BRIDGES 



121 



tiveness of both contending sets of 

 influences has been doubled, trebled 

 or quadrupled; and the ratio remains 

 constant. 



But a system of formulation differ- 

 ent from the ratio type has been 

 adopted by Goldschmidt in dealing 

 with the intersexes produced in the 

 course of his brilliant work on racial 

 crosses of Lymmitria dispar. To the 

 male tendency of a particular race he 

 assigns a positive value that is propor- 

 tional to the strength of the male-de- 

 termining gene or genes. To the fe- 

 male tendency he assigns another value 

 also proportionate to the strength of 

 the female-determining gene or genes. 

 He then assumes that when in an in- 

 dividual the male value is greater than 

 the female value the individual is a 

 male, and that, conversely, when the 

 female value is greater than the male 

 by this same number of units the in- 

 dividual is a female. The locus of the 

 male tendency gene {M) is in the 

 "Z-chromosome" of which two are 

 present in the male and one in the 

 female. The female tendency is strictly 

 maternally inherited; and hence the 

 locus of the F genes is in the W-chro- 

 mosome that descends from mother to 

 daughter. The F gene is supposed to 

 exert its influence on the cytoplasm of 

 the developing &2,^\ and hence, al- 

 though the male has no W-chromo- 

 some, he is supposed to have a definite 

 female tendency that was impressed 

 upon the cytoplasm of the tgg and 

 that persists throughout development. 

 For a "weak" race the value assigned 

 to F is 80, and to M, 60. In the^VZ 

 individual the cytoplasmic | F | of 80 



exceeds the M value of 60 by "the 

 epistatic minimum" of 20 units, and 

 the individual is a female. Likewise, 

 in the ZZ individual the | F | is 80, but 



the net M value is twice 60 or 120, 

 with an excess in the male direction of 



40 units. For a strong race both F and 

 M are higher, for example, 100 and 

 80, but the arithmetical relation be- 

 tu^een the values of F and A4 would 

 still govern the sex of the individual. 

 In a cross between a weak female and 

 a strong male the ZW individual re- 

 ceived an I F I of 80 from the mother 



and an M of 80 from the father. The 

 values are thus balanced midway be- 

 tw^een the excesses necessary for a fe- 

 male on the one hand or a male on the 

 other, and the result is an intersex. 

 This far the formulation is satisfac- 

 tory; but when extensive series of 

 crosses are compared, and an attempt 

 is made to give values to the F and the 

 M of each race that will hold through- 

 out the entire range of experiments, this 

 attempt is rather unsuccessful. It seems 

 to me that a reformulation on the ratio 

 instead of the algebraic basis would 

 give a series of consistent indices with- 

 out running counter to the very valu- 

 able physiological ideas that Gold- 

 schmidt has developed. 



In the table of sex-types of Dro- 

 sophila the haploid individual is en- 

 tered with the index 1.25, the same 

 as that of the 2N, 3N and 4N females. 

 Unfortunately the haploid individual 

 has not been discovered. But it is plain 

 from the view just given that the ex- 

 pectation for a haploid Drosophila 

 melmiogaster is that it would be fe- 

 male in sex character. Accordingly, it 

 is necessary to assume that the sex- 

 determining mechanism here is essen- 

 tially different from that in the bee 

 and similar forms in which the haploid 

 individual is a male. To me, sex-deter- 

 mination in the bee is the outstanding 

 unsolved puzzle, although before the 

 development of the idea of genie bal- 

 ance it seemed one of the clearest and 

 simplest of cases. If it is true that the 

 male is a haploid individual, then one 

 would suppose that the diploid in- 



