Exposures to farm animals has been associated with certain rare cancers. Simultaneously, using the same methodology and control group, we conducted a six-province incident, population-based study of Hodgkin's disease (HD), multiple myeloma (MM), and soft tissue sarcoma (STS). Farm residence or work was reported by 38% (n = 119) of HD, 45% (n = 178) of MM, 43% (n = 156) of STS cases and 45% (n = 673) of controls. We conducted conditional logistic regression analyses and report odds ratios (OR(adj)) and 95% confidence intervals. After adjustment for covariates, exposure to farm animals had minimal effect on risk. The independent risk factors after adjustment for covariates were a family history of cancer (MM, STS), occupational uranium exposure (HD), professional driving (MM), and personal previous cancer (MM) or shingles (HD, MM).