98 W. M. Smallwood 
Unio (Liuuıe, '01), Crepidula (CoNKLın, '02), or Haminea (SMALL- 
woop, ’04). There is surrounding the centrosome a dense sphere 
substance which renders it difficult to be certain how far the astral 
rays extend but in some of the faintly stained eggs, the rays can 
be traced to this central body. The centrosome becomes early 
differentiated into centroplasm and centriole. Within the sphere 
substance and close to the centrosome, a clear area begins to appear 
through which the rays and fibres do not pass. This is the first 
indieation of the centroplasm; it is apparently formed from the solid 
centrosome and increases in size until it is several times as large 
as the early stage shown in Fig. 3. The centroplasm fully formed 
takes a very heavy plasma stain (Figs. 5, 6). When stained with 
iron haematoxylin and before the subsequent differentiation with the 
iron, the deep (egg) centrosome frequently looks like a large, irre- 
gular preeipitate of haematoxylin which may be so tenaciously 
retained that in removing it all of the other structures are comple- 
tely de-stained. This preeipitate occurs but rarely in connection 
with the outer centrosome. Before diseussing the further structural 
changes that oceur in the centrosome, we will consider briefly the 
astral rays, spindle fibres, and sphere substance. 
A few astral rays are present around the centrosome in Fig. 2; 
these increase rapidly in number and prominence as the germinative 
vesiele disappears, until in Figs. 4 and 5, the typical astral appearance 
is produced. These rays seem to be of the same character and 
nearly of uniform length, there bing no sharp differentiation into 
primary and secondary rays in Haminea. 
The growth of the spindle fibres is interestingly shown in 
Figs. 2 and 3. The rays which are to unite the two centrosomes 
and become the central spindle proceed directly to this union (Fig. 4). 
The mantle fibres few in number in Fig. 2, are the most conspicuous 
of any of the rays at this stage. They extend into the meshwork 
of linin fibres where the isolated chromosomes are forming. More 
of these fibres are present in Fig. 3; while in Fig. 4, they have 
become very numerous and the linin meshwork has entirely dis- 
appeared, the two events being intimately associated. The completely 
formed spindle having the chromosomes in the equatorial region is 
shown in Fig. 5. 
An undifferentiated sphere substance appears in each of these 
Nudibranchs especially in the prophase of maturation but it does 
not become clearly defined into two layers as in Haminea and 
