Dose-response studies of pollutants or toxicants traditionally assume either a linear, a threshold or a quadratic model. In these typical models, low doses are assumed to display, respectively, either a proportional response to high doses, or no response, or a proportionally lesser response at low doses than at high doses. Hormesis, also described as the u-shaped dose-response, differs from the above models primarily in the low-dose range, where the response to toxicants is actually positive rather than minimal or nonexistent, while remaining negative at high doses. The ability to respond positively to toxicants at low levels may be due to an evolutionary adaptation to sudden, ephemeral changes in the environment (i.e. any stressor agent), and the need to compensate for these rapid changes; possibly leading to an overcompensation response. Using Bromus rubens, a grass native to Europe but also found along the Western coast of the United States, we conducted a greenhouse study testing the effects of serial dilutions of glyphosate (full recommended dose, 1 x 10-3, 1 x 10-4, 1 x 10-5, 1 x 10-6) on growth rate, growth patterns, root, shoot, inflorescence and seed biomass to determine if glyphosate did indeed cause a hormetic effect on Bromus rubens growth rates at low doses.
Results showed that apparent hormetic effects were present throughout the first three serial dilutions (1 x 10-3, 1 x 10-4, 1 x 10-5), to an extent in terms of height and leaf production, and more markedly in terms of biomass, with strongest effects at the 1 x 10-3 dose range, when compared to controls. The full dose treatment was effective in halting growth and causing plant death. Morphological observations throughout the study indicated thicker, wider leaves at doses within the hormetic range. In some cases, increased biomass production seems to have been accompanied by delayed seed production, which may be an indication of an ecological tradeoff. Overall results demonstrated the presence of a hormetic effect at low doses, and thus that a u-shaped dose-response model, rather than one of the traditional models, fits the data more accurately, at least in the low-dose range.