The objective of this work was to assess the potential of near infrared spectroscopy to predict the immunoglobulin G (IgG) content in bovine colostrum. Liquid colostrum samples (n=157) were collected from Holstein cows from 2 dairy farms in southern Chile. Samples were obtained within 1h of parturition and scanned in folded transmission (transflectance) in the visible-near infrared range. Multivariate regression models (modified partial least squares) were developed with spectral data against IgG content measured by radial immunodiffusion. The best calibration included a mathematical treatment of the spectra by a second derivative plus standard normal variate and detrending. The best equation explained a high proportion of the variation in IgG content (R 2 of 0.95 in calibration and 0.94 in cross-validation). Average (91.5g/L), standard deviation (37.6g/L), and range, as highest minus lowest values (171.9g/L) of reference values were 10.1, 4.2, and 19 times the value of the root mean square error of cross-validation (9.03g/L) respectively. Near-infrared spectroscopy, scanned in folded transmission, is an effective tool to predict the IgG content in liquid colostrum.