227 



without regard to whether the reaction is determined before 

 or after the addition of the „haemoglobin". 



As „haemoglobin" human blood was used in the first in- 

 vestigations in July 1918, a drop of which was spread over 

 the surface of an agar plate (of autoclaved agar). This is the 

 original method of Pfeiffer. It was natural to assume that 

 human blood might be better than that of other animal species 

 when it was a question of growing haemoglobinophilic bac- 

 teria obtained from man; but in my experience and in that 

 of practically all other authors this has proved not to be the 

 case. Pfeiffer's bacillus could be grown with greater certainty 

 on agar mixed with a liberal amount (20— 25%) of defibrina- 

 ted horse blood, but the growth was slight. 



For the inoculations from healthy persons in September 

 1918 I used agar mixed with haemolysed blood for the 

 first time on a large scale. Although the agar was autoclaved 

 this medium was very satisfactory, the growth being con- 

 siderably richer than on blood agar. I was more certain of 

 getting a good growth however when the agar was sterilised 

 at 100° as mentioned above, and such agar was therefore em- 

 ployed as the basis of the „haemoglobin" medium 1 in all the 

 subsequent experiments. 



Haemolysed horse blood was used as „haemoglobin" for 

 a long time with excellent results, prepared in the following 

 way: The flask containing the defibrinated blood was allowed 

 to stand for a couple of hours so that the blood corpuscles 

 could sink. The supernatant serum was then poured off as 

 completely as possible and replaced by distilled water, which 

 speedily dissolved the blood corpuscles. The haemoglobin solu- 

 tion was distributed in flasks with 20c.c. (corresponding to 

 about 6c.c. blood corpuscles) in each, which can be preserved 

 in the ice-box for several months before use. The freezing 

 completes the haemolysis. On the addition of the liquefied 

 contents of a flask of 600c.c. agar (liquefied and cooled to 

 50—40°) and then pouring the mixture into plates a moderately 

 red, transparent medium was obtained which for the first I1/4 

 years consistently produced a rich growth of Pfeiffer's ba- 

 cillus (isolated colonies 2—4 mm. in diameter) and which on 

 the whole, was ideal. I was rather surprised to find that during 

 the winter and spring 1919—20 it gradually became less satis- 



15* 



