Stable carbon isotope ratios (delta(13)C) were determined for alanine, proline, phenylalanine, valine, leucine, isoleucine, aspartate (aspartic acid and asparagine), glutamate (glutamic acid and glutamine), lysine, serine, glycine, and threonine from metabolically diverse microorganisms. The microorganisms examined included fermenting bacteria, organotrophic, chemolithotrophic, phototrophic, methylotrophic, methanogenic, acetogenic, acetotrophic, and naturally occurring cryptoendolithic communities from the Dry Valleys of Antarctica. Here we demonstrated that reactions involved in amino acid biosynthesis can be used to distinguish amino acids formed by life from those formed by nonbiological processes. The unique patterns of delta(13)C imprinted by life on amino acids produced a biological bias. We also showed that, by applying discriminant function analysis to the delta(13)C value of a pool of amino acids formed by biological activity, it was possible to identify key aspects of intermediary carbon metabolism in the microbial world. In fact, microorganisms examined in this study could be placed within one of three metabolic groups: (1) heterotrophs that grow by oxidizing compounds containing three or more carbon-to-carbon bonds (fermenters and organotrophs), (2) autotrophs that grow by taking up carbon dioxide (chemolitotrophs and phototrophs), and (3) acetoclastic microbes that grow by assimilation of formaldehyde or acetate (methylotrophs, methanogens, acetogens, and acetotrophs). Furthermore, we demonstrated that cryptoendolithic communities from Antarctica grouped most closely with the autotrophs, which indicates that the dominant metabolic pathways in these communities are likely those utilized for CO(2 )fixation. We propose that this technique can be used to determine the dominant metabolic types in a community and reveal the overall flow of carbon in a complex ecosystem.