384 
M. G. Sykes 
by Miyake 1 ) were found. Some such single appearances were iu- 
deed sometimes noticed, but subsequent careful examiuation led me 
to believe tbat even these structures were in reality double. Even 
in Fig. 22 the linin threads are seeu to be double and indeed in all 
cases, projecting double loops are easily to be found. 
III. Late Synapsis and the formation of the Spireme. 
The nucleus grows considerably in size during synapsis, which ap- 
parently extends over a long period of time. The tightly wound ball 
becomes somewhat loosened and a spireme is spun out, during the 
formation of which each pair of aggregations becomes smaller and 
is gradually broken down into small granules, which with the help 
of the linin are arranged along the double thread. At first the gra- 
nules are still arranged in masses of considerable size placed oppo- 
site each other, at uneven distances on the threads, (figs. 25 & 26. 
PI. VIII.). Soon these become again subdivided and acontinuous double 
spireme is thus formed, composed of two parallel threads of »linin« 
with small chromatin masses or chromomeres at equal intervals along 
its course; (Fig. 27. PI. VIII.) The chromomeres are distinctly seen to 
be in pairs, the members of each pair being arranged on the two 
parallel threads; it was not possible to resolve each chromomere into 
granules. Very thin sections of some preparations at this stage, 
(Fig. 28. PI. VIII) show minute ilbrils of linin projecting irregularly 
from the swollen parts of the spireme in which the chromomeres are 
embedded. These can also be seeu in Fig. 29, which represents a 
slightly more advanced stage in which the double spireme is seen 
to be less tightly coiled. 
The fibrils of linin gradually become less obvious, the double 
spireme begins to fuse in places, wliile at the same time its coils 
are gradually loosened. The linin fuses first and this brings the 
members of each pair of chromomeres into contact with one another, 
(figs. 30 & 31); they fiually fuse together and a single thick spireme, 
with single chromomeres at regulär intervals, is thus produced. All 
the various stages of fusion are generally visible in a single nucleus, 
so that it would appear that they do not extend over a very long 
period. It is occasionally possible to find the ragged bits of linin, 
above referred to, projecting from the chromomeres on a fused 
but still contracted spireme but these are never present in the 
next stage. 
i) Miyake. 1. c. 1906. Fig. 116. PI. V. 
