No abstract

The determination of fish species is an important task for food control, especially in the case of very similar and processed fish or fish products. Typically, this analysis is carried out by DNA-based methods. However, especially for small sample series the effort for sample preparation and the required analysis time are not satisfactory. We show that by means of a simple extraction and analysis of the 1H NMR spectra of these extracts a fast discrimination of fish species is possible. For this purpose, 69 samples consisting of five sample groups iridescent shark, European plaice, rock sole, witch flounder and common sole were analysed. The large differences in the 1H spectra could be further highlighted with a principal component and led already to a graphical differentiation of most sample groups. The performed classification of the samples using a subspace discriminant algorithm showed a recognition rate of 100% for all samples, regardless of additives like citric acid or lysine. These results show that 1H NMR spectroscopy is a robust, simple and rapid method for the recognition of the fish species studied.