Abstract
THE most accurate method of study of any disease is enlightened statistical analysis of a large number of cases. However, paroxysmal nocturnal hemoglobinuria is a rare blood dyscrasia; approximately 45 cases, not all of convincing authenticity, have appeared in the literature to this time.9 Consequently, no physician or group of physicians is likely to collect a number sufficient to be of statistical significance. Therefore, in this disease an older method, that of long-term observation of a few cases, must be relied on to add to the common stock of medical knowledge. This paper is based on clinical observation of two . . .