Plant populations growing at the margin of their range may exhibit traits that indicate genetic differentiation and adaptation to their local abiotic environment. Here, it was investigated whether geographically separated marginal populations of Arabidopsis lyrata ssp. petraea have distinct metabolic phenotypes within the plant foliage. Seeds of A. petraea were obtained from populations along a latitudinal gradient (49-64 N), namely Germany, Wales, Sweden and Iceland and grown in a controlled cabinet environment. Targeted metabolic profiles and fingerprints were obtained at the same initial developmental stage. The free amino acid compositions were population specific, with fold differences in arginine, aspartic acid, asparagines, glycine, phenylalanine, alanine, threonine, histidine, serine and gamma-aminobutyric acid (GABA) concentrations. Sucrose, mannose and fructose concentrations were also different between populations but polyhydric alcohol concentrations were not. Principal component analysis (PCA) of metabolite fingerprints revealed metabolic phenotypes for each population. It is suggested that glucosinolates were responsible for discriminating populations within the PCA. Metabolite fingerprinting and profiling has proved to be sufficiently sensitive to identify metabolic differences between plant populations. These findings show that there is significant natural variation in metabolism among populations of A. petraea.