Malignant ovarian tumor is regarded as the disease with the worse prognosis among obstetrical and gynecological malignancies. In the past, a great number of trials have been done with single or combined treatments. Because of the rather low incidence of malignant ovarian tumor in any one hospital over a short period of time and the numerous factors affecting the prognosis of patients, no definite procedures have been established so far. In order to evaluate the effects of chemotherapy, 460 cases were investigated retrospectively. These cases were treated primarily with surgery from 1974 to 1979 and adjuvant chemotherapy was selected freely by the doctors responsible. Cumulative 5-year survival rates were calculated as 88.8% for a low potential malignant group, 65.9% for stage I ovarian cancer, 40.6% for stage II, 9.7% for stage III, 0% for stage IV and 10% for metastatic tumor from other organs. The effects of chemotherapy were investigated especially for stage III and IV and the metastatic tumor group. Cumulative survival curves were justified by generalized Wilcoxon test. An MFC combination, a double-agent, a single-agent and a no-chemotherapy group were compared by survival curves. There were statistical significances between the MFC and no-chemotherapy group, and the double-agent and no-chemotherapy group. Significant tendency was present between the MFC and single-agent group. Several popularly-used combinations in this period such as MFC, VEM, METT, FAMT, METVFC and a group with other combinations were compared. There were no statistical significances among the 6 combinations, but an MFC combination seemed to be a better treatment for malignant ovarian tumors judging from accumulative survival curves.