Division in Mnium hornim. 
149 
intermediate stages were not very abundant. J udging from this the synapsis 
is of considerable duration, but the preceding stages are rapidly passed 
through. During the period of the greatest contraction the nucleus was 
found close to the periphery of the cell (Fig. 21), and, as this was of constant 
occurrence in different preparations, it may be concluded that the position 
is characteristic of the synaptic stage. 
The first indication of the emergence of the spireme from the contraction 
is given by the appearance of a number of thick indefinitely tangled threads 
within the deeply-staining mass (Fig. 22). This can be at once dis- 
tinguished from the somewhat similar stages immediately preceding the 
contraction on account of the thickness of the emerging thread characteristic 
of this stage. At this time the nucleus also begins to pass back to the 
centre of the cell. During synapsis there is a considerable increase in the 
amount of chromatin, and the spireme on its reappearance is at least twice 
as thick as the threadwork which formed the original reticulum. The spireme 
gradually becomes more loosely coiled and spreads out, and is soon found 
distributed throughout the nuclear cavity (Fig. 23, PI. XI). This second 
network obviously consists of a shorter length of thread than the presynaptic 
reticulum ; it appears that besides the increase in amount of chromatin an 
actual shortening of the thread goes on during the contracted state. 
Irregular masses of deeply-staining material are found in the course of the 
thread, but the body previously cut off from the nucleolus has now dis- 
appeared ; the nucleolus itself still shows no diminution in staining capacity. 
Although careful examination of the thread was made at this period, no 
definite evidence either of fusion or approximation was obtained. 
The second contraction now supervenes (Fig. 24) ; the spireme tends to 
become collected around the nucleolus, but the aggregation is not nearly so 
complete as at the first synapsis. Loops are found radiating outwards from 
the central mass, and in some of these longitudinal fission could undoubtedly 
be seen at this stage (see loop in upper part of Fig. 24). The loops, 
although for the most part consisting of undivided filaments, in some cases 
exhibit the splitting at frequent short intervals along their course. At the 
next stage the individual chromosomes can be recognized. As far as can 
be discovered these are not produced by the direct transverse division of the 
spireme, but rather by the aggregation of the chromatic material to certain 
lengths of the thread which are subsequently drawn asunder. The inter- 
vening portions of the spireme become thinner and take up the stain much 
less deeply. On the other hand, the lengths destined to form the chromo- 
somes increase in thickness and staining capacity. The previously mentioned 
masses of chromatic substance which are situated at intervals in the spireme 
appear to give up material to the chromosomes (Fig. 25) ; as the latter 
become more distinct the masses of chromatic substance gradually disappear. 
The whole process strongly suggests the transference of the material from 
