VII. CAUSES OF THE EVOLUTION OF VIRUSES 175 



thus become possible. 



Newly formed structure or a virus may be better maintained in 

 larger particles, while readily lost in smaller ones as mentioned 

 already (Part I, Chapter II), so that for the preservation of the struc- 

 ture larger particles may be more favourable. In addition, assimilase 

 action is considered to be directly proportional to the particle size, 

 that is, larger particles possess greater faculty for their multiplica- 

 bility as also pointed out already (Part II, Chapter II). Larger sizes 

 are, therefore, advantageous to the viruses for these two reasons, /. e., 

 for the maintenance of the specific structure and for the enhancement 

 of the assimilase action. Accordingly, viruses should have to secure 

 the property to form larger particles in order to continue their ex- 

 istence. 



This property like many other properties is determined by a 

 structure of the protein ; that is to say, the combination of many ele- 

 mentary bodies to form a large body, may be achieved when the ele- 

 mentary bodies have a peculiar structure favouring the large body 

 formation. Hence, the virus which has incidentally secured such a 

 structure by a variation may be in a better position to continue its 

 existence, and the structure may be transmitted in succession. Thus 

 the structure to form larger particles would be attained and streng- 

 thened, with the result that assimilases which can be called organisms 

 might always get possession of the property to form bodies larger 

 than ordinary virus particles. 



2. Difficult Situation of Newly Generated Viruses to 

 Continue Their Existence 



Viruses seem to be constantly generated from a variety of cells, 

 so that strains or kinds of newly formed viruses must be innumerable 

 and thus numberless kinds of viruses must be constantly being thrown 

 out on the earth. However, the kinds of fixed, typical viruses never 

 seem to be numberless, but limited to a rather small number. This 

 may indicate that the probability of newly formed viruses of becoming 

 fixed is extraordinarily small. Most of the viruses generated de novo 

 possibly disappear immediately after their generation, because they 

 have no instinct of self-preservation. 



If in some protoplasm a new structure is produced which is suffi- 

 ciently stable to be retained in the protoplasm particles into which the 

 protoplasm is decomposed, then such a particle may be able to transmit 

 the structure to another organism as a virus. In order to evolve into 

 a typical organism, however, the virus should be successively trans- 



