IMMUNOLOGICAL METHODS IN THE STUDY OF VIRUSES 535 



There are at least two distinct antigens on the surface of phage T2 (Lanni 

 and Lanni, 1953). One, wliich is present on the "donuts" developed in pro- 

 flavin-treated bacteria as well as on normal phage, provokes complement- 

 fixing and aggregating antibody but not neutraHzing antibody. The other 

 produces neutralizing antibody, is not present in donuts, but is liberated as 

 soluble material on bacterial lysis (Burnet, 1933a), as well as being present 

 on the virus surface, presumably in the terminal filaments of the tail. 



If we confine ourselves to phages of the C16-T2 group, except where other 

 phages are specifically mentioned, and to active antiserum produced by an 

 adequate course of immunization, the main features of the reaction between 

 phage and neutralizing antibody can be summarized in a few paragraphs. 



1. The antigens concerned in neutralization reactions are relatively specific, 

 but there is sufficient range of cross reaction to make serological relationship 

 the most satisfactory basis for the classification of bacterial viruses (Burnet, 

 1933b; Adams, 1952). When two related phages, e.g., C16 and D29 (Burnet, 

 1933c), or T2 and T4 (Hershey, 1946) are studied by cross absorption of 

 immune serum, the usual relation holds. Heterologous antigen removes 

 neutralizing activity against itself but leaves homologous activity little 

 reduced. 



2. The reaction proceeds logarithmically, with some evidence of a slight 

 initial lag and some slowing in the later phases (Burnet et al., 1937; Andrewes 

 and Elford, 1933; Hershey et al., 1943). The activity of neutrahzation in any 

 given system is most conveniently expressed as the first order velocity 

 constant K (Adams, 1950). Very little evidence for the attainment of an 

 equilibrium state has been presented but, in suitable experiments, it has been 

 shown that addition of soluble antigen can reactivate a proportion of neutral- 

 ized phage (Burnet, 1933a). This finding suggests that long exposure with 

 dilute reagents might allow the demonstration of an equihbrium state. The 

 kinetics of the reaction uidicate that the rate-limiting process is the combina- 

 tion of one antibody molecule with one phage particle. Minor discrepancies, 

 such as the initial lag, are probably due to "inhibited" phage in wliich ad- 

 sorption of bacterial products hinders free attachment to the host (Cann 

 and Clark, 1954). 



3. Virus inactivated by immune serum can be reactivated by papain 

 digestion (Kalmanson and Bronfenbrenner, 1943) or sonic treatment 

 (Anderson and Doermann, 1952). This points clearly to the denatured anti- 

 body globulin acting by interference with the interaction of virus and host 

 and not by any directly destructive effect. 



4. Despite the indication from the kinetics of the standard neutralization 

 process that neutralization normally results from the miion of a single anti- 

 body molecule, there is adequate evidence for the existence of partly neutral- 

 ized phage. The apparent rate of neutralization in the same phage-serum 



