An excess of mortality from stomach cancer has been found in Ontario gold miners (observed (obs) 104, standardised mortality ratio (SMR) 152, 95% confidence interval (95% CI) 125-185) and no excess of stomach cancer could be detected in other miners in Ontario (obs 74, SMR 102, 95% CI 80-128). The excess of stomach cancer appeared five to 19 years after the miners began gold mining in Ontario. In that interval, similar patterns of excess mortality from stomach cancer were found in miners born in north America (obs 14, SMR 268, CI 147-450) and in miners born outside north America (obs 12, SMR 280, 95% CI 145-489). Twenty or more years after the miners began mining gold, an excess of mortality from stomach cancer was found in gold miners born outside of north American (obs 41, SMR 160, 95% CI 115-218) but not in gold miners born in north America (obs 37, SMR 113, 95% CI 80-156). The excess of stomach cancer in gold miners under the age of 60 (obs 45, SMR 167, 95% CI 122-223) seems larger than the excess in gold miners between the ages of 60 and 74 (obs 59, SMR 143, 95% CI 109-184). Exposures to arsenic, chromium, mineral fibre, diesel emissions, and aluminium powder were considered as possible explanations of the excess of stomach cancer in Ontario gold miners. Exposure to diesel emissions and aluminium powder was rejected as gold miners and uranium miners were exposed to both agents but an excess of stomach cancer was noted only in gold miners. The association between the excess of stomach cancer and the time since the miner began mining gold suggested that duration of exposure to dust in gold mines ought to be weighted according to the time since the exposure to dust occurred and that an appropriate time weighting function would be one in the interval five to 19 years after each year of exposure to dust and zero otherwise. A statistically significant association between the relative risk of mortality from stomach cancer and the time weighted duration of exposure to dust in gold mines was found in miners under the age of 60. Time weighted indices of exposure to chromium and arsenic were formed for each gold miner by time weighting the product of the duration of exposure to dust in a gold mine and the percentages of arsenic and chromium in rocks in that gold mine. Exposure to mineral fibre was measured in terms of the time weighted duration of employment in those gold mines that contain mineral fibre. A statistically significant association between the excess of stomach cancer in gold miners under the age of 60 and the time weighted index of exposure to chromium occurred and not association was found between the excess of stomach cancer and either the time weighted duration of employment in mines containing mineral fibre. The excess of stomach cancer in gold miners under the age of 60 was better associated with the time weighted index of exposure to chromium than to the time weighted duration of exposure to dust in gold mines. Although the number of cases of gastric cancer that were classified according to the system of Lauren was small, the data suggest that for miners under the age of 60, exposure to chromium is associated with the development of the intestinal rather than the diffuse type of gastric cancer.