344 Action of the Genetic Material 



We return briefly to the former statement that the failure of the 

 exaggeration effect may be due to a position effect of the deficiency. 

 In discussing position effect previously, I pointed out that position 

 effect can be freely exchanged for genuine mutants in dosage experi- 

 ments, that the combination of two mutant loci plus one position 

 effect produced by a rearrangement near the same locus acts like three 

 doses of the mutant. In the same way a homozygous recessive position 

 effect is identical with the homozygous mutant effect. This shows at 

 once that we are not analyzing the effect of different quantities of 

 genie material, but that of additive action, of whatsoever change at a 

 mutant locus, upon the various kinds of reactions with which the 

 mutant locus interferes in one way or another. Thus the old discussion 

 of gene quantities is out of order here. It is even more so when we 

 accept the modern view that there is no normal gene and that all 

 mutants are position effects. Dosage then means, contrary to our old 

 views (1920a, 1927), not dosage of genie substance but dosage of 

 action. 



This conclusion must be kept in mind when we turn now to the 

 classic example of dosage, the increase of which enhances the mutant 

 effect away from normal, the case of Bar eyes in Drosophila. We 

 have already discussed the basic facts (see 1 S C c aa) and have seen 

 that the Bar effect is a position effect of a break involving a dupli- 

 cation; and that double Bar is a triplication in tandem, with two 

 position effects in one chromosome. Originally, when Bar was con- 

 sidered to be a dominant mutant locus reducing the eye-facet number, 

 the dosage series BB, BBB, BBBB could be considered as different 

 doses of a mutant gene. When Bar proved to be the position effect of 

 one tandem duplication, and double Bar that of a series of three 

 repeated chromosomal sections, it could be claimed that here the 

 dosage effect was actually studied with 1, 2, 3, 4 active sections of the 

 chromosome, that is, real doses of genie material. Then it turned out 

 that only position effects were involved, and the apparent real doses 

 of genie material became (as discussed above) additive effects. Again, 

 this dosage series does not shed light upon the nature of the genie 

 material (quantities of genie material as basis of the mutant effect) 

 but involves only the actions. In this case we have the type of action, 

 discussed above theoretically, in which increase of dosage (number) 

 of the causative agent increases the mutant effect, that is, lowers the 

 numbers of facets proportionally (though not in linear progression). 



In the bobbed dosage series, we had to conclude in a rather 

 general way that the amount of bristle-forming substance formed by 



