182 IN ACTIVATION OF COMPLEMENT BY LIGHT 



Examination of the horizontal columns in Table IV headed k 

 will show that the values in any one column are essentially con- 

 stant. This is what is to be expected in case the course of the 

 process is that of a monomolecular reaction, the equation of whose 



reaction curve or isotherm is - log = k, where a is the original 



t a—x 



concentration of the reacting. substance, x its concentration after the 



time /, and k a constant. When the amount of injury is still very 



slight the errors of titration may cause relatively great variations 



of this value such as are to be found in Experiments 62 and 70 



(Table IV). 



In Fig. 2 the probable error of titration is less than the diameter 



of the circles surrounding each point. It is at once apparent that 



errors in titration will not account for the irregularity of some of the 



observed curves. But it is evident that such deviations from the 



course of a monomolecular reaction are not an essential part of the 



process of photoinactivation, because the average of all the curves 



is almost exactly the theoretical monomolecular reaction isotherm 



(Table IV). Only during the first 2 minutes, when the influence of 



. . . . 1 a 



errors m titration is very great, does the observed value of - log 



t a—x 



for the average curve deviate from its mean value, 0.0289, by more 



than a fraction of its probable error, 0.0013. 



Photoinactivation is therefore a process which goes on at a rate 



proportional to the concentration of a single disappearing molecular 



species, which we may consider to be the substance responsible for 



the hemolytic property of serum. The deviation of the observed 



courses of photoinactivation from monomolecular curves also demands 



explanation, but may best be considered in connection with evidence 



which will be presented in a subsequent paper. 



SUMMARY. 



The photoinactivation of complement has been studied with a 

 view to determining if possible how many kinds of molecules dis- 

 appeared during the reaction. It was found that : 



