SEX DETERMINATION 225 



variation, such as would be explained by many randomly assorting 

 genes, rather than a clear division into two types. 



In Bonellia, there must be two alternative reaction systems which are 

 activated by the external conditions. In other organisms, true hermaph- 

 rodites, these two systems are present, but there is no differential 

 effect of external conditions, and no antagonism between the two types 

 of sexual differentiation, so that both are realized in the same animal, 

 sometimes even in the same organ (e.g. Mollusca). In some cases there 

 are separate male and female organs existing contemporaneously, in 

 others there is a definite time sequence, the animal functioning as a 

 male first in protandric types, as a female first in protogynic types, or 

 finally there may be a regular alternation between male and female 

 phases. 



We do not know much about the genetics of sex determination of 

 these complete hermaphrodites, but we understand fairly fully the 

 conditions in some animals intermediate between hermaphrodites and 

 animals with efficient sex determination. The best known case is in 

 frogs. ^ The female is homogametic MM FF and the male heterogametic 

 MMFf, but the difference between the X and the Y chromosomes 

 cannot be seen cytologically. In some races, the so-called undifferentiated 

 races, the F and / allelomorphs are nearly of the same strength and the 

 frogs show evidence of hermaphroditism, there being a period in early 

 hfe when even genetic males develop as females. It seems that all the 

 animals start development as females, and in males the development is 

 later switched over into the male direction at a time determined by the 

 relative strengths of the MM and Ff factors ; this depends chiefly on 

 the strength of the /, as the M and F factors are of nearly the same 

 strength in all races. In Drosophila we have seen that development 

 always starts in the male direction, in Lymantria in either the female 

 direction, giving female intersexes when switched over, or in the male 

 direction giving male intersexes when switched. The strengths of the/ 

 factors in frogs have been calculated by Witschi, on the hypothesis that 

 the sex determination is due to the difference between the male and 

 female factors rather than on their ratio; the data are hardly sufficient 

 to allow a final decision between the two hypotheses in this case. As in 

 Lymantria, it is found that the strength of the factor is vaguely corre- 

 lated with climatic conditions, weak / genes, with efficient sex deter- 

 mination, being absent in northern regions and at high altitudes. 



In the Amphibia, which lend themselves to embryological experi- 

 ^ Rev. Witschi 1934, cf. Witschi 1937, Burns 1938. 



