1912] GRIGGS—RHODOCHYTRIUM 147 
This cytoplasmic contraction appears to be a universal occur- 
rence, having been seen in all of the numerous cysts of this age 
observed. Nevertheless, in the writer’s judgment it is not to be 
interpreted as segmentation. That appears to be a distinct 
process of a different nature. Since the preliminary contraction 
occurs during mitosis, it. gives rise not to uninucleate but to 
binucleate segments. No indication of a constriction separating 
the daughter nuclei was seen in the telophases observed (figs. 63-65). 
The steps connecting this condition with what I take to be 
true segmentation could not be made out, but it would seem 
probable that the contraction disappears after mitosis is complete 
and the protoplasm of the cyst again becomes a continuous coeno- 
cyte. It will be understood that a regressive change of this char- 
acter would be difficult to demonstrate except in living material, 
which in Rhodochytrium is too thick and too deeply pigmented to 
permit the observation of details of this sort. If the zoospores 
were always the same size, or if segmentation always occurred after 
a given number of nuclear divisions, it might be possible to recog- 
nize those cysts which had passed through their last mitosis and 
were ready for the final segmentation, but both the size of the 
zoospores and the number formed in different sporangia vary to 
such an extent as to make it impossible to distinguish those spo- 
rangia which have completed the cycle of mitosis from those which 
have not. 
But whether the cysts again become continuous coenocytes or 
not, there is another sort of cleavage, which I take to be true 
segmentation, that appears to delimit the spores without reference 
to the separations brought about during the preliminary con- 
traction. This occurs by the precipitation of membranes around 
the protoplasmic units (fig. 31). Each nucleus with its quota of 
cytoplasm is cut off from the rest by a membrane which appears 
within the strands of cytoplasm after the fashion of free cell forma- 
tion in the endosperm of a seed plant. The membranes of the 
protospores are very delicate, but the method of their formation 
seems to be clearly indicated in the preparations. If one observes 
a protospore which is not yet completely surrounded, the terminal 
portion of the advancing membrane will appear simply as a heavy 
