RESUMO
Fourier transform infrared spectroscopy (FTIR), followed by multivariate treatment of the spectral data, was used to classify extra virgin olive oils (EVOOs) according to their genetic variety. EVOO samples corresponding to seven different genetic varieties (Arbequina, Borriolenca, Canetera, Farga, Hojiblanca, Picual, and Serrana) were analyzed. The wavelength scale of the FTIR spectra of the oils was divided into 20 regions. The normalized absorbance peak areas within these regions were used as predictor variables. Classification of the EVOO samples according to their genetic variety was achieved by linear discriminant analysis (LDA). A good resolution among all categories was achieved using a LDA model constructed with only nine predictor variables. With this LDA model, 88% of the EVOOs were correctly classified, with assignment probabilities higher than 95%. This method is helpful for olive oil producers because it provides useful information related to the genetic variety of EVOOs, which is required by European Community regulations.