One hundred sixty-five patients with surgically treated thymoma were followed over 28 years; 73% had myasthenia gravis at presentation. Invasiveness was based on macroscopic findings at operation. Postsurgical radiotherapy or chemotherapy were not routinely used. Overall survival was 84%, 79%, and 65% at 3, 5, and 10 years, respectively. Patients with invasive thymoma survived for a shorter period than patients with noninvasive tumors (67% versus 85% at 5 years); when radical excision was possible, no difference was detectable between the two groups. Patients with subtotally resected or only biopsied invasive thymomas survived 59% and 42% at 5 years, respectively. Lymphoepithelial cases had the worst prognosis of the histologic types considered. Myasthenia gravis did not adversely affect survival. Surgery is the basic treatment of thymomas. Macroscopic invasiveness and degree of excision judged by the surgeon have prognostic value and are reliable criteria of malignancy. Radiotherapy and chemotherapy may be effective, but their use should be limited to controlled trials.