The risk of chemical mixtures to ecosystems is often assessed by applying the model of concentration addition or response addition combined with species sensitivity distribution (SSD) curves. Mixture effect predictions have been shown to be consistent only when these models are applied for a single species, however, and not with several species simultaneously aggregated to SSDs. The more stringent procedure for mixture risk assessment would hence be to apply first the concentration addition or response addition models to each species separately and, in a second step, to combine the results to construct an SSD for a mixture. Unfortunately, this methodology is not applicable in most cases because the large data sets it requires are usually unavailable. Based on theoretical data sets generated, the authors aimed to characterize the difference that can exist between these 2 methodologies. Results show that the use of concentration addition on SSD directly may lead to underestimations of the mixture concentration affecting 5% or 50% of species, especially when substances present a large standard deviation in ecotoxicity data constructing their SSD. The application of response addition can lead to over- or underestimations, depending mainly on the slope of the dose–response curves of the individual species. When assessing the risk of mixtures, one must therefore keep in mind this source of error when applying concentration addition or response addition to SSDs directly.