ABSTRACTwith ER vesicles has not been ruled out (22). It is possible that ER-chloroplast associations which exist in vivo (2) may be maintained in certain chloroplast preparations.In earlier studies with Hippeastrum petal protoplasts (8), we found that 5 to 15% ofthe activities of key phenylpropanoid and flavonoid pathway enzymes was associated with a particultate subcellular fraction which was enriched in ER membranes. There is no published evidence to support the presence on or in the tonoplast of enzymes ofthese pathways or ofenzymes leading to other secondary products. No evidence was found to suggest the occurrence of UDPG-flavonoid glucosyltransferase (an enzyme near the end of the flavonoid biosynthetic sequence) on or in isolated tonoplast of Hippeastrum (24).To expand our understanding of the intracellular location of secondary product formation in plants, we fractionated petal homogenate-derived membranes on sucrose density gradients and utilized the EDTA shift method to demonstrate that various key enzymes ofphenylpropanoid and flavonoid metabolism are, at least in part, ER-bound.Biochemical and ultastructural studies have provided evidence to support the contention that phenylpropanoid and flavonoid biosynthesis occur on the ER (5,11, 22, 25). A number of investigators have shown that certain enzymes ofthese metabolic pathways are associated with microsomal membranes (11,20, 22) and Czichi and Kindl (3) demonstrated coupling between PAL2 and C4-H in microsomal preparations of potato and cucumber. Compelling micrographs of Douglas Fir callus cells suggest that 'tannin'-containing vesicles can arise from dilated regions of the rough RER, (16), and other ultrastructural studies (25) suggest that ER is the site of tannin synthesis in several tannin accumulating plants. Chloroplasts have also been suggested as being sites of phenylpropanoid and flavonoid metabolism, but the evidence is contradictory. Various flavonoids, certain cinnamic acid derivatives (9) and some phenylpropanoid and flavonoid biosynthetic enzymes were reported to occur in isolated plastids (14). Other reports conclude that plastids lack key enzymes ofthese pathways (9, 22). The possibility that plastid preparations which showed enzyme activities were contaminated ' The investigation reported in this paper ) is in connection with a project ofthe Kentucky Agricultural Experiment Station and is published with approval of the Director.