THE ROLE of the COMPONENTS IN FIXATION 233 



It thus became clear that complement fixation was a general reaction, liable 

 to occur when any antigen was allowed to react with its specific antibody in the 

 presence of complement. 



The study of the relation between complement fixation and precipitation has yielded 

 results of great theoretical importance. Gay ( 1905) showed that the precipitate formed 

 by the interaction of an antiserum with the corresponding antigen frequently had the power 

 of absorbing complement. This suggested that precipitation and complement fixation 

 might be two aspects of a smgle reaction. Muir and Martin (19066) found that, although 

 there was a close correlation between precipitation and complement fixation, the correlation 

 was not absolute. They showed that complement fixation might occur in the absence of 

 precipitation, and that, in the presence of a constant amount of antiserum, there was a 

 particular amount of antigen that gave maximal complement fixation, while amounts 

 much greater, or much less, than this might fix little or no complement. 



Dean (1912) elucidated the cause of these earher discrepancies. In constant-antibody 

 titration series he found that mixtures giving maximal complement fixation contained 

 less antigen than mixtures giving maximal precipitation ; in the one antibody was in 

 excess, in the other antigen was in excess. He showed also that complement was fixed 

 in the early stages of precipitation, before visible floccuh had formed, and concluded 

 that fixation was maximal in slowly aggregating mixtures of antigen and antibody, where 

 the relative persistence of small aggregates ensured a large total absorbing surface. The 

 two maxima were not due to the reactions of different antibodies but were diff"ereut 

 secondary results of a single antigen-antibody reaction. Goldsworthy (1928) confirmed 

 these results, but showed that the relation of the maximal complement-fixing and pre- 

 cipitating mixtures described by Dean were determined by the particular system he used. 

 With slowly reacting antisera, the maximal fixing mixture coincided with the optimal 

 ratio ; with rapidly reacting antisera, it was in the region of antibody excess. The deter- 

 mining factor appeared to be the exposure to complement of a maximal absorbing surface 

 of antigen-antibody particles for the maximum period. Later studies show that there 

 is no constant relation between the point of maximal fixation and the established reference 

 points of constant-antibody titrations. With an ovalbumin system, the point coincided 

 with the equivalence ratio (Maltaner and Maltaner 1940) ; with pneumococcus carbo- 

 hydrate, Goodner and Horsfall (1936) found it equal to the ratio for maximal precipita- 

 tion in the antigen excess region. Rice and Sickles (1942) and Rice (1943) in a more 

 extensive study of the pneumococcal carbohydrate system record a close relation between 

 the ratio for optimal fixation and the antibody content of the serum, and note that the 

 ratio usually falls in the region of antibody excess. Apparently the size of the antigenic 

 particle employed appears to aff"ect the ratio. Both Spooner and Bawden (1935) and 

 Piatt (1936), working respectively with tobacco mosaic virus and pneumococci, found 

 the ratio in the antigen excess region. They also noted that the ratio was higher with 

 low than with high serum concentrations. According to Piatt the result is due to the 

 fact that with a particulate antigen the complement-absorbing surface is determined by 

 the concentration of antigen, and a large number of cocci sub-optiraaUy sensitized for 

 agglutination would absorb more complement than fewer fuUy sensitized cocci. 



The Role of the Components in Fixation. 



During the fixation of complement very little protein is removed from guinea- 

 pig serum. Haurowitz (1940) found little was added to the weight of specific 

 precipitates formed in its presence. Heidelberger (1941) estimated that 0-25-0-40 

 mgm. per ml. of complement protein, mainly C'l, was added to specific precipitates. 

 Similar proportional weights are added when heemolytic antibody combines with 

 washed red-cell stromata (Heidelberger and Treffers 1941). In the sheep red- 

 cell hsemolytic system, from 2 to 7 molecules of C'l were fixed for every molecule 



