To assess photosynthesis as a bioindicator of toxicity from polycyclic aromatic hydrocarbons (PAHs), the response of chlorophyll-a fluorescence to creosote exposure was compared with effects on population-level plant growth. Large, outdoor, freshwater microcosms containing Eurasian watermilfoil (Myriophyllum spicatum) received either a single application or multiple applications of liquid creosote at nominal concentrations from 0.109 to 32.7 mg L(-1). For several weeks following treatment, photosynthetic electron transport was measured using pulse amplitude-modulated chlorophyll-a fluorescence. The maximum efficiency of photosystem II electron transport (Fv/Fm) and the quantum yield of photochemistry (deltaF/F'm) were diminished in plants shortly after the addition of creosote. In microcosms that received a single treatment of creosote only, the 50% effective concentrations (EC50s), expressed as the aqueous concentration of 15 of the most abundant PAHs, were 0.28 mg L(-1) for Fv/Fm and 0.30 mg L(-1) for deltaF/F'm. Chlorophyll-a fluorescence was diminished to a greater extent in microcosms that received multiple treatments of creosote, with EC50s of 0.13 mg L(-1) for Fv/Fm and 0.10 mg L(-1) for deltaF/F'm. Plant biomass accumulation was inhibited in a concentration-dependent manner in all microcosms treated with creosote, but this inhibition occurred to a greater degree in microcosms treated with multiple creosote applications. The response of chlorophyll-a fluorescence, measured only 8 d after creosote treatment, was similar to plant growth over the entire growing season, indicating that this technique can be used to estimate potential effects of contaminants before detrimental impacts on populations.