The mortality rates of steelworkers employed in different industry worksites were investigated through a study of 58,828 men employed from 1953 to 1961. Data included work site classification, race, age, nativity, and residence. Out of 53 work areas, the following five were associated with mortality rates significantly different than those expected: carpenter shop, janitors, machine shop, mechanical maintenance, and sheet finishing and for janitors and nonwhite workers employed in the cold reducing mills. Men employed in mechanical maintenance and white janitors exhibited a significant excess in deaths due to heart disease. Mortality rates in relation to length of employment were discussed and other incidences of mortality excesses and deficits were considered. The authors note that etiological differences may be understood better by studying populations at more than one point in time; further, that cause specific excesses and deficits in mortality may relate to health selection or to workplace factors.