Abstract: To assess the effect of crops on the utilization of carbon sources by soil microorganisms, we chose one piece of sloping agricultural land in Zhangjiachong, part of the Three Gorges Reservoir area in Zigui County, Hubei Province as the study site. An experiment was designed considering three types of crops, including tea, orange and peanuts, with or without hedges. In total, 18 soil samples were collected from 6 plots with 3 replicates in each plot. Soil physicochemical properties were determined, and soil microbial functional diversity was measured by Biolog ECO plates. Effects of crops and hedges on soil microbes, soil physicochemical properties and the relationships between them were explored by multivariate statistical analysis. The results indicated that hedge had no  significant effects on soil physicochemical properties, but crops did. Soil samples from the peanut plots showed lowest availability of nitrogen, phosphorus and potassium (P<0.001). Soil microbial activity was higher in the orange plot (without hedge) than that in the other plots (P<0.05). The functional diversity indices were in order of: orange plots > peanut plots > tea plots (P<0.05). Amino acids and polymers were the dominant carbon sources of soil microbial metabolism from all sample plots (P<0.05). The utilization of aromatic and amine carbon significantly differed among different crop lands (P<0.01), while hedge greatly affected the utilization of carbohydrates and carboxylic acids. The soil physicochemical variables had minor influence on the carbon source utilization by soil microbes in the orange plots. The utilization of carbon sources in the tea plots was highly affected by the soil factors such as soil pH, available nitrogen and available phosphorus, and that of soil microorganisms in the peanut plots was correlated with total nitrogen and available potassium. The results indicated that in  nutrientpoor soil, soil physicochemical properties, soil microbial activities and carbon source utilization were different among different crop plots, and soil microbial metabolism of carbon sources was correlated with soil physicochemical properties.