I42 BOTANICAL GAZETTE [FEBRUARY 
center. In elongated grains the crosses resemble those in legumi- 
nous starch, namely, a pair of Ys arranged bottom to bottom, 
indicating an elongated hilum. And in double grains, which are 
not infrequent, the stems of the Ys are sometimes divided, so 
that the very center of the grain appears bright. Such grains are 
of course unsymmetrical, and show the characteristic crosses only 
when the planes of polarization form the proper angle with the 
axes of the grain. There are also great differences in the brilliance 
of the grains; some are very beautiful objects, but others repolarize 
the light to such a slight extent that they are very faint and the 
dark crosses are difficult to see. Frequently, indeed, the grains 
become entirely black and vanish completely when the prisms are 
crossed. When this happened, I was inclined to suspect that I 
might have mistaken grains of some other substance for starch, 
but on running iodine under the cover the characteristic blue 
reaction promptly appeared to dispel all such doubts. 
ABSENCE OF PLASTIDS.—It is unsafe to assert, perhaps, that 
there are no plastids in Rhodochytrium, but it is certain that methods 
which bring them out clearly in such objects as old potato tubers 
failed to reveal them in Rhodochytrium. So far as could be deter- 
mined, the starch grains are formed directly in the cytoplasm with- 
out the intervention of plastids, pyrenoids, or other specialized 
protoplasmic bodies. There was only one feature which could 
be taken to give any indication of such bodies. Many of the 
grains do not stain uniformly throughout, but show a more deeply 
colored margin. This appearance is not confined to grains of any 
particular size, but is found from the smallest to the largest grains. 
Indeed, when present at all the border is usually wider and more 
conspicuous in the large grains than in the small. It occurs rather 
on certain slides or perhaps on certain pieces of material, being 
present in nearly all of the cysts of some slides while absent from 
others. The border appears to have the same crystalline structure 
as the rest of the grain, and seems definitely to be a part of it rather 
than a separate surrounding body. In no case did it present the 
granular appearance to be expected’ of a plastid. I have no satis- 
factory explanation to offer for this phenomenon, but I do not 
believe it is permissible to interpret it as a plastid. 
