196 Sargant . — Formation of the Sexual Nuclei 
spherical ball stained an intense green, and thus showed 
the presence of chromatin. The section cut on the third 
morning was particularly successful, and was preserved for 
future reference by laying it in much diluted Merkel’s solution 
for half an hour, and then staining with Mayer’s haemalum. 
I could not detect any alteration in the shape of the nuclei 
after this fixing and staining process. The fixed and stained 
section, though thin for one cut from fresh material, cannot be 
compared with a good hand-section from a hardened anther, 
and the structure of the nuclei is further obscured by the 
heavy stain ; but there can be no doubt that the nuclei are 
really in the condition of synapsis. Their outlines correspond 
perfectly well with that drawn from a nucleus in the fresh 
section before fixing. 
We may now proceed to examine the structure of the 
contracted nucleus in detail by means of sections considerably 
thinner than its mean diameter. The appearance of any 
particular nucleus in such a section will depend greatly upon 
how the contracted part lies with reference to the plane of the 
section. If the knife has passed through the area over which 
the chromatic ball is in contact with the boundary of the 
nuclear cavity, then the section of the nucleus will have an 
outline resembling that of Fig. 3#. But if the section is 
roughly parallel to the area of contact, the chromatic ball may 
appear as a circular disk within the nuclear cavity, but quite 
free from it. In this case the whole section of the chromatic 
ball commonly drops out, leaving the nuclear cavity as an 
empty space of irregular outline. It often happens also that 
the sections are injured by the knife, perhaps because the 
absence of a definite nuclear membrane weakens the structure 
of the whole nucleus. At any rate, fragments of the chro- 
matic thread and bits of nucleolar matter are constantly found 
outside the nuclear cavity, and the whole preparation is apt to 
look ragged and untidy. 
Neglecting such details as may be fairly attributed to 
imperfect preservation, we find that there are three features 
characteristic of synapsis — a more or less complete disap- 
