Several experiments were conducted to evaluate the behavior and performance of some potential endocrine disrupters (ECDs). Two in vitro screening assays, one based on MCF7-cell proliferation (E-screen test) and the other on estrogenic receptor activity [enzyme-linked receptor assay (ELRA)], were used for the tests, which were done in lysimeters 80 cm in diameter with depth of 30 cm (shallow) or 90 cm (deep). A sandy soil was used to fill in all lysimeters, which were spiked on the surface with either: (a) a sewage sludge (SS) at a dose equivalent to 20 tons ha-1; (b) a mixture of reference ECDs, comprising 17 alpha- and 17 beta-estradiol (E2), nonylphenol, octylphenol, and bisphenol A at doses 100 times higher than the maximum concentrations respectively found in the applied SS; or (c) a mixture of ECDs and SS. After percolation of the lysimeters with rain and/or artificial water, five leachates were sampled from each lysimeter during a period of 210 days. Immediately after the lysimeter percolation experiments, four and six soil fractions were dissected from, respectively, the 30-cm and 90-cm lysimeters and extracted by water. Both the leachate and soil extract samples were analyzed for their estrogenicity using the assays indicated above. The E-screen assay was highly sensitive only for some leachate and extract samples but gave no response for most leachates and soil extracts. The results of the ELRA assay suggests a significantly higher estrogenicity of leachate samples from shallow lysimeters compared with that of leachates from deep lysimeters. In contrast, the estrogenic effect measured for soil extracts of shallow lysimeters was lower than that measured for soil extracts of deep lysimeters. The results of the E-screen assay suggests the occurrence of a fast mobilization of applied ECDs and a moderate retardation effect of native ECDs contained in applied SS in the sandy soil used in the lysimeters. In lysimeters spiked with a mixture of SS and ECDs, the washing-out effect of ECDs in the first leachate fraction decreased, but the distribution of ECDs in the lysimeters increased. The relatively high estrogenic impact measured for soil water extracts suggests that the ECDs were mostly associated with water-soluble fractions of organic matter and/or water-suspended fractions of the mineral soil matrix. The application of SS to agricultural and forest fields may determine the immobilization of ECDs in soil or their movement to surface and/or groundwater. Therefore, an endocrine risk of exposure exists for the water and soil organisms.