IV. THE CRYSTALLINITY OF VIRUSES 63 



be separated in the same rod-shaped particles. 



Insects are generally affected by a group of viruses and the disease 

 is called polyhedrosis. Viruses causing polyhedrosis are claimed to have 

 also rod-like shapes similar to some plant viruses. For instance, ac- 

 cording to Bergold (33). the virus particles of silk-worm polyhedrosis 

 are 288 mjn in length (tobacco-mosaic virus is believed to have a length 

 of 280 m/0. and 40 m/.i in width, and consist of a nucleoprotein having 

 a high P content ; tobacco-mosaic virus has a similar chemical composi- 

 tion as referred to later. 



Such rod-shaped particles composed of nucleoprotein are not only 

 found in silk-worm, but also in many other insects infected with poly- 

 hedrosis. For example, the virus particles of polyhedrosis of the gyp- 

 symoth caterpillar {Porihetria dispar, L.) are 41x360 m/z ; western tent 

 csLterpillar (Malacosoma pluviale, Dyar.), 40x350 m/j; California oakworm 

 {Phryganidia California, Pack.), 30x270 m/u; western yellow-stripped 

 armworm {Prodenia praefica, Grote), 50x270 va{-i; spruce bud worm 

 {Christoneiira fumiferana, Clem.), 28x260 va/j. (34). 



The term polyhedrosis arises from the peculiar shape of inclusion 

 bodies appearing in cells of infected insects and it is stated that such 

 polymorphic inclusion bodies, in which viruses are contained, are them- 

 selves doubly refracting. 



For what reasons do some viruses exist in such rod-shaped particles 

 of a similar length not only in plants but also in insects ? Is it true 

 that a certain general directive force as Stanley suggested is operating 

 in the course of virus production ? 



This fact, however, can readily be explained from the writer's 

 view previously described as regards the protoplasm structure and the 

 mechanism of virus reproduction. According to the writer's view rod- 

 shaped particles are no more than the elementary bodies from which 

 lipids have been eliminated. Lipid expulsion may occur because of the 

 structural change of protoplasm protein caused by a virus, as the pro- 

 tein becomes thereby incapable of retaining the lipids. The length of 

 the particles, therefore, must be the length of an expanded elementary 

 body itself whose length in turn must be the length of protoplasm 

 protein molecule in a stretched form. As will be shown later, this 

 length appears to have no relation to the presence of nucleic acids. 



If we assume that the molecular weight of protoplasm, like that 

 of the ordinary euglobulin, is 150,000 and that the distance between 

 the two amino acid residues is 0.34 mn, and that molecular weight 

 around a single residue is 130, the length of a stretched thread-like 

 molecule is calculated to be approximately 0.4/^. This should be the 

 length of an expanded elementary body and, accordingly, the length 

 of a rod-shaped virus particle. The value thus estimated, however, is 



