88 BOTANICAL GAZETTE | FEBRUARY 
at the poles. There is a tendency for the polar cytoplasm to 
arrange itself radially. 
The short chromosomes of Peltandra, which are sometimes 
of a short Y form are closely arranged on the nuclear plate, evi- 
dently with the knee inward. In the vicinity of the knee are 
attached the spindle fibers. If there is then any noticeable sepa- 
ration, indicating double V or Y forms, this separation is in the 
equatorial plane, thus giving the normal nuclear plate arrange- 
ment with longitudinal fission. The form of the chromosomes 
during the separation into the daughter segments (fig. 20) sup- 
ports this view. The form of the short daughter segments soon 
after separation is somewhat oblong heart-shaped, since the 
angles of the V and Y forms seem to shorten and to fuse. 
If, however, my interpretation of the condition of things in 
Symplocarpus is correct, the separation at the nuclear plate stage 
of the first division brings about a transverse division. Then, as 
suggested to me by Professor Atkinson, the closely following 
second division would provide for the separation of the segments 
marked out by the first division. Korschelt’ states that in 
Ophryotrocha puerilis reduction is effected during the first 
division. 
There are about twenty-two chromosomes, and when they are 
arranged on the nuclear plate, a polar view may give the idea 
that they are variously fused. This appearance, however, is only 
due to the fact that the staining does not make evident the lines 
of contact, and a slightly oblique view aids the deception. 
Radiations as of extra spindle fibers, which in some way are 
set free during the splitting of the chromosomes, veil the central 
spindle, and reach out almost to the periphery of the cell in the 
middle diameter of the spindle. 
Cytoplasmic granules staining deeply are quite numerous in 
peripheral regions, and in Peltandra they are exceptionally large. 
In all stages of the developing and complete spindle, special 
attention has been given to the polar region, but in no case has 
there been any indication of bodies that might be taken for 
5 Zeitschrift fiir wiss. Zodlogie 60 : 543-688. 
