242 CELL HEREDITY 



The various patterns of nonchromosomal inheritance which have been 

 observed experimentally, such as those found in C hUimydo7nonas, yeast, 

 and Netirospora, provide no definitive information about the location or 

 nature of the determinants, or about their replication. The apparently 

 equal distribution of determinants to all progeny could be achieved by: 

 (a) random distribution of a large number of replicates; or (h) syn- 

 chronized replication and oriented transmission of one or a few repli- 

 cates. 



A nonchromosomal unit factor might exhibit unit factor inheritance, 

 mimicking Mendelian inheritance, but provide no evidence of linkage. 

 \n practice, the allelic pairs studied in Drosophila, maize, and Neuro- 

 spora have been mapable, but one may wonder whether there may be 

 nonchromosomal units which are indispensable or nonmutable, and 

 therefore undetected. Unit organelles like centrioles might be an 

 example. In Chlamydomonas, which contains just one chloroplast per 

 cell, a mutation which blocks chlorophyll synthesis has been found to 

 segregate only at the first division of meiosis. This result may be 

 interpreted either as close linkage to a centromere or as evidence of a 

 nonchromosomal location. No linkage with any other marker has yet 

 been observed. 



In the absence of direct identification of any permanent cell constitu- 

 ent as a nonchromosomal gene, a number of hypotheses have been de- 

 veloped to define these units in a testable manner. Before turning to a 

 more detailed consideration of the genetic evidence, it might be well to 

 consider briefly a few of the leading current proposals. 



1. The gene hypothesis states that nonchromosomal genes are molec- 

 ular entities with a permanence and stability comparable to that of 

 chromosomal genes. They are replicated with conservation of specificity 

 in a manner integrated with cell division, and their replication does not 

 depend upon their phenotypic expression. In other words, on this hypo- 

 thesis, nonchromosomal factors rank with chromosomal ones; they are 

 simply genes without chromosomes. 



2. The supertemplate hypothesis proposes that some structural ele- 

 ments, such as mitochondrial and chloroplast membranes with their con- 

 stituent macromolecules requiring precise orientation for activity, may 

 be formed by some copying process from a pre-existing surface of the 

 same kind. DNA replication seems to involve ordering in one dimen- 

 sion only, and the determination of protein specificity by arranging 

 amino acids in a proper linear sequence has also been considered as 

 a one-dimensional problem. The supertemplates postulated here might 

 be two-dimensional. 



