From Jan. 1958 to June 1983, all in-patients with gastric cancer, breast cancer and carcinoma of colon and rectum (as cancer group), and with cerebral hemorrhage (as control group) in our hospital were epidemiologically investigated in order to evaluate the relation between previous appendectomy and cancer incidence. Patients who had appendectomy more than three years before diagnosed named appendectomized (APP). In cancer group, out of 1,119 patients, 98 (8.76%) were APP, but in the control, only 14 (3.50%) of 400 patients were APP (P less than 0.001). As to cancer location, the ratio of APP were 9.47% (48/507) in patients with gastric cancer (P less than 0.001), 7.07% (21/297) in patients with carcinoma of colon and rectum (P less than 0.05) and 9.21% (29/315) in patients with breast cancer (P less than 0.01). As to sex, the ratio of APP, for the male were 9.82% (46/438) in cancer patients, while 4.47% (8/179) in the control (P less than 0.05), and for the female, were 8.08% (55/681) in cancer patients, while 2.71% (6/221) in the control (P less than 0.01). As to age, the ratio of APP was higher in various age groups of cancer than the control, but only in the 50-59 age group, the difference was significant (P less than 0.05). It shows that the ratio of APP in all cancer groups are significantly higher than those in the controls. The authors support the concept--appendectomy may influence the subsequent cancer risk, and deem it well worth to further investigate.