In a case-control study based on 859 consecutive non-Hodgkin's lymphoma (NHL) cases identified through a tumor registry between 1990 and 1998, the authors collected demographic, occupational, exposure, and education information. Exposures were identified through self-report, reported occupational history, and the use of a job-exposure matrix. Conditional logistic regression analyses of the 859 cases and 1,310 controls showed increased risks in workers exposed to gasoline (odds ratio [OR] 1.46; 95% confidence interval [CI] 1.04, 2.05), aliphatic or alicyclic hydrocarbons (OR 1.75; CI 1.03, 2.99), aromatic hydrocarbons (OR 1.45; (CI 1.13, 1.86), and solvents for mole than five years (OR 1.59: CI 1.11, 2.28), as well as automobile mechanics (OR 1.82; CI 1.18, 2.81) and painters (OR 1.77; CI 1.13, 2.76). Exposures to pesticides and farming were not associated with increased risk. Prior radiotherapy was associated with increased risk (OR 2.84; (CI 1.85, 4.37). Concordance between analyses based on self-reported exposures, occupations, and the job-exposure matrix supported the links between organic solvents and prior radiotherapy and NHL but did not support associations between farming or pesticides and NHL.