A closed-system algal toxicity test with no headspace was applied to evaluate the toxicity of chlorophenols to Pseudokirchneriella subcapitata. The dissolved oxygen production and the growth rate based on cell density were the response endpoints. Phenol and seven chlorophenols were tested using the above test technique. Median effective concentrations (EC50) range from 0.004 to 25.93 mg/l (based on DO production) and 0.0134 to 20.90 mg/l (based on growth rate). No-observed-effect concentration (NOEC) is within the range of 0.001-8.19 mg/l. In general, growth rate is a more sensitive response endpoint than the oxygen production, except for the case of pentachlorophenol. However, the differences in sensitivity between the two parameters were marginal. Furthermore, quantitative structure-activity relationships (QSAR's) based on the n-octanol/water partition coefficient (log P) and the acid dissociation constant (pK(a)) values were established with R(2) ranged from 0.90 to 0.96. From literature data also based on P. subcapitata, the new test method is 1.65-108 times more sensitive than the conventional algal batch tests. A completely different relative-sensitivity relationship among various aquatic organisms was thus observed. The results of this study indicate that the toxicity data of volatile organic chemicals derived by conventional algal toxicity tests may severely underestimate the impact of these toxicants. Our results show that alga is very sensitive to chlorophenols compared to other aquatic organisms such as the luminescent bacteria (the Microtox test), Daphnia magna, and rainbow trout.