Field fill material often contains gravel particles larger than the allowable limit for standard laboratory compaction tests. In such cases, the maximum dry density (ρd)max of the fill material compacted at a specified compaction energy level (CEL) must be estimated from the (ρd)max value via compaction tests at the same CEL on the sample obtained by removing large gravel particles. When the gravel mass ratio P is lower than 0.3 or 0.4, this estimate is usually made by the Walker-Holtz (WH) method, which assumes that the same CEL that was applied to the whole sample is transmitted to the portion other than gravel particles inside the whole sample. Research on this estimate for P higher than 0.3 or 0.4 validated by experimental data is rather limited. In this study, a series of laboratory compaction tests were performed changing the maximum particle size, CEL, P, and soil type to examine the validity of the WH method. Based on the test results, a method to modify the WH method is proposed to estimate the (ρd)max value more properly after adding or removing gravel particles compacted at a certain CEL.