Abstract: To achieve fast and non-destructive measurement of soluble solid content (SSC) in persimmon, a new method based on visible-near infrared reflectance (NIR) spectroscopy was put forward. A Field Spec 3 spectroradiometer was used for collecting 66 sample spectra data of the three kinds of persimmon separately. Then principal component analysis (PCA) was used to process the spectral data after pretreatment using the average Smoothing method, and 6 principal components(PCs) were selected based on accumulative reliabilities. These selected PCs would be taken as the inputs of the three-layer back-propagation artificial neural network (BP-ANN). A total of 66 persimmon samples were divided into calibration sets including 51 samples（17 samples of each variety） and validation sets including 15 samples（5 samples of each variety） randomly. The three-layer BP-ANN model was established with 6 nodes being 6 principal components (PCs) in input layer, 1 node being soluble solid content (SSC) in persimmon in output layer and 11 nodes in hidden layer. Then the model was used to predict soluble solid content of persimmon for the sample in the validation set. The results showed that a standard error of calibration (SEC) of the calibration model was 0.232, its prediction relative error below 3% was achieved, the decision coefficient (R2) between the predicted value and the measurement value was 0.99, and the forecast standard deviation (SEP) was 0.257. It can be concluded that PCA combined with BP-ANN is an available method for soluble solid content measurement of persimmon based on NIR spectroscopy.