NEW FLASHING LIGHT EXPERIMENTS 1473 



son and Arnold, a levelling-off of the yield vs. energy curve at a height cor- 

 responding to P = Chlo/2000, in disagreement with Tamiya's suggestion. 



Assuming that both the measurements of Emerson and Arnold (and 

 others) with Chlorella pyrenoidosa, and those of Tamiya with Chlorella 

 ellipsoidea, are experimentally rehable, it appears that the flash yield is 

 limited, in the first case, by a monomolecular back reaction with a rate 

 constant c^ 100 sec.~^, permitting only a single "round" of the stabilizing 

 reaction; and, in the second case, by a bimolecular back reaction allowing 

 time for several such rounds. This is not impossible; but it seems a re- 

 markable coincidence that the elimination (or slowing down) of the first- 

 order back reaction should be coupled with the emergence of a second-order 

 back reaction imposing almost the same ceiling on the maximum rate in 

 continuous light. (It will be recalled, however, that in chapter 28 we did 

 discuss the possibility that photosynthesis may contain several "bottleneck 

 reactions" adjusted to about the same maximum capacity.) 



In principle, the conclusion that both first-order back reactions (detautom- 

 erizations) and second-order back reactions (recombinations) can occur 

 in photosynthesis is plausible. The same is generally true of reactions in 

 which unstable intermediates are formed in the condensed state, e. g., the 

 decomposition of water by high-energy radiations. 



The observations of Strehler and Arnold (19ol), to be described in Chap- 

 ter 37C, section 5, may be worth mentioning in this connection. They 

 found a delayed re-emission of hght quanta by chlorophyll in vivo, probably 

 associated, as chemiluminescence, with a back reaction in photosynthesis. 

 The decay of this emission was a second-order process at 6.5° C, and had 

 an order of between one and two at 28° C. The absolute rate of the lumines- 

 cence decay was considerably slower than the decay of the dark reaction 

 derived from flashing Ught experiments: the period of half-decay was as 

 long as 1.5 seconds at 28° C. (but only 0.5 second at 6.5°). It is, therefore, 

 an open question whether this chemiluminescence is associated with the 

 same two (or, at least, one of the two) back reactions revealed by the ex- 

 periments of Emerson and Arnold, and of Tamiya and Chiba. 



More extended measurements of the flash yield vs. flash energy curve, 

 employing a variety of organisms undoubtedly are in order; the first need 

 is for a device permitting high intensity instantaneous flashes (such as can 

 be obtained from xenon discharge tubes) to be repeated at close intervals 

 (0.01 second or less). 



One fundamental question to be settled is whether it is true — as postu- 

 lated at the beginning of this chapter — that for sufficiently brief flashes the 

 integrated intensity (i. e., the energy, E) of the flash is all that matters, 

 while the true momentary intensity of illumination is irrelevant. Tamiya 

 and Chiba reported having conflrmed this assumption; but the figures they 



