Studies suggesting associations between asbestos (1332214) exposure and non-Hodgkin's lymphoma are reviewed. Among United States (US) Social Security disability claimants from 1959 to 1962, proportionate morbidity ratios were significantly raised for lymphoma and leukemia among stationary engineers, mechanics, and carpenters. An analysis of deaths in 1950 among US males demonstrated significantly increased standardized mortality ratios for lymphopoietic and hematopoietic malignancies, excluding leukemia, among stationary engineers, shipbuilders, and construction workers. A cohort study of 2,722 males employed in a facility manufacturing asbestos textile, friction, and packaging products, using 99 percent chrysotile (12001295) asbestos, is described. Seven deaths from non-Hodgkin's lymphoma versus 3.28 expected deaths are noted. An earlier study of 68 cases of asbestosis from this same facility identified five additional hematopoietic and lymphatic malignancies. Animal experiments reveal reticulosarcomas in rats administered chrysotile asbestos orally or by intratracheal injection with benzopyrene (50328). The possibility of an asbestos etiology for non-Hodgkin's lymphoma is supported by findings that thoracic lymph nodes are important points in the clearance pathway of asbestos fibers in animals and humans. The authors suggest that investigators who have followed up cohorts of asbestos workers determine expected and observed deaths from non-Hodgkin's lymphoma and submit pathologic material from these deaths to a central panel of pathologists for review.