Superficial (0 to 2 cm) sediments were sampled from 62 sites in Kattegat and Skagerrak during autumn 1989 and spring 1990, tested for toxicity to Daphnia magna and Nitocra spinipes (Crustacea) and analyzed for heavy metals (Cd, Cr, Cu, Hg, N, Pb, Zn), nutrients (N and P) and organic carbon. Whole sediment toxicity to Nitocra spinipes, expressed as 96-h LC50, ranged from 1.8 to > > 32 percent sediment (wet wt), which is equivalent to 0.63 to 53 percent dry wt. Sediment total metal concentrations (mg kg-1 dry wt) ranged from 0.01 to 0.32 for Cd, 8 to 57 for Cr, 3 to 40 for Cu, 0.03 to 0.86 for Hg, 3 to 43 for Ni, 6 to 37 for Pb and 21 to 156 for Zn. Analyzed concentrations of heavy metals were tested for correlation with whole sediment toxicity normalized to dry wt, and significant correlations (Spearman p < 0.05) were found for Cd, Cr, Cu, Hg, and Ni. However, the analyzed concentrations of these metals were below the spiked sediment toxicity of these heavy metals to N. spinipes, except for Cr and Zn for which analyzed maximum concentrations approached the 96-h spiked sediment LC50s. There was no improvement in correlation between the sum of heavy metal concentrations normalized to their spiked toxic concentrations (Toxic Unit approach) and the whole sediment toxicity. Calculated heavy-metal-derived toxicity based on toxic units and whole sediment toxicity ranged from 0.1 to 24 (mean value 2.3 and SD 4.2). Theoretically, a value of 1.0 would explain whole sediment toxicity from measured metal concentrations using this approach. Thus, in spite of the fact that the total concentrations of the heavy metals were sufficient to cause toxicity based on an additive model for most of these sediments, the observed toxicity of the sediments from Kattegat and Skagerrak could not exclusively be explained by the concentrations of heavy metals, except for Cr and Zn at their maximum concentrations. Therefore, other pollutants than these heavy metals must also be considered as possible sediment toxicants.