Information is lacking on the long-term impact of tillage systems on NO3 losses to surface and groundwater. An 11-yr (1982-1992) study was conducted to assess NO3 losses to subsurface, tile drainage for corn (Zea mays L.) grown with continuous conventional tillage (CT) and no tillage (NT) on a poorly drained Webster clay loam soil (fine-loamy, mixed, mesic Typic Haplaquoll) at Waseca, MN. Nitrogen was applied at an annual application rate of 200 kg ha(-1). Mean annual subsurface drain dow during the 11-yr period was 35 mm higher for NT (315 mm) compared with CT (280 mm). Flow-weighted nitrate-nitrogen (NO3-N) concentrations increased dramatically in the wet gears (1990 and 1991) following the dry period of 1987 to 1989. Flow-weighted NO3-N concentrations during the 11-yr period averaged 13.4 and 12.0 mg L(-1) for CT and NT, respectively. Although subsurface drain Bow was 12% higher with NT, NO3-N losses were about 5% higher with CT mainly due to higher NO3-N concentrations with CT in the last 2 yr. Corn grain yields and N removal were significantly higher in 6 out of 11 yr with CT compared with NT with no difference between tillage systems in the other 5 yr. Grain yields averaged 8.6 Mg ha(-1) with CT and 7.3 Mg ha(-1) with NT during the 11-yr period. Multiple regression equations showed that annual Bow-weighted NO3-N concentration is best predicted from residual son NO3 in the 0- to 1.2-m profile and spring rainfall while NO3-N flux can be predicted wed from May and June rainfall. Results from this long-term study indicate that on this poorly drained soil, CT had a positive effect on corn grain yield and N removal compared with NT, but tillage systems had minimal impact on NO3 losses to subsurface drain Bow. Higher drain Bow with NT does not necessarily result in higher NO3-N fluxes lost via subsurface drainage.