Growing more positive with age: The relationship between reproduction and survival in aging flies.

Populations of laboratory animals that are selected for increased lifespan often show negative correlated responses in early fecundity. However, late fecundity and/or total lifetime fecundity can be higher in the populations selected for increased lifespan. This has been interpreted by some as being at odds with the disposable soma theory, which predicts decreased lifespan to increase total reproductive output. Alternatively, the Y-model explores the effects of variation in resource allocation and acquisition on life histories. In this model, a negative relationship between lifespan and reproduction can be viewed as variation in allocation, whereas a positive relationship is the result of variation in acquisition. However, a frequently neglected complication of the Y-model is that older individuals often show a decline in resource acquisition. Therefore, differential allocation to maintenance and survival might affect this decline in late-life acquisition which will affect resource availability across the whole lifespan. In this paper we show that a model which incorporates the ideas of the Y-model, the disposable soma theory, and an age-related decrease in resource acquisition, i.e. feeding senescence, can explain how the relationship between fecundity and lifespan changes with age. Furthermore, by modeling environments with contrasting extrinsic mortality rates, we explored how the outcome of the model depended on the relative importance of early and late-life reproduction. In high mortality environments a relatively higher early fecundity, lower late fecundity, and lower lifespans were more optimal, whereas the opposite was true for low mortality environments. We applied predictions from the model to a cohort of individually-housed female Drosophila melanogaster flies for which we measured age specific fecundity and lifespan. Early fecundity was negatively associated with lifespan, while late fecundity related positively with lifespan in the same cohort. This verified that the mechanism of feeding senescence could explain patterns for age specific relationships between lifespan and fecundity.