In this study of cancer in offspring we demonstrate that factors linked to horticulture and use of pesticides are associated with cancer at an early age, whereas factors in animal husbandry, in particular poultry farming, are associated with cancers in later childhood and young adulthood. Incident cancer was investigated in offspring born in 1952-1991 to parents identified as farm holders in agricultural censuses in Norway in 1969-1989. In the follow-up of 323,292 offspring for 5.7 million person-years, 1,275 incident cancers were identified in the Cancer Registry for 1965-1991. The standardized incidence for all cancers was equal to the total rural population of Norway, but cohort subjects had an excess incidence of nervous-system tumours and testicular cancers in certain regions and strata of time that could imply that specific risk factors were of importance. Classification of exposure indicators was based on information given at the agricultural censuses. Risk factors were found for brain tumours, in particular non-astrocytic neuroepithelial tumours: for all ages, pig farming tripled the risk [rate ratio (RR), 3.11; 95% confidence interval (CI), 1.89-5.13]; indicators of pesticide use had an independent effect of the same magnitude in a dose-response fashion, strongest in children aged 0 to 14 years (RR, 3.37; 95% CI, 1.63-6.94). Horticulture and pesticide indicators were associated with all cancers at ages 0 to 4 years, Wilms' tumour, non-Hodgkin's lymphoma, eye cancer and neuroblastoma. Chicken farming was associated with some common cancers of adolescence, and was strongest for osteosarcoma and mixed cellular type of Hodgkin's disease. The main problem in this large cohort study is the crude exposure indicators available; the resulting misclassification is likely to bias any true association towards unity.