Long-term in vivo carcinogenicity tests of potassium bromate (KBrO3), sodium hypochlorite (NaClO), and sodium chlorite (NaClO2) have been conducted in Japan from 1977 to 1985. In these investigations, groups of approximately 50 male and 50 female F344 rats or B6C3F1 mice were given solutions of the compounds as their drinking water ad libitum at two dose levels determined on the basis of preliminary 13-week tests. Control animals were given distilled water. The carcinogenic potential of KBrO3 was tested by administering doses of 500 or 250 ppm to rats for 110 weeks. Significantly elevated incidences of renal cell tumors in males and females and mesotheliomas of the peritoneum in males as compared to controls were observed. When female mice were given KBrO3 at doses of 1000 or 500 ppm for 78 weeks, no significant differences in tumor incidences between experimental and control groups were apparent. NaClO was administered to male and female rats, respectively, at doses of 1000 or 500 ppm and 2000 or 1000 ppm for 104 weeks. In mice, NaClO was given at doses of 1000 or 500 ppm to either sex for 103 weeks. The incidences of tumors in NaClO-treated and control animals of both sexes were not significantly different in both rat and mouse studies. NaClO2 was given to rats of both sexes at a dose of 600 or 300 ppm for 85 weeks. No statistically significant differences were observed in the incidences of tumor formation between NaClO2-treated and control groups of both sexes. NaClO2 was administered to mice at a concentration of 500 or 250 ppm for 85 weeks. In males, the combined incidences of hyperplastic nodules and hepatocellular carcinomas of the liver in a low-dose group, and adenomas and adenocarcinomas of the lung in a high-dose group, were marginally increased compared to controls (p less than 0.05). However, these incidences in treated males were within the range of values of historical control data in our program. We concluded that KBrO3 was carcinogenic in rats of both sexes. NaClO was not carcinogenic in either rats and or mice under the conditions of the present studies. Although NaClO2 was shown to be noncarcinogenic in rats, the results for mice were evaluated as inconclusive. Also the results of two-stage mouse skin carcinogenesis using KBrO3, NaClO, and NaClO2 are presented. The necessity for further testing of oxidant chemicals to determine potential carcinogenic and/or promoting effects is suggested in view of the recently proposed role of active oxygen species in carcinogenesis.