Mortality findings in a 1964 to 1970 cohort of 1013 hourly-wage men exposed to methylene chloride were substantially unchanged after 4 additional years of observation through 1988. Mean exposure was 26 ppm (8-hour time-weighted average) for 23 years; median follow-up from first exposure was 33 years. A comparison with death rates in both general population and industrial referents showed nonsignificant deficits in observed-expected ratios for such hypothesized causes as lung and liver cancer and ischemic heart disease. Overall mortality from 1964 to 1988 (n = 238) was significantly decreased v both referent groups. The study had 90% power to detect relative risks of 1.7 and 1.3 for lung cancer and ischemic heart disease, respectively; power was inadequate for hepatic cancer. No pancreatic cancer deaths occurred since the 1984 follow-up; eight have been observed v 4.2 expected (P = .13). An analysis of dose response for selected causes of death demonstrated no statistically significant trend according to either career methylene chloride exposure or latency. Similar results were observed when the data were analyzed using Poisson regression modeling.