462 Smith.—Cytologic a l Studies in the Protococcales . Ill . 
primary cleavage. Fig. 24 shows this nuclear division after the formation 
of the first cleavage plane. This might be interpreted as a beginning of 
cleavage before the completion of the usual number of nuclear divisions 
were it not for the fact that fully 85 per cent, of the cells showing the 
primary cleavage furrow contained four nuclei, while about the same per¬ 
centage of cells in the cultures were producing eight autospores. In 
Scenedesmus ( 4 ) the single nucleus divides once just before cleavage, and 
there is a division of the nuclei before each successive series of cytoplasmic 
cleavages. Pediastrum (6), on the other hand, contains a relatively small 
number of nuclei that divide once or twice just before cleavage, but do not 
continue dividing after cleavage has started. Tetraedron, therefore, com¬ 
bines these two methods, since the cell is multinucleate before cleavage, but 
the nuclei divide after the first cytoplasmic cleavage. 
Two secondary cleavage furrows are now formed at right angles to the 
primary one. These start at the primary cleavage furrow and are apparently 
formed by a furrowing in of the plasma membrane (Fig. 26). These furrows 
run between the nuclei of each daughter-cell, but are not opposite one 
another in every instance (Fig. 25). Their formation is generally simul¬ 
taneous, but not always (Figs. 22, 23). The nuclear division which takes 
place after the completion of the first cleavage may be delayed until these 
second cleavage furrows have been completed or are well under way. 
Fig. 26 shows the completion of the second cleavage with the nuclei still in 
the resting condition, while the nuclei of Fig. 23 have only reached the 
prophase of division. Since there are eight nuclei at some stage after the 
completion of the second cleavage (Fig. 27), one further series of cleavages 
is necessary to make the uninucleate protoplasts. These last cleavage 
furrows are formed at an angle of forty-five degrees to the plane of the 
second series (Fig. 29). 
After the completion of the first cleavage there is a disappearance of 
the pyrenoid. Instances were never found where it disappeared before the 
first cleavage as in Pediastrum. The pyrenoid may disappear before the 
second cleavage (Fig. 22), but as a rule it remains until the completion of 
the cleavage of the mother-cell into four parts (Fig. 24). Apparently there 
is a gradual dissolution of this body, its contour changing from a spherical 
to a lens shape as the process progresses (Fig. 25). 
Eight angular protoplasts are found within the old mother-cell wall 
when cleavage is complete. These are the young autospores. Their shape 
now changes from the angular to the pillow shape that characterizes the 
mature cell of T. minimum. When a cellulose wall has been secreted the 
autospores are liberated by a gelatinization or a rupture of the old mother¬ 
cell wall. Shortly before this liberation a small pyrenoid is formed de novo, 
the new pyrenoid lying close to the nucleus (Fig. 36). 
In Pediastrum (6), an alga where the presence of zoospores has been 
