This investigation evaluated the mortality experience of workers at a dye and resin manufacturing plant in New Jersey. The retrospective follow-up study included 3266 workers employed for at least 6 months at the plant. Plant production areas were South Dyes, where anthraquinone dyes and intermediates were produced; North Dyes, where azo dyes and intermediates were produced; and plastics and additives, where various resins and additives for plastics were made. Analyses used standardized mortality ratio (SMRs) to compare the cohort's cause-specific mortality rates during 1952 to 1995 with the rates of the NJ population. There were fewer than expected deaths from all causes combined (728 observed vs 810 expected) and similar numbers of observed and expected cancer deaths (225 vs 232). Statistically significant work area-specific cancer excesses were limited to white men and included an excess of lung cancer in Maintenance workers (40 observed vs 26 expected; SMR, 153; 95% confidence interval [CI], 109 to 208) and in South Dyes workers (32 observed vs 19 expected; SMR, 168; CI, 115 to 237) and an excess of stomach cancer (5 observed vs 1.3 expected; SMR, 386; CI, 125 to 901), bladder cancer (4 observed vs 0.8 expected; SMR, 515; CI, 140 to 1318) and central nervous system cancer (5 observed vs 1 expected; SMR, 517; CI, 168 to 1206) in North Dyes workers. None of these increases was concentrated in work area subgroups with long duration of employment and long potential induction time. It was concluded that the excess of bladder cancer probably was due to exposure to carcinogenic arylamines at another facility, where some employees had worked before coming to the study plant. The other cancer increases may be attributable to chance, to uncontrolled confounding by smoking, or to an unidentified occupational exposure.