Our study objective is to examine the effects of arboraceous layer on the spatial pattern and morphological features of herbaceous layer in Elacagnus angustifolia-Achnatherum splendens community in Ningxia Autonomous Region. In a 50m×50m plot, we measured crown diameter, DBH (for E.angustifolia), height, and location of every individual by species. A 100 cm soil profile was divided into five strata at 0-10, 10-30, 30-60cm, and 60-100 cm in depth and sampled to measure the soil water contents, salinity, and physical and chemical properties. Wavelet analysis and buffer analysis were applied to examine the spatial relationship between the arboraceous and herbaceous layers. Additionally, three 24 m lines marked with A, B and C in the canopy projective region, and two 32 m lines marked with D and E in the open region were randomly placed for measuring individuals that touch the lines. The mean diameter, crown, height and density of A.splendens were recorded. We found that the E. angustifolia had a mean height of 7.9 m, a mean canopy diameter of 6.5 m, and the coverage up to 38.5 %, while the dominant herbaceous species, A. splendens, contributed 97.8 % of the total abundance. Soil moisture at 0-10 cm depth was only 4.67 % , but increased with depth with little variation in the 30-100 cm layers. The change in electrical conductivity was inconsistent with soil moisture and reached maximum value of 111×103 S/cm in the 60-100 cm layer. This suggests that the deeper soil had lower water potential and implied that arbors roots received water mainly from deep soil, causing an anti-gradient soil salinity and water potential, while herbaceous plants tended to use water in shallow soil. Wavelet transform showed that the intrinsic scale was 6.2 m for the canopy projective regions and was 8-12 m in the open area, which corresponded well to the canopy size in the canopy projective regions and 1.5 height of E. angustifolia in the open area. Buffer analysis further suggested a more complex relationship. In the arbor canopy vertical projection area, we found no obvious differences in diameter, crown size, and height of A. splendens, but from zone 6#, the average crown diameter and the height of A. splendens increased gradually along the transect. From zone 10# to 13#, the crown diameter decreased and rapidly reverted in zone 14# and 15#. These results suggested that the control of arboraceous layer on herbaceous layer on spatial pattern and morphological characteristics were influenced not only by canopy shading, but also by other causes such as root distribution pattern as the morphological characters not changed monotonically with distance.