Diagnosing senescence: inferring evolutionary causes from phenotypic patterns can be misleading.

Based on the predictions of two theories for the evolution of senescence, the 'antagonistic pleiotropy' and the 'mutation accumulation' theory, an age-specific increase in mortality and a decrease in fecundity are widely used criteria to diagnose senescence in natural, and laboratory populations. In this study we question the reliability of these criteria. Using a simple model we show that similar phenotypic patterns result from optimal life histories without senescence. With a tradeoff between reproduction and period survival, optimal life histories produce patterns of increasing mortality and decreasing fecundity as organisms age, even if the tradeoff does not deteriorate with age, so that we are not forced to invoke genetic effects such as antagonistic pleiotropy or accumulation of deleterious mutations to explain such patterns. Furthermore, if optimal life history theory is applied to senescent organisms, phenotypic patterns can result that are usually not associated with senescence. We conclude that the reliability of a diagnosis of senescence based on phenotypic patterns and the comprehension of the phenomenon senescence depends critically on understanding to what extent tradeoffs are determined by the effects of segregating genes.