Raised death rates have been reported for non-Hodgkin's lymphoma (NHL) and Hodgkin's disease (HD) among white male residents of Hancock County, Ohio, United States, for 1960-79. As a surveillance activity, to assess the possibility of workplace exposures contributing to these raised rates, a case-control study was conducted using death certificate records of white male Hancock County residents for 1958-83. There were 61 cases of NHL, 15 cases of HD, and 304 control subjects (chosen as a stratified random sample) in the study. Cases and controls were compared with respect to their usual occupation and industry statements on the death certificates, adjusting for age at death and year of death. Summary odds ratios (OR) and test-based 95% confidence intervals (CI) were calculated. NHL was associated with the occupation of farmer (OR = 1.6; CI = 0.8, 3.4; observed number of exposed cases (Obs) = 15). The association was restricted to 1958-73 (OR = 2.1; CI = 0.9, 4.8; Obs = 13). The three cases of HD among farmers occurred in a cluster in the 15-64 age range during 1958-63 (OR = 21.2). The results for occupations and industries other than farming were based on small numbers and were unremarkable. This small study adds to the growing body of reports linking farming and malignant lymphoma, particularly NHL.