The optimal treatment of early-stage ovarian borderline tumors is controversial. Only a few randomized trials evaluating adjuvant treatment for this disease have been published. Between 1970 and 1988 four consecutive randomized trials, including patients with ovarian borderline tumors, were conducted at the Norwegian Radium Hospital. After surgery, 253 stage I-II borderline tumors without residual disease were randomly allocated to these protocols. The adjuvant treatment in the four trials consisted of [1] external irradiation (Ext) combined with intraperitoneal instillation of radioactive gold (198Au) or Ext alone, [2] intraperitoneal radioactive therapy followed by thio-TEPA or no further treatment, [3] thio-TEPA or no adjuvant therapy, and [4] cisplatin or 32P treatment. The patients were equally distributed according to prognostic variables within the eight randomization groups. The overall corrected and crude survival were 99 and 94%, respectively. In 83% of the patients a hysterectomy, bilateral salpingo-oophorectomy, and omentectomy was performed. None of the patients with less extensive surgery relapsed. Adjuvant therapy did not seem to improve the overall corrected survival. On the contrary, toxicity was added with small bowel complications after radiation therapy, neurotoxicity after cisplatin treatment, and bone marrow toxicity after thio-TEPA therapy. It is concluded that stage I borderline tumors should not receive any adjuvant treatment.