Two- and three-year-old oysters Ostrea edulis (L.) were sampled in November 1995 from an area on the south coast of Ireland where the protozoan parasite Bonamia ostreae has been present for over ten years. Two methods of diagnosis were employed to determine the prevalence of infection in these animals: examination of ventricular heart smears and examination of haemolymph smears. Two animals out of forty (5%), which were diagnosed uninfected with Bonamia following examination of heart smears, were found to be infected following examination of haemolymph smears. Haemolymph and heart smears were examined from further oysters sampled in April and June 1997 but no differences in prevalence of infection were found in these samples using the two methods. However, results of the screening in 1995 raise a question as to the total reliability of heart smears as a method of diagnosing bonamiasis in situations where the result is of critical importance.