Beer. — On the Development of the Spores of Riccia glauca. 285 
plast can then be seen to be surrounded by a thin layer which gives 
ccllulose-pectose reactions (Fig. 36). It may be said, therefore, that in the 
formation of the second spore-wall the first lamellae which are deposited 
are comparatively loosely arranged together (outer region of second spore- 
wall), whilst the later ones become so firmly united that the lamellose 
nature of this portion of the wall becomes obscured (inner region of second 
spore-wall). It is these two regions of the wall — the loosely and the 
densely laminated areas respectively — which become more or less separated 
from one another by the interpolation of the dark material which was 
mentioned above. 
The endospore is formed comparatively late in the development of the 
spore. It gives the reactions of cellulose and pectose, and it is usually 
separated from the second spore-wall by a narrow space occupied by dark 
material similar to that which occurs in the more conspicuous gap between 
the two layers of the second spore-wall (Fig. 26). In spores which are 
nearly or quite mature the walls are deeply coloured brown and have 
become much denser and in consequence thinner. The lamellated structure 
of the wall is now obscured. Heated to redness on platinum foil with 
a drop of concentrated sulphuric acid the entire spore dissolves without 
leaving a silica skeleton behind. 
Nothing has been said above of the nutrition of the spore or of the 
sources of the material for the growth of its membranes, so that a few 
words upon this subject must be added here. That the protoplast of the 
spore is itself actively concerned in the growth of the membranes which 
surround it can scarcely be doubted. The new lamellae which are added 
to the second spore- wall are formed, as we have seen, in the most intimate 
union with the plasmatic membrane ; the nucleus of the spore also presents 
an appearance which strongly suggests that it is participating in metabolic 
activities. 
Unlike the usual resting nucleus of Riccia glauca, to which I have 
already referred, the chromatin is not confined to the nucleolus, but is also 
distributed along a rather thick filament which strongly recalls the spirem- 
thread of the dividing nucleus (Fig. 27^, 27 £). It is difficult to decide 
whether this thread forms one continuous structure or not. 
As the spore grows older and its walls become thickened we find that, 
although the thread long maintains its spirem-like arrangement, it gradually 
stains less and less deeply with the haematoxylin (Fig. 27 c). Somewhat 
similar spirem-nuclei have been described in various animal cells. Thus 
the well-known case of the salivary glands of Chironomus larvae, studied by 
Balbiani, may be recalled, or the ovarian eggs of Triton taeniatus , in which 
Born found a spirem-stage to precede a more diffuse arrangement of the 
chromatin. In most of the cases already known in which the chromatin is 
distributed through the nucleus in this manner, we are dealing with cells 
