252 BOTANICAL GAZETTE [APRIL 
are destined to form starch grains. The origin of the starch grains 
and the nature and fate of the pyrenoid segments are difficult questions. 
That the two problems are closely related, as maintained by T1M- 
BERLAKE, is the most suggestive conception that has so far been 
advanced, but the whole matter is less clear in Closterium than in 
Hydrodictyon. Further comparative studies on the pyrenoids of 
the Conjugatae will doubtless make the significance of the pyrenoid 
segments more clear. 
A culture of Closterium was kept in the dark for three days, and 
then when treated with iodin showed very little starch present around 
the pyrenoids. Specimens from this culture were imbedded and 
sectioned. It was found that in most cases the starch had entirely 
disappeared, but that in others there were still traces of it present 
in the form of a thin layer around the pyrenoids (jig. 24). Some 
stroma starch was also present. The stroma starch seems to per- 
sist longer than that formed more recently around the pyrenoids. 
Staining the pyrenoids after keeping them in the dark was rather 
difficult, requiring 10-20 minutes, where before it had not taken 
that number of seconds; the stain was not brilliant even then. It 
was noticeable too that the pyrenoids had diminished one-third to one- 
fourth in size, showing that some of the pyrenoid substance had been 
used up. 
The nucleus 
Only a brief description of the nucleus will be given here, my 
principal aim at this time being to correct the erroneous view as t0 
the shape of the chromatophore of Closterium, and to give further 
details as to the structure of the pyrenoids. The nucleus has in lon- 
gitudinal section the form of a double convex lens (figs. I, 29). Its 
body in both species is made up of a finely reticulate network of 
rather lightly staining fibers, forming meshes of almost uniform size. 
There seem to be no differentially stained chromatin granules on the 
strands. There are darker masses in the fibers which may be the 
net knots of the older authors; the extreme fineness of the network 
makes this point especially difficult to determine. The most striking 
part of the nucleus is the great mass of granules which lie in a group 
in the center of this reticulum and which stain a brilliant red in the 
triple stain. These bodies are globular or sometimes somewhat ang” 
lar granules of varying size, apparently unconnected and lying free 
