Adsorption and desorption isotherms of Simazine on a decomposed granitic (DG) soil, two natural loam soils, and selected soil particle size fractions were measured in batch experiments. Soil-water content (air-dry compared to oven-dry) markedly influenced the amounts of simazine adsorbed and desorbed on the low-organic DG soil and its soil size fractions while only slight effects were observed for the higher-organic loam soils and their soil size fractions. Decreasing adsorption and increasing desorption with increasing pH was observed for all soils. For both DG and loam, the largest soil size fraction (1.0-2.0 mm) was the most effective adsorber based on unit surface area. The Freundlich isotherm equation well described the measured adsorption and desorption data. The Freundlich adsorption and desorption coefficients (k_{f, a} and k_{f, d}) were both highly correlated with specific surface area for the DG and soil organic matter content for the loam. Also, k_{f, d} was highly correlated with k_{f, a} Three step desorption experiments gave a final desorption of simazine from the DG soil and its size fractions between 91-95 % of the total amount initially adsorbed compared to only 12-22 % for the loam soils and their size fractions. Minor sorption hysteresis was observed for DG. Large hysteresis effects were seen for the loam soils where the hysteresis index was controlled by the soil organic matter content. The difference in sorption characteristics and hysteresis for the two soil types was confirmed in micro-column transport experiments, where significant retardation and tailing of the simazine effluent concentration curve was observed only for the loam soil.