378 



the amniotic tubes liave increased until they fill up the entire 

 chorionic cavity. The common amniotic vesicle then atrophies and 

 disappears. 



It is not necessary to carry the account of development beyond 

 this point, for no difficulty will be experienced in connecting the 

 stage just described with the primitive-streak stage, of which a full 

 account is given in the article cited on the first page of this paper. 



General conclusions. 



It is clearly beyond the scope of this preliminary report to 

 discuss the apparent bearing these results may have on several im- 

 portant biological questions. Furthermore, the relation of polyembryony 

 to the question of the fixation of sex, to paedogenesis, to cyclical 

 parthenogenesis, and to the ordinary forms of agamic reproduction, 

 has been more or less fully discussed by several writers, and will 

 receive further consideration in the final paper. Here the writer 

 wishes to stress two or three points that have not heretofore been 

 emphasized. 



First of all it can be stated with much assurance that the close 

 series of stages in the possession of the writer, covering — as it does 

 the critical period of development, furnishes incontrovertable proof that 

 the four embryos of a litter of this animal are the product of a 

 single egg. 



One of the most suggestive discoveries reported above is the 

 one concerning the origin of the embryos by a process of precocious 

 budding. The writer has been one to urge that in all probability 

 each embryo is the lineal descendant of one of the blastomeres of 

 the four-celled stage. This conclusion seemed justified in view of 

 several lines of evidence: (1) the results of many experiments on 

 the separation of blastomeres, followed by a study of their potency, 

 seemed to support this idea; (2) many facts concerning the hereditary 

 control in the four embryos of a litter of the armadillo could easily 

 be explained on the basis of the assumption that each embryo was 

 the lineal descendant of one of the blastomeres of the four- celled 

 stage; (3) and finally, in the case of true or identical twins in the 

 human species it has been persistently urged that each embryo is the 

 product of one of the blastomeres of the two-celled stage. 



The discovery of the fact that in the armadillo, which we now 

 know to exhibit specific polyembryony, the embryos are the result of 



