90 THE ORIGIN OF GYNANDROMORPHS. 



this law even to the minute parts of a single organ, like the copulatory 

 organ, we find it also to apply, as will be demonstrated later. Now, 

 this is the fact which, in connection with the others, enables us to 

 formulate a definite physiological theory of sex-determination." 1 

 Goldschmidt sums up the situation in the following statement: 



"First, we recognized that the different effects of the same sex-factors in 

 different combinations can be understood only by assuming a quantitatively 

 different action; or, expressed in concrete terms, that the active substances, 

 which we represent as factors, are present in different but typical quantities. 

 Second, we were obliged to assume that these substances are distinct for each 

 sex. Third, we realized that in the action of these substances a time factor is 

 involved, which is definitely proportional to the quantities of the factorial 

 substances. From these facts only one conclusion can at present be drawn: 

 that the sex-factors are enzymes (or bodies with the properties of enzymes) 

 which accelerate a reaction according to their concentration " 2 



If the nature of the character is dependent on the relative quantities 

 of the male-producing enzyme called andrase and of the female-pro- 

 ducing enzyme called gynase, the question arises how intersexes that 

 are mosaics would ever arise, for there is no obvious reason why the 

 relative concentration should ever change in the course of development 

 as Goldschmidt must assume that it does change. Still less is it clear, 

 when the difference in the concentration is less than a given critical 

 difference (Goldschmidt's definite minimum value e), why the enzyme 

 that starts with a lesser concentration should always overtake the other 

 quantity, no matter which one starts below. Until this critical point 

 is explained all the speculation that Goldschmidt brings to bear on the 

 question only seems to cover up the difficulty rather than to clear it up. 

 Goldschmidt appears to have overlooked this difficulty and sets up the 

 opposite one, viz, that it is difficult to see why every gipsy moth is not 

 an intersex. He meets this supposed difficulty by the consideration 

 of the rate of development of the insect. Whether his answer to this 

 difficulty is valid or not, it does not seem to meet the difficulty which 

 to us seems the real one. 



Even were it established that many of the changes in embryonic 

 and larval development are due to enzymes a point that we are far 

 from wishing to dispute it need not follow that the segregating genes 

 that give rise to them are also these same enzymes. To treat these 

 half-way stages as the genes themselves is at present not without 

 danger, because even if the genes are enzymes it by no means follows 

 that the quantity of the gene is to be measured by the product of the 

 enzyme arising from it. 



In his latest communication Goldschmidt states his belief that the 

 sex-factors in the different races of gipsy moths are multiple allelo- 

 morphs and compares them to the series of factors that Castle has 



1 Goldschmidt. A Further Contribution to the Theory of Sex. (Journ. Exp. Zool., vol. 22, 

 No. 3, April 1917, p. 597). 



2 Ibid, p. 598. 



