NONCHROMOSOMAL GENES 235 



rediscoverers of N4endelism. It was his view that more than one system 

 of heredity might well exist, and that both MendeHan and non- 

 Mendehan patterns of inheritance merited investigation. In fact, his 

 own researches continued to include inquiries into both systems through- 

 out his life. Subsequently, other leading geneticists including Baur, 

 Winge, von Wettstein, and Renner in Europe, and Rhoades in the United 

 States carried out extensive studies of non-Mendelian heredity in higher 

 plants and firmly established the widespread occurrence of this phenom- 

 enon. Rhoades' analyses of the non-Mendelian inheritance of pollen 

 sterility and of damaged chloroplasts in maize, to be taken up later, 

 are unambiguous demonstrations of the existence of stable permanent 

 genetic determinants which do not segregate as chromosomal elements. 



A different impetus for further investigation of nonchromosomal 

 heredity has come from cytologists, remarking on the regularities and 

 precision of the organization of cytoplasmic structures, and also upon 

 the observed multiplication of cytoplasmic organelles by dividing, rather 

 than by forming de novo out of undifferentiated cytoplasm. This mode 

 of origin of organelles from pre-existing organelles might indicate the 

 existence of some kind of replication. As early as 1908, Meves proposed 

 that mitochondria might carry their own hereditary determinants, but 

 no genetical system was available to test the hypothesis. 



Recent advances in our understanding of nonchromosomal heredity 

 have come principally from work with microorganisms. The purpose of 

 this chapter will be to describe the methods employed for analysis, with 

 a few examples, and to summarize the present level of knowledge in this 

 field. 



DEFINITIONS AND CRITERIA 



In Chapter 1, we defined a hereditary determinant as a unit of inheri- 

 tance essential for the appearance of a particular character. We are now 

 in a position to refine this concept. On the basis of much indirect evi- 

 dence, we have come to associate heredity with particular macromolecu- 

 lar configurations. As a general definition, it is proposed that heredity 

 consists of the conservation of specificity during replication and, in this 

 sense, that a hereditary determinant is a genetically functional unit 

 which is replicated with conservation of specificity. 



Detection of hereditary determinants, on this formulation, requires a 

 method to identify genetically functional units, and a method to study 

 their replication. With respect to chromosomal factors, the classical 

 method of identification has been recombination analysis, including the 



