Taxonomic patterns of protozoan community structure for monitoring water quality were studied from May to September 2003 in Songhua River, northeast China. Protozoan communities were sampled monthly using PFU (polyurethane foam unit) method at the four stations with different pollution/ eutrophication levels. Physical-chemical parameters (e.g., water temperature, dissolved oxygen (DO), pH, biological oxygen demand (BOD 5 ), chemical oxygen demand (COD), ammonium nitrogen (NH 4 -N), and nitrate nitrogen (NO 3 -N) were measured synchronously for comparison with biotic parameters. A total of 53 protozoan taxa with nine common species were identified comprising 33 ciliates, 17 flagellates, 3 sarcodines. Multivariate analyses demonstrated that the spatial taxonomic patterns of protozoan communities significantly correlated with environmental conditions, especially related to the concentrations of COD, BOD 5 , NO 3 -N, and NH 4 -N, either alone or in combination with one another. Although species richness was found to be independent of physicalchemical parameters, the average taxonomic distinctness (D ? ) were positively significantly correlated with the concentration of dissolved oxygen and NO 3 -N, but negatively with COD and BOD 5 . It is suggested that spatial pattern of protozoan communities and taxonomic biodiversity can be used in assessing water quality of flowing river systems.