The aim of the studies reported in the Research Communication was to develop a rapid spectroscopic technique as an alternative method for the classification and discrimination of milk sources by Fourier transform infrared spectroscopy (FTIR). Cow, sheep and water buffalo milk samples were collected from various local milk producers in Istanbul, Turkey. In addition, various brands of packaged milk were purchased locally. Spectrums were obtained according to milk species origin and binary mixtures prepared in increments of 10% (10, 20, 30, 40, 50, 60, 70, 80 and 90%) for each sample analysed in FTIR spectroscopy. A successful milk species (cow, sheep, and water buffalo) discrimination and classification were achieved utilising Hierarchical cluster and principle component analyses (PCA) on the basis of Euclidean distance and Ward's algorithm. Amide-I (1700–1600/cm) and Amide-II (1565–1520/cm) spectral bands were used in the chemometric method. The results of the study indicated that adulteration of milk samples can be quantitatively detected by the FTIR technique in a short time with high accuracy. In conclusion, this method could be used as a new alternative technique for routine analysis in authenticity control of milk species origin.