The Cellular Basis 167 



the development of these male or female characters, which occurs 

 later, is influenced also by the external or internal environment (p. 

 208). Breeders of cattle are familiar with the fact that when 

 twin calves are of opposite sex, the male is sexually perfect, but 

 the female usually has many male characters and grows into a 

 steer-like animal which is sterile and is known as a "free martin." 

 In a recent paper Lillie has shown that in all such cases the twins 

 are connected by blood vessels at an early stage in utero and that 

 there is a more or less complete circulation of blood from one 

 foetus to the other, and he concludes that "sex hormones," which 

 are probably formed earlier in the male than in the female, are 

 carried from the male to the female twin, thus causing the de- 

 velopment of male organs in an animal which would otherwise 

 have been a female. Therefore the chromosomal or "zygotic 

 determination of sex is not irreversible predestination but a quan- 

 titative overbalance in the direction of one sex or the other" 

 which may later be changed. Somewhat similar conclusions had 

 previously been reached by Whitman and by Riddle regarding the 

 sex of pigeons, by Shull in the case of Lychnis, and especially by 

 Goldschmidt for the gypsy-moth. Goldschmidt supposes that 

 sex is determined by certain enzymes which he calls "andrase" 

 and "gynase"; an excess of the former leads to the development 

 of males, an excess of the latter to females, and varying mixtures 

 of the two to varying intergrades or "intersexes." He assumes 

 that these enzymes are present in the "sex determining" chro- 

 mosomes at fertilization as well as in later stages and thus he 

 attempts to 'identify all sex determining factors with these "sex 

 enzymes." The difference between determination by chromo- 

 somes and by internal secretions, that is, between heredity and 

 development, is found chiefly in the time at which these enzymes 

 act. Attractive as this hypothesis is, it is nevertheless largely 

 speculative at pVesent and cannot be regarded as final. 



Morgan has shown that "gynandromorphs" or "sex mosaics" 

 are due to the irregular distribution or loss of certain "sex 

 chromosomes" owing to abnormalities in fertilization or cleavage. 



