634 Davis .— Cytological Studies on Oenothera. II. 
which stage the chromatic bodies can be distinguished only with difficulty, 
if at all, since they lie in the meshes of the deeply staining strands of the 
network. A single large nucleolus is present and frequently one or more 
smaller nucleoli. It is important to note that this linin network 
develops as a reticulum from its earliest inception ; there is at this time no 
indication of a continuous spireme, much less two parallel spiremes such 
as have recently been described for a number of forms by Overton (’ 09 ), 
Lundegardh (’ 09 ), and Rosenberg (’ 09 #). 
The fate of the chromatic bodies in Oenothera as the nucleus approaches 
synapsis is a problem of great interest, but at present all that can be said 
is that their substance becomes apparently merged with that of the reticulum, 
from which they cannot finally be differentiated. It is, however, possible 
that in this reticulum there are regions, corresponding to the chromatic 
bodies, which are quite distinct from the linin substance of the network that 
in earlier stages (Figs. 1-4) is easily distinguished from the chromatic 
bodies. Such a chromatic framework, if present, may be responsible for 
the spireme which appears during synapsis. 
Synapsis. The approach of synapsis is indicated by the contraction 
of the reticulum away from the nuclear membrane (Fig. 6), and this 
contraction proceeds until almost all of the threads and meshes in the 
network are drawn into a close mass (Fig. 7) that lies generally near the 
large nucleolus. While the contraction is in progress the reticulum takes 
on the character of a tangle or ball of threads, constituting the synaptic 
knot, from which a few loops and very delicate strands usually extend into 
the nuclear cavity (Figs. 7-10). It is impossible to follow the individual 
threads in the denser portions of the knot, but the free loops at the edge 
give an opportunity for more detailed study. 
Although it is frequently possible to find threads that run closely 
parallel to one another for some distance the writer is unable to present 
any evidence that this condition is other than such a close association as 
would be brought about by the gathering together of a complexly looped 
thread or system of threads into a tight mass. The loops appear to be quite 
independent, as though they were parts of a much tangled system rather 
than portions of a split spireme, or of two independent systems of threads 
(maternal and paternal spiremes), which might be assumed to be associated 
with one another side by side. It is, however, impossible to say whether 
or not the coiled and twisted loops of the synaptic knot are parts of a 
single continuous thread ; if so, it must be a thread of great length. 
The seriation of such a group of stages, as is shown in Figs. 7-10, is ren¬ 
dered somewhat easier by the fact that the pollen mother-cells separate from 
one another during the process of synapsis, each one rounding off and coming 
to lie freely in the pollen chamber. Fig. 7 is undoubtedly the earliest 
stage of this group of figures, because the pollen mother-cells were still 
