Leukemia mortality at a nuclear naval shipyard was evaluated by examining the personnel records and death certificates of 24,545 white males employed at any time between 1952 and 1977. Each worker's total external radiation dose was calculated from a summary of individual radiation exposures. The study group was subdivided into those having a record lifetime cumulative radiation exposure of at least 1 millirem (Subcohort I), those unassigned to radiation areas (Subcohort II), and those assigned to radiation areas, but whose total lifetime cumulative dose was 0.000rem (Subcohort III). A total of 76 percent of the entire cohort was alive, 19 percent had died and 4 percent were unaccounted for at the time of the study. Total number of deaths was 4,762, while 5,361 were expected. Deaths due to leukemia, and to all lymphatic, hematopoietic and malignant neoplasms were slightly lower than expected. No risk trends by latency or by duration of employment were noted. In Subcohort I, mortality from all causes was significantly lower than expected, while mortality in Subcohort II was approximately equal to that expected. In Subcohort III mortality from all causes of death was significantly depressed. No significant difference in leukemia mortality was found between radiation and nonradiation workers.