THE LIMITS OF IIEREDTTY 177 



from tertiary Imclding. Some other species of 

 armadillo produce only one embr^-o from the e^g. 

 In the Texas species {Tatusia novcnicincta) Patterson 

 has shown that two buds grow out, each of which 

 immediately divides into two. In the species in 

 which eight young are produced, these four buds 

 presumably divide again, while tertiary budding of 

 the embryo produces a higher number. 



Patterson (191 3) found a " period of quiescence " 

 in the development of the young blastocyst, which 

 Newsman believes to be connected with the cause of 

 the budding. In the gastrula stage the embryo 

 remains quiescent for several weeks. Then placenta- 

 tion takes place, and the development is resumed. 

 Stockard (1921) suggests that while the blastoc^^st is 

 lying free and unattached in the uterine cavity, the 

 consequent absence of an oxygen supply {via the 

 blood) inhibits development. This condition of an 

 unattached blastocyst is found in no other mammal 

 except the deer. It is not known wh}'- the delay in 

 implantation of the blastoc3'^st occurs, but from 

 studies of other mammals it is suggested that it may 

 be due to some peculiarit}^ in the formation of the 

 corpora lutea.* In other animals, for example 

 birds, the deer, and lower animals, arrest of develop- 

 ment does not lead to pol^^embryony. But Stockard 

 finds from experiments with certain fish eggs that, 

 if development is arrested early (before gastrulation), 

 twins and double embryos are frequently produced. 

 Hence the interruption must occur at a critical time. 

 Also there must be present in the Qgg, as in that of 

 the armadillo, a decided tendcmcy to form buds under 

 conditions of arrest. .Ihus it appears that the 

 interaction of certain external and internal forces is 



* A yellow mass of cells in the ovary in place of a discharpjcd 



ovum. If the ovum is fertilised the corpus luteum persists and 



grows for several months. The number of corpora lutea therefore 



corresponds with the number of developing embryos. 



1 2 



