The Cytoplasm as Specific Substrate 227 



The main facts are these. Certain hnes of Paramecium aurelia 

 were found (rather rarely) which secreted a substance called para- 

 mecin, later proved to be a nucleoprotein. This substance kills other 

 paramecia called sensitives. It is formed only in the presence of an 

 allele K ( homo- or heterozygous ) , but not in the presence of kk. Since, 

 in normal conjugation, only nuclear material is exchanged, the excon- 

 jugants are genetically alike but diflFerent in regard to cytoplasm. All 

 sensitives remained sensitives and killers remained killers, showing 

 the killing capacity to be a purely cytoplasmic property, though 

 working only in the presence of K. The cytoplasmic property is called 

 kappa. If, under proper conditions, cytoplasm is exchanged between 

 the conjugants, the former sensitive becomes a killer, and so do his 

 descendants. Sensitives can further be transformed into killers by 

 exposing them to broken up bodies of killers; and killers can lose this 

 capacity after treatment with X rays or nitrogen mustard. All this sug- 

 gests the presence of kappa particles in the killer cytoplasm. The first 

 proof of this conclusion was found when it was shown that kappa could 

 be diluted by quick growth of the well-nourished cytoplasm with which 

 kappa could not keep pace. Or kappa could be accumulated by the 

 opposite process. Such experiments permitted a very ingenious way of 

 calculating the number of kappa particles present in the cytoplasm 

 (Freer). Finally it turned out that these particles are visible, can be 

 stained, consist of nucleoprotein (DNA), and look and behave like 

 viruses of the Rickettsia type. Also mutants have been found which 

 change the killing eflFect in a number of visible details in the victims. 

 The number of kappa particles in the heterozygote is half of what it is 

 in the homozygote. 



If we look at these general facts, in what I should like to call the 

 naive way, it is diJBBcult to see why the kappa viruses should be called 

 cytoplasmic genes, or why cytoplasmic inheritance would be involved 

 at all. All facts point to viruses, which produce a poisonous product, 

 deadly for non-carriers of the virus. The viruses are self-propagating 

 within the cytoplasm, and their propagation can be influenced by 

 environmental conditions such as temperature and food, and also by 

 genetically controlled chemical properties of the cytoplasm. They may 

 mutate, may be transferred artificially, in short, may show any known 

 property of viruses, including the chemical composition. It seems to 

 me that this "naive" view is the one we should apply in the interest 

 of clarity. Such a view, which does not detract in any way from the 

 importance and brilliancy of the work, is the only safe one. It is some- 

 thing different if we say, with Ephrussi (53), that, whatever the 



