This study was conducted to field-test the vegetation nomogram the US Environmental Protection Agency uses to estimate maximum expected concentrations of pesticide residues on plants when conducting ecological risk assessments. The objectives were to test whether residue concentrations were conservatively predicted by the nomogram and to determine the influence of vegetation structure on residue distribution and degradation. Azinphos-methyl was applied at 0, 0 88 and 3 61 kg/ha to 24 0 2-ha enclosures planted with alfalfa (Medicago sativa). Alfalfa in 12 enclosures was mowed in order to reduce vegetation height three weeks before the chemical was applied. Residue concentrations were measured on canopy-level spray cards, soil samples and alfalfa on days 0, 2, 6, 14 and 28 after spraying. Deposition of azinphos-methyl on spray cards was generally lower than the measured application rates. Residue concentrations increased proportionately with application rate. However, the nomogram underestimated mean concentrations on mowed alfalfa and unmowed alfalfa tops. Residue concentrations on the top 15 cm of unmowed alfalfa were ≥4 times greater than concentrations on the bottom 15 cm, indicating that much of the insecticide was intercepted in the canopy of unmowed alfalfa. Residue concentrations tended to be higher on mowed alfalfa than on unmowed alfalfa tops. The half-life of azinphos-methyl on alfalfa ranged from 1 7 to 5 1 days. Apparently, mowing did not affect residue persistence. The results indicate that the vegetation nomogram may underestimate pesticide residue concentrations on alfalfa and that variation in vegetation structure may substantially influence exposure of herbivores to pesticides.