4 F. M. BURNET AND W. M. STANLEY 



structure, since it allows the application of the method of X-ray crystal 

 analysis. This has had its most striking success in regard to tobacco 

 mosaic virus, for which Franklin's structure, based almost wholly on X-ray 

 diffraction, is now generally accepted (Franklin, 1955; Franklin et al., 1957), 

 It should perhaps be stressed that the accepted structure is simply a descrip- 

 tion of how two types of complex macromolecules are associated to produce 

 the hollow rods seen in electron micrographs. 



Although they have not been studied as intensively as tobacco mosaic 

 virus, some of the "spherical" plant viruses, such as tomato bushy stunt 

 virus and turnip yeUow virus, appear also to be simple geometrical arrange- 

 ments of protem and nucleic acid macromolecules. Crick and Watson (1956) 

 have discussed the theoretical aspects of symmetry in such structures and 

 Caspar's (1956) X-ray diffraction studies have indicated twofold and three- 

 fold symmetry axes in bushy stunt virus. Kaesberg (1956), from an examina- 

 tion of the shadows of heavily shadowed electron microscopic preparations, 

 believes that these viruses are probably very close approaches to the regular 

 icosahedron, a figure bounded by 20 equilateral triangles and with 12 corners. 

 Caspar has suggested that there may be 60 subunits, a group of 5 forming 

 each of the corners. 



The physical study of purified virus has always and necessarily been 

 associated with a chemical approach at whatever level was available at the 

 time. More than in any other aspect of virus work the range of viruses that 

 have been chemically studied is very small and there may well be surprises 

 in store when this is extended. The adenoviruses, with their curious type of 

 crystal structure within the host cell, and the larger less stable plant viruses 

 are two groups in which unexpected findings may emerge. However, it is 

 probable that what has been determined for the few prototype viruses, 

 tobacco mosaic and bushy stunt, phage T2, vaccinia, influenza A, and polio, 

 is broadly applicable in each case to a wide range of viruses of much the same 

 general quality. Perhaps the most important generalization that has emerged 

 is that all viruses contain nucleic acid and protein — the smaller animal and 

 plant viruses probably containing only these components. The remarkable 

 and very important work of Fraenkel-Conrat (1956) and of Gierer and 

 Schramm (1956) on infectious nucleic acid from tobacco mosaic virus nucleo- 

 protein indicates that the nucleic acids are the essential elements in the sense 

 of being the carriers of the genetic codes by which new virus is reconstructed. 

 There seems now to be no alternative to an acceptance of this view for 

 tobacco mosaic virus and there are only minor reservations in regard to the 

 bacteriophage T2. Preliminary work on the smaller annual viruses, while not 

 yet advanced to the point of certainty, seems to be pointing in the same 

 direction. The new concept of infectious nucleic acid seems destined to 

 become of the greatest importance to virology. Influenza and related viruses 



