N-acetylated and tertiary indolamines, some of which are possible neurotransmitter candidates in the CNS, cannot be visualized with the standard Falck-Hillarp histofluorescence method and very little is known about their cellular localization. The present investigation demonstrates that glyoxylic acid (GA), formaldehyde (FA) in combination with aluminum ions (the ALFA method) and trifluoroacetic acid anhydride (TFAA) are capable of forming fluorescent compounds from N-acetylated (e.g. melatonin and N-acetyl-5-hydroxytryptamine) and tertiary (e.g. bufotenin) indolamines in histochemical protein models. With GA and FA-aluminum more vigorous reaction conditions were required for demonstration of these compounds compared to those needed for optimal visualization of primary catecholand indolamines (prolonged reaction time and higher concentration of GA and FA and aluminum ions). The fluorophore formation from N-acetylated and tertiary indolamines, which represents a new reaction principle in amine fluorescence histochemistry, is proposed to proceed as follows. In the first step, the indole reacts in 2-position with the reagent. The intermediate formed is dehydrated in the second step, yielding a strongly fluorescent 2-methylene derivative, which either per se or as the corresponding autoxidized dimer constitutes the main fluorophore. TFAA and related anhydrides represent new and potent reagents for histochemical visualization of N-acetylated indolamines such as melatonin. In contrast to the GA and ALFA reactions the optimal formation of fluorphores with TFAA required only mild reaction conditions (2–10 min at 0–20° C). The main fluorophore formed from melatonin has been identified and the reaction with TFAA is proposed to proceed as follows. An unstable intermediate, the isoimidinium carboxylate, is formed in the first step and this compound is then cyclized to form the fluorophore, 6-methoxy-1-methyl-3,4-dihydro-β-carboline. The GA and ALFA methods are already widely used for visualization of catecholamine systems. The fluorescence microscopical and microspectrofluorometric analysis did not, however, veveal any specific structures containing N-acetylated or tertiary indolamines in the rat CNS. The TFAA reaction was highly specific for N-acetylated indolamines when applied to protein models. However, in tissue a disturbing background fluorescence appeared, which under all reaction conditions tested, developed concomitantly with the specific fluorescence from melatonin. The problem with this background reaction has to be solved before the TFAA reaction can be applied for demonstration of N-acetylated indolamines in tissue.