Extraction efficiency for metribuzin [4-amino-6-(l,l-dimethylethyl)-3-(methylthio)-l,2,4-triazin-5(4H)-one] from soil was evaluated using four solutions of methanol:water at ratios of 1:1, 7:3, 4:1, and 1:0 v/v. Two concentrations of metribuzin (0.216 and 2.44 ug/ g soil, unlabeled and I4 C-metribuzin, respectively) were added to surface and subsurface samples of a Dundee silt loam soil. Although the differences in extraction efficiencies were slight due to differences in methanol:water ratios, the metribuzin recovery varied with soil depth. Less metribuzin was recovered from surface soil extracted with 1:1 and 1:0 methanol:water solutions when compared to 7:3 and 4:1 solutions (80 and 81% vs 89 and 88% recoveries, respectively). About equal quantities, 81 to 85%, were recovered from the subsurface soil. Three extraction shaking times (0.5 h followed by another 0.5 h; 4 h:4 h; and 24 h:24 h) were also evaluated using the 4:1 extractant. No recovery differences were observed between the 0.5 h and 4 h shaking times. However, significantly higher recoveries occurred in both the surface and subsurface soil with 24 h shaking. The efficiency of this method was also determined on Eustis loamy sand, Sharkey clay, and Dorovan muck soils of diverse physical and chemical properties. The 4:1 extracting solution consistently 1.