The determinants of diet breadth are of interest to nutritionists, ecologists and evolutionary biologists. A recent synthesis addressing this issue found conflicting evidence for the relationship between diet breadth and mean individual fitness. Specifically, that analysis found that whilst on average a mixed diet does increase mean fitness, in some instances a single food provides equal (or higher) fitness than a mixed diet. Critical to ecological and evolutionary considerations of diet, however, is not only average, but also variance in fitness. We combine contemporary meta-analytic methods with models of nutritional geometry to evaluate how diet affects between-individual variance in fitness within generalist consumers from a range of trophic levels. As predicted by nutritional geometry, we found that between-individual variance in fitness-related traits is higher on single-food than mixed-food diets. The effect was strong for longevity (57% higher) and reproductive traits (37%), and present but weaker for size-related traits (10%). Further, the effect became stronger as the number of available foods increased. The availability of multiple foods likely allows individuals with differing nutritional optima to customise intake, each maximising their own fitness. Importantly, these findings may suggest that selection on traits correlated with nutritional requirements is weak in heterogeneous nutritional environments.

Data.ObjectsAn R data file, which contains all of the data used for the analyses in "An overlooked consequence of dietary mixing: a varied diet reduces inter-individual variance in fitness". All the data objects are stored in a list. There are details of each data object in the accompanying R code.CodeAll the code to recreate the analyses in "An overlooked consequence of dietary mixing: a varied diet reduces inter-individual variance in fitness". The necessary data are available in the accompanying R data file.