262 
Sidnev I. Kornhauser 
While this is true for only the early zygotene stage, yet it argues agaiust 
the origin of the double threads by a longitudinal division. Soon the 
loops shorten to the zygopachytene condition and the nucleus goes 
through the stage of synizesis. The loops remain distinct, although 
they are closely crowded together and very deeply stained. Gradually 
the loops loosen out and the bouquet stage is entered upon. 
As in E. hinotata, there is present one very long loop, which can 
be seen in side views of the bouquet stage. It doubtless represents the 
pair of niacrochromosomes, which are easily distinguished in the later stages 
of the growth period. The macrochromosome loop often bends so that 
its apex (middle of the bend) conies into dose approximation with the 
apex of one of the shorter loops. These loops, seen in profile, then show, 
where they approach each other, a light line, which at first niight be 
interpreted as the point of union of two chromosomes (Fig. 80) and used 
as an argument for metasyndesis. Careful focussing, however, shows 
that this is not the case, but that the light line, such as seen in figure 80, 
is between two independent loops, not a point of approximation of half 
loops. In polar views one can foUow the contours of each loop, and their 
apices do not show any evidence of end to end conjugation. Moreover, 
if this light hne seen in side views were the line of fusion between two 
chromosomes, it would be necessary to explain why the conjugants are 
very frequently of such different lengths. 
The total number of autosome loops can be got by counting the ends 
of the loops at the positive’) pole of the nucleus. There are eighteen such 
ends, which correspond to nine double chi'omosomes. The longitudinal 
or interchromosomal cleft is often plainly visible in these polar views 
(Plate XX, Fig. 81). 
As the loops give up their regulär polar arrangement, the cytoplasm 
and nucleus both increase in volume; the threads beconie narrower and 
less deeply stained, and they do not show the interchromosomal hght 
line very distinctly. Figures 82 — 84 give three successive optical sections 
of a cell at this stage in which all nine autosome loops could be foUowed 
from end to end. These sections are combined in the diagram Figure A. 
I have not attempted in this diagram to indicate the doubleness of the 
threads or the distribution of the chromatin granulös, but merely represent 
the extent of the loops and their relation to the nucleolus and to the 
.c-chromosome. 
7) I designate as “positive” pole of the nucleus the pole toward wliicli the free 
ends of the loops are directed. 
