152 CELL HEREDITY 



even if as infrequent as 10~*, provided that mutations were so rare that 

 thev would not be confounded with recombinants. If in a cross no re- 

 combinants between two genes are found, it is still impossible to say that 

 no recombinants would be obser\ed if more offspring were examined. 

 It is easier to define nonallelism than allelism in terms of the transmis- 

 sion unit, for the distinction between two ver\ closely linked units two 

 pairs of alleles and one unit one pair of alleles' becomes a technical 

 one, limited by our abilitv' to anal^'ze large numbers of progen\ . 



In addition there is the unit of mutation. Now. a chromosome carry- 

 ing a mutant gene is in a sense a mutated unit and so is a mutant func- 

 tional unit- It was in this broad sense that the mutable unit was con- 

 sidered in Chapter 2. But a more specific definition refers to that smallest 

 part of the genot\pe which, when changed, is reproduced in the changed 

 form. The transmission unit and the functional unit depend for their 

 recognition on allelic differences initiated by mutation. Considera- 

 tions from the chemistry of DNA lead us to belie\e that the functional 

 unit and the smallest unit of mutation are not identical; there are other 

 arguments which lead to the same conclusion. 



^^ hen genes are not closely Unked. the classical assumption that the 

 three units are the same leads to no confusion; the\ mutate indepen- 

 dentlv. their effects on the organism are distinguishable, and their re- 

 combination is frequent enough to obser\e. On the other hand, when 

 two genes are closelv linked, diflBcuIties with the classical concepts arise. 

 First, there do occur simultaneous mutations of recombinable genes with 

 separate functions; furthermore, genes that mutate independently and 

 have distinguishable effects may not recombine: finally, genes that 

 mutate independently and recombine may have indistinguishable func- 

 tions. In the first case, the unit of mutation seems larger than the other 

 tvi'O; in the second case, the transmission unit, and in the third case. 

 the functional unit, seem to be largest. 



Simultaneous mutation of closely Unked genes does not necessarily 

 demonstate the existence of a mutable unit containing smaller functional 

 and transmission units, for it could be due to the simultaneous mutation 

 of two neighboring genes, brought about, for example, by a minute 

 deletion or chromosomal rearrangement. The independent mutation 

 of different functional units in a chromosomal region where crossing 

 over has not occurred does not necessarily mean that it cannot occur. 

 Such genes may be ver\ closely linked or some parts of the chromosome 

 mav cross over less frequently than others of equal physical size. But 

 when units that mutate independently and recombine appear to be 

 allelic, we may be confronted with the situation called position pseudo- 

 allelism. 



