34 BOTANICAL GAZETTE [JULY 
also is not uncommon. The rays are usually simple, but may be 
branched especially during the earlier stages. 
It is worthy of note that the radiations are most pronounced 
and stain most deeply with gentian violet, while the nucleus is 
elongating and its poles are separating from each other; and, 
further, that during this period many of the radiations connect 
the poles with the Hautschicht. The explanation which we ven- 
ture to suggest is that the radiations take an active part in sep- 
arating the poles from each other. The fact that the radiations 
disappear as soon as the poles have reached their widest separa- 
tion supports this hypothesis. The reappearance of the rays in 
the telophase does not seem to be so definitely concerned with 
movement, because they again disappear before the nucleus 
has perceptibly changed its position: still, it is possible that 
there may be a slight movement in the nucleus toward the center 
of the new cell. The reappearance, however, takes place as 
the nuclear membrane begins to be formed, and it may be an 
expression of kinoplasmic activity during the formation of the 
Hautschicht surrounding the nuclear membrane, or the rays may 
be contributing to the formation of the nuclear membrane itself, 
which, we believe, is largely kinoplasmic in its nature. 
THE SECOND MITOSIS IN THE GERMINATING SPORE, WITH REMARKS ON 
APICAL CELL, ANTHERIDIA, NUCLEOLI, AND CHROMOSOMES. 
The second mitosis is remarkably easy to fix and stain; so 
that, while the first mitosis, if equally well prepared, might 
show the early prophases with a little more clearness, our 
material afforded a better study of these stages during the sec- 
ond mitosis. 
In studying the second mitosis, special attention was devoted 
to the centrosphere and to the origin of the achromatic struc- 
tures. The terms “centrosome” and ‘“centrosphere” are fre- 
quently confused. Until much more is known about the origin 
of these structures and their relation to each other, it is hardly 
worth while to attempt any definitions. A typical centrosphere 
—as the term is used in this paper—is shown in fig. J2. The 
centrosphere consists of the same substance as the astral rays 
and the spindle fibers. The elongated body toward which the 
