240 Rickett.—Fertilization in Sphaerocarpos . 
sometimes they are present throughout the nucleus, so that the latter has 
the appearance of a rough wheel with a large hub and irregular spokes 
(Fig. 7 ). Finally, the whole male nucleus may resolve itself into a compact 
reticulum which fills the whole of the space within the membrane, no part 
being noticeably denser than any other. This structure is illustrated in 
Fig. 12 , a very early stage of the next phase. The reticulum is composed 
of granules linked together by threads which are almost as thick as the 
granules, so that the whole structure is still fairly dense and compact. The 
chromatic material of the nucleus now takes a violet colour in the triple 
stain, whereas in former stages it stained more readily with the safranin. 
This series of changes is not found in succession throughout the period 
occupied by this phase. It seems probable that each nucleus must pass 
through these steps in order to attain the reticular structure ; but some of 
the later stages are found in some nuclei at the beginning of the period, 
while others, perhaps in the same plant, still resemble those in the preceding 
phase. On the other hand, the earlier stages may be found throughout the 
period. 
It is in this phase of the process, in fact, that the times at which the 
various changes take place begin to be variable. Series A and B furnish 
most of the cases in which the chromatic mass of the male nucleus is con¬ 
tracted and dense, while in Series C the more open reticular type of structure 
is more common. This variation between series suggests not only that the 
time necessary for the different phases may vary, but that the extent of the 
changes undergone by the nucleus may differ in different cases. It may be 
that in some cases the male nucleus never forms an open reticulum. The 
cause of this variation may lie in temperature or other external conditions, 
for each of the series was fixed at a different time of year from any of the 
others. And again, it is possible that the procedure of fixing varied suffi¬ 
ciently to account for these differences. 
Another complication enters in the fact that in Phase 5 , next to be 
described, the male nucleus seems to contract again, and presents an 
appearance similar to what is seen during Phase 3 (PI. IV, Fig. 14 ). This 
makes it difficult to say, in the case of such a contracted nucleus occurring 
during the present phase, whether it is a delayed or an advanced stage. In 
one case, unfortunately torn in the preparation, the male nucleus seemed to 
contain a single much-coiled thread, along which were scattered granules in 
a single row. 
Some variation begins to appear also in the female nucleus. In general, 
the chromatic substance is still massed in the centre, and little can be made 
out of its structure; but occasional nuclei are found in which more of the 
chromatin is distributed through the nuclear cavity in the form of threads 
and granules, with larger masses here and there. Typical nuclei of this 
phase are shown in Figs. 6 and 9 . Sometimes also the changes characteristic 
