Refractory ceramic fibers (RCF) are man-made vitreous fibers used primarily in industrial high-temperature applications, especially for insulation of furnaces and kilns. Because of their increasing use and potential for human exposure, a chronic toxicity/carcinogenicity inhalation study was conducted in Fischer 344 (F344) rats. Five groups of 140 weanling male F344 rats were exposed via noseonly inhalation to either HEPA-filtered air (chamber controls) or 30 mg/m(3) (approximately 220 fibers/cm(3)) of three types [kaolin-based, high-purity, and aluminum zirconia silica (AZS)] of "size-selected" RCF fibers (approximately 1µ in diameter and approximately 20 um in length) and an "after-service" heat-treated (2400°F for 24 h) kaolin-based fiber for 6 h/day, 5 days/wk for 24 mo. They were then held unexposed until approximately 20% survival and then sacrificed at 30 mo. A positive control group of 80 F344 rats was exposed to 10 mg/m(3) chrysotile asbestos. Croups of 3-6 animals were sacrificed at 3, 6, 9, 12, 15, 18, and 24 mo to follow the progression of lesions and to determine fiber lung burdens. Additional groups of 3 rats were removed from exposure at 3, 6, 9, 12, and 18 mo and were held until sacrificed at 24 mo (recovery groups) for similar determinations. Lung burdens increased rapidly for all RCFs, appearing to plateau by about 12 mo. By 24 mo, lung burdens ranged from 2.6 to 9.6 × 10(5) fiberslmg of dry lung tissue for the RCFs tested. Treatment-related lesions were restricted to the lungs. To some extent all types of RCF resulted in macrophage infiltration, bronchiolization of proximal alveoli, and microgranuloma formation by 3 mo of exposure. Interstitial fibrosis was observed at 6 mo for all types of RCF, except the "after-service" fiber where fibrosis was not seen until 12 mo. The lesions progressed in severity until 12-15 mo, after which they plateaued. A minimal amount of focal pleural fibrosis was first observed at 9 mo and progressed to a mild severity by the end of the study. Fxposure-related pulmonary neoplasms (bronchoalveolar adenomas and carcinomas combined) were observed with all 4 types of RCF [kaolin, 16 of 123 (13%); AZS, 9 of 121 (7.4%); high-purity, 19 of 121 (15.7%); and "after-service,"4 of 118 (3.4%)], compared to 2 of 120 (1.5%) in the untreated air controls. Pleural mesotheliomas were observed in two kaolin, three AZS, two high-purity, and one "after-service" exposed rats. A comparable but slightly greater amount of fibrosis was observed in the lungs of the positive (chrysotile asbestos) controls. The incidence of bronchoalveolar neoplasms in the chrysotile exposed rats was 13 of 69 (18.8%), and a mesothelioma occurred in 1 (1.4%) animal. The results of this study showed that the four types of RCF studied had carcinogenic activity in rats at the maximum tolerated dose.