STUDIES ON HJEMOLYSINS. 91 



stances, provided the receptors fitting these are present exclusively 

 in- vitally important organs, e.g., the central nervous system. This 

 perhaps .explains the circumstance that it is exceedingly difficult 

 io produce an antitoxin in mice and guinea-pigs with unchanged 

 tetanus poison, while this is easily effected when toxoids are used. 

 On the other hand, an immunization of rabbits by means of unchanged 

 tetanus poison is very easy to attain, because in these animals, 

 as is shown by the investigations of Donitz and of Roux, the greater 

 part of the receptors lies outside of the poison-endangered central 

 nervous system. 



However, even without any development of illness it is not at 

 all necessary that antibodies should be produced in every case in 

 which an anchoring occurs. Metchnikoff, for example, has called 

 attention to the fact that with frogs in whom every trace of illness 

 is avoided by keeping them cool (as we know from Courmont's beau- 

 tiful observations) it is impossible to produce any tetanus antitoxin. 

 Investigations by Morgenroth have confirmed this result and shown 

 further that even by treatment with toxoids under various conditions 

 it is impossible to produce a trace of immunity. Probably in this 

 particular case these results indicate that the regenerative powers 

 of the frog's tissues are not equal to these extraordinary demands. 



Such an explanation for failure of the antibody to develop is, how- 

 ever, much less probable in the case of warm-blooded animals; and 

 as the number of experiments increases, these cases are becoming 

 more frequent. Probably all who have busied themselves with the 

 subject will have found, particularly with the artificially .produced 

 cell poisons, that in some cases it is extremely difficult if not impos- 

 sible to effect the production of anti-immune bodies. Thus, 

 Metchnikoff injected a series of guinea-pigs with specific spermo- 

 toxin, a substance which certainly finds receptors in the guinea-pig's 

 organism. Despite this, he found but two cases in which even a 

 suggestion of antispermotoxin could be demonstrated. In the 

 case of a dog injected with a specific dog blood immune body derived 

 from a sheep, we have failed despite long-continued treatment to 

 produce any anti-immune body. With this series of phenomena 

 must also be classed the fact that it is extremely difficult if not impos- 

 sible in a number of animal species to produce antienzymes by the 

 continued injection of certain enzymes. 



The explanation of these facts presents two possibilities: First, 

 the receptors concerned in the particular case may be of peculiar 



