Gelatin is a collagen derivative, which has a large application in the pharmaceutical, food and adhesive industries as well as photography. The large similarity in structure and properties of gelatins from different origins makes their differentiation difficult. Certain chemometric methods, such as principal component analysis (PCA), can help to classify and characterize gelatin components. In this study 14 bovine and 5 porcine gelatins were examined. The analysis procedure involved complete hydrolysis of samples by classic acid hydrolysis in order to release their amino acid residues. Separation and determination of amino acids was achieved by reversed-phase (RP) HPLC following pre-column derivatisation. Orthophtaldialdehyde (OPA) and 4-chloro-7-nitro benzofurazane (NBD-Cl) were used as derivatisation reagents. From the 20 peaks detected by HPLC analysis, one was very typical in bovine gelatin. Peak height, area, area percentage and width were used to make matrixes. Principal component analysis with the MATLAB program was used to differentiate these gelatins. PCA on matrix of height, width and total matrix were resulted in good differentiation between bovine and porcine gelatins.