224 Carter . — Studies on the Chloroplasts of Desmids. /. 
In the case of species having axile chloroplasts, the pyrenoids are 
usually found in those parts of the cell where the chloroplast is most 
massive. This is frequently the axis of the chloroplast itself, but in some 
cases, e.g. the larger species of Euastrum , the actual axis of the chloro- 
plast is very slender, and the pyrenoids are practically confined to the 
more massive peripheral parts of the chloroplast (Fig. 73). The number of 
pyrenoids embedded in the axis of axile chloroplasts depends on its size. 
Thus where there is a long narrow axis, e.g. Closterium , the pyrenoids 
occur in a line, and may be quite numerous (Figs. 6, 10, 23, 35, 38, and 47). 
Where the axis is larger and broader, e.g. some species of Micrasterias , 
the number of pyrenoids present may be more than 100, and they are 
scattered throughout. In many species where the axis is much more 
limited in its extent, there is typically only one pyrenoid in the axis of each 
chloroplast. 
The general condition of the cell, however, probably greatly influences 
the amount of food-reserves stored in the form of pyrenoids, and at any 
time two or more pyrenoids may occur in such positions wfiere only one 
would have been expected. This is true both in the case of axile and 
parietal chloroplasts, which were formerly supposed to have only one 
pyrenoid. 
When division of a pyrenoid occurs, usually by budding, or the gradual 
constriction of one to form two, the newly formed pyrenoids separate from 
each other almost immediately, where this is possible, e.g. the extensive 
chloroplast-plates of Xanthidium armatum , (Breb.) Rabenh. (Figs. 129 and 
130), or Micrasterias denticidata , Breb. Where the space is more limited, 
however, the migration of the pyrenoids is impossible. For example, some 
species of Cosmarium have two axile chloroplasts in each semi-cell, each 
chloroplast containing originally a single pyrenoid. If either of these 
pyrenoids divides, the products of its division could not possibly separate, 
because the chloroplast only forms a minute film round it, and the two 
pyrenoids remain in the position of the original one. Further division may 
result in the formation of a group of three or even more pyrenoids in the 
axis of one or more of the chloroplasts in some individuals. In the living 
condition such a group of pyrenoids would be apparent simply as one 
refractive mass, and it would be almost impossible to distinguish it from 
a single pyrenoid. This is probably why the number of pyrenoids has 
been thought to be definite in so many Desmids. The examination of 
stained material, however, has shown that this is not so, although the 
relative position of the pyrenoid groups is really quite constant in such 
cases. 
Lutkemuller ( 1893 ) noted such variations in the number of pyrenoids 
in several species of Cosmarium , and also in species of Arthr ode smus, 
Staurastrum , and Euastrum , whilst W. and G. S. West ( 1898 ) have also 
