To evaluate human exposure, experimental data are preferred, especially when these data mimic the desired conditions closely, are well-controlled, and are measured reliably. However, if the conditions of the experiment do not match the desired scenario as closely as one would like, limitations and uncertainties may exist. In such cases, mathematical models are appealing as one can construct them to address parameters or measures that may not be possible in a real-world experiment. For this analysis, the experimental results from the Maine study (Stahler et al., 2008) are preferable to a simple mathematical model because the scenarios being addressed closely resemble the experimental data. These data can be compared to selected human health toxicity benchmarks.