Genetic and Non-Genetic Variation 281 



for such problems; and microbial genetics and especially bac- 

 terial genetics is certainly a border-line problem. Further, 

 cybernetics, which deals with transmission of information, 

 seems especially relevant to genetical problems. We propose 

 to define genetical information theory as the problem of the 

 mechanism of transfer of genetical information from parents 

 to offspring. In higher plants and animals with normal sexual 

 reproduction, the genetical information, or rather the infor- 

 mation specificity (or perhaps even better the information 

 potentiality) is carried in particulate genes, located in the 

 chromosomes and passed on to the sexual offspring through 

 the meiotic cycle. However, the information which is con- 

 tained in the chromosomes of the gametes and zygotes must 

 be decoded, or translated to be expressed as the offspring's 

 phenotype. 



Such an approach to the genetical systems provides us with 

 a three-phase model: (1) the information; (2) a decoding 

 device which transmits the information to (3) a receptor. 

 I find such a three-phase model attractive, because it supple- 

 ments the classical two-phase model embodied in the concepts 

 genotype-phenotype, which seem inadequate for the analysis 

 of physiological and developmental genetics. I certainly do 

 not claim any priority with this model, and some courageous 

 colleagues have already proposed identifying the information 

 system with the DNA, the decoding device with the RNA, 

 and the receptor system with the proteins (for instance 

 Gamow, 1955), and they have discussed the problem of how to 

 translate the four-digit language of the DNA molecule into the 

 twenty-digit language of the proteins. However, such ad- 

 venturous spirits have been properly dealt with (Delbriick 

 and Stent, 1956), and we shall here leave the chemical aspect 

 aside, although there will probably be general agreement that 

 we must identify the receptor system with the proteins; that 

 the information system is in the nucleus and that the trans- 

 lation system is to be found in the cytoplasm. 



If we accept the above model, any phenotypic event, includ- 

 ing the appearance of drug-resistant colonies of bacteria, may 



