THE NATURE OF THE GENE 315 



only known substances with the necessary properties are the 

 intracellular proteinases, which both hyclrolyze and synthesize 

 the protein molecule. At this point Bergmann pronounces the 

 hypothesis that the proteinases are themselves proteins. "If 

 the proteinases themselves are proteins and at the same time 

 have the ability to synthesize other individual proteins then 

 there must exist proteinases which have the ability to synthesize 

 replicas of their own structural pattern and therefore are able to 

 multiply in suitable surroundings." 



It is astounding to see how these chemical facts and hypotheses 

 fit the requirements for a chemical theory of heredity, as postu- 

 lated in our previous discussion. Let us assume that the indi- 

 vidual chromosome actually is a single immense chain molecule 

 and a proteinase. (This means the essential part of the chromo- 

 some — the so-called gene string — to which is added nucleic acid, 

 which makes in some way the stability of the long chain possible. 

 All modern hypotheses regarding chromosome structure have 

 reckoned with these two constituents. 



This proteinase then has different effects according to its 

 special surroundings. Either it may produce its own replica, 

 which amounts to a division of the chromosome, or it may 

 synthesize other proteins from the parts present, or it may 

 hydrolyze proteins. The latter two activities would constitute 

 what is usually conceived as being the function of the gene. 

 The immense specificity of this proteinase is based upon its 

 typical structure. This, however, represents a most complicated 

 though regular pattern, composed of the superimposed rhythms 

 of the different amino acid residues. This again means that any 

 breakage or rearrangement in the chain leads to a destruction 

 or impairing of the specificity and therefore to other reaction 

 products, which have to be assumed to control the phenotypic 

 changes. Thus it seems that the latest developments of genetics 

 and of protein chemistry permit the statement of a reasonable 

 hypothesis regarding the chromosome, which could easily be 

 elaborated much further. 



Among the facts which will have to be explained on the basis 

 of this new model there is the question how the relations between 

 quantities of what is called a gene and a proportional effect can 

 be explained in terms of steric changes. There is no difficulty 

 in the cases where duplications or deficiencies are involved, as 



