THE CARBON DIOXIDE FACTOR 



1897 



alkaline or acid suspension medium) and then through an electrode vessel 

 filled with dilute bicarbonate solution ; pH changes were recorded in this 

 vessel, and CO2 changes in the gas calculated from them. In experiments 

 of both types, CO2 saturation was found to require much higher CO2 con- 

 centrations than anticipated from earlier data. 



[CO2], moles/liter x 10^ 





>- 

 <0 



5 

 o 



2 



P 0/ 



' CO2 » /o otm. 



Fig. 37D.4. Carbon dioxide curves of Scenedesmus in 2.5 X 10 "^ M KH2PO4 at two 

 light intensities (after Rosenberg 1954). 28° C. Suspension density 1 volume percent. 

 The curves are of "Bose-type" (fig. 26.3) rather than of "Blackman type" (fig. 2G.2), 

 as they should be if diffusion were the cause of [C02]-dependence (p. 923). 



Fig. 37D.4 shows typical CO2 curves obtained by Rosenberg (in bicar- 

 bonate-free medium). The relation of the curves obtained at two differ- 

 ent light intensities argues against the attribution of the slow increase to 

 diffusion limitations such as may occur in a vessel stirred only by the gas 

 stream. (According to Chapter 27, diffusion limitation must produce curve 

 families of the "Blackman type," rather than such of "Bose type," appar- 

 ent in fig. 37D.4; compare above the remarks of Steemann-Nielsen on the 

 measurements of Harder.) Very similar [C02]-curves were obtained also 

 in 2 X 10~^ M NaHCOs; at the higher bicarbonate concentrations satura- 

 tion w^as reached earlier. 



In Gaffron and Rosenberg's experiments, the saturation rate, once 

 reached, was the same in "COj-adapted" and "bicarbonate-adapted" algae 

 (c/. fig. 37D.2); this result does not support the suggestion of Warburg 



