Epidemiological aspects of beryllium (7440417) induced chronic lung disease (berylliosis) were investigated. Cases of this nonmalignant disease included in the national case registry were examined. Effects of occupational exposure and exposure duration and dose were studied. Conclusions drawn in a 1951 review of the epidemiology of berylliosis were reexamined. Reports of 224 cases of acute chemical pneumonitis attributable to beryllium exposure were registered; all but 16 occurred in beryllium production facilities. Of 622 cases of berylliosis, 557 were associated with occupational beryllium exposure in fluorescent lamp and neon tube manufacture, beryllium extraction, research laboratories, and the ceramics industry, or to beryllium copper in foundries or cold working facilities. Sixty five cases were due to community or household air pollution. Only 32 cases developed in workers exposed after 1950, and 18 in those exposed after 1962. Dose/response relationships among cohorts were erratic or nonexistent. For example, 1 percent of residents living within 0.25 miles of a beryllium production facility developed berylliosis; only 1.3 percent of workers in the same facility developed the disease, although their exposure was estimated to be more than 1,000 times greater. Additionally, 86 percent of berylliosis cases were employed 6 months or less; only 2 percent of those developing berylliosis at other facilities were employed for so short a time. Berylliosis incidence for fluorescent lamp manufacturers was comparable to that of production facilities despite concentrations of exposure lower by orders of magnitude. The mean and maximum latency period gradually declined from 16 years for persons first exposed from 1922 to 1937, to 10 years for the period 1937 through 1959, to 6.6 years since then. The authors conclude that berylliosis among fluorescent lamp workers may be due to free beryllium-oxide (1304569) rather than to the phosphor. The unusual epidemiological features of berylliosis are still present.