Abstract: Several different brands powdered milks, as well as adulterated powdered milk, were measured by the electronic nose (E-nose) which was composed of ten metallic oxide sensors.Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) were used in the data analysis and pattern recognition.The results demonstrated the powdered milk could be identified according to their bands.The model was useful for predicting the bands of the powdered milk.Meanwhile, the adulterated powdered milk could be discriminated by E-nose.Furthermore, the concentration of adulterant in the powdered milk could also be acquired in general.In conclusion, the E-nose could be applied to identify the adulterated powdered milk.