Acute toxicity of pesticides in water was assessed singly and in mixtures using the responses of the luminescent bacterium Vibrio fischeri (BioTox(TM)), the aquatic invertebrate Daphnia magna (Daphtoxkit(TM)), and the MitoScan(TM) assay. The latter utilized fragmented mitochondria to enzymatically convert beta-nicotinamide adenine dinucleotide (NADH) to its oxidized form, NAD+. The rate of the conversion being sensitive to type and concentration of toxicants. The pesticides tested were Carbofuran (2,3-dihydro-2,2-dimethylbenzofuran-7-yl methylcarbamate), Cyromazine (N-cyclopropyl- 1,3,5-triazine-2,4,6-triamine), Fenamiphos (ethyl 4-methylthio-m-tolyI isopropylphosphoramidate), and Formetanate (3-dimethylaminomethyleneinlinophenyl methylcarbamate). The toxicity bioassays were characterized in terms of relative sensitivity, reproducibility, range of the linear response, and the ability to reveal synergistic/antagonistic interactions among toxicants. The D. magna assay was the most sensitive and best able to detect toxic interactions of mixtures. Also, unlike the other assays used, the response of the daphnid system was linear over a 10-fold change in pesticide concentration. Relative to the BioTox(TM), the MitoScan(TM) was 2- to 11-fold more sensitive for the compounds and mixtures tested. The EC50 reproducibility of all tests was within 20% coefficient of variation; however, the lowest observable effect concentration (LOEC) were only reproducible to +/-35% on average. Cyromazine was the least toxic of the pesticides tested. To test the predictive value of the concept of concentration addition, toxicities of binary and quaternary mixtures of four different pesticides were analyzed. Synergistic/antagonistic responses were most frequently observed in testing with D. magna. Synergistic/antagonistic effects were seen only in 25 and 50% of the cases with the BioTox(TM) and the MitoScanT(TM) assays, respectively, (C) 2002 Elsevier Science B.V. All rights reserved.