Our previous cytogenetic biomonitoring of a group of inhabitants in a village (Mellery, Belgium) where exposure to a mixture of toxic environmental pollutants, (probably originating from a neighbouring chemical waste disposal site) was suspected, showed that difference in the SCE and HFC bioassays was more pronounced for children. The results of follow-up study in 1992 confirmed this surprising conclusion by an even higher incidence. As very few studies have been performed on the levels of children's biomarkers, this group of exposed populations needed to be explored further. Do children residing in the vicinity of hazardous waste sites indeed represent a population at higher risk? In the present study, we compare the performance of various bioassays (SCE, HFC, SSB and MN) in extended exposed and reference children's groups. Simultaneously, in the exposed group, we followed variation in the lymphocyte SCE frequencies as a function of time. Reversibility of the latter biomarker was ascertained subsequent to a preliminary technical remediation of the disposal site. We compared these data with those obtained from a synchronous cross-sectional study on a group of children living near a similar chemical disposal site. The two exposed populations did not differ from the reference population regarding to the SCE and HFC mean levels. Comparisons of the mean levels of the two other biomarkers, SSB and MN, showed no difference between the Mellery exposed children and the reference group from Wavre whereas significant differences appeared when the Hensies group is compared either to the Mellery or to the Wavre reference group.