Correct species identification is a precondition for many ecological studies. Morphologically highly similar, i.e., cryptic, species are an important component of biodiversity but particularly difficult to discriminate and therefore understudied ecologically. To find new methods for their rapid identification, thus, is important. The cuticle's chemical signature of insects often is unique for species. Near-infrared spectroscopy (NIRS) can capture such signatures. Imaging NIRS facilitates precise positioning of the measurement area on biological objects and high-resolution spatial capturing. Here, we tested the applicability of imaging NIRS to the discrimination of cryptic species by using the ants Tetramorium caespitum and T. impurum. The classification success of Partial Least Squares Regression was 98.8%. Principal Component Analysis grouped spectra of some T. impurum individuals with T. caespitum. Combined with molecular-genetic and morphological evidence, this result enabled us to pose testable hypotheses about the biology of these species. We conclude that discrimination of T. caespitum and T. impurum with imaging NIRS is possible, promising that imaging NIRS could become a time- and cost-efficient tool for the reliable discrimination of cryptic species. This and the direct facilitation of potential biological insight beyond species identification underscore the value of imaging NIRS to ecology.