A GENERAL RESUME OF IMMUNITY. 523 



unity in the laws which govern the action of antibodies; it would 

 be to distribute them into divisions, separated by impassable bar- 

 riers, and I have already mentioned in speaking of Ehrlich's classifi- 

 cation how inacceptable such a system is. Certain antigens are toxic 

 and certain others have a greater tendency toward precipitation 

 than others; such differences between the antigens, however, do 

 not prevent us from supposing that their antibodies are very similar 

 and that their action is subject to the same interpretation in the 

 various instances. 



A phenomenon which has justly attracted the attention of experi- 

 menters is the increasing stability of complexes of antibody-antigen 

 following their formation. It would seem as if these complexes, 

 which are easily dissociable in the beginning, subsequently become 

 consolidated in some manner so as to resist decomposing influences 

 better. Facts of this sort have been noted by Landsteiner and 

 Jagic with agglutinins, and by von Dungern, Sachs, and Otto 

 and Sachs with the antitoxins of certain poisons such as the toxin 

 of botulismus, arachnolysin and the like. Facts of this nature, it 

 seems to me, should be interpreted in accordance with the adsorp- 

 tion theory. As Nernst has noted, certain dyes as they unite 

 more and more intimately with the object which they stain, lose 

 correspondingly their adhesion for the surrounding fluid; when 

 they have become fixed they show a distinct tendency to become 

 insoluble. It seems that we may treat as an analogy the fact 

 that various albuminous fluids treated by such agents as alcohol 

 give precipitates which redissolve easily in water when they have 

 been recently formed, but become more and more insoluble when 

 kept. The adhesion which unites the albuminous molecules with 

 one another and gives them a solid state becomes preponderating 

 and opposes the dissemination which the contrary effect, the 

 adhesion for water, would tend to produce. The consolidation of 

 complexes of antibody and antigen is apparently a phenomenon 

 of the same class. 



The quantities of antibodies which antigens can fix depending on 

 their concentration, have been determined particularly with aggluti- 

 nation phenomena by the important researches of Eisenberg and 

 Volk. The greater the concentration of the antibody the more the 

 fixation of the antigen, but the quantity fixed does not increase so 



