232 Carter . — Studies on the Chloroplasts of Desmids. I. 
ridges round the central axis in these complicated chloroplasts is continually 
changing, the position of any particular ridge in relation to the other vary- 
ing from time to time, and it is this which probably accounts for the 
repeated bringing together and separation of neighbouring ridges which 
causes the reticulate appearance seen in the external view. 
In species like Cl. costatum , Cl. striolatum , Cl. regulare , and Cl. 
angustatum , therefore, transverse sections of different individuals may 
present striking differences of structure, according to the relative size of the 
ridges and axis. If the ridges of the individual in the external view are 
nearly parallel, then in section the axis will probably be very broad, 
occupying most of the interior of the cell, whilst the ridges will be compara- 
tively low, but quite distinct from each other. If, on the other hand, the 
ridges are anastomosing to any great extent, in such cases the axis in 
transverse section will be much smaller, the ridges will be more extensive 
and also more or less branched, the chloroplast as a whole being much more 
irregular in outline. 
As regards the finer structure of the chloroplast, it was noticed that in 
sections of Closteriinn showing branched ridges the protoplasmic network of 
the chloroplast is much closer and more compact than that of sections 
showing a broad axis and low ridges. This is particularly noticeable in 
those species in which there is the possibility of very great variation in the 
relative size of the axis and ridges, those specimens with large or branched 
ridges showing an extremely fine reticulation of the chloroplast, and more- 
over the ridges in such cases are usually very thin, but the protoplasmic 
network of the whole chloroplast is very dense and very sharply defined 
from the rest of the protoplast (Figs. 26-9, 32, and 41-2). In specimens of 
these species which have low ridges and a very massive central axis, the 
chloroplast is seen in transverse section to have a very coarsely reticulate 
appearance, which on careful examination is found to be due to the presence 
of numerous granules which do not take the stain (Figs. 30 and 34). Test- 
ing with iodine shows that the chloroplast is full of tiny starch-grains 
which are packed within the meshes of the reticulum of the chloroplast, 
either in the axis only, or in the ridges as well. 
If sections of individuals having branched ridges be stained with iodine, 
it is seen that no starch is present, with the exception of that forming the 
starch-sheaths of the pyrenoids. This tends to prove that in these species 
the size of the axis depends largely on the condition of the cell. If little 
starch is present, then the axis will probably be slender, but if photosyn- 
thesis has been going on very rapidly, starch-grains begin to accumulate in 
such large numbers that they stretch apart the meshes of the reticulum and 
cause the axis to become very swollen. Later, the ridges also may be 
similarly distended. 
Apparently the presence of large quantities of starch in the low-ridged 
