132 Nature of the Genetic Material 



additional meaning that the biochemical action of the position effect 

 is identical with that of the mutant it imitates, and further that 

 variegation or lack of it is not an essential difference for position effect. 

 If we accept here for argument's sake the one gene — one biochemical 

 action hypothesis (to be discussed below), it shows that the po- 

 sition effect works the same way though the "gene" has not changed, 

 or (if the word "gene" is used, as it frequently is, to mean a mutant 

 locus ) though no "gene" or mutant locus is present. 



3. Mutants which affect general features like growth, fertility, 

 sterility, and viability frequently cannot be isolated and localized 

 like the standard mutants of elementary Mendelism. Many rearrange- 

 ments produce such generalized effects ( Muller and Altenburg, 1930 ) 

 in Drosophila. The same has been shown by Roberts (1942) for a 

 large number of maize translocations, which in a statistical study 

 were found to produce deviations in growth, considered as position 

 effects. Jones (1944) found chromosome rearrangements in maize con- 

 nected with growth defects. 



4. Mutants appear in all well-studied loci as more or less large 

 series of multiple alleles. Position effects may be found also as series 

 of multiple alleles, sometimes large numbers of them, as in the scute 

 and brown loci of Drosophila. In this case mutants and position effects 

 form together, indiscriminately, the series of multiple alleles; that is, 

 a compound may consist of a pair of mutants, or a pair of position 

 effects (where homozygous rearrangements are viable, as in the above- 

 mentioned yellow and white inversion), or one mutant and one 

 position effect. 



5. Many mutants — rarely localized but sometimes so, as in the 

 so-called vestigial dominigenes of Goldschmidt (1937) — are known 

 to have no visible effects of their own, but to act as enhancers or 

 suppressors of other mutant actions, which in a general way must 

 mean a quantitative effect in shifting developmental conditions upon 

 which the expression of a mutant depends. The same is true for 

 some modifier effects. We (Goldschmidt, Gardner, and Kodani, 1939; 

 Goldschmidt and Gardner, 1942; Gardner, 1942; Goldschmidt, 1952a) 

 found specific actions of inversions, (themselves without known, i.e., 

 visible effect) upon dominance and expressivity of other loci (Beaded, 

 Beadex, and plexus ) , that is, a specific modifier action. We noticed also 

 that this position effect modifying dominance or penetrance of a 

 heterozygous (Bd) or homozygous (Bx) dominant can be used for dis- 

 covering unexpected rearrangements, as was shown for an inversion 

 in a wild-type stock. Recently this effect was rediscovered by Lewis 



