IMMUNITY 223 



distinct components. If a large piece of filter-paper is dipped in 

 a rather dilute solution of a dye it is faintly stained ; if the 

 piece is cut into small pieces and these are immersed in turn 

 for a certain time, we find that the first pieces take up the colour 

 and leave almost none for the last. Substitute for the dye the 

 antitoxin and for the paper the toxin ; if the mixture is made at 

 one blow the toxin is attenuated in its whole bulk : if, however, 

 the mixture is made by several additions, the first portions of 

 the toxin take up the antitoxin and the later portions, not being 

 neutralized, remain much more toxic. 



Further, the fact that toxin and antitoxin mixtures (and in 

 general antigens and antibodies) become in time less dissolvable 

 and more stable is also explained by adsorption phenomena. 



When a piece of cloth is placed in a dyeing vat the dye 

 loses its attachment to the dissolving fluid and adheres more 

 and more intimately to the cloth until it can no longer be 

 redissolved. A similar example may be quoted in connection 

 with the precipitates produced by alcohol in certain albuminous 

 fluids, precipitates which are fairly easily redissolved in water 

 immediately after their precipitation, but which are no longer 

 soluble when a certain time has been allowed to elapse for 

 their aggregation. When three substances exist together, two 

 of them may compete with each other in the flxation of the 

 third ; it is thus that the protective action of certain substances 

 is to be explained, as, for example, the albuminous substances 

 of the blood which protect blood corpuscles against the action 

 of soap (Meyer) or of eel-serum (Frouin).i 



One thing that the physical theory has not yet explained is 

 the specificity of the reactions of immunity, but it is not 

 incapable of explaining even this. 



It is not difficult to imagine that slight physical modifica- 

 tions may change the affinities on which depend the molecular 

 attraction ; that is certainly no more difl5cult to imagine than 



1 Citrate of soda protects blood corpuscles against the agglutinating and 

 hfemoly tic action of sulphate of barium. The lecithin of ox-serum is held 

 in check as regards its action on guinea-pigs' corpuscles by some albuminoid 

 material. 



