THE TOXINS 171 



errors may be avoided by employing young bacteria from 

 cultures of twelve to eighteen hours on solid media. 



2. But in many cases difficulties are met with when one 

 attempts to determine the relations between this endotoxin and 

 the soluble poisons secreted by the same bacteria and 

 apparently true toxins. For example, the cholera vibrio cer- 

 tainly contains an endotoxin in Pfeiffer's signification, but it is 

 none the less true that the cultures contain an excreted poison, 

 the poison studied by Roux, Metchnikoff, and Salimbeni, on 

 which the hope of a serotherapy in cholera has been founded. 

 It is this poison which is absorbed into the body and produces 

 the cramps, the chilling, and death, whereas the vibrios, however 

 numerous they may be, remain in the intestine and only very 

 rarely invade the blood and the tissues. " The free poison,'' 

 says Pfeiffer, " does not exist in cultures except when these are 

 already old and contain many vibrios already destroyed and 

 autolysed." The cholera toxin of Roux and Metchnikoff is, 

 however, chiefly secreted during the first days of a culture 

 kept under the requisite conditions (a virulent vibrio and a 

 well-aerated culture). This example is not in favour of 

 Pfeiffer's opinion, at least in the case of cholera, for although 

 he admits that the endotoxic bacteria can also secrete other 

 soluble poisons, he maintains that the latter are quite different 

 from endotoxins and do not possess their specificity. 



3. Do the endotoxins give rise to antitoxins in the body of 

 an immunised animal, as do the soluble toxins ? This is the 

 most disputed question of all. Without denying the existence 

 of anti-endotoxins in principle, Pfeiffer's school considers that 

 hitherto none have been obtained which have passed satis- 

 factory tests, and that the sera prepared against plague and 

 typhoid fever have not hitherto been successful precisely 

 because they do not contain anti-endotoxin. 



To prepare an anti-endotoxin, as in the preparation of an 

 antitoxin, it is necessary to inject several times into an animal, 

 for example, the horse, the toxic substance, in this case the 

 bacterial bodies, entire or broken up. Good results are not 

 obtained when endotoxins are injected subcutaneously ; intra- 



