SMALLWOOD : MATUEATION OF HAMINEA SOLITAKIA. 269 



alone constitutes the centrosome of the second maturation spindle. 

 Lillie (:0l), it will be recalled, found in Unio a similar mass of granules 

 at the beginning of the formation of the second maturation spindle ; and 

 one of these bodies became the centrosome, while the others formed a row 

 of microsomes that marked the outer limit of the medullary zone, Lillie's 

 " inner sphere," and finally passed into the cytoplasm as indistinguish- 

 able granules. Such is not the case in Haminea, — at least another ex- 

 planation is possible : a gradual reduction in size and a fusion of the five 

 bodies into one that becomes smaller than any of the five in the earliest 

 observed stage. The question as to what becomes of the disappearing 

 substance is partly explained by observing the changes that take place 

 in the stages noted. These stages form a connected series, as can be 

 shown by noting the sequence based on the known changes in the 

 chromatin, and on the formation of spindle fibres connected with the 

 chromosomes. The reaction to stains does not seem to furnish sufficient 

 evidence to decide in favor of either a physical or chemical explanation 

 of the change. While it is barely possible that these bodies may break 

 up and the parts become indistinguishable from the deeply staining 

 cytoplasm surrounding them, it seems more probable to me that they 

 change chemically, and then gradually become united into one bodys 



While the egg is still in the ovotestis, the centrosome begins to be 

 differentiated into two concenti'ic regions. These can be seen in Figure 

 16 (Plate 3), a very narrow zone enveloping the deeply staining central 

 part. This outer region takes such a deep plasma stain that it is difficult 

 to distinguish the two. A further description of these parts will be 

 taken up in the next section. 



Cortical layer. The centrosome is surrounded by a finely granular 

 mass, which shows neither yolk spheres nor the reticulate structure of the 

 cytoplasm which occupies a still more peripheral position (Plates 1-3, 

 Figs. 4-6 and 9-13). This mass takes a deeper stain than the surround- 

 ing cytoplasm, and has a decided affinity for plasma stains, especially in 

 the earlier stages. Van Beneden et Neyt ('87) named this the " couche 

 corticale." Conkliu (:01, p. 287) says, " This substance is not self- 

 propagating, tut arises anew in each cell generation, being composed of 

 nuclear sap and of dissolved oxychromatin from the nucleus, and of 

 hyaloplasm from the cell body." The cortical layer nearly envelops the 

 germinative vesicle, as can be seen in favorable preparations by focus- 

 ing. In some stages I have found a thin layer of it (Plate 2, Fig. 11) 

 extending entirely around the germinative vesicle. This is still more 

 evident in a cross-section of the equatorial region of the spindle (Plate 



