The lung cancer mortality of welders was investigated using a cohort of 3,247 males from the International Brotherhood of Boilermakers, Iron Ship Builders, Blacksmiths, Forgers and Helpers in Seattle, Washington. The subjects had been union members for at least 3 years. The vital status of the men was determined as of January 1, 1977, and lung cancer deaths were coded according the Codes 162 and 163 of the Sixth, Seventh, and Eighth revisions of the International Classification of Diseases. A modified life table computer model was used to determine the standardized mortality ratios. The ratios were compared to the United States death rates for white males specific for age and calendar time and also to death rates among 5,432 nonwelders from the same union. A total of 16 percent of the cohort was deceased at the time of the study cutoff date. Fifty of these deaths were due to lung cancer as compared to an expected 37.95 deaths due to lung cancer in the general white, male population. Age of first employment and calendar time of employment were both positively correlated with cancer mortality. The welders showed a significantly greater incidence of lung cancer relative to nonwelders in the same union but only after 20 years from first exposure. The overall increase in risk for mortality attributable to lung cancer among welders relative to the general population was 32 percent, but this increased to 74 percent at 20 or more person years from first employment. The authors conclude that it appears that welders are exposed to carcinogenic fumes from their work, but this requires further study for possible confounding factors.