Soil nitrogen mineralization potential (Nmin) has to be spatially quantified to enable farmers to vary N fertilizer rates, optimize crop yields, and minimize N transfer from soils to the environment. The study objectives were to assess the spatial variability in soil Nmin potential based on clay and organic matter (OM) contents and the impact of grouping soils using these criteria on corn grain (Zea mays L.) yield, N uptake response curves to N fertilizer, and soil residual N. Four indicators were used: OM content and three equations involving OM and clay content. The study was conducted on a 15-ha field near Montreal, Quebec, Canada. In the spring 2000, soil samples (n = 150) were collected on a 30- x 30-m grid and six rates of N fertilizer (0 to 250 kg N ha-1) were applied. Kriged maps of particle size showed areas of clay, clay loam, and fine sandy loam soils. The Nmin indicators were spatially structured but soil nitrate (NO3) was not. The N fertilizer rate to reach maximum grain yield (Nmax), as estimated by a quadratic model, varied among textural classes and Nmin indicators, and ranged from 159 to 250 kg N ha-1. The proportion of variability (R2) and the standard error of the estimate (SE) varied among textural groups and Nmin indicators. The R2 ranged from 0.53 to 0.91 and the SE from 0.13 to 1.62. Corn grain N uptake was significantly affected by N fertilizer and the pattern of response differed with soil texture. For the 50 kg N ha-1 rate, the apparent Nmin potential (ANM) was significantly larger in the clay loam (122 kg ha-1) than in the fine sandy loam (80 kg ha-1) or clay (64 kg ha-1) soils. The fall soil residual N was not affected by N fertlizer inputs. Textural classes can be used to predict Nmax. The Nmin indicators may also assist the variable rate N fertilizer inputs for corn production.