An isotope dilution method for protein quantification is presented in the context of matrix-assisted laser desorption/ionisation time-of-flight mass spectrometry (MALDI-TOFMS) and mass fingerprinting experiments, revealing an unappreciated high reproducibility and accuracy of relative peak intensity measurements. Labelled proteins were generated by growing cells in a medium containing (15)N-enriched amino acids, and were mixed with proteins of natural isotopic composition from control cells in ratios of approximately 0:1, 1:7, 1:2, 2:1, 7:1, and 1:0 (labelled/unlabelled). Mixtures were separated by two-dimensional gel electrophoresis and analysed by MALDI-TOFMS using typical experimental conditions. A linear relationship is demonstrated between the relative isotopologue abundances (RIA values) for particular peaks in the isotopic distribution of tryptic peptide fragments of the proteins, and the mole fractions of labelled proteins in the mixture. Analysis of RIA values (ARIA quantification) for peptides of six typical silver-stained protein spots for the various mixtures could reproduce the experimentally contrived ratios with approximate errors between 4% (2:1 mixture) and about 18% (1:7 mixture). A consideration of error and its propagation is discussed. ARIA does not require complete separation of the isotope patterns of labelled and unlabelled peptides, and is therefore advantageous in combination with all kinds of labelling experiments in biological systems, because it is compatible with minimal metabolic incorporation of labelling reagent. Simulations indicate that the minimum required (15)N enrichment of the total amino acid pool sufficient for ARIA is less than 4%. In an accompanying paper in this issue, we apply ARIA to proteins differentially labelled with isotope-coded alkylation reagents.