7.3g melatonin was refluxed in 100mL ethylene glycol (distilled from antifreeze) with a 1.25 molar equivalent of KOH (do the math; I can't recall the mass), for fortyfive minutes, by which time the solution was a very menstrual sort of red-orange. The mixture was allowed to cool and then cut with water and extracted with DCM. The extracts were pooled and solvent removed to yield a dark red, slightly fishysmelling oil.
The oil was dissolved into ~100 mL acetonitrile, at which point chunks of white powder precipitated at the bottom of the vessel (unhydrolyzed melatonin?). This liquid was poured into a 500mL flask and more solvent added to bring the total volume up to ~475 mL. Isopropyl bromide and K2CO3 were added in a 5:2:4 molar ratio to the theoretical amount of mexamine. At this point, the experimenter was disturbed to note that the carbonate salt really did not want to go into solution. The mixture was stirred for four hours, after which time the carbonate still had not all gone into solution. Hmmmm...
This was not a problem the last time this procedure was attempted, but then, proportionately more solvent was used that time. The mixture was allowed to stir for about twenty more hours at room temperature, at which point dissolving the entire amount of salt was given up as hopeless. The experimenter hopes that he merely misweighed the amount of salt he added.
As per drone's second breath of hope, the reaction mixture was poured over an equal amount of water and extracted with ethyl acetate, which did not emulse as much as DCM usually does (in the experimenter's experience) with aqueous tryptamine solutions. The pooled EtOAc extracts were then greatly reduced in volume by a combination of evaporation, boilover, and combustion. The former process is preferable to the latter two.
The more-concentrated EtOAc sol'n was treated with a few drops of [HCl] in water, and the whole mess evaporated, giving an ichorous red tar of a product. This tar was dried to give off-white crystals and chunks of darker tar, m = 3.3g. Recrystallization from isopropanol gave red (?) crystals, m = 2.1g and a very red mother liquor.
At this point, the yield is obviously terrible and a bioassay has not yet been conducted as the experimental animals are all on holiday. The results will be submitted as soon as the data are gathered.
Sorry I'm such a lousy chemist.