A GENERAL RESUME OF IMMUNITY. 497 



when it would force facts? What is perhaps the most obscure and 

 important question with which immunity confronts us? Evidently 

 the question of the specificity of immunization as shown in the 

 specificity of sera. The solution of this question, however, is 

 probably wholly unsuspected. Anyone who should attempt to 

 furnish a solution at the present moment, or to give a distinctive 

 explanation as a result of reasoning on the subject, would neces- 

 sarily be led to support it by undemonstrated facts. It is better, 

 then, to seek for the truth without wishing to define it before we 

 have found it. 



My objection to too generalized conceptions do not in the least 

 prevent me from recognizing the ingenuity and the genius of those 

 who, like my esteemed colleague Ehrlich, have proposed a general 

 interpretation of specificity. Ehrlich's theory has influenced too 

 many minds, has become too generally known on account of the 

 well merited reputation of its author, not to deserve thorough con- 

 sideration. My own impression of the theory has not been favor- 

 able, and it has seemed my duty to combat it. Its principal fault 

 to my thinking is that it is not, strictly speaking, a theory, but 

 rather an assertion of a certain number of undemonstrated facts. 

 According to Ehrlich, antibodies are produced as follows: in the 

 first place the antigen when introduced into the animal body meets 

 with a substance with which it unites. So far we all agree. It is 

 quite certain that foreign substances which lead to the formation 

 of antibodies are taken up by the tissues and produce a reaction 

 with certain substances in the body to which, for convenience, the 

 name of receptors may be given. The body restores these destroyed 

 receptors by producing new ones identical with the original. But 

 according to Ehrlich, these new receptors are over-produced to such 

 an extent that they flow over into the fluid of the body, retain their 

 essential property of uniting with the antigen, and are then to be 

 designated as the antibodies of the serum. In short, the antibody 

 is identical with the receptor affected by the antigen. To draw 

 such a conclusion is, however, to affirm a fact that has never been 

 demonstrated. Wholly different hypotheses might as legitimately 

 be offered. We might suppose, for instance, that the body of the 

 animal that is being immunized, instead of reproducing old recep- 

 tors in large amount without changing them, builds up substances 



