90 F. McAllister 
body of Anthoceros. Based upon these comparisons the pyrenoid 
of Anthoceros would in reaUty be a compact group of minute pyrenoids, 
that is to say a "multiple pyrenoid." " 
The pyrenoid of Anthoceros, then, though differing strikingly in 
structure from those of the green algae, has nevertheless much in 
common with them. The published work indicates that the structures 
which have previously been termed "pyrenoids," though apparently 
all kernel-like protein bodies, are nevertheless not all alike in their 
relation to starch formation. While those present in the chloroplasts 
of the diatoms, in certain of the Rhodophyceae, and the Euglenidae 
'may be concerned in the formation of other carbohydrates, true 
starch has not been demonstrated in any members of these groups^ 
According to Schmitz (19) and Schimper (18) the Floridian starch 
appearing in many of these forms seems to be formed with no apparent 
connection with the pyrenoids or even with the plastids. In the 
Chlorophyceae the work of Timberlake has shown that the pyrenoid 
takes a direct morphological part in the process of starch formation, 
at least in the forms studied by him. Segments of the pyrenoid are 
split off to form the rudimentary starch grains (20) or the entire 
pyrenoid, as cited above, may form a single large starch mass (21). 
And while Lutman (11) was unable to arrive at a definite conclusion 
as to the mode of starch formation in Closterium, his work at least 
ofifers no support to the earlier conclusions of Schmitz (19) and of 
Schimper (18) that the pyrenoid although closely related to starch 
formation, does not itself take any direct, morphological part in the 
process. It is probable that more weight should be attached to the 
more recent work because of the great advances in microscopic tech- 
nique in the last thirty years. 
There are no visible structures in the chloroplasts of the other 
Bryophyta that are analogous to the pyrenoid bodies of Anthoceros. 
As described above their chloroplasts are considerably smaller and do 
not show in any case any visible products of photosynthesis previous 
to the appearance of the starch grains. No bodies are visible which 
serve as the beginnings of the starch grains, either scattered or aggre- 
gated in a mass. The smallest stainable bodies of the plastid stain 
the same color as the large starch grains and are therefore quite 
probably starch. As in Anthoceros the size of the chloroplasts remains 
apparently the same whether they are free from starch or partly 
filled with it. 
