IMMUNOLOGICAL METHODS IN THE STUDY OF VIRUSES 533 



in contrast to mouse lung virus, which reacted with antiserum to normal 

 mouse tissiies. In both instances, specific aggregation by antiviral sera could 

 also be demonstrated and Knight's conclusion was that perhaps 20-30 % of 

 the surface of the virus particle carried host antigen patterns. Influenza virus 

 in the filamentary form is agglutmable by immune serum, producing woolly 

 aggregates easily visible by dark ground microscopy (Burnet and Lind, 1957). 

 Belyavm (1955, 1956) has suggested the use of the agglutmation reaction in 

 routine serological work with influenza virus strains. The hemagglutinin in- 

 hibition reaction, however, is so convenient that it is not likely to be sup- 

 planted by another test givmg essentially the same information. 



Smith et al. (1956a) have shown that concentrated poho virus preparations 

 are flocculated specifically by immune rabbit sera. The reaction can be read 

 either macroscopically or by dark ground examination imder the microscope 

 (Smith et al., 1956b). It has been observed with human convalescent serum. 



B. Complement Fixation 



Despite extensive recent work on the nature of the four components of 

 complement, complement fixation still remains essentially an empirical means 

 of indicating that an antigen-antibody reaction has occurred. Orthodox in- 

 terpretation is that the forming aggregates of an antigen-antibody reaction 

 actively adsorb complement, rendering it unavailable to' hemolyze the 

 sensitized red ceUs added as indicator. The conditions for optimal fixation 

 of complement vary with the system and empirical adjustments will usually 

 be needed to obtain the most useful results. Reactions vnth both soluble and 

 particulate antigens will fix complement; in most of the earher work with 

 viruses no distinction was made between reactions due to virus particles and 

 those due to soluble products of virus multiphcation. 



For obvious reasons, comj)lement fixation has been much more extensively 

 applied to the study of animal viruses than with plant or bacterial viruses, 

 but some apphcations have been reported in both the latter fields. A subject 

 of major interest at the present time is the significance of soluble protein 

 with viral specificity that is produced in the course of infection. Such material 

 can often be detected by complement fixation reactions after particulate 

 virus is removed by centrifugation, adsorption to red cells, or other appro- 

 priate method. Extensive discussion of this theme m relation to influenza 

 virus will be found in Chapter Ill(b) of Volume 3. If the findings with in- 

 fluenza can be generahzed, it appears that soluble complement-fixmg antigen 

 is produced early in the course of infection, that the corresponding antibody 

 is stimulated only by infection, and that it is of no significance in conferring 

 protection against reinfection. Positive complement fixation tests following 

 infection tend to become negative much sooner than positive neutralization 



