Stochastic variation in food availability influences weight and age at maturity.

Variation in mean food availability, and in the variance around the mean, affects the growth rate during development. Previous theoretical work on the influence of environmental quality or growth rates on the phenotypic traits age and size at maturation assumed that there is no variation in growth rate or food availability within a generation. We develop a stochastic dynamic programming (SDP) model of the foraging behaviour of aphidophagous syrphids, and use this model to predict when syrphids should pupate (mature) when average food availability changes, or varies stochastically, during development. The optimal strategy takes into account not only the availability of food, but also the timing of its availability. Food availability, when small, influences developmental time, but not weight at pupation. Food availability, when large, influences weight at pupation, but not developmental time. When the food supply is low, the optimal strategy adjusts the size at pupation downwards for stochastic as opposed to deterministic availability of food. The conclusions reinforce the need for life-history studies to consider state dependence and short-term variability in growth rates.