CLEAVAGE IN TOXOPNEUSTES VARIEGATUS. 143 



be distinguished. With the cutting off of the micromeres there 

 is separated from the rest of the embryo that protoplasm which 

 later goes to form the mesenchyme. These cells, or cells derived 

 from them, must pass from the wall of the blastula into its 

 cavity; and as early as the sixteen-celled stage some of the first 

 four micromeres exhibit this tendency to migrate inward. Sec- 

 tions passing through the poles of a sixteen-celled stage show 

 the micromeres visibly crowded. Figs. 13 and 14 are sections 

 of two different individuals in this stage, and the micromeres at 

 the vegetal pole are projecting slightly into the blastoccele with 

 their inner ends bulging as a result of pressure from the adjacent 

 cells in the wall. The difference in the number of the micro- 

 meres in the two figures is due to the planes of the sections. 

 Fig. 13 is cut parallel to the short axis of the group of micromeres, 

 while Fig. 14 is cut parallel to the long axis of the group. 



A later stage is shown in Fig. 15 which is a somewhat oblique 

 section through the micromeres. In this case the micromeres 

 have multiplied and the process of pushing in has continued 

 until one cell of the particular section in question has lost all 

 connection with the surface. 



Figs. i6a, i6b and 160 represent three serial sections of one 

 specimen which is slightly older than the preceding one. Here 

 the migration is quite evident. From this time on the micro- 

 meres are not readily distinguishable by their small size as hereto- 

 fore, since by an unequal rate of division, all the cells have become 

 more nearly the same size. The fact that in earlier stages it was 

 the micromeres, and they alone, which passed inward, and the 

 fact that throughout the stages the migration is restricted to a 

 limited area in the wall, afford proof that in the later stages the 

 mass of cells which has passed inward represents the offspring 

 of the original micromeres. 



A more pronounced wandering in of cells is shown in Figs. 

 i"ja, i~?b and ijc serial sections of one embryo. Here, cer- 

 tainly, pressure from without has ceased to operate as the cause 

 of migration, for the wandering cells are free to a great degree 

 from contact with cells of the wall. 



There have been various theories of the cause of the separation 

 of the mesenchyme cells from the wall of the blastula in other 



