Considerable effort is focused presently on the detection and comprehensive assignment of post-translational modifications of proteins. Obviously attention must be paid to the possibility of chemical modifications that may occur to protein samples during sample handling and manipulation prior to analysis by tandem mass spectrometry. This is of particular concern when a modification is isobaric with the mass differential in common with a known post-translational analog. Here we provide evidence that silver staining protocols that use formaldehyde can result in epsilon-formylation of lysine residues. This modification is in fact isobaric with the important product of methyltransferases, epsilon,epsilon-dimethyllysine. Without exercising proper caution the analysis of silver-stained protein samples by mass spectrometry looking for dimethylation of lysine will yield a significant number of misassigned sites of modification. High accuracy measurements of the mass of the precursor ions and their fragments are required to eliminate this uncertainty. The occurrence of dimethylation of the epsilon-amino function of lysine residues has been reported often in histones. For histone samples excised from silver-stained gels, we found that most sites initially assigned to be dimethylated by automatic search engines under standard search parameters (100 ppm error tolerance) are actually in fact formylated. Caution must be exercised when data obtained from instruments unable to perform high accuracy mass measurements (better than 5 ppm) are to be interpreted.