STUDIES OF THE DEVELOPMENT OF PEPEROMIA HISPIDULA 365 
form of a rather loosely looped spireme, with distinct chromomeres 
scattered along it (fig. 62). These chromomeres are about half a mu 
in diameter and occur at intervals of 1.5 or 2 along the thread. The 
intervening portions of the thread are slightly narrower and stain 
much less deeply. The thread at this stage is evidently double (figs. 
62, 120-19. 2-30. 6). Soon after this spireme appears it begins to 
segment, by constriction, to form looped and twisted chromosomes 
(fig. 63). These chromosomes are at first 15 or 20 times as long as 
thick but later they become considerably thickened or may be nearly 
globular in form (figs. 65, 66). 
The nucleolus retains its original size and staining character up to 
the time of segmenting of the spireme. When lightly stained one or 
more vacuoles or lighter staining areas are often visible in it (figs. 
59, 63). Between the time of formation of the chromosomes and the 
completion of the spindle the nucleolus, as a unit, disappears though 
fragments with the same staining qualities are still seen within or 
without the nuclear area (figs. 65, 66, 67). It may be recalled here 
that similar dark stained granules are visible in the cytoplasm while 
the nuclear wall is still intact (figs. 60, 62, 64). Hence it is not certain 
that any of the dark bodies seen in the cytoplasm during mitosis are 
of nucleolar origin. During the progress of synapsis a peculiar concave 
sheet or disk is found lying against the inner surface of the nuclear 
wall, often on the side opposite the synaptic knot (figs. 59, 60, 61, 62). 
This disk is .5 thick and often 10 /x across. This body is first seen 
when the nucleus is just entering synapsis (fig. 60) and was not seen 
after the segmenting of the spireme (fig. 63). Structures which may 
be remains of these disks were occasionally found lying near the poles 
of the spindle, but the identity of these is not certain (fig. 65). This 
disk stains a deep black with iron haematoxylin and a dark gray with 
Flemming's triple stain, and in structure it may appear in section as a 
clear black line or, in other cases, it is rough in outline (figs. 59, 60, 62). 
The morphological identity of this disk has not yet been determined. 
Further work on the origin and fate of this disk is planned as part of a 
detailed study of the whole process of reduction in the embryo sac 
mother-cell. 
A marked segregation, of the at first nearly homogeneous cyto- 
plasm of the mother-cell, forming 2 concentric layers around the 
nucleus, usually occurs during the progress of synapsis. The outer 
layer, of 4 or 5 in thickness, is highly vacuolated, stains weakly and 
