This study addresses the question of whether positive mutual fund performance persists from year to year. It uses returns from substantially all equity mutual funds over the period 1976 through 1988, and uses a contingency table test to detect persistance in raw returns and risk adjusted returns. We find that, when winer is defined as being in the top helf or the top quester of the sample, mutual fund winners repeat. Bootstrapping tests show that the effect exists on a montly level as well as at mulitple year horizons. Breaking the results down by high-variance and low variance funds shows some evidence that volatility, and hence survivorship may play an explanatorly role.