GENE CONVERSION 26-26 



course of a few hours in the absence of cell division. Winge and 

 Roberts (1948) have studied a culture of S. Chevalieri which fer- 

 ments galactose by "long-term" adaptation which occurs after cells 

 have grown from 3 to 12 days, in a medium containing galactose. 

 After the adaptation the fermentation itself occurs at a rapid rate. 

 They state that "long-term" adaptation of S. Chevalieri to galactose 

 does not involve a shift in the population (mutation followed by se- 

 lection) but is an adaptation of the entire original population. They 

 consider the delay an indication of the time required by the cells to 

 build up galactozymase to a sufficiently high concentration to make 

 it demonstrable. They made hybrids between these two species and 

 found a regular segregation of rapidly fermenting (1 to 2 days) to 

 "long-term" adapting (3 to 12 days) spores from each ascus pro- 

 vided the ascospores were isolated in beer wort, a medium con- 

 taining considerable amounts of maltose, but lacking galactose. If 

 the ascospores were isolated in galactose broth, all four spores 

 from each ascus fermented galactose immediately (1-2 days). 

 However, they used only asci in which all 4 spores germinated and 

 the possibility exists that some nonutilizers of galactose may have 

 failed to grow. They described a series of experiments in which 

 they deadapted and readapted galactose fermenting cultures and 

 denied the significance of mutation as a factor in this process. 



I have pointed out that our so-called "nonfermenter" (g) clones 

 .(Chapter 11) are capable of limited fermentation of galactose under 

 conditions in which cell division does not occur, indicating that the 

 recessive allele is capable of producing small amounts of enzyme. 

 The fermentation by the g cultures differs from the delayed fermen- 

 tation by S. Chevalieri since the g clones ferment at a very low rate 

 after adaptation without cell division, while S. Chevalieri ferments 

 as rapidly as normal S. cerevisiae after a "long-term" adaptation 

 involving growth. We have discovered occasional cultures in our 

 pedigrees which resemble the cultures of S. Chevalieri described 

 by Winge and Roberts, by fermenting galactose after a long expo- 

 sure involving growth. Mr. Mundkur and I (Chapter 11) have 

 shown that this involves mutation from the g to the G allele follow- 

 ed by selection of the mutant G population in the presence of galac- 

 tose. This process is made especially effective since the g cells 

 are capable of using galactose slowly, thereby building up the popu- 

 lation to a point at which a low mutation rate of g to G could pro- 

 duce a sufficient number of rapid fermenters to take over. 



These facts have established the existence of three phenotypes 

 (1) rapid fermenters (2) so-called "nonfermenters", and (3) "long- 

 term" fermenters. These phenotypes will be designated G, g, and 

 gg. The designation of gg is used by Winge and Roberts for a gene 

 wnich they consider an allele of G. We shall use the same designa- 

 tion, but we interpret this phenotype as a clone composed original- 



