Ill 



larity which was first noticed, of the hiemolysis being 

 nearly proportionate to the square of the quantity of 

 toxin is only correct to a certain extent; that is to say 

 for the blood dilutions, varying from 7,5 % to 10,8 %. 

 In the case of sodium hydrate this regularity is only 

 found in a very small portion of the series. We shall 

 later on return to the explanation of this phenomenon. 



Experiments were also undertaken to find the action 

 of bases on blood corpuscles diluted with a solution 

 of cane sugar. 



The results are given in the following table, in which 

 a in cc. signifies the added amount of the base which 

 heads the table. The ciphers in parenthesis belong to 

 an older series of experiments with blood from another 

 bleeding. 



Allowing for errors of experiments, equivalent 

 amounts of the three alkalies give similar values. Hut 

 the action of the hydrates of the alkaline earths is much 

 smaller (excepted perhaps the case of very diluted solu- 

 tions). This seems to rise from the fact, that on the 

 test tubes with calcium and baryum a precipitate is 

 formed, which seems to consist of faint red crystal 

 needles. The product of haemolysis seems in this 

 case not to be so easily dissolved as that of alka- 

 lies. The hnemolysis is therefore nearly constant, and 



23 



