The capacity of the triose-phosphate shuttle and various combinations of glycolytic intermediates to substitute for the ATP requirement for fatty-acid and glycerolipid biosynthesis in pea (Pisum sativum L.) root plastids was assessed. In all cases, ATP gave the greatest rates of fatty-acid and glycerolipid biosynthesis. Rates of up to 66 and 27 nmol.(mg proteinfl.h 1 were observed for the incorporation of acetate and glycerol-3-phosphate into lipids in the presence of ATP. In the absence of exogenously supplied ATP, the triose-phosphate shuttle gave up to 44 and 33% of the ATP-control activity in promoting fatty-acid and glycerolipid biosynthesis from acetate and glycerol-3-phosphate, respectively. The optimum shuttle components were 2 mM dihydroxyacetonephosphate (DHAP), 2 mM oxaloacetic acid and 4 mM inorganic phosphate (referred to as the DHAP shuttle). Glyceraldehyde-3-phosphate, as a shuttle triose, was approximately 82% as effective as DHAP in promoting fatty-acid synthesis while 2-phosphoglycerate, 3phosphoglycerate, and phosphoenolpyruvate were only 27-37% as effective as DHAP. When glycolytic intermediates were used as energy sources for fatty-acid synthesis, in the absenc~ of both exogenously supplied ATP and the triose-phosphate shuttle, phosphoenolpyruvate, 2phosphoglycerate, fructose-6-phosphate and glucose-6phosphate each gave 48%, 17%, 23% and 17%, respectively, of the ATP-control activity. Other triose phosphates tested were much less effective in promoting fattyacid synthesis. When exogenously supplied ATP was supplemented with the DHAP shuttle or glycolytic intermediates, the complete shuttle increased fatty-acid biosynthesis by 37% while DHAP alone resulted in 24% stimulation. Glucose-6-phosphate, fructose-6-phosphate and glycerol-3-phosphate similarly all improved the rates