An understanding the relationships between N-fertilization rate, crop-N recovery, residual inorganic N in soil and crop yield can assist in making N fertilizer recommendations which balance crop value and environmental risk. Nitrogen was applied to three plantings of broccoli at 0, 125, 250, 375, 500 and 625 kg N ha(-1) as a broadcast application of ammonium nitrate, 50% at transplanting and 50% 2 wk later. Crop-marketable yield, aboveground dry matter yield and N accumulation, and extractable-soil inorganic N to 75 cm depth were determined at harvest. Crop-marketable yield, measured as fresh weight of heads, increased with increasing N rate to a maximum between N rates of 435 and 560 kg N ha(-1). Nitrogen accumulation in the aboveground portion of the plant increased with increasing N rate and reached a maximum of 345-465 kg N ha(-1). Apparent fertilizer-N recovery in the aboveground portion of the plant decreased linearly from between 46 and 93% at a N rate of 125 kg N ha(-1) to between 20 and 44% at a N rate of 625 kg N ha(-1). Apparent fertilizer-N recovery in the harvested portion of the plant decreased linearly from between 14 and 25% at a N rate of 125 kg N ha(-1) to between 8 and 14% at a N rate of 625 kg N ha(-1). Soil profile nitrate-N to 75 cm depth at harvest increased with N application rate in a curvilinear fashion, and averaged 3 kg N ha(-1) with no N applied to 420 kg N ha(-1) at the maximum N rate. The potential to develop a soil nitrate test to predict the fertilizer-N requirement of broccoli is discussed.