144 Wilson. — On Spore Formation and Nuclear 
and in the preparations forms a lightly-stained irregularly-shaped area 
around the nucleus, sharply limited from the remaining protoplasm (Fig. 2). 
At first sight, this suggests improper fixation, but examination of the other 
parts of the cell leads to the conclusion that this represents a real structural 
difference. The nucleus is large in proportion to the cell, and contains one 
large deeply-staining nucleolus. The remainder of the nuclear cavity is 
filled with a very fine homogeneous network, somewhat resembling the cyto- 
plasm in structure ; chromatin is very scanty or almost absent in the 
resting nucleus, a few granules sometimes occurring scattered in the inter- 
sections of the alveolar walls. 
A nucleus of the type can be considered to be characteristic of the 
Muscineae. It has been described by Beer (4) in Atrichum undulatnm , by 
Arens (1) in the spermatogenic cells of Poly trichum juniperinum and Mnium 
hornum , and by Drs. Leeuwen-Reijnvaan (20 and 21) in several species of 
Polytriclmm. In the case of Atrichum undulatum , Beer has also noted the 
presence of compound nucleoli, and he has described similar nucleoli in the 
nuclei of the vegetative cells of Ricciaglauca (4). These, he states, ‘consist of 
a lightly coloured matrix, in which are embedded a number of grains of chro- 
matin.’ Such nucleoli have been described by other observers, and they have 
been met with in the resting nuclei, both of the archesporial cells and of the 
spore-mother-cells of Mnium , but it is open to question whether this 
appearance is due to a real structure or is an effect brought about by the 
fixing reagent. These nucleoli were found much more frequently in pre- 
parations fixed with acetic alcohol than in those which had been treated 
with Flemming’s Mixture. Fischer (11) has emphasized the fact that the 
former reagent causes the precipitation of some of the proteids of the cell 
in a soluble form, and it is possible that this appearance is due to partial 
solution of the nucleolus at the time of fixing or during subsequent treatment. 
The archesporial cells rapidly increase in size, and division soon takes 
place. Material preserved at various hours of the day was examined, but 
no definite relation could be found between the time of preservation and the 
progress of division. On the whole, divisions were found to be most 
frequent during the morning, but this, no doubt, was partly due to the rise 
of temperature ; various stages could generally be found in the same section. 
The first dividing wall is generally periclinal, and this is soon followed by 
two anticlinal divisions at right angles, so that each of the original cells 
becomes divided into octants. In some cases no further divisions occur 
before meiosis, but in others each octant may again divide by a periclinal 
wall ; in a few cases this is followed by yet another division resulting in 
the formation of a wall transverse to the axis of the sporogonium. Thus, 
groups of 8, 1 6, or 32 cells may be produced from each original cell of the 
archesporium. Each group retains the outline of the original cell, and its 
limits can be distinguished up to the time of spore formation. Nuclear 
