MECHANISM OF ANTIGEN- ANTIBODY REACTIONS 215 



cohesive force is unaffected, ag'glutination occurs wlicnever the 

 charge is reduced by electrolytes to a point below a critical level 

 of about 15 millivolts; that is, the unaffected cohesive force now 

 becomes relatively greater than the force of repulsion. (4) Salt 

 in high concentration depresses the cohesive force of unsensitized 

 bacteria so that no agglutination occurs even thougli tliere may 

 be no measurable charge ; i.e., the cohesive force is now so small 

 that it is always less than the repelling force. (5) When bacteria 

 are treated with immune sera, their cohesive force is in some 

 manner protected from this depressing effect of strong salt (e.g., 

 physiologic salt solution, etc.) and agglutination is determined 

 solely by the charge; that is, whenever the potential of the sensi- 

 tized bacteria is I'educed by electrolyte to a point below 15 milli- 

 volts, the suspension agglutinates." 



Shibley continues by saying that ''this explains the observation 

 of Bordet, confirmed by Northrop and DeKruif in the course of 

 the work being cited, that electrolytes are essential for specific ag- 

 glutination. That is to say, the salt, routinely used in the ordinary 

 reactions, reduces the charge on the bacteria so that this charge 

 comes to lie in the 15 millivolt agglutination zone, and tlie cohesive 

 force of the sensitized bacteria, being insusceptible to the depress- 

 ing effect of the electrol;^i;e, is now relatively greater than the 

 repelling force and flocculation occurs. (6) Their results refute 

 the idea that combination of antibody and organism is caused by 

 difference of sign of the charges carried by the two substances ; but 

 are in agreement witli the assumption that the agglutinin forms 

 a film on the surface of the organism." 



Effect of Sex.sitiz.\tion on Charge.— Shibley then proceeds to 

 offer more experimental evidence in support of the hypothesis that 

 the union of antigen and antibody consists in the coating of the 

 former by the latter. He reports that A\'hile both normal and im- 

 mune sera depress the charge in the pneumococcus, the effect of 

 normal serum is slight when compared with the marked reducing 

 effect of specific immune serum. He studied the effect of specific 

 immune serum for Type I pneumococcus, a strain of a hemolytic 

 streptococcus, Esch. coli, E. tiiphosa, S. paratijphosi A, Myco. tu- 

 berculosis and for the Flexner, Shiga and J\It. Desert strains of dysen- 

 tery bacillus. All of tlie specific imnume sera possessed the charge-re- 

 ducing effect except the three for dysentery bacilli. The three 



