Abstract: Objective: To investigate the determination of soluble solids and total acidity in Gannan navel orange by using near infrared (NIR) diffuse reflection spectroscopy. Methods: NIR spectroscopy was used to determine the contents of soluble solids and total acidity in 150 Gannan navel oranges, including 120 navel oranges for model establishment and 30 navel oranges for model validation. The NIR spectra ranged at 350－1800 nm of navel oranges were collected. After the moving of the spectra, the first-order differential D1log (1/R) and the second-order differential D2log (1/R) pre-treatments were used to process. Principal component regression (PCR) and partial least square (PLS) regression were used to establish a mathematical prediction model. Results: The PLS model with D1log (1/R) data treatment could exhibit better prediction efficiency. The correlation coefficients of soluble solids and total acidity using this prediction model were 0.9263 and 0.9562 and the root mean standard errors of soluble solids and total acidity were 0.4102 °Brix and 0.018%, respectively. Conclusion: NIR spectroscopy can provide an accurate, reliable and nondestructive method for evaluating internal quality index through soluble solids and total acidity of Gannan navel orange.

Key words: near infrared diffuse reflection spectroscopy, non-destructive determination, Gannan navel orange, soluble solid content, total acidity