NIR spectroscopy was used to assess two physical/chemical parameters in white wines during alcoholic fermentation. NIR calibration models were developed using a set of 64 samples scanned in a rectangular quartz cuvette with a 50 mm path-length in the 700-1060 nm region, in a miniature fiber optic NIR spectrometer system working in transmission mode. Principal component analysis (PCA) and partial least squares (PLS) regressions were used to interpret spectra and develop calibrations for wine composition. The correlation coefficient (r) and the standard error of cross-validation (SECV) were 0.99,4.22 g/dm(3) for volumic mass and 0.99, 10.44g/l, for reducing sugars, respectively. Suitable wavelengths for volumic mass and reducing sugars were also proposed according to x-loading weights and regression coefficients. The results obtained suggest that the miniature fiber optic NIR spectrometer is a promising tool for monitoring the white wine fermentation process. (C) 2008 Elsevier Ltd. All rights reserved.