242 IV. THE PRINCIPLES OF LiFE PHENOMENA 



SO that each unit structure in a particle can behave occasionally, like 

 in genes, as each determinant site. For example, if a virus with 

 three unit structures, A, B, and C, affects a cell together v^ith another 

 virus having unit structures. A', B', and C, and if the degree of 

 strength of each pair of the structures is as follov^s : A>A', B<B', 

 and C<C', then nev^ly formed virus will be A, B' and C, whereas if 

 the former virus combines with a virus. A", B", and C", and the 

 degree of strength of the structures, being A>A", B>B" and C<C", 

 then the new virus will be A, B, and C" . In such a manner, each 

 unit structure of the virus can be revealed separately appearing as if 

 it were composed of many genes. 



It may, however, be unreasonable to assume that in a gene particle 

 only a single unit structure can always have the supremacy, directing 

 a property without being influenced by other structures. There is a 

 good reason to suppose, as will be mentioned later, that it is difficult 

 to change a restricted portion in a protein molecule without altering 

 other portions. Moreover, it is possible that the manner of arrange- 

 ment or of combination of different structures may sometimes deter- 

 mine a character. A normal organism is said to be the resultant of 

 action of some thousand of genes ; for instance, in Drosophila about 

 eight thousand of genes are believed to be in action (23). Presence of 

 such a great number of genes cannot be explained by the assumption 

 that a component character is always directed by only a single unit 

 structure. 



Antigens, like viruses, are capable of producing a sort of replica 

 of their own in the protoplasm of certain cells, though the replicas 

 produced by them are not so perfect as those produced by viruses, 

 and moreover though the cells in which the replicas are to be formed 

 are the cells which are generated for the purpose of antibody produc- 

 tion. Anyhow, there seems no doubt that antigens are a replica- 

 producer just as viruses or genes, and in this respect genes may be 

 regarded as the strongest antigens in the cell protoplasm. 



By the famous researches of Landsteiner and his coworkers (24) 

 it has become evident that the specificity of antigenic character of 

 proteins resides in the chemical structure of their molecules. The 

 average size of a specific group is said to be of the order of 600 to 

 1,000 in molecular weight, but occasionally very simple artificial che- 

 mical groupings may act as specific determinant (25). It seems, how- 

 ever, that the specificity of natural proteins is due not to a single 

 type of group but to a specific arrangement of different polar groups 

 in the surface of the protein molecule (26). The same may hold true 

 for genes. 



In general, the substances able to behave as antigens are proteins. 



