92 
F. McAllister 
This orange stained zone he beUeves to be less complex chemically 
than the starch. He shows that this orange stain is characteristic of 
immature cell plates and of cell walls which are being dissolved. It is 
probable that at times the pyrenoid bodies of Anthoceros, as well as 
parts of the pyrenoid of Hydrodictyon and probably other algae, may 
be regarded as intermediate products of starch formation. This inter- 
mediate product reacts microchemically as protein and therefore prob- 
ably has a complexity considerably greater than that of either the 
soluble or the insoluble carbohydrates. If it can be shown that in the 
case of the leucoplast the intermediate product is really a carbohydrate 
not greatly different from starch then it would seem that starch 
formation in the presence of a pyrenoid is quite a different process in 
which more complex reactions involving a protein stage seem to be 
necessary to the production of the carbohydrate. Since starch may 
be deposited in chloroplasts as well as in leucoplasts in the absence of 
light, it is quite possible that normally it is formed from soluble 
carbohydrate without the intervention of photosynthesis. 
The phenomenon of starch formation from the pyrenoid bodies of 
Anthoceros seems to support the view held by Schimper (i8) and 
Eberdt (7) that starch is formed by a transformation of protoplasm. 
Certainly starch which is formed by the direct transformation of a 
protein body can hardly be said to be secreted from the protoplasm, 
as is held to be the case by Meyer (13) and by Salter (15). 
The formation of new pyrenoids in Anthoceros is normally by a 
separation of a preexisting pyrenoid into two halves during the 
fission of the chloroplast. New pyrenoid bodies are apparently 
formed similarly, that is, by the division of other pyrenoid bodies. 
In the embryonic tissue of the sporophyte however we have what 
clearly seems to be a formation of pyrenoid bodies and pyrenoids 
de novo. Formation of pyrenoids de novo is not uncommon in the 
green algae and probably should not be unexpected here. The forma- 
tion of pyrenoids here is however by the aggregation of scattered pyren- 
noid bodies which have themselves apparently arisen de novo. The 
origin of the pyrenoid bodies from scattered parts of the chloroplast is 
suggestive and may indicate that their "Anlagen" have been present 
from the beginning of the sporophyte but scattered and not readily 
stainable. Such a hypothesis as this might be used to explain the 
formation of starch in the sporogenous cells where no visible pyrenoid 
is present. 
