Abstract: In order to simplify the quality evaluation of Korla fragrant pear and its bud mutants as well as hybrid varieties (strains), the main quality traits of 13 fruit samples from Korla fragrant pear and its bud mutants as well as hybrid varieties (strains) including single fruit weight, flesh firmness, fruit shape index, core/fruit ratio, soluble solid content, soluble sugar content, titratable acidity, stone cell content, vitamin C content, and sugar/acid ratio and juice content were analyzed by principal component analysis (PCA) and cluster analysis. The results showed significant differences in fruit quality parameters among these varieties (strains). The variation (CV) coefficients of stone cell content and single fruit weight were larger, whereas the CV of juice content was smaller than that of other quality parameters. There was a simple correlation between some quality traits. In PCA, 5 principal components were extracted and explained cumulatively 88.964 5% of the total variance. These principal components were sufficient to represent the quality characteristics of Korla fragrant pear. Five varieties (strains) were selected based on their principal component scores, including 84-1-1, 84-1-15, Xinli No. 9, Xinli No. 8 and 84-4-72. The five most important fruit quality parameters were single fruit weight, fruit shape index, stone cell content, soluble solid content and juice content according to cluster analysis. In conclusion, principal component analysis and cluster analysis are useful for the evaluation and selection of fruit quality traits of Korla fragrant pear and its bud mutations as well as hybrid varieties (strains).

Key words: Korla fragrant pear, fruit quality, principal component analysis, cluster analysis, evaluation factor