Soil particle-size distribution (PSD) is the most important physical attributes due to its influence on soil properties related to water movement and soil management, erosion. Therefore, characterizing variations in the PSDs of soil are an important issue in environmental research. Till now, the fractal model could well characterize soil particle-size distribution, and its calculation is a function of the subdivisions of soil fractions (clay, silt, and sand). But few studies are carried on to analyzed the effects of soil particle-size classification (PSC) systems on calculating single volume fractal dimensions. Therefore, top soil across different stands, which composed of Pinus sylvestris var. mongolica, Pinus tabuliformis and Populus were selected. The study was undertaken in the Magu forestry farm, China. Soil PSD was measured with a laser diffraction technique, and fractal dimensions (D) were calculated. The experimental results showed that: (1) The PSD was distributed within the sand classification followed by clay and silt particle contents. The order of D values was USDA (1993)> IS014688 (2002)> ISSS (1929) PSC systems. (2) Fractal dimensions exhibit a significant negative linear correlation with the sand content and a significant positive correlation with the content of clay and silt. (3) The D values of three PSC systems are significantly positive correlated, which indicated that they could achieve mutual transformation. (4) According to the fractal model and descriptions of soil fractions under different soil PSC systems, refining clay and sand fractions could increase D values. From above, it was suggested that USDA (1993) soil PSC system is highly recommended.