Abstract: Abstract: Wheat yield formation is mainly dependent on dry matter accumulation, which is influenced by irrigation and nitrogen input. It will be very helpful for improving yield to understand how irrigation and nitrogen affect wheat dry matter accumulation. Field experiments over 2 years with three-level irrigation and four-level nitrogen input for winter wheat were conducted during 2012-2014 in Guanzhong Plain located in Shaanxi Province, China. Dry matter accumulation with growing degree day was fitted by Richards function, and 5 derived characteristic parameters were analyzed using the univariate analysis of variance (ANOVA) and the multivariate analysis of variance (MANOVA), the result of which was more appropriate when there were significant correlations among dependent variables. None of the 5 derived characteristic parameters of dry matter accumulation was significantly influenced by the interaction of irrigation and nitrogen fertilization. None of irrigation, nitrogen and year had a significant effect on the function parameter b and c. Irrigation significantly prolonged the thermal time required for the entire dry matter accumulation process and then led to an increase of dry matter at maturity from 10 831 kg/hm2 under rainfed condition to 13 813 kg/hm2 with irrigation at wintering and jointing stages. Nitrogen fertilization significantly enhanced the average and maximum rate of dry matter accumulation, and shortened the thermal time achieving the maximum rate of dry matter accumulation since sowing. Dry matter at maturity significantly increased from 8 001 to 14 112 kg/hm2 under the nitrogen input from 0 to 210 kg/hm2 in this study. No more dry matter was gained from more nitrogen input. Weather conditions over 2 years had a significant effect on the thermal time entering the rapid growth stage and achieving the maximum growth rate since sowing, but had no significant effect on the maximum of dry matter. The effect of irrigation on yield and yield components varied with weather conditions in study period. Irrigation significantly increased the yield and the weight of 1 000 grains in 2012-2013, however, it had no significant effect on those in 2013-2014. The weight of 1 000 grains evolved from 35.8 under rainfed condition to 41.7 g with irrigation at wintering and jointing in 2012-2013, which resulted that the corresponding yield increased from 4.74 to 6.29 t/hm2. In 2013-2014, irrigation significantly enhanced the grain number per square meter from 13 833 under rainfed to 15 749 with irrigation at winter and jointing. Yield and grain number per square meter were significantly influenced by nitrogen, while nitrogen had no effect on the weight of 1 000 grains over 2 years. Grain number per square meter increased from 10 414 to 15 911 as nitrogen input increased from 0 to 210 kg/hm2, which contributed to the corresponding increase of yield from 4.42 to 6.96 t/hm2. No more yield and grain number were gained from more nitrogen input, which indicated the nitrogen rate of 210 kg/hm2 would be sufficient for wheat growth and yield formation. The interaction of irrigation and nitrogen had no effect on yield and yield components during 2012-2014. The yield had a higher correlation with grain number per square meter than with the weight of 1 000 grains, which indicated that yield was largely determined by grain number. Both of yield and grain number per square meter had a strong correlation with the average and maximum rate of dry matter accumulation, which indicated improving the rate of dry matter accumulation would be a promising measure to raise yield in the future. This also indicated that nitrogen was the limited factor to wheat grain yield improvement in this study area. The results obtained in this research provide scientific basis and valuable information for selecting the optimum irrigation and nitrogen regimes in semi-arid regions of China.