Objective: To establish quantitative models of paeoniflorin and water in samples of Paeoniae Radix Alba using near infrared (NIR) spectroscopy. Method: 108 batches of Paeoniae Radix Alba samples were collected from different regions of China. Near infrared spectrum correction models for determination of paeoniflorin and water in samples were developed with the application of partial least squares regression. The reference analyses were performed with oven-drying method and ultra performance liquid chromatography method respectively for determination of water and paeoniflorin. Correlation coefficient and predicted root mean square error were also used for evaluating the estimated performance of the models. Result: For paeoniflorin and water

the correlation coefficients for calibration set were 0.9676 and 0.8731

respectively;root-mean-square errors for calibration set were 0.406 and 0.264

respectively;root-mean-square errors for prediction set of the two analytes reached 0.599 and 0.252

respectively. Conclusion: This method is rapid

simple

reliable and has no pollution

which can be used for rapid and accurate determination of paeoniflorin and water in Paeoniae Radix Alba.