We evaluated the efficacy of alternate day, high dose prednisolone for the treatment of paroxysmal nocturnal hemoglobinuria (PNH). Nineteen patients were included. Thirteen were men and six were women, aged between 13-56 years. Eleven patients improved, eight with good response and three with fair response. Eight patients were non-responders. Responders had gradual improvement in the hemoglobin level, but none achieved a normal hemoglobin level. Age at diagnosis, sex, initial hemoglobin, white count, and percentage of a positive Ham's test had no apparent bearing on treatment outcome. A prolonged interval from diagnosis to prednisolone treatment decreased the chance of a favorable hematologic response to therapy. Age at the treatment in non-responders was higher than responders. Responders had higher numbers of colonies derived from BFU-E and CFU-GM both in the blood and bone marrow than non-responders although the differences did not achieve statistical significance. These data indicate that alternate day, high dose prednisolone therapy is effective in some patients with PNH.