One explanation for unexpectedly widespread ground water contamination from atrazine (2-chloro-4-ethylamino-6-isopropylamino-s-triazine) may be the occurrence of colloid-facilitated transport, whereby the dissolved herbicide becomes adsorbed to mobile colloids that migrate through preferential flow-paths in the soil zone and into the ground water. The objectives of this study were to determine the extent of adsorpton of atrazine to bulk soil and to soil colloids and to determine the extent of colloid-facilitated transport of atrazine at a field site in Virginia during simulated rainfall events. Equilibrium batch adsorption experiments were performed over a concentration range of 0.05 to 10.0 mg atrazine L-1 on bulk soil samples and on colloidal suspensions of 75 mg L-1, a concentration comparable with those observed at the field site. Linear partition coefficients ranged from 0.496 to 2.48 L kg-1 for the bulk soil and from 70.8 to 832 L kg-1 for the soil colloids. In the field, gravity lysimeters were insured at a depth of 25 cm below the surface of six 0.25-m2 undisturbed plots. Mass recovery of surface-applied atrazine in the lysimeters was not significantly affected by rainfall rate and was, on average, 2.7% for plots receiving 25 mm h-1 simulated rainfall and 3.6% for plots receiving 50 mm h-1 simulated rainfall. Of the total atrazine collected in the lysimeters, the fraction that was colloid-associated ranged from 4.9 to 30% (mean of 15%), indicating that a measurable portion of mobile atrazine is transported via association with colloids.One explanation for unexpectedly widespread ground water contamination from atrazine (2-chloro-4-ethylamino-6-isopropylamino-s-triazine) may be the occurrence of colloid-facilitated transport, whereby the dissolved herbicide becomes adsorbed to mobile colloids that migrate through preferential flow-paths in the soil zone and into the ground water. The objectives of this study were to determine the extent of adsorption of atrazine to bulk soil and to soil colloids and to determine the extent of colloid-facilitated transport of atrazine at a field site in Virginia during simulated rainfall events. Equilibrium batch adsorption experiments were performed over a concentration range of 0.05 to 10.0 mg atrazine L-1 on bulk soil samples and on colloidal suspensions of 75 mg L-1, a concentration comparable with those observed at the field site. Linear partition coefficients ranged from 0.496 to 2.48 L kg-1 for the bulk soil and from 70.8 to 832 L kg-1 for the soil colloids. In the field, gravity lysimeters were installed at a depth of 25 cm below the surface of six 0.25-m2 undisturbed plots. Mass recovery of surface-applied atrazine in the lysimeters was not significantly affected by rainfall rate and was, on average, 2.7% for plots receiving 25 mm h-1 simulated rainfall and 3.6% for plots receiving 50 mm h-1 simulated rainfall. Of the total atrazine collected in the lysimeters, the fraction that was colloid-associated ranged from 4.9 to 30% (mean of 15%), indicating that a measurable portion of mobile atrazine is transported via association with colloids.In the Muddy Creek watershed, VA, the extent of atrazine adsorption to the immobile soil matrix and mobile soil colloids was examined in an agricultural silt loam soil, and atrazine transport during simulated rainfall events of different intensities was assessed. Results from batch equilibration experiments revealed that atrazine adsorption was significantly greater on the colloids than on the bulk soil, which was attributed to the higher specific surface area and organic content of the colloids. Despite the mobilization of colloids, however, the mass recovery of colloid-associated atrazine was low for all plots. Precipitation rate did not significantly affect atrazine mass recovery.