246 BOTANICAL GAZETTE [APRIL 
variation among them; some of them are so dark green as to be almost 
opaque, while others are much lighter green and are semi-trans- 
parent. It is clear that the density of the cytoplasm between the 
ridges determines in a large degree the external appearance of the 
plant. DrBary has described the chromatophore as being homo- 
geneous, except for the pyrenoids, in surface view, but a careful 
examination shows that the larger vacuoles of the cytoplasm between 
the ridges appear as lighter areas in the darker green of the chro- 
matophore. 
The material of the chromatophore itself is in the form of a very 
fine reticulum, the meshes of which run out into the ridges in such 
a manner that in cross section the latter appear to be almost radially 
striate. The meshes are quite small in the outer layer of the chro- 
matophore, but its central region shows a much coarser reticulum. 
The meshes here vary much in coarseness, and their structure, in 
connection with the cytoplasm in the grooves, determines whether 
the plant has a dark or light green color. 
The number of ridges on the chromatophore varies from 12 to 
18; this is more than is given in the systematic descriptions of this 
form, but it is practically impossible to count the number of ridges 
accurately except in sections. ee 
Toward the ends of the Closterium the ridges seem to withdraw 
from the walls, and a cross section at these points shows a chromato- 
phore with a somewhat star-shaped form and a rather thick layer of 
cytoplasm between the outer edges of the ridges and the cell wall 
(fig. 3). The ridges are fewer in number also, owing to the fact 
that some of them do not extend to the ends, and their outer 
edges are acute instead of blunt and rounded as is the case nearer 
the center. 
In C. monilijerum (figs. 5-7) the structure of the chromatophore 
is practically the same except that the number of the ridges in my form 
1s always 10; NAGELI’s figures show 10 and 6. This gives an easy 
additional means of identifying the species in section, as the ridges 
of C. Ehrenbergii are always more numerous. The central core of 
the chromatophore is always larger than NAGELI figured it. In 
optical cross section it gives the appearance he shows, but actual 
cross sections show clearly that it is essentially similar to that of C. 
