112 THE BIOLOGY OF TWINS 



If we were able artificially to subdivide a fertilized 

 egg into two or more parts and the individuals develop- 

 ing from the parts of a given egg were always of the 

 same sex, we should consider the theory of zygotic sex- 

 determination as proven. Our nicest technique, how- 

 ever, has not been adequate to carry out so crucial 

 an experiment, and we are therefore forced to rely 

 upon an equivalent experiment in nature. It has been 

 shown conclusively for two species of armadUlo, and by 

 analogy for man, that an egg, divided at an early 

 period into two or more embryonic primordia, produces 

 individuals all of the same sex. In hundreds of sets 

 of quadruplets in the Texas armadillo there has occurred 

 no exception to this rule, in spite of the fact that in some 

 cases there are marked differences in size due to unequal 

 environment factors. In one case two fetuses of a set 

 are nearly twice the size of two others, yet the sex of all 

 is the same, showing that the zygotically determined 

 sex is incapable of alteration through any ordinary 

 environmental change. Similarly, in man twins that 

 are monochorial and in other ways bear evidences of 

 monozygotic origin are always of the same sex. Con- 

 joined twins, which are unquestionably monozygotic, 

 are also same-sexed, although one half of an individual 

 may be much better developed than the other. 



Not only in mammals but also in other animals 

 exhibiting polyembryony it is true that all individuals 

 derived from a single germ-cell are same-sexed; several 

 investigators have shown this for various genera of 

 parasitic hymenoptera. Silvestri was the first author 

 to discover polyembryony in these animals. He found 

 that in the genus Litomastix, in which the eggs are laid 



