226 Cytoplasm as Seat of Genetic Properties 



enzymatic action can be located as some particulate constituents of 

 the cytoplasm which can be centrifuged and fractionated; in the latter 

 example we do not know the place in the cytoplasm where kynurenine 

 is formed, though in Ephcstia (Kiihn, 1948) we know the sites where 

 it takes place. In other words, we face a general relation between a 

 mutant locus and a biochemical function, which is found in all ap- 

 propriate studies. With due respect for the brilliant analysis, I con- 

 clude that serious misunderstandings and misuse of the facts as 

 support of the plasmagene hypothesis would have been avoided if 

 tlie experiments had been described in the following terminology. 

 Absence of a group of respiratory enzymes in yeast can be caused by 

 a recessive mutant locus (I have explained elsewhere why I consider 

 unjustified the extrapolation upon a dominant locus responsible for the 

 normal function). The same phenotypic effect as a phenocopy can be 

 produced by the action of euflavine, which removes the carriers of the 

 enzymes. So far this is the only case in which we know how a pheno- 

 copy is produced: by removal of a formed cytoplasmic particle from 

 the cell via an unequal division in regard to this particle, while in 

 the mutant the synthesis of the enzyme or a needed coenzyme or any 

 other irreplaceable stuflF for synthesis or for the function of the enzyme 

 is probably prevented. In addition, the phenocopy is reproduced by 

 vegetative propagation because of the special feature of loss of some- 

 thing. The fact reported by Ephrussi that recessive mutant loci exist 

 which influence the frequency of these abnormal divisions is inter- 

 esting, but it does not change the general feature. After all, the oc- 

 currence of the abnormal divisions, like any other organismic variation, 

 may be due to chance variation (i.e., environment) or to genetic 

 change. 



c. The killer effect 



One of the most brilliant pieces of genetic analysis has been per- 

 formed by Sonneborn and his school (see 1947, 1951a) for the killer 

 effect in Paramecium. This work is mainly responsible for the re- 

 awakening of interest in cytoplasmic heredity in the United States, 

 while in Europe it has been ever present since the early work of 

 Correns, Bateson, Goldschmidt, and von Wettstein, quoted previously. 

 It seemed that here a clear case of particulate cytoplasmic effect had 

 been established, until Sonneborn himself found decisive facts which 

 do not favor such an explanation. Sonneborn's own position will be 

 discussed in the conclusions, at the end of this chapter, on cytoplasmic 

 heredity. 



