Ill 



All these series agree with the above rule, allowing 

 for errors of experiment. (To avoid mistake we mention 

 here, that a great many series, which will be dealt with 

 below, deviate from this rule). 



Is it not by accident that the constant Kj in the so- 

 dium series increases quicker than in proportion to i/x, 

 while ammonia follows closely this rule, and lysin shows 

 a deviation in the inverse direction as NaOH (see p 15). 



If the rate of reaction at each moment were propor- 

 tionate to the square root of the amount transformed, 

 then in the case of small values of x, when 100 x could 

 approximately be considered as constant, the velocity 

 of reaction would be expressed by the equation. 



dx 



in which a stands for the quantity of toxin; from this 

 follows 



V/x = 2 Kat 



the amount transformed should then be proportionate 

 to both the square of the time of reaction and to the 

 square of the amount of active toxin. Both these rela- 

 tions are only approximately right, especially the first. 

 This gives for the three last mentioned velocities of 

 reaction the following values for t: yx. 



mean 



ammonia 1.7 2.0 2.2 2.5 2.1 

 sodium 6.4 6.4 5.2 4.7 5.7 



lysin 1.9 2.2 2.4 2.7 2.3 



In all three cases variations from the mean of about 

 20 percent occur. At all events these rules may have a 

 certain value for the purpose of interpolation. 



This furnishes a good example of a velocity of reaction 



32 



