Mehlich 3 method of soil analysis is widely used in soil analytical laboratories throughout the world. It is suitable for measuring P in low to high pH soils and is capable of determining multiple elements simultaneously. However, this method has not been statistically validated among laboratories. Two laboratories using Mehlich 3 could report different extractable P values. The objective of this study was to determine the reproducibility and robustness of Mehlich 3 method. An in house homogeneity test was conducted for each of the ten soils which were then replicated 3 times and sent to 26 domestic and international laboratories primarily for P analysis. Samples were either scooped or weighed. Preliminary results indicate a wide range of repeatability relative standard deviation (RSDr) values. Scooped samples analyzed by colorimetric method varied between 0.4-33.4 percent. A 0.3-30.8 percent was found for scooped, ICP analyzed; 0.0-62.3 percent for weighed, colorimetric analyzed; and 0.0-81.1 percent RSDr for weighed, ICP analyzed samples. The reproducibility relative standard deviation (RSDR) values ranged between 9.4-57.7 percent for scooped, ICP analyzed; 9.1-58.2 percent for weighed, ICP analyzed; 8.6-51.2 percent for scooped, colorimetric; and 9.8-35.6 percent RSDR for samples that were weighed and analyzed colorimetrically.