A multigenerational study on the long-term efects of three dose levels of DDT (2, 20 and 250ppm in the diet) and one dose level of N-nitrosodimethylamine ( D M N ) (0.0003 % in the drinking water) was started in 1968. Each treatment was administered to two separate colonies of BALB/c mice. Treatment ofparent mice started at 4-5 weeks of age. The results observed in the first two generations are presented in this report.Survival in males was poor because of fighting. In females, the highest dose level of DDT produced liver-cell tumours in 44% of the parent mice and 74% of the first generation mice, whereas no liver-cell tumours were found in mice given the control diet, or lower dietary concentrations of DDT. Liver cell tumours did not give metastases, but grew after transplantation to syngeneic mice. The ability of DDT to induce liver-cell tumours was identical in both colonies of mice given 250ppm DDT. Malignant lymphomas occurred in about SO % of the mice in all colonies given 0, 2 or 20 ppm DDT. In one of the two colonies given 250ppm DOT, the incidence of malignant lymphomas was 14 %, whereas in the other it was 36 %. The incidence of lung adenomas was not affected by DDT treatment. Tumours at sites other than liver, lymphatic system and lungs were highest in control mice and lowest in mice given 250ppm DDT. However, no linear dose-response was found, and the decrease was more obvious in the parental than in the first generation. A study on the combinations of tumours in mice with more than one tumour and an analysis of litter distribution of liver-cell tumourbearing mice did not reveal any particular trend in the distribution of liver-cell tumours. The quantitative aspects of tumour induction by DDT in BALBIc mice are compared with data from previous studies in mice. DMN consistently produced lung adenomas, liver blood cysts and haemangioendotheliomas.