A number of cell cultures of Peganum harmala were initiated to check for a correlation between the barman alkaloid content of seedlings and cell lines derived therefrom. Despite a poor correlation between callus or suspension culture lines and parent plants, the mean alkaloid contents of strains derived from seedlings with higher alkaloid yields were nevertheless higher than the mean contents of strains derived from low yield plants.Generally, alkaloid accumulation decreased with the numbers of transfers. By permanent visual selection for fluorescent areas of the caDluses, however, a mean content of 0.1% harman alkaloids and 0.1% serotonin could be maintained, which was 10 times higher than in unselected callus cultures.The effects of medium constituents on harman alkaloid and serotonin accumulation were measured for a low yielding, faster growing suspension culture strain and a slowly growing, but high yielding cell line. This led to the development of a production medium without 2,4-dichlorophenoxyacetic acid and phosphate, and with Ca, Mg, and nitrate as sole macronutrients. When this production medium was used, the accumulation of harman alkaloids and serotonin was increased from 0.1 to 1% in the low yielding cell line and from 1 to 2% in the high yielding strain.Plant cell cultures offer a convenient system for studying biochemical aspects of secondary metabolism of higher plants. This is best illustrated by the progress in the enzymology of flavonoids (8) and indole alkaloids (26). The formation of secondary compounds is rather low in most plant cell cultures, however, and conditions allowing higher synthesis of these compounds often have to be established first. This may be achieved by analytical and biochemical selection for highly productive strains from established cultures (3,22,27), or by altering the medium composition for higher production (10,27,28). Another strategy applied to obtain high producing strains is to start cultures from selected, high producing plants (12,22,27 Cultures were incubated at 26°C in the dark on a gyratory shaker (120 rpm) and subcultured every 3 weeks. For media experiments, 0.5 g vacuum filtered 12-day-old cells of the stock suspension cultures were inoculated into 25 ml medium in 50-ml Erlenmeyer flasks.Analytical Methods. Freeze-dried material was extracted with methanol for analysis. Harman alkaloids were determined fluorimetrically either directly in phosphate buffer (0.2 M, pH 5.0), allowing distinction between aromatic and 3,4-dihydro-fl-carboline alkaloids, or after separation by HPLC (20). After separation on silica gel plates in CHCl3/MeOH/25% NH3 (10:4:1 and 5:4:1) alkaloids were quantified directly by their absorbance (harmine, harmol 320 nm; harmaline, harmalol 380 nm) and fluorescence (harmine, harmol: excitation 313 nm, emission 340 nm; harmaline, harmalol: excitation 365 nm, emission 415 nm) using a Shimadzu TLC-Scanner. Concentrations were expressed as harmine HCI and harmaline-HCI equivalents.Serotonin was also assayed fluorimetrically (excitatio...