147 
Meiotic Mitoses of Osmuncia. 
shading in the beading, whilst at still other times the sides (threads) are so 
closely appressed as to form a single row of beads. P'ig. 50 shows a Her- 
mann fixed superficial section of the loosening of synapsis. In this nucleus 
every portion of the spireme filament exhibits fission (i. e. separation) and 
that of a very beaded kind characteristic of Hermann preparations, whilst 
in strong Flemming preparations the sides (threads) are more homogeneous 
(Fig. 51). The spireme filament distributes itself throughout the nucleus 
(Fig. 51), the -so-called ‘ hollow spireme ’ stage, and as it invades the nuclear 
cavity it becomes thrown into loops (Fig. 52), and these are often orientated 
round the nucleolus (Fig. 53). It is exceedingly difficult to decide whether 
free ends are present, but the evidence points to an affirmative, and at 
a slightly later stage they are certainly to be found. From this stage 
onwards the spireme filament becomes more uniformly homogeneous, and 
fission rapidly closes. Figs. 53 and 54 have been taken from sporangia in 
which the nuclei of half of the sporangium show fission, whilst in those 
of the other half fission has already closed (PI. X, Fig. 59). 
Perhaps it may not be out of place here to make a slight digression 
and allude to the widely separated stages in the progress of nuclear division 
that may be found in a single sporangium, testifying to the rapidity with 
which many of these phases are passed through. In a single sporangium 
some nuclei may be in the pre-second contraction phase, whilst others are 
in diakinesis (PI. XII, Fig. 101) ; or half of the nuclei may be in second con- 
traction and the other half in metaphase. These rapid transitions increase 
the difficulty in interpreting the complicated stages leading to second con- 
traction. Wilson Smith ( 18 ) has given some interesting data on this subject, 
and it is evident that in his plants growth was not so rapid as in the material 
collected for the present research. His material of O. regalis was fixed in 
northern Indiana, and he calculates that the resting and early spireme 
stages last for about two weeks, synapsis for three or four days, whilst the 
two meiotic divisions are effected in quick succession, within a period of 
two or three days. 
Later Hollow Spireme to Second Contraction. 
To describe clearly the series of nuclear phases between hollow spireme 
and second contraction is a difficult task, for not only do the spireme fila- 
ments show great diversity according to the fixative used (with strong 
chromic the swelling is considerable), but even adjacent nuclei in a sporan- 
gium reveal strikingly different characters. 
The principle to be borne in mind is, that as the nuclei pass into 
second contraction an arrangement for the conjunction in pairs of entire 
univalent spireme segments (i.e. of filaments) is evolved. This conjunction 
may be in all stages of completion before second contraction supervenes, 
but it is finally consummated in the massing of second contraction, during 
