SMITH: TETRADESMUS, A NEW COENOBIC ALGA 1g 
throughout the cytoplasm and is not in the form of a definite 
chromatophore. Every cell possesses a single, eccentrically placed 
pyrenoid. In preparations stained with the safranin-gentian 
violet combination, the pyrenoid takes on a brilliant red coloration 
and the surrounding layer of starch is colored blue. The pyrenoid 
appears to be an entirely homogeneous body and not composed 
of lighter and darker regions, as Lutman (10) has shown to be the 
case in Closterium. On the other hand, the minuteness of the 
pyrenoid may possibly account for the failure to observe these 
details in its structure. A method devised by E. M. Gilbert, but 
as yet unpublished, in which a differentiation of the structure of 
the pyrenoid of Sphaeroplea was obtained by the use of safranin 
and Lichtgriin, was also tried, but no differentiation in the pyrenoid 
of Tetradesmus was observed. Surrounding the pyrenoid is a 
hyaline region which extends from the surface of the pyrenoid 
to the starch grains. This possibly may be a region that is com- 
posed of intermediate compounds in the formation of the starch 
grains, but this region failed to take any of the different stains 
used, although Timberlake (15) found that at times a part of it 
(in Hydrodictyon) stained with orange G. The starch granules 
surrounding the pyrenoid form a layer which is circular in outline 
and not angular as in Hydrodictyon (Timberlake) and Closterium 
(Lutman). Fic. 6 shows that there is some variation in the width 
of the starch layer and also that the outline of the starch ring may 
approach the ovoid but never the angular. The drawings also 
bring out the fact that the pyrenoid is not always in the center of 
the starch ring but may be towards one side. The starch layer is 
not a single unbroken ring but is composed of curved plates. 
Fic. 8 shows this segmentation of the starch layer. The deter- 
mination of this point is an exceedingly difficult matter, although 
it may be noted again that the individual starch plates in Scenedes- 
mus are easily observed. In the case of some of the older cells 
starch granules may be also found in the cytoplasm some distance 
from the pyrenoid (FIG. 7). These granules have been given the 
name of “stroma” starch by Klebs (8), but Timberlake (15) has 
shown that they are of pyrenoidal origin. The existence of the 
‘‘stroma”’ starch may be taken as evidence of a somewhat patho- 
logical condition, since it is never found (at least in 7 etradesmus) 
