ox TTTF, (JASTRULATTOX TX PETROMYZOX. 31 



junction line : n number of tlie rancromeric elements at this part 

 on either side is pressed out of the archenteric border, and a 

 new connection is established lu'twcen the micromeric epithelium 

 and the macromeric part. Such a particular process is 

 perhaps indispensable, because a broad surface of the macro- 

 meric hemisphere is converted into a narrow passage of the 

 blastopore. The cells thus pushed out assume a looser epithelial 

 arrangement which stands on both sides in connection with the 

 " dorsal 2^hite," as if tlic}^ were its lateral continuations (Figs. 

 27-30, /. mac. ep.). It is this structure which much attracted the 

 attention of observers and misled especially W. Scott^' to assume 

 it as the mesoderm split off from the main mass of the macro- 

 meric elements. For the covenience of reference, T will call it 

 the lateral irregular epithelium, although it is by no means to 

 be regarded as a definite structure. 



Anteriorly the " dorsal plate " is more or less broad and 

 passes over into an irregular epithelium of the archenteric wall 

 brought about by early invagination of the macromeric elements. 

 At its posterior extremity, the " dorsal plate " is divided at every 

 stage after its appearance into two arms which embrace the 

 blastopore from the dorsal side forming its lateral lips {Fig. 

 28). In the comjilete gastrula, the lateral lips represented by 

 the folding of micromeric epithelium are extended to the 

 ventral median part, the two arms meeting with each other at 

 the ventral blastoporic lip ; but the folding in is here still very 

 slight {Fig. 22). After all, the oval depression of the blastopore is 

 for the first time converted into a circular pore which con- 

 stitutes the definitive blastopore. 



1) W. Scott : Beiträge zur Entwickeliingsgeseliiclite der retromyzonten. Morpliol. 

 Jahrb., Bd. VII, 1882. 



