This was a concern, because the large extrapola tion beyond the data adds uncertainty to the BMDL estimate and the large extrapolation had the potential for calculating a BMDL above the threshold for co critical effects that were not amenable to modeling. 3. 3. Step 3: areas of uncertainty in deriving an RfD The U. S. EPA has suggested five different uncertainty fac tors to address issues of variability and uncertainty when deriving RfDs and the panel deliberations surrounding each of these five areas are provided below. 3. 3. 1. Interspecies variability Because no human health effects data or comparative data on toxicokinetics or toxicodynamics between rats and humans is available, a factor of 10 was considered by the panel to be appro priate for UF A, for all degradates. 3. 3. 2.

Intraspecies variability The panels evaluation suggests that not enough information was available to modify the value of 10 for UF H and information does not exist that Opioid Receptor would allow the development of a chemical specific adjustment factor for human variability. The factor of 10 for human variability was considered adequate, since the lim ited absorption and metabolism of the degradates suggest that var iability in toxicokinetics might be lower than for other chemicals. This latter conclusion was consis tent with the fact that an FQPA factor of 1 fold was applied to two related herbicides, indicating that the toxicity database is complete. For acetochlor, an FQPA safety factor has been applied for acute exposures only, for the lack of develop mental neurotoxicity and immunotoxicity tests.

However, Opioid Receptor there is no indication of selective toxicity in either rats or rabbits to in ute ro or post natal exposures for either parent chemicals or degradates. 3. 3. 3. Subchronic to chronic Because only subchronic studies are available for these degra dates, but data were identified that suggest a lack of progression with exposure duration, the factor of 10 was reduced by the panel to a factor of 3. Moreover, this factor was reduced in the context of applying multiple UFs as described below. 3. 3. 4. LOAEL to NOAEL extrapolation Each degradate had at least one adequate study from which a NOAEL was selected as the point of departure. Therefore, a LOAEL to NOAEL extrapolation is not necessary. Thus, a value of 1 was considered by the panel to be appropriate for the UF L. 3. 3. 5.

Database This factor was reduced in the context of applying multiple UF as described below. 3. 4. Selection of chemical specific combined uncertainty factors The UF for subchronic to chronic duration and the UF for data base completeness were reviewed in significant detail, since none of the degradate studies was longer than about 90 days and none of the individual RAF Signaling Pathway degradates had a complete database as defined by U. S. EPA. Possible combinations of these two factors were con sidered since both address issues related to deficiencies in the overall database. The impact of lack of a reproductive toxicity study on the selec tion of the UF for database insufficiency was also carefully consid ered.

Organ weight and histopathology findings reported in the available subchronic studies for these degradates indicated that the reproductive organs were not targets for any of the degradates. In the absence of data on functional reproductive capacity PLK for these degradates, the available studies on the parent chemicals were used to inform the potential impact of this data gap. This approach was considered, with the caveat that the modes of action for the parent chemicals are apparently not the same as for these degra dates, and thus direct comparisons are somewhat limited. On the other hand, such comparison might be considered as overestimat ing the impacts of missing studies, since the parent chemicals were considered more likely to be biologically available and reactive than the degradates.

Comparing the effect levels for reproductive toxicity versus the most sensitive systemic effects for the parent chemicals, the NOAEL for reproductive toxicity in rats for alachlor was 30 fold higher than the chronic dog NOAEL, but no reliable NOAEL from subchronic rat study was available for PARP com parison. For acetochlor, the critical reproductive toxicity NOAEL of 65. 6 mg/kg day in male rats was 30 fold higher than the chronic dog NOAEL of 2 mg/kg day and was about 7 fold lower than the NOAEL of 10 mg/kg day from the subchronic rat study. Thus, for acetochlor, reproductive effects in rats were not the critical effect. Reproduc tive organ effects in dogs were a co critical effect for acetochlor and such effects remain a possibility for these degradates. For a related acetanilide, metolachlor, reproductive, or developmental toxicity was not the critical effect, and the available studies in rats and dogs for the alachlor and acetochlor degradates did not suggest a concern for testicular effects. Because alachlor produced ocular effects, the panel considered the potential for the degradates to produce ocular effects as well. The panel did not judge the ocular and periocular effects observed for alachlor ESA in the drinking water study to be treat ment related.