Abstract: In this study, five walnuts varieties were used as research objects, and the phenotypic data（size, weight and component） and oxidative characteristics were gathered, then the differences of walnuts phenotypic characteristics were investigated using principal component analysis（PCA） and hierarchical clustering method. Results showed that three principal components were extracted by PCA, and the contribution rate of cumulative variance reached to 93.078 9%. Based on the combination of phenotype（size, weight, composition） and oxidation characteristics, systematic cluster analysis was used to intuitively reflect the indices differences of different walnuts varieties through clustering heat maps, which could effectively distinguish different varieties of walnuts.