Igor | GAMETOGENESIS AND FERTILIZATION IN ALBUGO 239 
The latter is typically very dense and alveolar, does not stain 
with gentian-violet but takes the orange G. The former stains 
darkly with the gentian-violet, and is filar, not alveolar. The 
processes leading to zonation may indeed be characterized pro- 
visionally as a differentiating of the oogonium into two regions, 
a periplasm rich in kinoplasm, and an oosphere rich in tropho- 
plasm. This statement is borne out by all the positive charac- 
ters of the trophoplasm and by the stain reaction and structure 
of the kinoplasm. 
The nuclei are in mitosis and do not lose their membrane 
until zonation is sharp, nor do they reenter the ooplasm until 
their membrane is lost. The kinoplasmic nuclear membrane 
is apparently left in the periplasm, and its absence is evident 
during the second mitosis, thus resulting in that marked differ- 
ence in character between the first and the second mitosis, which 
is illustrated in the figures, a difference which was noted in A. 
Buti in my earlier paper (1899, p. 231). The nuclear membrane 
in the second mitosis is very thin or perhaps absent; the achro- 
matic figure is weak and consequently often distorted and irreg- 
ular. Every indication is that of an absence of kinoplasm. 
Thus the behavior of both oogonial nuclei and cytoplasm 
confirms in a striking way the view that kinoplasm is important 
in sexual differentiation, and suggests that the nuclei pass to the 
periphery to rid themselves of superfluous kinoplasm, possibly 
to prevent parthenogenetic development in the oosphere. If this 
be the true reason for the migration of the nuclei, it logically 
follows that kinoplasm is not readily convertible into tro- 
Phoplasm, at least not in the conditions that prevail in these 
oospheres. 
In the antheridium a behavior complementary to that exhib- 
ited in the oogonium is seen. The antheridial protoplasm stains 
intensely with the gentian-violet before fertilization, but after 
fertilization the cytoplasm left in the antheridium fails to give 
this reaction. The antheridial nuclei as they lie in the tube pos- 
sess a heavy membrane and stain darkly, giving every indication 
that they are rich in kinoplasm. In A. candida zonation is not 
