The Cytoplasm as Specific Substrate 223 



point is that the study of fractions of homogenates shows that the 

 mutant action is confined to a particulate structure of the cell. (See 

 discussion below.) These "mutants" are vegetative mutants because 

 they arise at a vegetative division (bud formation). The "mutation 

 rate" may be small or rather high. In the presence of some acridines, 

 like euflavine, it increases immensely, up to almost 100 per cent, 

 meaning that each bud is a "mutant." This was checked for individual 

 buds of single cells; an explanation by selection is thus excluded. 

 These changes remain irreversible in the vegetative cycle. The high 

 spontaneous "mutation rate," the occurrence of the "mutation" in 

 several yeast species, the occurrence at a cell division, and the fact 

 that the "mutation" can be induced in haploid and diploid cells lead 

 to the conclusion that the mutant character is due to the loss or 

 inactivation of an extranuclear, particulate, and self-reproducing fac- 

 tor required for the normal synthesis of the respiratory enzymes (see, 

 above, Slonimski's fractionation). As a result the "mutants" are de- 

 pendent upon accidental non-inclusion of the hypothetical particles in 

 a forming bud, and the mutagenic action of the acridines is due to an 

 elective toxic effect on these particles, which may then be excluded 

 from the bud. ( Ordinary genie mutations resembling these "vegetative 

 mutants" also have occurred.) 



It is rather interesting that the efiFect of euflavine in producing 

 the "mutated" buds lasts over a number of generations even if the cells 

 are returned to normal medium, which is not unexpected for a cyto- 

 plasmic effect. 



The check for possible gene mutation is rather difficult because 

 the enzyme-defective cells do not undergo meiosis. However, a normal 

 yeast and a vegetative "mutant" of opposite mating type can be 

 crossed with simultaneous introduction of markers. A normal 2:2 

 segregation follows for the markers, but no segregation for the respira- 

 tory effect, thus proving that the latter is of a cytoplasmic nature. 

 Ephrussi therefore speaks (with an unfortunate terminology) of a 

 cytoplasmic mutation based upon a cytoplasmic factor that is self- 

 reproducing and particulate. If these particles are few in number, the 

 possibihty exists that a bud does not receive any of them, and is 

 therefore a "loss mutant." (See the terminology below.) Euflavine 

 would simply favor such an unequal division. ' 



In a series of interesting experiments, Ephrussi further analyzed 

 the way in which the acridines act in inducing the "mutation," 

 whether an unequal distribution of particles is involved, or rather a 

 poisoning of some of them; finally, whether the experiments could be 



