lo THE BIOLOGY OF TWINS 



This is approximately a ratio of i : ^ • ^i^ • 

 Now if all twin births in man are dizygotic and the sex 

 is predetermined at the time of fertilization, as there 

 is every reason to believe, the sex-ratios of twins should 

 be: J : 2 * I • There are, however, nearly twice 

 as many same-sexed twins as there should be on this 

 basis, and the only satisfactory explanation of this 

 discrepancy between the observed and the expected 

 ratios appears to be that nearly half of all same-sexed 

 twins are monozygotic and hence morphologically stand 

 for hut one individual to the pair. It would appear 

 then that in Nichols' data the difference between the 

 actual numbers of same-sexed twins and half the number 

 of opposite-sexed twins will give the probable number of 

 monozygotic twins of each sex; this would be 102,448 

 monozygotic male twins and 87,263 monozygotic female 

 twins. About one-fourth of all human twins then, if 

 this reasoning holds, are monozygotic. My own observa- 

 tion and that of biologists with whom I have discussed 

 this point agree very closely with this conclusion. 



The form of the human uterus and the intra-uterine 

 relations of twins serve as another line of evidence 

 favoring the existence of monozygotic human twins. 

 The human uterus is of the simple type, resembling 

 that of the armadillo Dasypus, and is not adapted for 

 ordinary multiple gestation. Such a uterus, however, 

 is apparently as favorable for polyembryony (the 

 production of plural embryos from one egg) as that of 

 our armadillos, in which this kind of reproduction 

 occurs normally. 



The evidence furnished by the data collected by 

 obstetricians of twins in utero also favors the existence 



