In this study, UV based treatments were implemented at pilot scale to assess their ability to remove hormones from treated wastewater, especially with the view to equip small and medium size Wastewater Treatment Plants (WTPs). To this end, the degradation of a mixture of estrogenic hormones (Estrone (E1), β-Estradiol (E2), and 17α-Ethinyl Estradiol (EE2)) in waters by UV photolysis and UV/H2O2 process was investigated in real conditions. A particular attention was paid at designing a well validated laboratory scale pilot in order to optimise oxidant concentrations and UV fluence. A Low pressure lamp (254 nm) was used in a flow through commercial reactor. The effects of water matrices (drinking water and treated wastewater) and H2O2 concentrations (10, 40, and 90 mg/L) on the pilot efficiency were first determined. Only E1 could be partially degraded by UV photolysis whereas hormones were all well removed by UV/H2O2 process in both matrices. The second part of the study focused on a chemical and biological assessment of UV photolysis and UV/H2O2 process (30 and 50 mg/L). Degradation rate constants of hormones as well as changes in estrogenic activity (YES bioassay) and toxicity (Vibrio fischeri) were followed at the same time. UV photolysis could not remove neither estrogens nor estrogenic activity at relevant UV fluence in waters. However 80% of initial estrogenic compounds and estrogenic activity could be removed from treated wastewater by combining UV fluence of 423 and 520 mJ/cm(2) with 50 and 30 mg/L of H2O2, respectively. No high estrogenic or toxic by-products were detected by the two bioassays following UV photolysis or UV/H2O2 process. Operating costs were estimated for a full scale pilot. H2O2 was the major cost. By combining the appropriate concentration of H2O2 and UV fluence, it could be possible to design a cost effective treatment for treating estrogens in small and medium size WTPs