The effect of in vivo treatment with levamisole (LMS) as cancer immunotherapy was studied on mammary cancer (female) and transplanted MH-134 hepatoma (male) in C3H/He mice. LMS 1.15mg/kg was administered s.c. at various intervals and various stages of cancer growth. In experiment I, sterile saline was injected s.c. in the scapular region of mice bearing mammary cancer every other day ten times as a control group. LMS was injected until the animal died in the third group with the same method. The treatment with LMS in the second group resulted in a most extended survial period and a tumor-free survivor, compared with control and third group. In experiment II, sterile saline was injected s.c. in the gluteal region every other day 10 times from the day the hepatoma cells were inoculated as a control group. LMS was injected s.c. in the same region every other day 10 times from 0, 7, 10, 14, 21 and 28 days after hepatoma cells inoculation. In the fourth group injected 10 days after tumor inoculation, treatment with LMS resulted in a most extended survival period and two tumorfree survivors. These experiments demonstrated that the effect of LMS was different according to the administration time under this dose and LMS prevented tumors from growing when it was used in mice bearing moderately grown tumors. LMS did not apear to posses any significant antitumor effect in both cases (mammary cancer and hepatoma), when it was used in early or far advanced stage. This suggests that LMS has no antitumor effect when cell-mediated immunity is within normal limit, and that LMS has remarkable effect as a cancer immunotherapy when cell-mediated immunity comes to be depressed. So, it can be said that LMS may be immuno-normalizer rather than immuno-potentiator, because LMS is effective in moderately advanced stage of cancer, compared with other immuno-therapeutic agents which are more effective in early stage.