Mcioiic Mitoses of Osmunda . 
151 
Coming out of Second Contraction and Diakinesis. 
From the stage of second contraction, and onwards up to the close of 
the homotype division, the nuclear phases are comparatively straight- 
forward, are simple of interpretation, and afford little ground for debate. 
There is no difficulty in distinguishing between the nuclei issuing from 
second contraction, and those about to enter that phase ; for in working 
through the post-second contraction events, bivalent chromosomes in vary- 
ing stages of evolution will be seen. Further, the univalent chromosome 
segments often issue from second contraction in detached portions, and 
these speedily join end to end (Fig. 73). Moreover, the spore mother-cells 
now begin to separate. It has been found in these preparations, in which 
the method of fixation has largely eliminated artificial contraction, that the 
rounding-off of the spore mother-cells occurs at a considerably later phase 
than would appear to be the case when studying material which had not 
been so well fixed. 
As second contraction loosens, the bivalent segments begin to sort 
themselves out (Fig. 73). Some of the univalents (filaments) are still closely 
conjoined, whilst others are more or less separated, and it may be possible 
in isolated instances to distinguish the separate bivalent combinations, i. e. 
the future heterotype chromosomes. Some second contraction figures 
loosen directly into more or less organized thick, concentrated chromosomes 
(Fig. 74), whilst in others the univalents are more slender, and are in process 
of disjoining (Fig. 75). 
As the univalent chromosomes of each bivalent combination separate, 
they thicken (Fig. 7 6) and stain as more or less homogeneous bodies. The 
nuclear surface then increases considerably and the chromosome segments 
are dispersed at its periphery. Some of the bivalent chromosomes are 
clearly individualized, whilst others are not completely organized. Fission 
can only rarely be detected at this stage (Fig. 77). This is the reappearance 
of the fission which was evident in the univalent post-synaptic spireme, 
parting the threads which had associated during synapsis, and it will 
eventually result in the separation of the daughter univalent chromosomes 
on the homotype spindle. 
The chromosomes concentrate rapidly (Fig. 78) ; the small granules 
and globules which are often seen scattered in the nucleus and even 
attached to the chromosomes are probably fragments of dissolving 
nucleolus. 
In fortunate preparations spindle fibres are seen to radiate towards the 
nucleus from four poles which appear as cones in the cytoplasm (Fig. 79), as 
in Polypodium vulgar e ( 7 ) and other ferns. Strasburger ( 17 ) has figured 
a multipolar spindle origin (PI. IV, Fig. 178). Wilson Smith ( 18 ) occa- 
N 
