8 S. G, WILDMAN 



are equally infective but it is a rare event for a virus particle to be present 

 at the exact location and the precise time when a susceptible site is produced. 

 However, Rappaport and Siegel (1955) have obtained evidence to suggest 

 that difficulty in creating susceptible sites is not the whole answer. Their 

 findings indicate that a fraction of the particles identified as TMV in the 

 extracellular state are simply not infective. The evidence comes from in- 

 activation of the infectious activity of TMV by antisera. When the ratio of 

 antiserum to virus is held constant, but the concentration of virus plus anti- 

 serum is varied over a ten-fold range, they find that the percentage of virus 

 activity that survives the antibody treatment remains relatively constant. 

 However, centrifugation experiments applying gra\ntational forces insuffi- 

 cient to sediment virus infectivity in the absence of antiserum revealed that 

 the greater the total concentration of virus and antiserum, the more infect- 

 ivity was sedimentable. Varying the total concentration of virus and anti- 

 serum led to the formation of different sized virus-antibody aggregates. They 

 obtained evidence that the larger aggregates consisted of hundreds of virus 

 particles, yet the fraction of infectious units m these mixtures was the same 

 as in those mixtures containing the smaller aggregates. They believe that 

 these results indicate that the number of infectious entities in a population 

 of TMV is less than the total number of rod-shaped particles. Although some 

 investigators do not concur with this interpretation of these particular 

 results, a similar situation of low specific infectivity is known to exist in the 

 cases of pohomyeHtis and influenza viruses, due presumably, in part, to 

 unavoidable thermal inactivation. It would be of interest to determine if 

 similar results would be obtained with animal viruses as with TMV in 

 controlled aggregation experiments. 



D. What Cells Serve as EiUry Ports for Virus? 



As mentioned previously, the surface of an N. glutinosa leaf is composed of 

 a heterogeneous collection of cell types. The question of what cells serve as 

 the starting point of virus infection is still largely unresolved. Arguments 

 have been presented which would appear to eliminate the leaf hairs as entry 

 ports (Boyle and McKumey, 1938). Yet, as Sheffield (1936) showed more 

 than twenty years ago, delicate puncturing of individual hair cells in the 

 presence of virus would sometimes cause lesions to develop in the region of 

 the injured hair cell. Unfortunately, the efficiency of starting infections by 

 this means is very low. In our repetition of her experiments, we have seldom 

 been able to get more than 5 % takes by this method, and I believe this to 

 be the experience of others who have tried the same method. However, the 

 conclusion that it is sometimes possible to start the infection in leaf hairs 

 can scarcely be doubted. Whether or not the leaf hairs are a significant entry 



