THE PYRENOID OF ANTHOCEROS 85 
practically all of the cells have these densely aggregated pyrenoids. 
In such cases the plastids are very thin and the pyrenoid bodies 
closely aggregated (figure 3). 
The pyrenoid bodies in these dense aggregates are usually less 
angular than those in the looser, active pyrenoids. The bodies, 
though still of the shape of a flattened disc, have rounded edges (figures 
I and 19). Those in the active pyrenoids are loosely aggregated and 
are flatter, with sharp, angular edges. So loosely are they aggregated 
in some plastids that they can only be distinguished from the inner 
starch grains by their staining reaction and possibly by their smaller 
size (figure 8). 
This difference in the shapes of the active and quiescent pyrenoid 
bodies shows very clearly that the shape of the bodies is not due to 
the mutual pressure of plastic structures in the compact aggregates, 
for if this were the case we should of course find the greatest angularity 
in the compact masses. 
Due to their minuteness I have been unable to find conclusive 
evidence of the formation of new pyrenoid bodies by the fragmenta- 
tion or fission of preexisting bodies. They however often lie very 
closely together and overlap in such a manner as to suggest very 
strongly such an origin (figure 7). Indeed, the form and the arrange- 
ment of the pyrenoid bodies and the starch grains as a whole suggests 
very strikingly an origin by fission. Their form is very commonly that 
of an elongated disc or spindle, convex on the outside and concave on 
the inside. The general form of the mature starch grains as well as that 
of the loosely aggregated pyrenoid bodies is the same as that of com- 
pact masses frequently seen among active plastids (figure 7). These 
compact pyrenoids among looser, active ones have apparently been 
slower in arriving at the proper conditions to deposit starch. Their 
pyrenoid bodies are brilliant red and very similar to those in active 
pyrenoids with abundant surrounding starch. It is difficult to conceive 
that bodies overlapping one another in such definite order as do the 
pyrenoid bodies and starch grains could have arisen de novo in the 
cytoplasm of the plastid. Nevertheless the fact of their being invisible 
in the very young assimilative cells in the area of intercalary growth of 
the sporophyte would seem to point to the conclusion that there, at 
least, the pyrenoid bodies are lacking entirely and later arise de novo. 
The direct transformation of pyrenoid bodies into starch grains is 
easily followed. In many plastids there is a gradual transition of the 
