Quantitative structure activity relationships (QSARs) can be used to predict biological responses to a chemical on the basis of physicochemical descriptions. Since polycyclic aromatic hydrocarbons (PAHs) are highly photoactive by virtue of their fused aromatic ring structures, the degree and rate of photoreactivity can be significant modifiers of their toxicity. In general, photoinduced toxicity of PAHs is expressed through two mechanisms: photosensitization, which results in production of highly reactive singlet oxygen, and photomodification, which alters the structure of the molecules changing their toxicity. We developed a preliminary model, based on summation of a photosensitization factor and a photomodification factor, to predict the chronic toxicity of six PAHs at 2 μg mL-1 to the duckweed, Lemna gibba G3, under simulated solar radiation. The model, which was significant (p = 0.004) and had a high correlation coefficient (r = 0.99), shows growth inhibition to be proportional to the sum of the normalized photomodification rate and a factor for the photosensitization parameters of each chemical.