SEROLOGY OF VIRUSES 73 
suspension of high concentration; such are the viruses of vaccinia and 
the Shope papilloma which can be obtained in considerable quantities 
in the natural lesions. 
The actual neutralization of the virus by its specific antibody is 
thought to be due to a reversible union between antibody molecules 
and certain determinant groups on the virus surface. When a certain 
amount of antibody has united, the virus particle becomes non- 
- infective—the amount necessary may differ according to the method 
by which the infectivity of the particle was tested. If the concentration 
of virus is high enough, and sufficient time is given, collisions of the 
partially coated particles result in the formation of aggregates (Burnet, 
Keogh and Lush, 1937). 
Chester (1936) prepared neutralized mixtures of tobacco mosaic 
virus juice and immune serum by titrating the serum with the juice 
until the supernatant fluid after centrifuging contained an excess of 
neither serum nor virus, as determined by precipitin tests. 
When such mixtures were subjected to a number of chemical, 
physical, and serological treatments, no free virus nor antibody was 
recovered, but when the mixtures were partially digested with pepsin, 
the antibodies were destroyed and a large portion of the virus was 
recovered, as determined by precipitin testing. 
Tests for virus-neutralizing antibodies to animal viruses have been 
employed for the following purposes: (1) to assess the potency of 
therapeutic antisera with a view.sto their standardization; (2) for 
detecting the presence of virus-neutralizing antibodies in human or 
animal sera for diagnostic or epidemiological purposes; (3) to identify 
Viruses or to investigate their antigenic structure. 
The general principle on which these tests are based is the same: 
for the majority of instances the serum to be tested is mixed with the 
virus in suitable proportions, incubated for a short time, and inoculated 
into a suitable animal or an egg. In some tests, the amount of virus 
used is kept constant and different dilutions of serum are added; in 
others, the amount of serum is kept constant and the virus diluted 
instead (van Rooyen and Rhodes, 1948). 
In considering the serology of plant viruses, we find an entirely 
different situation. Since, so far as we know, there is no antibody 
formation in plants, there is no acquired immunity to virus diseases 
in the accepted sense. The only type of acquired immunity in plants 
is of the non-sterile type. In other words, there is acquired immunity 
between related viruses only; thus, a plant infected with a tobacco 
