Sources of variability in plant toxicity testing.

Published literature is investigated regarding the response of plants to various substances to determine the sensitivity of agricultural plants versus other species, the similarity of effects seen at different taxonomic levels, sensitivity of plants growing outdoors versus in a greenhouse, and the sensitivity of different measurement endpoints. We find that agricultural species are not consistently more or less sensitive to the herbicides tested than non-crop species. Genus and family taxonomic groupings may show similar responses among species, but this similarity quickly decreases as the comparison progress between orders and classes. Results from field and greenhouse studies are less in agreement between studies than data from the other topics. Shoot length will be affected at concentrations lower than for other vegetative endpoints for most species tested for inorganic substances, but for organic substances root and shoot mass were more sensitive. Overall, there is no one species or endpoint that is consistently the most sensitive for all species or all chemicals in all soils, and differences in bioavailability among compounds may confound comparison of test results. Therefore, species sensitivity distributions, adjusted for bioavailability when possible, should be considered in order to better evaluate effects to non-target terrestrial plants.