314 BOTANICAL GAZETTE [APRIL 
that is, from a resting nucleus as conventionally described. 
Observations here lead to agreement with SHARP (24), as follows: 
While one easily gains the impression of separate chromatin granules 
connected by threads with another substance . . . . it seems more probable 
that the “chromatin granules” are merely the heavier portions of the alveolated 
and reticulated chromosomes, and that the lighter “supporting network” 
consists simply of the thinner portions of the same, together with the delicate 
anastamoses. 
In portions of the thread as it enters synizesis, a curious 
partly paired, partly vacuolate-split appearance is discernible 
(fig. 3). It is stages of this sort which lend themselves as con- 
veniently to one philosophy of nuclear division as another, and 
which should not be taken as pivotal until all other means of 
explanation have been exhausted. The optical difficulties attendant 
upon the close state of aggregation are very great here of course, 
necessitating thin sections and special technique. 
Synizesis culminates in an extremely dense ball whose compo- 
nents are without doubt filiform. The embryo sac mother nucleus 
in fig. 4 is typical in every respect save that of size, being larger 
than usual. The loosening thread (dolichoneme) is fairly uniform 
at first (figs. 5, 6), and Jurt (12) is doubtless right in stating that 
any apparent nodes are optical effects due to foreshortening or 
crossing. 
After the thread becomes rather evenly distributed through 
the nuclear cavity, its uniform appearance is altered by the advent 
of changes which are hard to explain except as fissions. Certainly 
they are quite different from (1) accidental or other juxtaposition 
of whole threads, or (2) the twisting together of limb and bight 
into a loop. Both (1) and (2) are to be seen in figs. 7 and 8, where 
they may be compared with the seeming fissions. Whatever the 
change that gives this appearance of duality, it is clearly not 
simultaneous throughout the thread. A priori, is there any reason 
why it should be? The unevenness of its origin perhaps may 
explain the failure of other students of parthenogenetic species of 
Taraxacum to observe anything suggesting a “double” thread. 
Careful inspection of Osawa’s fig. 46 (17) shows that the phe- 
nomenon is by no means precluded there. 
