The ability of the commercial polybrominated diphenyl ether (PBDE) preparation Bromkal 70-5 DE to alter thyroid hormone and vitamin A levels as well as microsomal enzyme activities was compared with that of the commercial polychlorinated biphenyl (PCB) preparation Aroclor 1254 in orally exposed female rats (Sprague-Dawley) and mice (C57BL/6 N). Additional mice were exposed to the PBDE congener 2,2',4,4'-tetrabromodiphenyl ether (DE-47), or to the PCB congener 2,3,3',4,4'-pentachlorobiphenyl (CB-105). For 14 days the animals were given approximately isomolar daily oral doses of Aroclor 1254, CB-105 (both 10 mg/kg body weight), Bromkal 70-5 DE or DE-47 (both at 18 mg/kg body weight). In addition, further groups of rats and mice received a higher dose of Bromkal 70-5 DE, 36 mg/kg body weight. Bromkal 70-5 DE and DE-47 decreased plasma free and total thyroxine (T4) levels in both rats and mice, although with lower potency than that of Aroclor 1254 and CB-105. By contrast, thyroid-stimulating hormone (TSH) levels were not significantly changed in any of the groups. Reduction of hepatic vitamin A levels was seen in rats after Aroclor 1254 and Bromkal 70-5 DE exposure. A similar tendency was seen also in mice, but the effects were significant only for concentration data and not the total amount. Induction ofmicrosomal phase I enzymes, measured as ethoxy, methoxy and pentoxy resorufin O-dealkylase (EROD, MROD, PROD) activities, was greatest after exposure to Aroclor 1254/CB-105 but were also significant in the Bromkal 70-5 DE/DE-47-treated groups. However, induction of uridine diphosphoglucuronosyl transferase (UDPGT) was small and for most groups insignificant. In conclusion, the PBDE compounds studied, although having a lower potency than the PCB compounds, decreased thyroxine and vitamin A levels and induced microsomal enzyme activities. Rats were more sensitive to the observed effects than mice. Microsomal phase I activity might be related, directly or indirectly, to the T4 and vitamin A effects, whereas several factors (such as weak enzyme induction and lack of correlation with altered T4 and vitamin A levels) argue against any UDPGT-related effects.