The metal content of a number of sparkling wines was determined by atomic spectrometry techniques. Al, Ba, Ca, Cu, Fe, K, Mg, Mn, Na, P, Sr and Zn by using inductively coupled plasma atomic emission spectrometry (ICP-AES); Cd, Ni and Pb by graphite furnace atomic absorption spectrometry (GFAAS) and As from hydride generation AAS (HGAAS). Two kinds of sparkling wines were studied with D.O. trademark: cava and champagne. 18 samples of "brut" cava and 17 samples of "brut" champagne of different brands were analyzed following the procedure described in the paper. By using the metal concentrations as chemical descriptors the two classes of samples (cava and champagne) are perfectly discriminated, when applying pattern supervised learning recognition techniques such as linear discriminant analysis (LDA) and soft independent modeling of class analogie (SIMCA). The number of false positives and negatives were zero, which indicates a remarkable authentication power of the descriptors used.