In a lifelong experiment Sprague-Dawley rats received cyclophosphamide (CP) i n drinking water. Daily doses of 2.5 (I). 1.25 (II), 0.63 (Ill), and 0.31 (IV) mg/kg body weight were given t o groups of 40 male and 40 female rats five times a week. The median survival time (MST) of treated rats proved to be dose-dependent. In rats treated with dose I the MST was 638 days for males and 642 for females, i n those given dose IV it was 906 days for males and 934 days for females. The percentage of control animals with malignant tumors was 11% for males and 15% for females. The corresponding figures for CP-treated animals ranged from 32 t o 43% for male rats and from 30 t o 35% for females. A t doses I, II, and 111 CP proved t o be carcinogenic (p vs control<0.05). No carcinomas of the urinary bladder were found i n controls but one female showed a benign papilloma of the bladder. However, 17 CP-treated rats died with carcinomas of the urinary bladder and, remarkably, only one of these was a female. In addition, 18 male and 10 female animals showed benign tumors of the same organotropy. No leukemia was diagnosed in controls, but in CP-treated rats leukemia was found in 8 male and 11 female animals. These findings support the evidence emerging from case reports in humans showing an increased risk of cancer of the urinary bladder as well as of leukemia i n patients treated with CP.