Soil erosion on agricultural land has become a global environmental problem and information on rates of soil loss is an important requirement both for quantifying the problem and for developing improved land management and soil conversation practices. The use of fallout ¹³⁷Cs measurements can overcome many of the difficulties facing traditional approaches to erosion monitoring and provide an effective means for obtaining estimates of soil redistribution rates on cultivated land. There are two key problems calling for scientific solutions to practical utilization of this technique. One is the determination of reference value of¹³⁷Cs content of soil that has never experienced erosion and deposition, and the other is the establishment of the quantitative model between the amount of  ¹³⁷Cs loss from the soil profile and the rate of soil erosion. At present, a number of empirical functions and theoretical models have been developed to establish the relationship between the change in the soil  ¹³⁷Cs inventory relative to the local ¹³⁷Cs fallout input and the rate of soil erosion. They are reviewed and evaluated in this paper. The physically based mass balance models are relatively reliable because they involved considerations of many factors related to the erosion process. The mass balance models include Walling model, Zhang Xinbao.model, Yang Hao.model, and Zhou Weizhi.model. Although they all belong to the mass balance model, they were established by different ways. This paper will mainly discuss the difference and similarity among these four models.