PLATE XXXV. 



Fig. 1. Ctenolabrus. [16-32-cell stage. Somewhat older than stages shown in Plate XXXII, 



(71) Fig. 2]. 9 drawings of a single cap (successive) X 280. [be = blastocoele; pb = periblast). 

 Corresponds to two hours 35 minutes after fecundation. 



The second section has not reached nuclei but has the clear areas. 



Periblast: — The periblast (subgerminal plate) is seen between the cells in the angles formed 

 by the adjacent cells. It is quite thick (about .004 mm.). It looks like a thick dense mem- 

 brane, is very sharp in outline, and stops abruptly in the angles of the cells. It is thus con- 

 tinuous with the lower surface of the cells. There is a strong contrast between it and the 

 protoplasm. The latter is gi-anular but the periblast is more homogeneous and membrane like. 

 It is quite as strongly stained with carmine as the protoplasm. Very small open triangular 

 spaces are seen above the periblast ic parts. 



The third section shows larger spaces in angles of cells and reveals more extensive position 

 of periblast. One pole of three nuclei is seen. The periblast is a little less thick. 



The fourth section shows the other poles of two [of these t hree] cells. The periblast terminates 

 in the angles of two peripheral cells, and is entirely free from the central ones, except perhaps the 

 right central, under which no distinct membrane can be seen but a light colored substance 

 extends beneath this cell connecting the periblast. The periblast would seem to form by growth 

 from the superficial yolk. This explains why it is often wanting near the center of floor, anil 

 often appears continuous with a superficial lighter colored layer of the yolk, as seen in section 

 eight. 



The eighth section cuts two marginal and two central cells. (The four central cells are blacker 

 than the marginal ones as was seen both before and after sectioning). Here the plane of division 

 is marked. In the central cells observe the inclination. Notice the nuclei; they are in edge of 

 light areas, and a poorly defined small area, about size of nucleus or a little larger, is seen in 

 each case on the outer pole of the nucleus. 



The periblast is much thinner, becomes less dense, lighter, and poorly defined under central 

 cells. Yolk [is indicated] below these cells to show that the forming periblast merges into it. 

 The ninth section = same cells. 



The tenth section = right cell, at middle, in passing to next cell. 

 The twelfth section corresponds to section eight. 

 The nineteenth section corresponds to section four. 



The periblast forms a continuous sheet in sections seventeen, eighteen, and nineteen, but is 

 interrupted in the twentieth. It is very thin centrally in section seventeen, plainer in eighteen, 

 plainest in nineteen. 

 Fig. 2. 32-cell [stage section; 6c = blastocoele). The central cells are browned more than the 



(S3) marginal, the marginal are redder than the central. 



The interesting fact is that the lower central cells are also more brown, ju.sf like the upper 

 central cells, thus appearing to have the same value. 

 F'g- 3. Transvcr.se section, near middle \ph = periblast|. On the right [lower end of figure) is 



(84) seen a nucleus in the periblastic swelling or ri)n. Perenyi swells the material a little and renders 

 it easier to trace the periblast. The periblast stretches across in a very thin evanescent layer. 

 Along the middle, it is not seen, except between two cells, where it appears to be continuous 

 with them. Cap two layers deep in central part, but these central cells are all browner than 

 the marginal in eggs treated with osmic and Merkel. 



