VIRUS PARTICLES AND THEIR FUNCTIONAL ACTIVITY 349 



favor of the theory that one virus particle can initiate infection is the observa- 

 tion that each phage particle counted by electron microscopy can produce a 

 plaque (Luria et al., 1951; Section IV, B) although there is, at present, no 

 such direct evidence from the quantal-response type of animal virus assay 

 data. Lauffer and Price (1945), upon examination of a variety of virus 

 infectivity data drawn from the literature, concluded that they were, for the 

 most part, consistent with the "one-particle" theory and noted that the 

 alternative theory requires the improbable assumption that the variability 

 in the response of all virus-host systems is of the same order, since the slopes 

 of their titration curves are almost always similar. Discrepancies that do 

 occur (i.e., where the dose-response curve is flatter than the "one-particle" 

 curve) may be attributed to variability of host response as a secondary 

 factor. For example, Parker et al. (1941) found that titrations of vaccinia 

 virus of moderate virulence yielded somewhat flattened curves when carried 

 out in several rabbits, in contrast to the "one-particle" type dose-response 

 curve obtained with virulent strains. However, when this same strain of 

 moderate virulence was titered by the inoculation of all dilutions into 

 individual rabbits, the titration curve obtained from each rabbit was ade- 

 quately described by the Poisson equation, assuming that a single particle 

 initiates infection, although the individual 50 % end points varied consider- 

 ably. 



Armitage and Spicer (1956) also have shown from statistical analysis of 

 dilution counting experiments that departure of the quantal response curve 

 from a "one-particle" curve may occur, particularly at higher virus concen- 

 trations, in the form of fewer positive responses than expected as a consequence 

 of variation of host susceptibility. The over-aU results is a general flattening 

 of the curve. 



B. Observed Ratios between Virus Particle Count and Infectious Units 



The tentative conclusion reached by many on the basis of statistical 

 inference that a single virus particle can initiate infection has awaited 

 confirmation by direct observation. It must be admitted that, except for 

 bacterial viruses, confirmatory evidence for a 1 : 1 correspondence between 

 countable characteristic physical particles and infectious units is scant. The 

 difficulty may lie in finding the optimal conditions for measuring infectivity. 

 If the assumption that one particle is sufficient to initiate infection is correct, 

 the observed ratios of successful to total particles, which are usually less 

 than one, can be considered estimates of the probability of infection by a 

 virus particle under the conditions of the titration. Failure to observe 1 : 1 

 ratios may be due, in part, to the selection of host and/or route of inoculation 

 or may be inherent in the particle itself. The methods employed for particle 

 counting are incapable of distinguishing between fully infectious particles 



