C3H mice were injected subcutaneously (s.c.) with a tumorigenic dose (150 micrograms/mouse) of 3-methylcholanthrene (MC), followed by a 24-h injection and subsequent weekly injections of Corynebacterium parvum (CP) i.p. for a total of 100 days. Basal and CP-augmented NK cell activities were measured in controls and treatment groups during pre-tumor and tumor development stages. Basal NK activity in spleen, peripheral blood and lung tissue was enhanced by CP, but was suppressed by MC. A resulting transient MC induced suppression of splenic NK activity at 10 days was partially restored and sustained by CP treatment and immunosuppression was again observed in tumor-bearing compared to control mice. Mice treated with MC alone showed a higher tumor incidence than animals treated with MC + CP at 45-60 days, while there was no difference in tumor incidence in these two treatment groups at 100 days post injection. The mechanism of the observed transient immunosuppression induced by MC appears to be related to an early toxic effect on large granular lymphocytes (LGL) which was decreased at 10 days and again at 100 days in tumor-bearing mice. Although MC did not appear to exert an effect on effector:target cell conjugate formation, an early suppression in the lytic activity of LGL, may have predisposed the animal to malignant transformation of susceptible cells at the site of MC injection.