitness. In this article, we develop an integrated measure of fitness based on the philosophical underpinnings provided by the notion of propensity fitness, combined with classical demographic methods. We show how life-history data, collected for an individual, can be formatted as an age-structured population projection matrix, A((m)). The dominant eigenvalue, lambda((m)), of this matrix is an estimate of that individual's propensity fitness. Using life-history data sets on European sparrowhawks and blue tits, we show that the interpretation of analyses of selection can shift radically using the integrated propensity fitness measure instead of components of fitness, such as lifetime reproductive output. Individual fitness, as formulated here, provides an integrated measure of performance that should prove useful in empirical studies in ecology and evolutionary biology.