A microcosm approach was designed to study the combined effects of tributyltin (TBT) from antifouling paints and ultraviolet-B radiation (UVBR: 280-320 nm), on a natural planktonic assemblage (<150 microm) isolated from the St. Lawrence Estuary at the end of the springtime. Microcosms (9l, cylindrical Teflon bags, 75 cm heightx25 cm width) were immersed in the water column of mesocosms (1800 l, polyethylene bags, 2.3 m depth) and exposed to two different UVBR regimes: natural ambient UVBR (NUVBR), and enhanced level of UVBR (HUVBR). During consecutive 5 days, effects of TBT (120 ng l -1) and enhanced UVBR (giving a biologically weighted UVBR 2.15-fold higher than natural light condition) were monitored in the samples coming from following treatments: (i) NUVBR light condition without TBT (NUVBR), (ii) NUVBR light condition with TBT-added (NUVBR+TBT), (iii) HUVBR light condition without TBT (HUVBR) and (iv) HUVBR light condition with TBT-added (HUVBR+TBT). Each treatment was conducted in triplicate microcosms. Different parameters were then measured during 5 days, including TBT analysis, bacterial abundance and productivity, phytoplankton abundance, cellular characteristics and growth rates, as well as in vivo chlorophyll a (Chl a) fluorescence. Following TBT addition (NUVBR+TBT treatment), Chl a concentrations never exceeded 1 microg l-1 whereas final values as high as 54 microg l-1 were observed in TBT-free treatments (NUVBR and HUVBR). TBT addition resulted also in the lost of fluorescence signal of the maximum efficiency of the photosystem II in phytoplankton assemblage. TBT toxicity caused on phytoplankton <20 microm an increase of mean cell size and changes in shape reflected a drastic disturbance of the cell cycle leading to an inhibition of the apparent growth rate. These negative effects of TBT resulted in a final abundance of phytoplankton <20 microm of 591+/-35 cells ml-1 in NUVBR+TBT relative to NUVBR treatment (i.e., 31,846+/-312 cells ml-1). Moreover, when cells were submitted to TBT under enhanced UVBR (HUVBR+TBT treatment), final abundance of phytoplankton <20 microm was only 182+/-90 cells ml-1, with a significant interaction between TBT and UVBR during the last 2 days of the experiment. The same type of interaction was also observed for bacterial abundance in NUVBR+TBT and HUVBR+TBT with stimulation of 226 and of 403%, respectively due to TBT addition relative to NUVBR treatment. When considering bacterial productivity, TBT addition resulted in an inhibition of 32%, and this inhibition was significantly more pronounced under dual stresses (i.e., 77% in HUVBR+TBT). These results clearly demonstrate that the combination of TBT and UVBR stresses have synergistic effects affecting the first trophic levels of the marine food web.