Smith—The Organization of the Colony . 
1183 
have very short “horns” (Fig. 38), or wholly lacking (Fig. 32), 
may be found in a single uniaigal culture. One not familiar 
with the history of the cultures would say that Figures 34 and 
35 represent different species. Such is not the case, however, 
since both of these colonies came from the same uniaigal cul¬ 
ture. System atists have not paid enough attention to this 
variation in the length of the horn, which is usually correlated 
with variation in the shape of the cell. The length of the 
horns is usually the same for the cells of a single coenobe, but 
Figure 21 shows a colony one of whose cells has quite long 
“horns,” the other cells have short “horns” and are polyhedral. 
This variation was not found in four-celled ceonobia, but was 
found several times in coenobia containing more than four cells. 
The systematic bases for the classification of species of Pedi- 
astrum are rather indefinite, but probably the surest points for 
the classification of species are the shape of the cells and the 
size of the horns. As a result of my cultures, I am convinced 
that a good many variations that have been considered marked 
enough to warrant specific differentiation are merely normal 
variations within the limits of another species. The exact 
cause of this variation I am unable to explain. A possible 
explanation is that differences in the length of horns of differ¬ 
ent coenobia are due to somewhat the same environmental con¬ 
ditions that cause different cells in a coenobe to form daughter 
colonies with different numbers of cells. This explanation, 
however, does not account for the variation in length of horns 
in different cells of the same coenobe. This variation in the 
length of horns is another proof of the independence of the 
individual cells. 
In cultures of Pcdiastrum Boryanum grown on agar slants 
there are more irregularities than in those grown in a liquid 
medium. In the former, the cells of the coenobe are rarely 
all in one plane, but are usually in a more or less flattened mass. 
These colonies fragment easily, so that in examining them in a 
mount in water one is sometimes in doubt as to whether one is 
examining a normal four-celled coenobe or a four-celled frag- 
