Genie Control of Development 341 



bb. Dosage within allelism 



Dosage effects may permit us to draw conclusions concerning the 

 action of the genie material if it can be shown that different actual 

 numbers of mutant loci exercise a proportional quantitative effect 

 upon the mutant phenotype. To make possible such a demonstration, 

 genie action must be of a type which permits quantitative shifts; that 

 is, it must be of a kinetic type. If we think of the one gene — one 

 synthetic step hypothesis of Beadle, it is obvious that genie actions of 

 this type cannot show dosage effects. If the mutant locus inhibits one 

 synthetic step and thus prevents a definite substance from being 

 present, with the consequent effect, whatever it is in individual cases, 

 then the presence of three or four of these mutant loci cannot make 

 any difference, for the effect is of the all-or-none type. But if the 

 mutant effect were to change the speed of a reaction or, perhaps, any 

 other imaginable quantitative feature of development and differentia- 

 tion, different doses of the mutant should, ceteris paribus, produce 

 proportional quantities of the effect up to a saturation point (thresh- 

 old for maximal action). The actual existence of a dosage effect for 

 many mutant loci is therefore additional evidence against the assump- 

 tion that the one gene — one synthetic step theory covers more than a 

 group of special cases. 



If we speak of genuine dosage experiments, we exclude dosages 

 assumed on the basis of our former view that mutants are different 

 quantities of genie substance, a concept which was later given up; 

 though we may still hold that a mutant is a change in regard to a 

 quantitative feature of genie action. This idea also permits our con- 

 sidering the relation between normal and mutant action, for example, 

 dominance as a dosage effect, not necessarily dosage of genie material 

 but also (or always?) dosage of action. We shall return to this later; 

 now we are dealing with actual doses of genie material at the mutant 

 loci. In all experiments of this kind an uncertainty factor is involved 

 because more or less large chromosomal sections are always used for 

 the production of different quantities of one locus. But it is a fair con- 

 clusion to attribute clear and orderly dosage effects to the dosages 

 and not to interference by other chromosomal parts. 



Dosage differences may be produced by means which are in no 

 way completely comparable. In the first type, available without special 

 experimental conditions, sex-linked mutant loci are present in one dose 

 in the heterozygous sex and in two doses in the homozygous sex. How- 

 ever, this is clearly not a simple dosage situation, since the one-dose 



