IMMUNOLOGICAL METHODS IN THE STUDY OF VIRUSES 527 



involved to account, for instance, for the relatively normal course of viral 

 infections in children with agammaglobulinemia (Good and Varco, 1955). In 

 all probability the immune process is a complex one and involves more than 

 a simple production of circulating antibody in the form of a modified y-globu- 

 lin. There is no doubt, however, that classic circulating antibody does play 

 an important part; in all the apphcations of serology to be discussed, serum 

 from animals that have either been infected with a virus or appropriately 

 inocidated with a suitable preparation of virus particles is used. In most 

 instances, the rabbit is the animal used, both because of its availability and 

 convenience and because of the specific and reactive type of antibody it 

 produces. With some animal viruses much more satisfactory sera are obtained 

 if the animal is susceptible to infection by the virus than if a nonsusceptible 

 animal is used. This holds particularly if antibody reacting with soluble com- 

 plement fixing antigen is required. On the other hand, Gard et al. (1956) 

 have considered that for such a standard task as assessing the antigenic 

 potency of a killed virus polio vaccine, there are positive advantages in 

 using a nonsusceptible animal, in their case the guinea pig. The absence of 

 antigenic damage during inactivation can then be verified by showing that 

 the antibody response is equivalent to that given by fully viable virus. 



In the present context the production of antibody can be regarded as a 

 simple response to the entry of foreign antigen into the Ijrmph and blood 

 stream of the rabbit, or whatever other vertebrate is being used. We are con- 

 cerned only with the reaction of virus particles or virus products with already 

 formed classic antibody. No attempt will therefore be made to discuss either 

 the general problem of antibody production or the controversial question of 

 the persistence of virus in the immune host. 



As is found with most other types of antibody, that produced in mammals, 

 either following infection with live virus or by immunization with inactivated 

 material, is present in the y-globulin fraction (Koprowski et al., 1947). 



For our present purpose, antibody may be characterized as specifically 

 patterned -globulin molecules which, in virtue of that pattern, unite pre- 

 ferentially with antigenic determinants forming part of the chemical structure 

 of the virus used for immunization. The various sections of this chapter will 

 be concerned essentially with the nature of the union between antibody and 

 virus, and with the physical and functional results of that union. 



II. The Concepts of Immunological Specificity 



The development of concepts of immunological specificity, due largely to 



Landsteiner (1946), has been based almost wholly on the use of precipitin 



or complement fixation reactions, both of which are essentially aggregation 



reactions in which a complex of antigen and antibody units either builds 



vol. I — 35 



