Danazol was administered in a daily dose of 100, 200, 400, or 600 mg for 6 months in a double-blind fashion to 27 women with pelvic endometriosis. Symptoms and pelvic findings were observed and recorded monthly. Laparoscopy and laparoscopic biopsies were performed before and on the last day of treatment to evaluate the extent of endometriosis and the effect of the drug. The findings were documented with drawings and photography. The degree of clinical improvement varied with the daily dose used, from just over 50% on a regimen of 100 mg/day to 83% on a regimen of 600 mg/day. Laparoscopic improvement in the extent of endometriosis was noted in all patients but residual disease was common. The degree of laparoscopic improvement appeared to be related to the dose of danazol and to the effect of the drug on the menstrual cycle. The highest, 81%, laparoscopic improvement was observed in patients who developed amenorrhea during the study. After the completion of treatment, 6 patients required operation for residual endometriosis or for its early recurrence. The recurrence of endometriosis during 24 months of follow-up was observed in 29% of patients. Six of 15 infertile patients conceived spontaneously within 6 months after treatment in spite of mild (5) or moderate (1) residual disease. There was no difference in the extent of endometriosis between infertile patients who did or did not conceive. There was, however, a statistically significant difference in the adhesion score between these 2 groups. The authors conclude that danazol may be less effective in doses lower than the standard 800 mg/day. However, downward adjustment of the individual dose may be attempted on the basis of the development of amenorrhea and clinical improvement.