I examined size-selective predation by small (159–239 mm, total length, TL) piscivorous walleyesStizostedion vitreumfeeding on three sizes of yellow perchPerca flavescens.The size distributions were presented in three experiments with yellow perch size standardized to walleye size by using prey/predator TL ratios (PPTLRs). Experiment 1 included PPTLRs of 0.2, 0.3, and 0.4; experiment 2 included PPTLRs of 0.3, 0.4, and 0.5; experiment 3 included PPTLRs of 0.4, 0.5, and 0.6. Attack and consumption rates did not differ among prey sizes in experiment 1, but they were greater on small prey in experiments 2 and 3. For prey sizes common to more than one experiment (PPTLRs 0.3, 0.4, 0.5), selectivity was experiment-dependent. There was increased selection for a specific prey size as it became relatively smaller compared with the sizes of alternative yellow perch. Active selection, as measured by attack rate, was an important contributor to patterns of yellow perch consumption by walleyes.