A field experiment was carried out to study the effect of nitrogen (N) fertilizer rate (0, 120 kg N hm^-2, and 240 kg N hm^-2) on summer maize (Zea mays L. var. Shaandan 911) and soil water-nitrogen dynamics. The dryland study was conducted in the southern part of the Loess Plateau during the warm and rainy summer of 2004 on a soil with high amounts of residual mineral N. The results showed that appropriate N application rates increased crop biomass and grain yield. Over application of N fertilizer did not increase crop biomass or yield. In the over-fertilized plots, N use efficiency was only 39% and N residual efficiency was as high as 87.2%. Water distribution in the soil profile was significantly influenced by the N rate. The 240 kg N hm^-2 treatment contained 593 mm of water at harvest and the water had moved to a depth of 200 cm. In comparison, the 0 kg N hm^-2 treatment contained 561 mm of water and the 120 kg N hm^-2 contained 553 mm of water. In both treatments, water had moved to a depth of 180 cm. Nitrogen rates significantly affected the accumulation and distribution of nitrate N in the soil profile, but had no significant effect on the amount or distribution of ammonium in the soil profile. At harvest, total nitrate N in the 0-200cm depth was 78 kg hm^-2 in the 0 kg N hm^-2 treatment compared to 148 kg hm^-2 in the 120 kg hm^-2 and 290kg hm^-2 in the 240 kg hm^-2. Nitrate N had moved down to a depth of 60 cm in the 0 and 120 kg hm^-2 treatments and 140cm in the 240 kg hm^-2 treatment due to percolation during the summer rainfall. The results suggest that appropriate N application rates could promote the utilization of soil water and mineral N by summer maize and increase its biomass and grain yield. In contrast, over application of N results in the accumulation of a large amount of nitrate N in the soil profile and increases the risk of nitrate N leaching to the subsoil. In addition, it was observed that nitrate N leaching is significantly slower than the downward movement of soil water.