Ill 





9 15 19 min. 



C. 1 cc. l /io ii NH a \ 100 -x 91.5 80 57 44 perc. 



\ 100 x 

 I K t 0, 



,0077 0,0107 0,016:5 0,0187 



( i 2.25 6.2 8.2 min. 



D. 2 cc. Vto n NH 3 - 100 x 88 70 59 perc. 



[ K! 0,026 0,025 0,028 



It appears from these examples that K, usually in- 

 creases very rapidly during the process. Only in the 

 last case Kj is nearly constant and this is probably owing 

 to errors of experiment. It is easy to imagine, that this 

 increase of the velocity of reaction, as time goes on, 

 depends on the circumstance, that the red blood corp- 

 uscles' cellular membrane must be destroyed before 

 haemolysis can occur. In the beginning no haemolysis 

 at all takes place. Little by little the most accessible 

 alkali or lysin-bound corpuscles are haeroolysed, and later 

 on the most resistent, whose membranes will be destroyed 

 only after a prolonged action of the toxin. In chemistry 

 several similar phenomena occur, especially when solu- 

 tions act upon solid bodies; and it is then said, that a 

 certain time of induction is necessary, before the attack 

 occurs. Physicists often find it difficult to understand 

 what this lime of induction means. In the present case 

 it is easily explained why there should be a time of 

 induction, this being here equivalent to the time requi- 

 red for the destruction of the cellular membrane. 



In order to clear up the connection between the con- 

 centration of toxin and the velocity of reaction, we must 

 therefore proceed so that we ascertain if a certain dose 

 of toxin produces the same degree of haemolysis during 

 a certain time as half that dose of toxin e flee Is in double 

 that lime. From the above figures this would seem to 



28 



