Genie Control of Development 335 



I am completely in agreement with the trend of the ideas of 

 this "relation-minded morphologist," but we need also the "pictures" 

 of the substance-minded one, because, after all, the relational or 

 kinetic processes lead to real substances at a definite place, in a 

 definite structure. We do not know what, for example, facet-forming 

 substances are. But if injection of a definite substance into a Dro- 

 sophila larva may control facet formation, even through a kinetic proc- 

 ess, the substance-minded morphologist and the relation-minded one 

 become the same person. 



a. Dosage of genie material 



The most obvious method for such quantitative studies, destined 

 to throw light on genie action, is to compare the actions of the same 

 genie material in different doses. If the normal gene is assumed to be 

 a lump of an active substance, we might conclude from the mass law 

 of chemical kinetics that the product of its action is proportional to its 

 dose (Goldschmidt, 1917fl, 1920a, 1927). If a mutant gene is a different 

 quantity of the same substance, it should produce more of the end 

 product if it has a larger quantity, and less if the mutant is a smaller 

 quantity, than the normal locus. (The consequences will be taken up 

 below. ) If, however, the mutant is only a position effect and its normal 

 allele the normal chromosomal structure, it cannot be predicted what 

 the change of action of the mutant will be in regard to the end prod- 

 uct. It might be the same quantitative relation as in the theory of 

 quantitative mutation of a gene in the classic sense; but the effect 

 might also be a purely qualitative change, which in dosage experi- 

 ments would hardly lead to results of a simple quantitative type. 



aa. Dosage in sex determination 



The idea of dosage was to a certain extent contained in the 

 presence-absence theory of Bateson, which may be formulated in 

 terms of gene quantities. However, dosage first seriously entered 

 genetical theory when the mechanism of the sex chromosomes was 

 discovered: the visual demonstration that two doses of genie mate- 

 rial, actually two whole chromosomes, determined one sex; and one 

 dose, one chromosome, determined the other sex. But all the attempts 

 to connect this dosage relation with genie action led to impossible 

 consequences until it was found (Goldschmidt, 1912, 1915, 1916c, 

 1920c) that the dosage was not acting by means of its own direct 

 effect, but by setting up a quantitative or dosage method for deciding 

 a developmental alternative. This was the balance theory of sex de- 



