Modern maize (Zea mays L.) hybrids are generally regarded as strongly population dependent because maximum grain yields (GYs) per area are achieved primarily in high-density populations. is study was conducted to analyze changes in density independence with plant density based on the response of GY, dry matter (DM) accumulation, and the harvest index (HI) to changes in plant density. Two modern cultivars, ZhengDan958 and ZhongDan909, were planted at 12 densities ranging from 1.5 to 18 plants m -2 . e experiment was conducted for 3 yr, with drip irrigation and plastic mulching, at the 71 Group and Qitai Farms located in Xinjiang, China. With increased plant density, DM accumulation per area increased logarithmically, the HI decreased according to a cubic curve, and GY per area increased quadratically; the optimum density was 10.57 plants m -2 . Further analysis showed that the response of GY per area, DM per area, and the HI to changes in plant density could be divided into four density ranges: Range I (£4.7 plants m -2 ), in which DM per area, the HI, and GY per area were signi cantly a ected by density; Range II (4.7-8.3 plants m -2 ), in which the HI was una ected by density but DM per area and GY per area were signi cantly a ected; Range III (8.3-10.75 plants m -2 ), in which GY per area was una ected by density but DM per area and the HI were signi cantly a ected; and Range IV (³10.7 plants m -2 ), in which DM per area was una ected by density but the HI and GY per area were signi cantly a ected. ese results indicated that Range II is a density-independent range and Range III is a GY-stable range.

Good canopy structure is essential for optimal maize (Zea mays L.) production. However, creating appropriate maize canopy structure can be difficult, because the characteristics of individual plants are altered by changes in plant age, density and interactions with neighbouring plants. The objective of the current study was to find a reliable method for building good maize canopy structure by analysing changes in canopy structure, light distribution and grain yield (GY). A modern maize cultivar (ZhengDan958) was planted at 12 densities ranging from 1.5 to 18 plants/m2 at two field locations in Xinjiang, China. At the silking stage (R1), plant and ear height increased with plant density as well as leaf area index (LAI), whereas leaf area per plant decreased logarithmically. The fraction of light intercepted by the plant (F) increased with increasing plant density, but the light extinction coefficient (K) decreased linearly from 0.61 to 0.39. Taking the optimum value of F (95%) as an example, and using measured values of K for each plant density at R1 and the equation from Beer's law, the corresponding (theoretical) LAI for each plant density was calculated and optimum plant density (9.72 plants/m2) obtained by calculating the difference between theoretical LAIs and actual observations. Further analysis showed that plant density ranging from 10.64 to 11.55 plants/m2 yielded a stable GY range. Therefore, taking into account the persistence time for maximum LAI, the plant density required to obtain an ideal GY maize canopy structure should be increased by 10–18% from 9.72 plants/m2.

SUMMARYCrop nitrogen (N) status is an important indicator of crop health and predictor of subsequent crop yield. The present study was conducted to analyse the relationships between nitrogen nutrition index (NNI), nitrogen biomass difference (ΔNB) and spectral indices in wheat, and then attempt to improve field N management. Spectral indices and concurrent sample N and biomass parameters were obtained from the Shihezi University experimental site in Xinjiang, China during 2009 and 2010. The results showed that all spectral indices were significantly correlated with NNI. Regression functions with the highest determination coefficient (R2) and the lowest root mean square error (RMSE) were used to improve prediction of NNI, and then the selected spectral index was used to estimate NNI and ΔNB. The strongest relationships were observed for the products of modified normalized difference 705 × biomass dry weight (BND705) and the enhanced vegetation index 2 (EVI2) for estimating NNI. There were also strong relationships between the NNI and the normalized NNI (ΔNNI) as well as between ΔNNI and ΔNB, with a linear relationship between ΔNB and the spectral index BND705 and a linear relationship between ΔNB and the spectral index EVI2. These results indicated that BND705 and EVI2 can be used to improve the accuracy of NNI estimation, and the correlations of ΔNB and NNI with BND705 and EVI2 can be used to further improve field N management in wheat.