130 PHYSIOLOGICAL GENETICS 



differences, in which the underlying mutant genes have been 

 analyzed, viz., the case of Papilio polytes (De Meijere, 1910; 

 Fryer, 1913), P. dardanus (Poulton, Ford, 1936), Argynnis 

 paphia (Goldschmidt and Fischer, 1922), and Colias edusa 

 (Gerould, 1911). Here mutant genes and their recombinations 

 produce the phenotypes of the immensely different mimetic- 

 females, with normal Mendelian segregations, whereas the males 

 are phenotypically constant but genetically identical with the 

 females. In case of gynandromorphism (Goldschmidt and 

 Fischer, 1927), the female parts may show one or the other sex- 

 controlled type, whereas the male side is not influenced (see also 

 Cockayne, 1935). 



It is obvious that these facts also require for their explanation 

 a developmental system, as described before in this chapter, 

 a system in which the time relations of determinative processes 

 in the development of one or the other sex are fixed in such a way 

 that the action of the mutant genes reaches an effective thresh- 

 old, or the production of the necessary amount of something, 

 either in time for taking part in the following developmental 

 processes or too late for this. In more detailed form, this idea 

 is developed in Goldschmidt and Minami (1933) and Gold- 

 schmidt (1927c). 



2. Dosage through Sex Chromosomes. — Now we may turn 

 to the main subject of this chapter, the effect of different doses 

 of one and the same gene, as produced in the two sexes in the 

 case of sex-linked genes. As mentioned before, there is usually 

 no difference of effect, and this may be for a number of reasons 

 which are discussed below: 



1. Sex-linked genes may be genes of such a high potency of 

 action that the maximum effect that is physiologically possible 

 is produced in the simplex condition. Haldane has pointed out 

 that such a condition in the Wild type might have been brought 

 about by selection in the past. But if mutant genes are involved, 

 or even series of multiple allelomorphs, such an interpretation is 

 not very probable. 



2. Dissimilar effects of one and two genes in the respective 

 sexes may be prevented by a system of dosage compensation, 

 according to the idea of Muller (1932), who assumed that within 

 the X-chromosome a number of modifying genes exist, which 

 produce a compensating effect, which makes up for the lessened 



