ENEKGY EFFICIENCY IN PHOTOSYNTHESIS 273 



carbon dioxide which persisted for 2 or 3 min after the light was turned 

 off, and then decreased to a minimum before rising again to give a steady 

 evokition of oxygen from photosynthesis. McAlister (1937), using an 

 infrared absorption method that is specific for carbon dioxide, found that 

 carbon dioxide continues to be given off from an algal suspension just 

 after the light is turned on. 



The manometric method calls for very careful control and impar- 

 tial treatment of data, particularly in view of the fact that the total 

 manometer-reading change is small and that the experimental error of 

 reading the manometer, perhaps 0.5 mm, introduces a large error into 

 the calculated value of the quantum yield. Kok (1948) gives a graph in 

 his Fig. 20 showing many determinations of the photons per molecule 

 $~i from manometric measurements, and it is clear that there is a large 

 spread in the values. 



Inouye (1951) carried out careful experiments with a two- vessel 

 manometer duplicating, as closely as possible, the conditions of Warburg 

 et al. (1950). The same apparatus and the same conditions of culturing 

 and algal treatment were used. The results were very much like those 

 obtained by these workers, and it was possible to get some <S>^^ values of 

 4 photons per molecule, but if all the data are used, the calculations give 

 <I>~^ values ranging from 2 to 12. If calculated in the same way, the 

 results of the investigators whose work was being duplicated have a 

 similar spread. 



Brackett (personal communication, 1952), as well as Emerson and 

 colleagues (Emerson and Nishimura, 1949; Nishimura et al., 1951), has 

 made a thorough study of the errors inherent in manometry. He con- 

 cludes that, at 120 oscillations per minute, over 3 min may be required 

 for equilibration. Higher shaking rates reduce the time constant, and 

 smaller amplitudes and denser suspensions tend to increase the time 

 required for equilibration through the liquid-gas interface. There are 

 dead spaces at the bottom and along the walls from 0.1 to 1 mm thick 

 where shaking is not effective. When dense cultures are illuminated 

 from below, a substantial fraction (up to about 15 per cent of the yield) 

 may be developed in these dead spaces from which the gases must diffuse. 

 This introduces a second slow process of increased importance during 

 illumination. 



According to Brackett (personal communication, 1952), if the size and 

 shape of the vessels are different, as in some of the differential Warburg 

 methods, the stirring efficiency of the two vessels can be widely different. 

 Hence the equilibrium time for the small vessel may be longer than that 

 for the large vessel, and the ratio 7 disturbed. 



Values obtained during the period of transient equilibrium show spuri- 

 ously high values at the beginning, then decay logarithmically toward the 

 correct value. 



