We studied metabolism of monochlorobenzene (MCB), 1,2-dichlorobenzene (1,2-DCB) and 1,4-DCB in liver microsomes from untreated male and female Wistar rats and B6C3F1 mice or in those after the induction of CYP3A or 2E1 as well as in human male liver microsomes. MCB and 1,2-DCB were oxidised mainly by rat and human CYP2E1. It was found that 1,4-DCB was oxidised by rat and human CYP2E1 at a several-fold lower rate than 1,2-DCB, but a greater part to covalently binding products. In contrast to previous studies showing rat CYP3A1 as the main CYP form oxidising both DCBs, our experiments indicate only a certain role of rat and human CYP3A in MCB, 1,2-DCB and 1,4-DCB oxidation to covalently bound products. The relative roles of human liver CYP2E1 and 3A4 in the metabolism of 1,4-DCB seem to be individually different. Metabolic rates of MCB, 1,2-DCB and 1,4-DCB correlated with CYP2E1 immunochemical level in microsomes from 11 different human livers and with metabolic rates of CYP2E1 substrates. These rates in different human livers were up to 10-fold different and were generally several-fold higher than those in untreated rats or mice. Metabolic activation of MCB and 1,2-DCB to products binding covalently to microsomal proteins and to calf-thymus DNA, respectively, mostly corresponded to production of water-soluble metabolites. Significant species and sex differences in the oxidation of MCB, 1,2-DCB and 1,4-DCB were reflected in a markedly higher oxidation in male mice than male rats and higher oxidation in male than female mice. The formation of covalently bound products generally corresponded to production of soluble metabolites, but female rats formed significantly less covalently bound products of 1,4-DCB (and also of 1,2-DCB and MCB) than male rats and mice of both sexes, in possible reflection of the fact that 1,4-DCB is not carcinogenic in female rats despite its carcinogenicity for male rats and both sexes of mice.