Starvation rates of fish larvae living in patchy prey environments can have an important impact on cohort survival and recruitment. Despite this, little is known about how fluctuations in feeding experience influence starvation resistance and how this changes with ontogeny. Fish previously exposed to fluctuating food densities may not respond to long periods without food in the same way as fish previously exposed to a constant prey density. In a series of laboratory experiments with larvae and juveniles of yellow perchPerca flavescens, we tested the effects of continuous and intermittent feeding on times to starvation and on mass loss up to death from starvation for fish with initial total lengths of 10, 15, and 20 mm. Results indicated that proportional mass loss up to starvation was independent of fish mass, but that it did depend on feeding history. Fish that fed continuously before starvation all died after losing the same proportion of body mass (55%), but intermittent feeders died when they were slightly heavier (51–46% of body mass lost). Times to 50% mortality followed a different pattern; there was no significant difference in times to 50% mortality for fish that had fed continuously or intermittently for the same number of days before starvation. We conclude that short-term fluctuations (≤4 d) in food availability do not appear to affect times to starvation but do influence mass loss during starvation in young yellow perch.