14 RALPH E. WAGER. 



egg. On the other hand, the process of adding cells in which an 

 enlargement and regressive change has taken place has no simi- 

 larity to amoebic activities. It appears, then, that the egg grows 

 not only by processes comparable to amoebic activities, but also 

 by those which have no possible similarity to them. 



In all cases examined, the egg, just before breaking through 

 the ectoderm, had appropriated all of the nourishing cells between 

 it and the layer of large ectoderm cells (Fig. 14, PI. III.). At 

 either side of the egg mass a considerable amount of unused 

 material may still remain after the egg has broken through the 

 ectoderm but it is probable that it is used by another egg or, if 

 conditions do not favor such a process, may be resorbed. 



In many of the preparations the nuclei of the nourishing cells 

 were possessed of several nucleolar bodies in addition to the 

 nucleolus. These were sometimes darkly staining, but more 

 often of a yellowish color. Often some were closely applied to 

 the nucleolus and others to the nuclear membrane. Extremely 

 irregular masses are often thus formed. Such nuclei are shown 

 in Fig. 20, PI. III. More will be said about them below. 



When brought into the egg, and, as above noted, occasionally 

 outside of it, a change is noticeable in the appearance and staining 

 of the cells. In the case of whole cells these changes are as 

 follows : The nucleus approaches the periphery of the cell ; the 

 nucleoli become peripherally arranged within the nuclear mem- 

 brane and often appear to be pushing outward as the membrane 

 is frequently extremely irregular and sometimes broken though 

 this phenomenon is not always to be observed (Fig. 21, a, PI. 

 III.). The chromatin filaments are visible for a time but are 

 much less distinct than in the normal nucleus. Whether or not 

 the chromatin collects in the form of small spherules which re- 

 semble the nucleoli, I have been unable to determine. In some 

 cases it seems to do so. Allen ('00) describes such a process 

 for Tubiilaida crocea. 



In the next stage the nucleolar bodies appear to coalesce in 

 part, as their identity can no longer be recognized. The result- 

 ing mass is irregular in outline and very darkly staining. At the 

 same time the region about it begins to react slightly to the stain 

 (Fig. 21, b, PI. III.). The chromatin filaments are now entirely 



