Genie Control of Development 367 



proposed ( 1920 ) calculating such a coefficient by the formula Co = 



AA - AAi 



-7-7 7—7— X 100. (AA and AiAi are the quantities found for the 



AA — AjAi 



homozygote; AAi, for the heterozygote. ) Hersh had found that the rate 

 of change in mean facet number at any temperature is proportional to 

 the mean facet number at that temperature, expressed as an exponen- 

 tial function of the formula y = ae"^* in which r is the relative rate of 

 change; t, the temperature; a, a constant; and e, the base of natural 

 logarithms. The instantaneous change in facet number at a given 



dy 

 temperature is then -7- = yr. This value is calculated for different 



temperatures and combinations, and from these the coefficient of 

 dominance is derived. The results show that the degree of dominance 

 for the number of facets formed at a given moment changes regularly. 

 It decreases, for example, in B/+, with increase of temperature; it 

 increases for dominance of BB over wild type, and so on. Hersh 

 interprets this result as meaning that the more dominant "gene" has 

 relatively greater or less effect proportional to the changes of growth 

 by temperature action, and thus he attributes the results of his meas- 

 urements to the relative activities or potencies of the alleles. We shall 

 not go into further details as, for example, the strange discovery by 

 Luce that Bar and Infrabar act in the opposite direction in tempera- 

 ture experiments. The main point is that such exact dominance experi- 

 ments demand the type of explanation exemplified in our model 

 (fig. 18), while it is of secondary importance whether the shift in the 

 experiments is attributed to one or the other of the variables. The 

 decisive conclusion for genie action is then, as before, that a major 

 role is played by the control of the kinetics of developmental reactions 

 in their relation to others which are simultaneous and, perhaps, com- 

 peting. It is to be hoped that one day this may be expressed in terms 

 of known substances and exact formulations of the kinetics. I have 

 already emphasized that the action of dominance modifiers, which 

 have been studied so much, must be understood in the same terms. 

 A large number of such cases can be found in my book of 1938, and 

 many new ones do not give additional information. 



We have used here repeatedly the term "potency of alleles," in- 

 troduced in early Mendelian days by Davenport and Castle, given a 

 definite meaning in my work on intersexuality, and used later by 

 Wright and Hersh. The term has changed its meaning often. As 

 far as dominance is concerned, we saw that no necessity exists for as- 

 suming different potencies of the mutant locus itseff, though there are 



