NONCHROMOSOMAL GENES 243 



3. The steady-state hypothesis proposes that a system of interlocking 

 enzymatic reactions can be set in motion by a particular environmental 

 influence, and have such intrinsic stability as to be perpetuated in- 

 definitely under suitable environmental conditions. A sudden shift in en- 

 vironment may induce a metabolic shift to a new steady state, mimicking 

 a mutational change. Since the degree of suceptibility to environmental 

 stress will vary with the particular system, no general prediction about 

 stability can be made. However, it is just this ill-defined aspect of the 

 hypothesis that weakens it by making it virtually incapable of disproof, ex- 

 cept by demonstrating what is the actual mechanism. Since an essential 

 feature of the steady-state system is its regulation by environmental 

 influences, it is likely that stable nonchromosomal determinants are not 

 of this type, in particular those maintained for long periods under diverse 

 environments in a latent state, with their phenotypes unexpressed (e.g., 

 streptomycin resistance in the absence of streptomycin ). 



4. The episome hypothesis of Jacob and Wollman has been proposed 

 only with respect to dispensable functions, those without which an 

 organism can still survive (see Chapter 5). The term was initially 

 applied to prophages and to the F factor in E. coli which can exist in 

 two alternative states: either integrated with the chromosome and not in- 

 fectious or unintegrated in the sense of being unmapable, infectious, and 

 not dividing synchronously with the cell. Episomes should be mapable 

 in the integrated state, and therefore readily distinguishable from non- 

 chromosomal genes. However, if one postulates that the factor always 

 becomes unintegrated in mitosis and meiosis, a most unlikely possibility, 

 then, of course, mapability is lost and only the unintegrated state can 

 be studied. 



No factors have been found as yet in other systems which fit the 

 criteria for episomes, but several have been described with some common 

 features. For example, C02-sensitivity and sex ratio in Drosophila, the 

 killer trait in Paramecium, barrage and senescence in Podospora, all de- 

 pend upon the presence of an infectious unintegrated nonchromosomal 

 factor. 



The method of choice for distinguishing among these alternatives 

 would be transformation or infection: that is, the establishment of the 

 new determinant in cells of strain A by introducing a defined cell consti- 

 tuent from strain B. In most systems, this experiment has not succeeded, 

 and only indirect criteria have been available for judging the nature of 

 the nonchromosomal genes. Because of the complexities of the problem 

 and the indirect methods of analysis, diverse systems have been grouped 

 as nonchromosomal. Probably they include examples consistent with 



