Nitrogen fertilizer need of corn can vary widely within fields. Accurate diagnosis of N need can ensure full grain yields while reducing fertilizer bills and environmental impacts. Canopy reflectance is related to N status of corn, and can be measured rapidly and with high spatial resolution. Our objective was to calibrate reflectance measurements to predict N fertilizer need. Small-plot experiments were conducted in production corn fields. N rates from 0 to 330 kg N ha-1 were surface-applied as ammonium nitrate. Reflectance measurements were taken with a Cropscan MSR87 passive radiometer at three or four timings in each experiment, starting at V6-7 and ending by V13. Economically optimal N rate (EONR) was calculated from yield response data and related to reflectance using regression analysis. Highly significant relationships were found between relative reflectance (refelectance from unfertilized or moderately fertilized plots divided by reflectance from highly fertilized plots) and EONR at all stages. Relative green/near-infrared reflectance appears to be promising for predicting EONR. This calibration may be useful for controlling variable-rate N fertilizer applicators equipped with reflectance sensors.