When cyclic photophosphorylation is inhibited in Chlorella vulgaris cells by carbonylcyanide-trifluoromethoxy phenylhydrazone, photosynthetic CO0-fixation under anaerobic conditions exhibits a distinct lag. Under the same conditions, the light-dependent formation of ribulose diphosphate shows also this lag. It is concluded that cyclic photophosphorylation is required to fill up the pools of phosphorylated intermediates of the Calvin cycle at a time when noncyclic photophosphorylation cannot yet efficiently operate. Under aerobic conditions, the initial energy demand can be accommodated by respiratory ATP or cyclic photophosphorylation or both. Evidence for stoichiometric participation of cyclic photophosphorylation in photosynthesis is still lacking. Whereas evidence for the existence of cyclic photophosphorylation in vitro and in vivo has accumulated from many laboratories, no agreement has been achieved up to now on the role of this photophosphorylation. Energy generated by cyclic photophosphorylation can be used under various experimental conditions by autotrophic cells for uptake and assimilation of organic molecules (5, 18, 20), uptake of ions (4, 9), and for protein synthesis (11, 14). It has been questioned, however, for several reasons, whether ATP generated by cyclic photophosphorylation is stoichiometrically required for photosynthetic CO,-fixation (16, 17), as has been suggested by Arnon (1) and accepted by most textbooks of physiology and biochemistry. In addition, it has even been suggested that cyclic photophosphorylation only occurs under anaerobic conditions and that its physiological importance, therefore, seems doubtful (3). Recently, Schurmann et al. (13) have published data which they claim to prove the participation of cyclic photophosphorylation in CO,-fixation of isolated chloroplasts. In reality, these data only show that cyclic photophosphorylation is required for the initial phase of photosynthetic CO-fixation; e.g. for filling up pools of phosphorylated intermediates of the Calvin cycle.' This fully agrees with previous suggestions from in vivo experiments (6). In this paper, evidence is presented that cyclic photophosphorylation is obligatorily required in vivo under certain experimental conditions during the initial phase of CO,-fixation. It is used for example to increase the level of RuDP. 1Abbreviations: RuDP: ribulose diphosphate; CCP: carbonylcyanide-trifluoromethoxy-phenylhydrazone; Calvin cycle: reductive pentose phosphate cycle. MATERIALS AND METHODS Chlorella vulgaris (strain 211/1 lh, G6ttingen) was grown as described previously (15). The rate of CO,-fixation was determined by adding NaH1'CO0 at a f.al concentration of 756 pg/ml (= 0.272 uc/ml) to an algal suspension containing 165 ,ug chlorophyll/ml. At the times indicated, samples of 0.3 ml were added to 1 ml ethanol-acetic acid v/v (95:5). Aliquots were counted on metal planchets at an efficiency of 40%. The determination of Pi and of RuDP was as described before (7).