The easiest way to take harmel is to grind up three to five grams of seeds in a coffee grinder, put the powder in gel caps, and let your stomach extract the alkaloids. If you dislike being dyed yellow on the inside, it is easy toe xtract the alkaloids. The following method, modified from Hafenfratz, produces crystals of high purity.
Grind 100 grams of esfrand seeds. A coffee grinder is good for this job. Soak the seeds for several days in 300 ml of 3% acetic acid. A little arithmetic will tell you that 180 ml of white vinegar, which is 5% acetic, diluted with 120 ml distilled water gives you what you need. Squeeze out the seeds, and extract the mass again with another
100 ml of 3% acetic. A potato ricer with a strip of cheese cloth is effective. Combine the two extracts and elt them settle for 24 or 48 hours. Decant off as much f the clear solution as you can-it helps immensely with the filtering. If you don't have a Buchner funnel, a filter flask, and vacuum, the procedure will be a little slower-try to do as much as you can without decanting. THe more solid matter that can be removed at this stage, the easier will the subsequent precipitations. I filter it twice, about a hundred ml at a time, changing the filter papers when they clog.
Heat the filtered solution to near boiling, and stir in 90 grams of NaCl (salt). Don't use iodized salt. When the salt is dissolved, let the solution cool slowly-finish with 24 hours in the refrigerator. Both the harmine and harmaline will precipitate. Filter the precipitate and discard the filtrate.
Redisolve the precipitate in 150 ml of hot water and add another 25 grams of salt. Stir to dissolve the salt and let the solution col as before, and filter out the crystals. Repeat this step once or twice more to get long, clean, yellowish crystals. Sometimes there is a little insoluble gunk that filters with the crystals. It can be removed by filtering the hot harmel solution before adding the salt.
Average yields are about 3%. Dr. Alexander Shulgin kindly submitted one of my samples for analysis by GCMS, and it proved to be of high purity-(there was no viscine). The ratio of harmine to harmaline was about 2:1.
Hafenfratz has instructions for separating the harmine and harmaline, if you can't resist the lure of chemistry.