Abstract: Plants represent uncommon targets to evaluate pharmaceuticals toxicity. In this work, Lemna minor was employed as a plant model to determine the toxicity of selected pharmaceuticals, and to assay if such toxicity could be predicted by QSAR models based on green algae. Among eight compounds, measurable toxicity was determined for ketoprofen (EC₅₀ = 11.8 ± 1.9 mg/L), fluoxetine (EC₅₀ = 27.0 ± 8.7 mg/L) and clindamycin 2-phosphate (EC₅₀ = 57.7 ± 1.7 mg/L). Even though a correlation of r² = 0.87 was observed between experimental toxicity towards algae and L. minor, QSAR estimations based on algae data poorly predicted the toxicity of pharmaceuticals on the plant. More experimental data for L. minor are necessary to determine the applicability of these predictions; nonetheless, these results remark the importance of measuring experimental ecotoxicological parameters for individual taxa. The toxicity of pharmaceutical binary mixtures (ketoprofen, fluoxetine and clindamycin) revealed in some cases deviations from the concentration addition model; nonetheless these deviations were small, thus the interactions are unlikely to be of severe biological significance. Moreover, the EC₅₀ concentrations determined for these pharmaceuticals are significantly higher than those detected in the environment, suggesting that acute effects on L. minor would not take place at ecosystem level.