1922] RANDOLPH—MAIZE 367 
successively smaller stages to proplastids which lie just within the 
lower limit of visibility, there appears to be no justification for the 
view that more than one kind of initial granule is present in the 
meristematic cell. It is only through a study of these minute 
initial stages that many of the principal questions involving chon- 
driosomes and plastids can be answered. These earliest stages 
seem to have received but scant attention from previous workers. 
It is to be emphasized that the evidence at hand leads directly to 
the conclusion that the various structures observed in later stages 
(green and white plastids) have arisen by a process of differentiation 
from bodies of one kind, rather than from bodies initially unlike. 
The cause of such a difference in behavior on the part of the pri- 
mordia can only be conjectured. 
INDIVIDUALITY OF PLASTIDS.—The proplastids have been found 
to be a constant feature of cytoplasmic organization in maize, and 
the question arises as to the origin of these bodies. The definiteness 
of their behavior in relation to the evolution of chloroplasts indicates 
that they are at least concerned in the elaboration of certain 
products of cell activity. Whether or not they may also be found 
to be associated with other vital functions remains as a problem 
for further research. 
Plastid primordia have been considered by previous workers 
as essential constituents of the cytoplasm which retain their 
individuality and persist throughout the life cycle (FORENBACHER 
21, PENSA, CAVERS 5, GUILLIERMOND, MorrieR, EMBERGER). The 
evidence from primitive plants (ALLEN, SAPEHIN, SCHERRER), in 
which well developed plastids are present throughout the life 
cycle, may be considered as evidence favoring this view. The 
occurrence of division stages in partially developed chloroplasts 
suggests also that the smaller bodies may divide, but my obser- 
vations indicate rather that such an assumption is hardly war- 
ranted. Appearances have frequently been found which suggest 
division stages, but definite proof of division is very difficult to 
obtain. The fact that a frequent association and subsequent 
separation of the proplastids occur in embryonic cells, as herein 
described, renders it unsafe to draw conclusions on this point. 
Even if it were shown that these minute bodies multiply by 
