Abstract: The changes of fruit quality (physical indexes, reducing sugar, sucrose, total acid, V_C, polyphenols and flavonoids) of three Phyllanthus emblica L. varieties ('Cuizhu', 'Yingyu' and wild) at different picking stages were analyzed to determine the best picking date of different Phyllanthus emblica L. varieties. The change trend of each quality index was analyzed, and the fruit quality index of Phyllanthus emblica L. was comprehensively evaluated by correlation analysis and discriminant analysis. The analysis showed that the physical index increased first and then tended to be stable. The total acid content in the chemical components generally showed a downward trend and then tended to be stable. Except that the sucrose content of wild Phyllanthus emblica L. increased first and then decreased, and the sucrose content of 'Cuizhu' decreased and then tended to be stable, the reducing sugar content of the three varieties increased and tended to be stable throughout the picking period. The contents of V_C and flavonoids in active components increased first and then decreased, and polyphenols increased first and then tended to be stable. Finally, it was determined that the best picking period of 'Cuizhu' was from the end of October to the end of November, and the best picking period of 'Yingyu' and wild was from the end of October to the end of December. The appearance, size, chemical components and active components of Phyllanthus emblica L. fruit at this stage were excellent. The correlation analysis among the indexes showed that the order of correlation among the indexes of the three varieties of Phyllanthus emblica L. was wild>'Yingyu'>'Cuizhu'. The Bayesian discriminant function of three varieties of Phyllanthus emblica L. was verified by cross discrimination, and the positive judgment rate was 100%, which could correctly distinguish different varieties of Phyllanthus emblica L. The results of this study would provide a theoretical basis for Phyllanthus emblica L. picking, value-added processing and commodity classification.