Gregory. — Spore-Formation in Leptosporangiate Ferns. 451 
not very favourable for observation. In preparations stained with aniline- 
safranin, gentian violet, orange G 1 the nucleolus of the young spore-mother- 
cell is brilliant red, while the reticulum has a purple tinge. At a later stage 
the spireme shows increasing affinity for the safranin. 
The nucleolus is always in close contact with the thread, with which in 
some cases it appears to be in direct continuity (see Wager’s figure, loc. cit., 
PI. V, Fig. 16) 2 . When this is the case the part of the thread in the 
immediate vicinity of the nucleolus stains a brighter red, and is without 
the purple tinge which characterizes the remaining portions. The con- 
traction of the spireme invariably takes place in such a way that the 
nucleolus is closely surrounded by the aggregated thread. The nucleolus 
becomes vacuolated, and its staining capacity diminishes, while the chromo- 
somes continue to take the stain strongly. The nucleolus is visible for 
some little time after the segregation into chromosomes has taken place, as 
a spherical body lying free in the nuclear cavity, but it disappears com- 
pletely at about the time that the spindle-fibres first become visible. 
No nucleolus has been found in the daughter-nuclei, so that all trace 
of this body is lost from shortly after the formation of the chromosomes of 
the heterotype division until its reappearance in the nucleus of the young 
spore. This absence is not surprising, since the daughter-nuclei scarcely 
enter the resting condition (see p. 449) ; the manner of its reappearance in 
the young spore-nucleus scarcely suggests its direct formation by a fusion 
of the chromosomes into nucleolus-like masses 3 . The appearance is rather 
that of a fusion of the chromosomes resulting in the formation of a thread, 
which is becoming reticular at the time when the nucleolus (or nucleoli) 
reappears in close connexion with it. 
In the Ferns the relative proportion of the nucleolus to the reticulum 
as regards size and staining capacity differs considerably from that in 
Phaseolus and certain other plants 4 . The transference of stainable matter 
from the nuclear thread to the nucleolus in the resting condition of the 
nucleus would appear, therefore, to vary in the degree to which it is carried 
out in different plants ; and the reverse process, which always precedes 
mitosis, seems to indicate that in the fission of the chromosomes some 
provision is made for the distribution of nuclear matter in a way which 
would not be accomplished by a direct division. The work of Sutton, 
already quoted (p. 450), apparently points to an individuality of the 
1 This was found to be more satisfactory than haematoxylin owing to the differential staining 
of the nucleolus and spireme. 
2 The figures illustrating this paper were all completed before the appearance of Wager’s paper 
directed special attention to the relations of the nucleolus, so that the figures specially illustrating 
these points must unfortunately be omitted. 
3 Wager, 1. c. 
4 The structure of the nucleus in e.g. Lathyrus odoratus is very like that of Phaseolus as figured 
by Wager. 
