120 
L. Digby 
It will be gatherecl from this descriptioii that the appearaiice and 
disposilion of associated univalent Segments as they enter, and as they 
come out of, second contraction is often very siniilar. 
The decision arrived at from the study of the phases of the hetero- 
type division conforms to the view first expressed by Farmer and Moore 
(11), that the association of univalent Segments, initiated m previous 
stages, is completed during second contraction. 
The phases concerned with second contraction must be quickly 
passed through, as nuclei üi very different progressive stages may be 
found in the same loculus, and even in the same section of a loculus. 
AMien the nuclear contents have come out of second contraction 
they are extremely viscous in character, and the separated univalent 
Segments may be dra^vn out as fine Strands (fig. 79). The various parallel 
lengths are connected by concentrated niasses (fig. 80). It is generally ini- 
possible, at this stage, to individuahze the future three bivalent chromo-, 
somes, for, hke the soniatic chromosomes, they originate m portions 
and it is only immediately prior to the appearance of the spmdle fibres 
that these fuse together to form the mature heterotype chromosomes 
(figs. 81 and 82). The chromosomes are in fact evolved out of what appears 
to be a tangle of viscous nuclear contents. 
The typical heterotype chromosomes are of three distmet sizes (PI. X, 
fig. 83) as described by Rosenberg (46). MTien they have completed 
their growth the ahnost colourless nucleolus collapses, and its substance 
is distributed in the form of globules. 
Metaphase of the heterotype division. 
The dlfference in size of the chromosomes, when on the spindlc (fig. 83), 
is very appareut. The large one is generally C shaped, the middle sized 
one niore or less V shaped, whilst the small one resembles a dyad. 
Anaphase of the heterotype division. 
The smaU chromosome is the fh’st to separate, and its rounded univa- 
lent Segments may have already arrived at the poles, whilst those of the 
larger chromosomes are stUl in contact (fig. 84). Fissiou of the two large 
univalent chromosomes almost cleaves theh- daughter halves, causing 
them to be V or U shaped, their apices being directed towards the spindle 
poles (figs. 84 and 85). ln the polai’ view of a late anaphase the chronio- 
somes sometimes appear to be more than three in number (fig. 87). This 
is due to a transverse division of the chromosome segments. Rosenberg 
