882 LOCAL AND TISSUE IMMUNITY 



The discovery of antibodies, more specifically of antitoxins, in the year 1890, was 

 naturally followed by an attempt to explain the mode and locus of origin of these 

 significant reactive substances which are so characteristic of acquired immunity. The 

 earliest observations showed that in spite of the very direct reaction between antigen 

 and antibody, a marked disproportion between the amount of antigen employed and 

 antibody formed is evident.' Repeated bleedings and consequent mechanical removal 

 of a large proportion of antibody gives only a slight and temporary decrease in this 

 substance,' and indicates that the reservoir of the antibody lies outside the blood- 

 stream; and again, pilocarpin, a cell stimulant, increases antibody production.^ 



In fact no one, from the beginning, has seriously disputed that the cells of the 

 body form the background of immunity. In the process of phagocytosis the more 

 mobile cells, the polymorphonuclear leukocytes, are actively engaged in body de- 

 fense, although in the case of virulent infections their action to be efficient must be 

 preluded by a sensitizing or tropinizing effect on the invading bacteria, by the action 

 of the reaction antibodies of the body fluids. And these antibodies in turn must 

 originate in cells. In other words, the striking and important new properties of the 

 blood serum in conditions of immunity have to an undue extent distracted our atten- 

 tion from the mechanism through which these properties arise. Too little attention 

 has been paid to cells of the body if we except the polymorphonuclear leukocytes 

 whose function Metchnikoff^ and his pupils have so ably exploited. The antibodies 

 of the plasma, important as they are in diagnosis and at times in serum therapy, are 

 at best only a reflection of more fundamental processes that have gone on in the 

 cells. 



Studies have of course been undertaken from the very beginning which bear on 

 the origin of antibodies, but it now seems evident that progress has been in no small 

 measure inhibited by the too f ufly explanatory theory of Ehrlich which for many years 

 has dominated the field. This theory, as is well known, rests on certain assumptions 

 which are probably erroneous, or at all events unproved. The first of these assump- 

 tions is that those cells or parts of cells which are particularly fitted to unite with, or, 

 in the case of poisonous substances, be injured by any foreign protein, are the particu- 

 lar "receptors" which through their overgrowth constitute the specific antibodies. 

 We know that every attempt to prove this fundamental proposition has failed. The 

 central nervous system of animals susceptible to tetanus toxin is not equivalent to, 

 nor does it give rise to, tetanus antitoxin as Wassermann and Takaki^ thought. 

 Metchnikoff found, indeed, that an antispermotoxin from rabbits which neutralizes 

 the spermotoxic antiserum of guinea pigs immunized against rabbit spermatozoa is 

 formed as readily in castrated as in normal rabbits, thus proving that the antispermo- 

 toxin is not formed in those cells on which the antigen (spermotoxin) has a specific 

 and harmful effect. In short, Ehrlich 's theory, whfle claiming to offer a complete 



' Knorr, A.: Fortschr. d. Med., 15, 657. 1897. 



2 Roux, E., and Vaillard, L.: Ann. de VInst. Pasteur, 7, 65. 1893; Salomonsen, C. J., and 

 Madsen, T.: ibid., 8, 763. 1898. 



3 Salomonsen, C. J., and Madsen, T.: loc. cit. 



^MetchnikolT, E.: Immiinily in Infective Diseases. Cambridge University Press, 1905; Ann.de 

 rinsl. Pasteur, 13, 737. 1899. 



5 Wassermann, A., and Takaki, T.: Berl. klin. Wchnschr., 35, 5. 1898. 



