668 ALICE JOHNSON. 



time that the hypoblast has been perforated by the invagination, 

 the differentiation of the layers has extended as far back as the 

 blastopore. Therefore, when a neurenteric canal is formed, 

 this exists at the front end of the now reduced primitive 

 streak. 



In Birds, the invagination blastopore occurs comparatively 

 late in development, e. g. it is most fully developed in the 

 Duck with twenty-six mesoblastic somites and a medullary 

 canal closed except at the extreme hind end. A neurenteric 

 canal is found at a later stage. The primitive streak exists 

 only behind the invagination blastopore and its corresponding 

 structure, the neurenteric canal. At the latter, the hypoblast 

 is fused with the epiblast and mesoblast, but remains separate 

 from these two layers throughout the rest of the primitive 

 streak. 



In Mammals, the invagination blastopore begins as a pit in 

 the epiblast at the front end of the primitive streak.^ It then 

 extends downwards and perforates the blastoderm completely. 

 When the medullary groove is formed, it constitutes a neuren- 

 teric canal, piercing the floor of the hind end of the groove, 

 but, before the medullary folds close, its ventral opening into 

 the archenteron has become obliterated, and its upper part 

 alone remains. 



The view that the primitive streak represents part of the 

 original blastopore is now so generally accepted that it may be 

 assumed here for purposes of argument. 



No reason has been suggested for the various behaviour of 

 the blastopore of the Chordata in these different cases. It is 

 sometimes a simple opening which gives rise to a neurenteric 

 canal and then vanishes altogether. In other instances, it 

 is elongated and composed of a primitive streak with an 

 opening (which becomes a neurenteric canal) generally at its 

 front end, but in one form (the Lizard) in the middle, or the 

 opening may be (in the Newt) at the hind end of the primitive 

 streak and persist without having any connection with a neur- 



1 W. Heape, "On the Germinal Layers and Early Development of the 

 Mole," 'Proc. Roy. Soc.,' 1881. 



