448 BOTANICAL GAZETTE [DECEMBER 
coming together side by side of parts of the spirem that before were 
arranged end to end. “The chromosomes then contract and show 
the twisted appearance characteristic of the heterotypic division 
(fig. 11). 
After considerable contraction they have the appearance of two 
irregular rounded masses lying together. Sometimes these are seen 
to be connected by strands (fig. 13), and just before this, when the 
constituent halves are about twice as long as wide, they are sometimes 
placed end to end with a constriction between them (jig. 12). 
It would seem from this that the two halves originally placed end 
to end in the spirem, then side by side in the loops, again come to 
lie end to end, and that each half probably represents a chromosome 
(cf. fig. 16). The strands connecting the two halves are often seen 
to be double (figs. 12, 13), and they might always be so if seen in the 
right plane. Besides this, the halves sometimes show evidence of 
being double. This may be due, as is often supposed, to the split 
previously described. Before the spindle is formed the two halves 
come together, producing somewhat elongated chromosomes (fig: 7 4). 
About this time the nucleolus begins to fragment and to be thrown out 
into the cytoplasm. After the spindle is formed (fig. 15) the chromo- 
somes divide transversely to their long axes (fig. 16). From what has 
been said it seems evident that this division separates parts of the 
spirem which were originally placed end to end and that it is there- 
fore a transverse division. This is then the heterotypic and reducing 
division. . 
As the chromosomes approach the poles they become crowded 
together (fig. 17) and surrounded by a clear space. While this 15 
going on, the chromosomes lose their distinct outlines and a nucleolus 
makes its appearance in their midst. While the chromosomes lose 
their distinct outline and probably also some of their substance, they 
seem nevertheless to be represented by irregular masses during most 
if not all of the period between the first and second divisions (figs: 
18-20). Between the succeeding divisions they seem to go to ~ 
to a much greater extent (cf. figs. 22-25). Before the formation ¢ = 
spindle, the chromosomes appear as double structures, consisting 
two rods lying side by side. It may be that these represent i 
two halves of the spirem seen in the prophases of the first — 
