Identification of cattle, llama and horse meat by near infrared reflectance or transflectance spectroscopy

Visible and near infrared reflectance spectroscopy (VIS–NIRS) was used to discriminate meat and meat juices from three livestock species. In a first trial, samples of Longissimus lumborum muscle, corresponding to beef (31) llamas (21) and horses (27), were homogenised and their spectra collected in reflectance (NIRSystems 6500 scanning monochromator, in the range of 400–2500 nm). In the second trial, samples of meat juice (same muscle) from the same species (20 beef, 19 llama and 19 horse) were scanned in folded transmission (transflectance). Discriminating models (PLS regression) were developed against “dummy” variables, testing different mathematical treatments of the spectra. Best models indentified the species of almost all samples by their meat (reflectance) or meat juice (transflectance) spectra. A few (three of beef and one of llama, for meat samples; one of beef and one of horse, for juice samples) were classified as uncertain. It is concluded that NIRS is an effective tool to recognise meat and meat juice from beef, llama and horses.

► We tested near infrared spectroscopy to discriminate meats of beef, llama and horses.
► Ground meat and meat juice, scanned in reflectance and transflectance, respectively.
► Reliable VIS–NIRS models recognised meat species with both techniques.
► Most samples were correctly classified, with a few uncertain and none misclassified.