466 ItADIATlON JJIULOUY 



to a light oi coufstant intensity, for short exposures the number is rela- 

 tively higher than for longer exposures, so that the photoreactivation 

 curve shows a deviation from linearity (Sect. .3-5a). If the same curve is 

 obtained by giving all doses of light as series of short flashes separated by 

 long dark periods, the deviation from linearity disappears. 



The two types of experiments just described show that the amount of 

 photoreactivation obtained with a given light dose depends not only on 

 the light reaction, but also on dark reactions connected with it. Bowcn 

 was able to demonstrate that at least one dark reaction involved in photo- 

 reactivation must precede the light reaction, because photoreactivability 

 of bacteria infected with ina('ti\'e phage is affected by a temperature 

 change from 40° to 0°C' if the temperature of the system is changed acvcral 

 minuU'fs before a short light flash (of o sec length). If the temperature is 

 changed immediately after the flash, photoreactivation is not affected. 



The results of these experiments can be understood on the basis of the 

 following model. The probability for a photoreactivable particle to be 

 reactivated is proportional to the time integral of the concentration of a 

 photoproduct X*. Production and destruction of X* are determined 

 by the following reactions. 



The first two reactions establish an ecjuilibrium between two bacterial 

 components M and X"^ which may or may not be combined with the 

 infecting phage; they are both temperature dependent. The third reac- 

 tion is the photochemical reaction, the fourth is the reaction by which X* 

 is destroyed and eliminated, and the fifth is the reaction of X* with phage, 

 which constitutes photoreactivation. 



After a long dark period, N has an e(iuilibrium concentration 



[X] = ^^. (12-1) 



and X* has concentration zero. 



After a long period of contiinious light. X has an e(iuilil>rium con- 

 centration 



and X*, 





which satisfies the hyperbolic i-elation found experimentally 



