Spatial variability in economic optimum N rate (EONR) for corn (Zea mays L.) is not often considered as the primary criterion for site-specific N management. Our objective was to characterize the spatial variability in EONR for corn, considering the implications of soil water characteristics that affect EONR variability along a hillslope. Ten locations were selected along a 300-m toposequence in central Pennsylvania. At each location, two reps of six N treatments were applied at planting. Soil water content (0-90 cm) was recorded weekly at each location between 23 June and 1 September. Grain yield was determined at harvest. A quadratic-plateau model was used to describe the yield response to N and determine EONR at each location. A response to N fertilizer was observed at nine of 10 locations. The EONR ranged from 47 to 188 kg N ha^-1 among the nine locations, while EONR for the mean response was 137 kg N ha^-1. At four of nine locations, observed EONR deviated from field-mean EONR by 40 to 50 kg N ha^-1.  Although EONR increased from 68 to 141 kg N ha^-1 as grain yield observed at EONR increased from 11.1 to 13.1 Mg ha^-1 (r²=0.43), the relationship between EONR and the change in soil profile water content (0-90 cm) between 30 June and 25 July (representing the driest and wettest soil conditions early in the growing season) was quite strong (r²=0.92). Soil water redistribution during this critical part of the growing season was implicated in the observed differences in EONR. Successful site-specific N management will depend on an evaluation of the spatial variability in EONR and the corresponding causal factors of EONR variability, which may or may not be related to the spatial yield variability