DIPHTHERIA. 127 



fact is contrary to the assumption that the chemical poison 

 of the diphtheria hacillns is a ferment or en/yme." 



The fluid was tested for basic products, I nit with wholly 

 negative results, except that small amounts of kreatinin and 

 cholin were found. It was also distilled at from 20 to 

 3~) in a vacuum, and the distillate was found to be inert. 



The poisonous substance was found to be insoluble in 

 alcohol, soluble in water, and non-dial v/able. It was pre- 

 cipitated by saturation with ammonium sulphate. 



The substance; was obtained by allowing the germ-free 

 filtrate, after being rendered feebly acid with acetic acid, to 

 fall into a large volume of absolute alcohol. It was puri- 

 fied by repeated solution in water and precipitation with 

 alcohol. It contains a large amount of sulphur, and re- 

 sponds to the biuret and Millon tests. It is, therefore, 

 classified among the albumins. Since it is not precipitated 

 by saturation with magnesium sulphate at 30, it cannot 

 belong to the globulins. The fact that it is precipitated by 

 saturation with ammonium sulphate, and that it does not 

 dialyze, shows that it is not peptone. It is, therefore, 

 classified by BRIEGER and FRANKEL among the albumins, 

 and is designated as a " toxalbumin." 



The special reactions and the results of an ultimate 

 analysis of this substance have already been given (page 20). 



This proteid induces in animals all the symptoms and 

 post-mortem appearances which have been mentioned as 

 following the administration of the filtered cultures. It is 

 to be noted that the injection of small quantities of this 

 proteid (2 milligrammes per 1 kilogramme of the body- 

 weight of the animal) does not produce its effects until after 

 the lapse of weeks, and possibly months. This peculiarity 

 in action distinguishes this class of substances from all other 

 chemical poisons, and it has received as yet no satisfactory 

 explanation. There is no reason for l>elieving that the body 

 obtained by BRIEGER and FRANKEL is chemically pure, 

 and until it has been obtained in this condition we can 

 only sp:-eulate concerning its true nature. 



It should be remarked that the LoFFLER bacillus shows 

 not only marked morphological variations, but that it is 



