Wolfe .< — Cy to logical Studies on N emotion . 62 l 
best observed at metaphase, at which stage the shape of this structure 
in mitoses occurring in the early development of the gonimoblasts presents 
a strong contrast to that assumed by it in the vegetative phase, since 
in the latter it is relatively long and narrow, with the fibres meeting 
at the poles at a fairly acute angle, whereas, on the other hand, in the 
cells of the gonimoblasts its length remains about the same, while its 
width has become nearly doubled, and consequently the angle at the 
poles is now markedly obtuse (cf. Fig. 75 with 65, for example). It 
should be mentioned, perhaps, that the formation of the cell-wall which 
has been described above as occurring after mitosis is quite independent 
of these fibres. The central portion of the spindle could be observed 
for a short time after the separation of the daughter-chromosomes 
(Fig. 66), while its cones remain intact, somewhat shortened perhaps, 
for a considerable time (Fig. 69). 
Nucleolus. 
The nucleus, as has been already briefly pointed out, consists of 
a relatively large cavity bounded by a delicate wall and filled with 
nuclear sap, in the centre of which appears a single, large, heavily staining 
'nucleolus, 5 attached to the wall by a considerable number of radiating 
fibrillae (see figs.). The dimensions of the nucleus, and nucleolus as 
well, vary to some extent, as may be seen from the figures, its increase 
in size being apparently connected with preparations for division. The 
nucleolus is not homogeneous in structure, and consists of a peripheral 
region of denser material surrounding a central substance which shows 
a much less marked capacity for absorbing stains. This condition of 
affairs is represented in Fig. 51, which illustrates a favourable section 
passing through the body in question. As mitosis approaches, this 
peripheral substance aggregates in heaps, leaving thinner areas between 
(Fig. 52) ; and in some cases these aggregations are of such uniform 
character as to suggest that they may be the units from which the 
chromosomes are to be formed, their number being decidedly in excess 
of, and probably about double that of the chromosomes (Fig. 57). 
At this period the nucleolus in many cases assumes a lenticular shape, 
its short axis probably coincident with the long axis of the future spindle. 
The correctness of this assumption is indicated by the conditions present 
in later stages (Figs. 59, 61, and 62), in which the chromosomes are 
gathered chiefly in the equatorial region of the dividing nucleus. 
At this period the radiating fibrillae stain somewhat more con- 
spicuously, the nucleolus becomes irregular in outline and often slightly 
drawn out along them, while distinct, deeply staining masses appear at 
the points where they intersect the nuclear wall (Figs. 54-56). Inasmuch 
as previously no stainable material whatever could be detected in this 
U u 2 
