system and the recorded rates of changes in gas tensions, 

 we calculated the total change in these gases as a function 

 of time. Then we added 0^*02 to the system and took sam- 

 ples of algae every few seconds for the first few minutes and 

 then less frequently up to an hour. Each sample of algae was 

 killed immediately and a portion analyzed as described 

 earlier. A part of each sample was reserved and was dried on 

 a planchet to determine the rate of appearance of C^* in all 

 stable nonvolatile compounds. This rate proved to be the 

 same as the externally measured rate of uptake of CO2 and 

 C^^ between 20 and 60 seconds after the addition of C^^. If 

 unstable or volatile intermediates do precede these stable 

 compounds, they are equivalent in micromoles of carbon 

 to no more than 5 seconds photosynthetic fixation, according 

 to the shape of the fixation curve during the first 20 seconds. 

 We analyzed each sample by paper chromatography 

 and determined the radioactivity in each compound in each 

 sample. On the basis of the externally measured uptake rates, 

 at least 85 per cent of the carbon was found to be incorpo- 

 rated into individual compounds on the paper chromato- 

 grams during the first 40 seconds. At least 70 per cent of the 

 total carbon uptake rate could be accounted for by the ap- 

 pearance of C^** in compounds apparently derived from the 

 RuDP carboxylation reaction of the carbon reduction cycle 

 via the pathways shown in Figure 2. Another 5 per cent or 

 more was found to be incorporated via C1-C3 carboxylation. 

 About 5 per cent was found in unidentified compounds or 

 in glutamic acid, whose photosynthetic pathway is not defi- 

 nitely known. Of the 15 per cent not accounted for, some 

 may be in nonextractable polysaccharides, whose sugar phos- 

 phate precursors become labeled very quickly. More of the 

 unaccounted-for radiocarbon is undoubtedly in a large num- 

 ber of unmeasured compounds on the chromatograms. Each 

 of these compounds contains by itself too little C^"* to be 

 readily determined. In any event, it is clear that the known 

 fixation pathways are the only quantitatively important 



22 



