The sorption of unformulated (98% pure) and formulated (86% pure + adjuvants) metolachlor to peat, rubber, and steam-exploded wood was studied. The concentration of pesticide ranged from 50 mg/L to 400 mg/L. Equilibrium concentrations for the batch reactors were reached within 24 hr. The sorption data for both unformulated and formulated metolachlor were best described by the Freundlich isotherm model. Metolachlor was preferentially sorbed in the following order: rubber > peat > wood fibers, with removal efficiencies of 80-85%, 70-80% and 50-65%, respectively. Differences in slopes of the isotherm lines and K values were negligible for each sorbent-type tested for both unformulated and formulated metolachlor. These results suggest that the presence of surfactants did not affect the sorption capacity of the sorbents at the concentration levels tested. However, isotherm plots showed S-shaped curve behavior for all of the sorbents in the presence of unformulated metolachlor, while a C-shaped curve was seen when the sorbents were mixed with formulated metolachlor. This suggests that different bonding mechanisms may be involved for the sorption of unformulated and formulated metolachlor to the sorbents. Various procedures were studied to improve removal efficiencies of formulated metolachlor. Sorption of metolachlor to peat was enhanced by hydrating the peat and pre-treating the peat with HCI. Circulation of formulated metolachlor through a rubber-packed column showed the greatest removal, with only 6 mg/L of the initial 400 mg/L remaining in solution. Removal efficiencies of steamexploded wood fibers were not improved by any of the methods investigated. / Master of Science