Abstract: The methods to determine the true shrinkage limit of clay are proposed by updating both the testing procedure and the data processing of the shrinkage limit and shrinkage test methods in the Chinese standard for geotechnical testing method (GB/T 50123—2019). The shrinkage limit tests on slurry specimens and the shrinkage tests on reconstituted specimens of six types of clay are carried out. A systemic comparison is carried out between the specified shrinkage limits in GB/T 50123—2019 and the true shrinkage ones yielded by the proposed methods. The conclusions are drawn as follows: (1) The true shrinkage limits obtained from the slurry specimens (shrinkage limit tests) almost coincide with those obtained from the reconstituted specimens (shrinkage tests). This agreement indicates the true shrinkage limit is independent of testing methods, which is an inherent property denoting the consistency limit of clay. (2) The shrinkage limits given by the original methods in GB/T 50123— 2019 are about 2.7 times larger than the true ones. (3) Upon the true shrinking limit, the clays are all unsaturated with the saturation between 22 % and 57 %. The reasons of the overestimation of the original standard-specified methods are speculated as: The method for the shrinkage limit tests assumes that the soil is saturated at the end of shrinking, while that for the shrinkage tests uses the intersection of the initial best-fitting inclined straight line and the final horizontal straight line of the shrinkage curve as the shrinkage limit.