WILCOX : SPERMATOGENESIS OF CALOPTENUS FEMUR-RUBRUM. 201 
while the spindle fibres are colored orange. The distinctness of the 
centrosomes is not exaggerated in Figure 25. No one could think them 
to be simply the points to which the spindle fibres converge. The pres- 
ervation of the histological condition is, moreover, in other respects very 
good ; there has been no local condensation of the protoplasmic structure 
sufficient to account for them. In some stages of spermatid metamor- 
phosis the centrosome is the most deeply stained and most distinctly con- 
toured body in the spermatid, not excepting the chromatic mass of the 
head. Moreover, Biitschli’s theory seems to me insufficient to explain 
all protoplasmic structures and all organs of the cell. 
Reinke (’94, p. 273) takes, as it seems to me, a much more reason- 
able position: “Ich sehe nun in meinen Praparaten alle drei Dinge: 
Korner, Faden, die zum Theil Netze bilden und schliesslich Waben oder 
Schiume.” It is very difficult for me to believe that spindle fibres are 
simply the lines along which the walls of the minute cells of a honey- 
comb structure meet. How, if this theory is valid, could we get cross 
sections of fibres such as are seen in Figure 42 (Plate 2)? Again, if the 
apparent spindle fibres are due to the much elongated form into which 
the honeycomb cells are compressed, I have difficulty in understanding 
how are to be explained the fibres bridging over the space between two 
cells, which, except for the fibres, are already completely separated 
(Plate 2, Fig. 31). If we grant that during division there is a mechani- 
cal force of sufficient intensity and definiteness to produce apparent fibres 
from the honeycomb structure of the protoplasm, the same force must 
remain in operation during the metamorphosis of the spermatids, in 
order to keep the remnant of the spindle fibres and their final product, 
the Nebenkern, in the condition of a distinctly modified portion of the 
protoplasm throughout this long period. Furthermore, how could we, 
on this assumption, account fer the intercellular ligament described by 
Lee (95), which persists through several generations ? 
These structures are, to my mind, something more than mere “con- 
figurations” of the protoplasmic honeycomb, as Eismond would have us 
believe. They rise to the dignity of cell organs. I do not wish to 
maintain that the centrosome, Nebenkern, and spindle fibres preserve 
their individuality indefinitely. They may not be directly concerned in 
the transmission of hereditary substance, but they do possess a special 
chemical nature, and they are of some morphological significance. We 
are not justified in considering these structures explained by the simple 
statement “dass wir es mit etwaigen endokinetischen Erscheinungen zu 
thun haben.” If the centrosome and attraction sphere have no morpho- 
