Abstract: There are many methods for determining particle-size distribution of soil sample. As the principle of each method is different, the results are not the same. In order to measure the particle size precisely, it is very important to reveal the reasons for the different results obtained by different methods, to select proper methods, and to establish the conversion relationship. In this study, we used five methods to measure particle-size distribution of five soil samples. The results showed that the average particle size measured by Sieve-pipette method (SPM) is the smallest among all the methods, with the largest portion of fine particles; while the average particle size measured by Scanning Electron Microscopy (SEM) is the biggest, with the largest portion of coarse particles. Except the soil sample with high CaCO₃ content, the particle size distributions of all other four samples are very similar when measured using different methods. The conversion relationship could be established between SPM and LD, while the conversion relationship could be established between S-LD1 and SEM only for sand and silt fractions. Generally speaking, the results measured by S-LD1 are the most reliable in study of wind erosion. The results measured by SEM are suitable for samples with narrow particle size distributions, and the results measured by laser diffraction method (LD) are reliable for soils with lesssoil aggregate.

Key words: soil particle-size distribution, sieve-pipette method, sieving-laser diffraction, scanning electron microscopy, laser diffraction