DEVELOPMENT OP FLUSTRELLA HISPIDA. 455 



The Formation of Encloderm. — PI. 24, fig. hi, illus- 

 trates a somewhat later stage than that just described. The 

 small aboral cells and the ring of larger cells underlying 

 these have divided, and the latter are now seen to enclose 

 about two thirds of the four large central oral cells within 

 the segmentation ca.vity. These four large oral cells have 

 also again divided, but this time the plane of cleavage 

 has been a hoi'izontal one ; and the four larger upper seg- 

 ments resulting from this division lie within the segmentation 

 cavity, and represent the primitive endoderm. The four lower 

 segments retain their original oral position. 



The Ectoderm. — From this time onwards, cell division 

 becomes less regular, and for a time at least, it takes place 

 more rapidly in the transverse than in the longitudinal direc- 

 tion, so that the larva tends to become spherical in form. The 

 small aboral cells divide repeatedly, forming the aboral ecto- 

 derm ; while the ring of larger intermediate cells, w^hich were 

 shown to have been initially derived from the large oral cells, 

 A I, B I, C \, D I, in like manner give rise to the oral ecto- 

 derm. A study of the living embryo and of sections (PI. 24, 

 figs. 53-60) shows that the cells of the aboral ectoderm tend to 

 remain, throughout embryonic development,smaller than those 

 of the oral ectoderm, the difference becoming more maiked 

 as development proceeds, but it is not possible at this, or at 

 any later stage, to distinguish any definite ring of cells which 

 can be correlated with the ciliated ring of the mature larva. 

 The true origin of this structure will be dealt with later, but 

 it may not be out of place to say here a few words in order to 

 explain how it has come about that the existence of such a 

 ring of cells has been supposed by Barrels and other authors 

 to occur at this stage. It is true that observations made on 

 the entire ^g^ at this period, especially when it is viewed 

 from the aboral surface, give somewhat the appearance of 

 there being an equatorial ring of cells, but this appearance 

 is a deceptive one. As stated above, an equatorial ring of 

 twelve cells actually did exist at the thirty-two-cell stage, 

 but the cells of this ring, as has already been described, have 



