『Abstract
　Nitrogen (N) and precipitation are the main limiting factors in food production under rain-fed cropping systems in arid and semiarid regions. A long-term field experiment was conducted from 1984 to 2009 in the Changwu State Key Agro-Ecological Experimental Station of the Loess Plateau, Chins, to study interactions between precipitation and N fertilization. Winter wheat (Triticum aestivum L.) was grown with N fertilization at 0, 45, 90, 135, and 180 kg N ha^-1. With 25-year N fertilization, soil organic carbon and total N were increased by 18％ and 26％, respectively. Mean yields and precipitation use efficiency (PUE) levels with N fertilization were increased by onefold to twofold compared with unfertilized soils. Generally, during the period of 1984-2009, grain yields with 45-180 kg N ha^-1 increased at the rate of 38-49 kg ha^-1 year^-1 (p＜0.05), whereas PUE increased at the rate of 0.071-0.088 kg mm^-1 year^-1. These increases over time are largely attributable to improvements on soil organic carbon and total N. Variations in yield were proportionally reduced by N fertilization, but heavily depended on the fluctuation of precipitation. Grain yields with 45-180 kg N ha^-1 decreased by approximately 10％-45％ (p＜0.05) in the dry years (drought index＜-0.35) compared with the normal years, whereas significant increases were not detected in the wet years (drought index＞0.35). Grain yields with 90-180 kg N ha^-1 were positively correlated (p＜0.05) with fallow season precipitation, but insignificantly correlated with growing season precipitation. The optimum N fertilization rates in the dry, normal and wet years increased following an increasing sequence of 45, 135 and 180 kg N ha^-1. These results have significant implications for optimizing N fertilization and maximizing yield and PUE in rain-fed cropping systems in arid and semiarid regions.

Keywords: Grain yield; Precipitation distribution; Precipitation use efficiency; soil organic carbon; Long-term N fertilization』

1. Introduction
2. Materials and methods
　2.1. Experimental site
　2.2. Experimental design
　2.3. Sampling and analysis
　2.4. Data analysis
　2.5. Statistical analysis
3. Results
　3.1. Annual precipitation and seasonal patterns
　3.2. SOC, TN and N fertilization rates
　3.3. Grain yield and N fertilization rates
　3.4. PUE and N fertilization rates
4. Discussion
　4.1. Changes in SOC, grain yield and PUE under N fertilization
　4.2. Interannual variations in grain yield under N fertilization
　4.3. Potential implications for optimum N fertilization rate and precipitation utilization
5. Conclusions
Acknowledgments
References