Sediment from 27 river sites in Northern Spain were tested with the aquatic oligochaete Tubifex tubifex (Annelida, Clitellata) 28-day chronic bioassay. Sampling sites were chosen from those established by regulatory water agencies for water-quality surveillance networks in rivers of the Basque Country and the Ebro basin. Inclusion of this test in an assessment programme with chemical and benthic community data currently collected by the water agencies will enable a more comprehensive ecotoxicological assessment. Cocoon and adult biomass were used as end points in addition to percent mortality, number of cocoons, young per adult, and percent of hatch end points as proposed in the American Society for Testing and Materials (ASTM) guidelines. Ecotoxicity assessment of the sediments was carried out by comparing mortality, growth, and reproduction in test sediments with their respective experimental control. Cluster analysis resulted in four groups of sediments that were compared using multidimensional scaling ordination (MDS), and the contribution of toxicity end points and the relationship of chemical variables to the MDS ordination space were assessed by principal component correlation (PCC). A gradient of sediments from nontoxic (4 controls and 6 sediments) to severely toxic (8 sediments) was observed, with all of the original biological end points contributing significantly to the ordination. The most toxic sediments could be separated into 2 groups based on the effects of pollutants on mortality (4 sediments) or on reproduction and growth (4 sediments). Remaining sites were grouped together as toxic sediments but showed a variable degree of sublethal effects. The acceptability criteria for validation of the bioassay, as recommended by the ASTM, measured in the control batch of each bioassay were achieved for survival and cocoon production (coefficient of variation [CV] and variability with regard to their average calculated through laboratory control charts). However, the ASTM criteria were found to be strict for the CV for total number of young, and a more realistic criterion is proposed. New criteria are also suggested to improve quality assurance of the bioassay, namely, a minimum number of cocoons per adult in the control group.