<p>Many hoverfly species (Diptera: Syrphidae) are Batesian mimics of stinging Hymenoptera, although their degree of similarity to their putative models can vary considerably. Some species are considered excellent mimics, while others are considered poor mimics. Previous research has shown that larger hoverfly species are often better mimics, possibly because there is stronger selection for mimetic fidelity in larger, more profitable prey. However, research has long shown that predators can use body size to discriminate between mimics and their models. Therefore, an alternative hypothesis is that mimetic hoverflies that differ in size from their hymenopteran models are more readily identified as mimics and so face weaker selection to improve their colour pattern similarity. To compare these contrasting hypotheses, we tested whether mimetic fidelity was better explained by the absolute body size of the mimic or by the size of the mimic relative to its model while controlling for shared phylogeny. Mimetic fidelity was assessed by human ranking of 49 mimic-model pairs, while size discriminability of each pair was determined by estimating the standardized difference in intertegular distance between the hoverfly mimic and its closest hymenopteran model. We found that the body size of a mimic was positively correlated with the body size of its model and that the fidelity of the mimic was positively associated with the size of its model. However, we found no strong evidence of a relationship between mimetic fidelity and hoverfly body size or its size relative to its model.</p>