Soil classification systems organize soil variability into useful groupings that can be identified by field investigation and documented in soil survey, and form the basis for the exchange and extension of soil science research and soil resources management. Fuzzy clustering analysis may be used whenever a composite classification of soil incorporates multiple parameters. In this paper, seventy-seven topsoil samples were collected from Qinghai and Heilongjiang of China, and the element contents of topsoil were detected by wavelength dispersive X-ray fluorescence spectroscopy. In fuzzy clustering analysis, all data were standardized, and then a fuzzy similarity matrix was established and the fuzzy relation was stabilized. The results showed that topsoil samples of Qinghai and Heilongjiang were completely grouped into two clusters according to their districts, when given a suitable threshold λ= 0.7580. This work supplied the quantification classification method of alpine soil (Qinghai) and unsaturation siallitic soil (Heilongjiang).