THE BIOCHEMISTRY OF PLANT VIRUSES 103 



is essentially much the same regardless of the age of the infection and of 

 the general cultural conditions. Matthews (1958) has made an intensive 

 investigation of this problem and has found a ratio of two nucleoprotein 

 particles to one protein particle (by number) over a wide range of conditions, 

 and he has suggested that the top component is the empty shell from which 

 the nucleic acid has emerged to give rise to two new virus particles (he 

 actually says three, but this is probably an error). On the other hand, 

 Jeener (1954) favors the idea that the top component is a precursor of the 

 virus nucleoprotein. His reasons for this are that the toj) component is 

 much more rapidly labeled when the host plants are exposed to C^^Og, than 

 are the ^^rus nucleoprotein particles. These two observations are very 

 difficult to reconcile, but a fuller and more systematic investigation of the 

 carbon labeling may throw further light on the matter. It may prove that 

 the top component is not a homogeneous population (in its origin at least), 

 and contains both virus particles in the making and the shells of virus 

 particles which have lost their nucleic acid. 



There is no reason to think that the top component is an artifact produced 

 during the isolation procedure, because it may be demonstrated in pre- 

 parations which have been isolated in a number of different ways. It is also 

 possible to take healthy plant sap, add the nucleoprotein to it in a pure form 

 and reisolate without any top component being formed in the process 

 (Markham, 1953a). 



The crystals were first examined by Bernal and CarHsle (1948) who found 

 that both top and bottom component crystals gave powder diagrams which 

 fitted a face-centered cubic lattice with a unit ceU of about 710 A side. 

 (The top component crystals have a rather larger unit ceU.) This was, of 

 course, much larger than could be filled with single virus particles, and a 

 number of reflections were found to be missing, which suggested that the 

 lattice was in fact a diamond lattice with 8 virus particles per cell, and an 

 interparticle distance of 30.6 m/x. This was in agreement with smaU crystalline 

 aggregates of the virus observed under the electron microscope by Cosslett 

 and Markham (1948). 



Some time later larger crystals became available and still photographs 

 of these were taken by Bernal and CarHsle (1951). The imit cell of the nucleo- 

 protein was foimd to have a 703-A side, and the protein a 715-A side; 

 Bernal and Carlisle suggested that the presence of the nucleic acid caused the 

 virus particles to be more compact by holding them together (this could be 

 due to attraction between the negative charges of the nucleic acid acting 

 on the interior of the virus sheU, which probably has a large number of 

 positive charges on it). 



The problem was investigated somewhat later by Klug and associates 

 (1957a;b) who actually used some of the same material as did Bernal and 



