212 THE ANTIGEN-ANTIBODT REACTIONS 



of the antigen-antibody compound ; but it is possible that some of them, for instance the 

 lessened absorption of agglutinins by typhoid bacilli at a pH below 40 observed by de 

 Kruif and Northrop, may be due to the destruction or inactivation of the bacterial antigen 

 concerned (see Duncan 1935). 



We are, in the study of such reactions as these, dealing with systems of great 

 physical and chemical complexity, about which we know relatively little. We 

 should therefore be very cautious in ascribing any observed phenomenon to the 

 influence of a particular physical or chemical factor until other possibilities have 

 been satisfactorily excluded. 



Before considering the effect of electrolytes, and other factors, on the actual 

 flocculation of bacteria, it will be convenient to note certain quantitative data in 

 regard to the absorption of agglutinins by bacterial cells, when these ancillary 

 factors are kept constant. Here, as elsewhere, the observed relationships are 

 inexplicable on the hypothesis of firm union in simple multiple proportions. For 

 instance, Eisenberg and Volk (1902) showed that, when a constant amount of 

 bacterial suspension was allowed to react with varying concentrations of an aggluti- 

 nating serum, proportionately more agglutinin was absorbed from the more dilute 

 serum, though absolutely more was absorbed from the more concentrated serum. 

 When a constant dilution of antiserum was absorbed with varying amounts of 

 bacteria, the amount of agglutinin bound did not bear a linear relation to the 

 absorbing dose of bacterial cells. With increasing doses of bacteria, the amount 

 of agglutinin removed became proportionately less. Such a relationship is analo- 

 gous to that observed in any adsorption process. Craw (1905) noted, in studying 

 the absorption of agglutinin by bacteria, a phenomenon analogous to that described 

 by Danysz in the toxin-antitoxin reaction (see p. 216). 



The Zone Phenomenon in Agglutination. 



The system with which we are dealing in the agglutination reaction diflFers sharply 

 from that concerned in precipitation in that the floccules consist mainly of the antigen- 

 carrying material, the concentration of which in the initial mixtures can be judged from 

 measures of opacity, or from actual counts, of the bacterial suspension. For this reason 

 quantitative measures of the agglutination reaction are most conveniently performed by 

 the constant-antigen method, varymg the concentration of antisera. The constant- 

 antibody technique is possible, but has the disadvantage that the initial heavy opacity 

 of concentrated antigen suspensions may mask moderate degrees of flocculation. Titration 

 of bacterium-agglutmin systems by both techniques yields optimally flocculating zones, 

 though they are less well defined than precipitin zones. The failure of some bacterial 

 suspensions to agglutinate in the higher concentrations of the dilution series of homologous 

 antiserum, the " pro-zone," has been noted by many workers. In other systems, the 

 inhibition of agglutination is not absolute but can be observed, particularly with light 

 bacterial suspensions (Heuer 1922, da Costa Cruz 1929). 



Duncan (19326) and Miles (1933) studied the constant-antigen optimum and its relation 

 to the constant-antibody optimum. As in precipitin systems, the antibody content of 

 the constant-antibody optimal mixtm-e was greater than that in the constant-antigen 

 optimal mixture. In one system, Duncan found the difference to be six-fold. Equiva- 

 lence zones cannot be fuUy defined, since removing a precipitate by centrifugation in the 

 zones of antigen-excess also brings down unagglutinated antigen, but the equivalence 

 zone appears to cover a wide region of the constant-antigen series, including the optimal 

 mixture. 



Shibley (1929) observed the pro-zone phenomenon in antisera that had been exposed 

 to moderate heat, and attributed it to a modification of antibody, whereby it was prefer- 

 entially absorbed by the bacteria, but had lost its power to sensitize. The probable 



