对Bliss、Finney法作为“最严谨,唯一的求算LD_(50)的对照标准”进行再评价;在理论与计算实践层面上比较Bliss、Finney法与非线性拟合法。结果分布态的未知性、权重系数的不可靠性,回归中的数值转换不精确性,Bliss、Finney法的最佳性,普适性质的商讨。LD_(50)的非线性拟合求算法更为精确、准确、实用、有效,值得在新药研究与开发中使用。 The Bliss and Finney methods were reevaluated as “the most rigorous and the only control standard to calculate LD_ (50)”. Bliss, Finney and non-linear fitting methods were compared theoretically and computationally. The unidentified distribution of results, the unreliability of weight coefficients, the inaccuracy of numerical conversion in regression, the best of Bliss and Finney methods, and the discussion of universal properties. The LD_ (50) nonlinear fitting method is more accurate, accurate, practical and effective and worth to be used in the research and development of new drugs.