THE MUTATED GENE 69 



thing that determines the position of the threshold for the suc- 

 cessful action of the M-genes within the X-chromosomes. It 

 may be added, also, that the other example mentioned in the last 

 chapter, Oehlker's Oenotheras, show the same threshold 

 phenomenon (see Fig. 22). At the threshold limits (26 for 

 normal versus low-grade sepaloid), a fluctuation of some indi- 

 viduals across the border line occurs. Furthermore, a number of 

 genotypes are found of different quantitative value, pheno- 

 typically alike because located beyond the threshold, and strictly 

 comparable to the hypermales and females in Lymantria and the 

 hyperwild in vestigial flies. 



3. Penetrance. — The facts reported on pages 65 to 68 brought 

 us again into contact with a problem that had been occasionally 

 mentioned before, the discussion of which finds its proper place 

 here. Timofeeff (1926) introduced the term penetrance to 

 describe the fact that the phenotypic effect of a gene does not 

 always take place in 100 per cent of the individuals homozygous 

 for the gene. In the vestigial case (see Fig. 14), the genes vg ni /vg ni 

 had zero penetrance; and the nicked-notched compound, 

 0.2 per cent penetrance. This term might be used for descriptive 

 purposes, provided one realizes that penetrance is a phenotypic 

 result, which may have many underlying reasons such as the 

 absence of genetic modifiers necessary for the production of the 

 effect, the existence of a threshold for visible effect which is con- 

 trolled genetically or by outside conditions, or a weak potency 

 of the gene in question, combined with a threshold that is not 

 reached. Of these possibilities the first one may be omitted 

 here, for if Mendelian segregation of modifiers, present in the line, 

 leads .to definite percentages of visible effects, it amounts to a 

 question of Mendelian segregation with genes of otherwise 

 invisible effect. If such modifying genes, however, control 

 threshold conditions, the problem is the same as all threshold 

 problems in regard to the details of explanation. But the numer- 

 ical values of penetrance are again the result of a Mendelian 

 segregation, this time of modifiers for the threshold of visible 

 action of the main gene. 



In the preceding section, we mentioned the cases in which 

 obviously a low percentage of penetrance was the result of the 

 presence of a threshold in such a position that the always present 

 fluctuation of the decisive process allowed only a small number 



