Abstract-A series of experiments was conducted to assess the level of variability in phytotoxicity testing and to investigate factors that may explain some of the observed uncertainties and inconsistencies. The work was conducted in greenhouse or growth chamber environments with plants growing individually in pots and harvested 28 d after spraying with two herbicides, glyphosate and atrazine, as formulated products. Between six and 10 doses were used on five or six replicates, necessitating over 4,500 individually growing plants.In the first set of experiments, several ecotypes (originating from different areas of the world) of eight wild plant species were tested. Significant differences in sensitivity to atrazine and glyphosate were found among ecotypes of most species tested. In the second suite of experiments, the reproducibility of results during different seasons (when growing conditions vary) was investigated using three crops and four wild plant species. Results showed that seasonal variability elicited a pronounced discrepancy in response between plants tested at different times of the year. It was found that no consistent effects could be attributed to the biotic or abiotic factors investigated. Several ecotypes of the same species differed in their seed size, percentage germination, or germination requirements, as well as in growth patterns, but these differences could not explain differences in herbicide sensitivity. Likewise, differences in phytotoxicity could not be attributed to factors such as temperature, light intensity, and sunlight duration. The present study supports the inclusion of an uncertainty factor in risk assessments to account for the intrinsic variability in plant sensitivity to herbicides. Environ. Toxicol. Chem.