Dissolved organic matter (DOM) from five different origins decreased the bioconcentration of benzo[a]pyrene (BaP) in the nematode, Caenorhabditis elegans. The decrease became more pronounced with increasing concentrations of DOM, but the effect per mg l(-1) DOC was largest at low levels of DOM, indicating the lack of a simple direct relationship between DOM concentration and the bioconcentration factor (BCF). We tested the hypothesis that the quantitative relationship between DOM concentration and BCF can be described by a theoretically derived equation based on the assumption that only freely dissolved contaminants ape bioavailable (BCF = control BCF x 1/(1 + partition coefficient x DOM concentration)). This equation was used in non-linear regression procedures to fit curves to the experimental data. The resulting regression curves for data from this study (correlation coefficients (r(2)) ranging from 0.80 to 0.94), and for data from the literature (r(2) ranging from 0.62 to 1.00), showed that the model equation was able to correctly describe the relationship between DOM concentration and BCF. The slope of each curve resulted from the 'biologically determined' partition coefficient (K-DOC) that had been estimated by the regression procedure. Thus, the data set for each DOM source was reduced to a single K-DOC value (range: 20 +/- 4 x 10(4) to 49 +/- 6 x 10(4) l kg(-1) DOC (mean +/- S.E.), which allowed to compare different types of DOM, regarding their ability to reduce the bioconcentration of BaP. A comparison of 6,,, values showed that there were clear differences between the effects of DOM from different sources. In summary, we conclude that distinct effects of DOM on the bioconcentration of contaminants can occur at environmentally representative concentrations of DOM, but only for combinations of very hydrophobic contaminants (e.g. BaP) and DOM with a high binding capacity. (C) 1999 Elsevier Science B.V. All rights reserved.