Fluoxetine, the active ingredient of many antidepressants, was identified as specifically toxic toward algae in a quantitative structure-activity relationship (QSAR) analysis with literature data for algae, daphnia, and fish. The goal of this study was to elucidate the mode of action in algae and to evaluate the toxicity of the major human metabolites of fluoxetine using two different algae tests. The time dependence and sensitivity of thedifferenteffectendpointsyield information on the physiological mode of action. Baseline toxicity was predicted with QSARs based on measured liposome-water partition coefficients. The ratio of predicted baseline toxicity to experimental toxicity (toxic ratio TR) gives information on the intrinsic potency (extent of specificity of effect). The metabolite p-trifluoromethylphenol was classified to act as baseline toxicant Fluoxetine (TR 60-150) and its pharmacologically active metabolite norfluoxetine (TR 10-80) exhibited specific toxicity. By comparison with reference compounds we conclude that fluoxetine and norfluoxetine have an effect on the energy budget of algal cells since the time pattern of these two compounds is most similar to that observed for norflurazon, but they act less specifically as indicated by lower TR values and the similarity of the effect pattern to baseline toxicants. The mixture toxicity of fluoxetine and its human metabolites norfluoxetine and p-TFMP can be predicted using the model of concentration addition for practical purposes of risk assessment despite small deviations from this model for the specific endpoints like PSII inhibition because the integrative endpoints like growth rate and reproduction in all cases gave agreement with the predictions for concentration addition.