1184 Wisconsin Academy of Sciences , Arts and Letters. 
ment of a larger colony. Ordinarily the cells of four-celled 
colonies, or fragments of larger colonies, are in a plane but the 
long axes of the cells have no definite relationships (Figs. 20, 
23, 25, 26, & 31). On the other hand, the cells of coenobia of 
more than four cells grown on agar slants are usually not in 
one plane but in an irregular mass 
When the culture medium is a liquid, the cells are more 
likely to be regularly arranged. In four-celled coenobia hav¬ 
ing the regular nonparallel concurrent coplanar arrangement,, 
there are the same differences as to the contact of the cells at 
the center of the colony that were described for Pedistrum 
tetras (Figs. 33-35). All gradations between the condition 
shown in Figure 33 and that in Figure 35 can be found. At 
other times the arrangement shown in Figure 33 is so exagger¬ 
ated that the nonpairallel concurrent coplanar arrangement is 
entirely lost, and the cells are arranged in an alternating man¬ 
ner that approaches the parallel nonconcurrent coplanar struc¬ 
ture found in Scenedesmus (Fig. 19). In comparing Figures 
19 and 33 it should be borne in mind that this classification is 
based on the position of the long axes of the cells, and while 
there seems to be no great difference between these two colonies 
on a casual examination, the analysis of the relationships of 
the long axes show3 there is a marked difference. In still 
other cases, the cells are not at all in the same plane but one or 
two of them may be in a different plane from that of the other 
three or two. The cells that are not in the same plane may 
either be parallel to> the plane of the coenobe (Figs. 26 & 38), 
or in a plane at an angle with the plane of the coenobe (Figs. 
18 & 37). Occasionally the cells of the coenobe have a regular- 
concurrent non coplanar arrangement (Fig. 36) so that the 
colony might easily be taken for one of Sorastrum rather than 
Pediastrvm, did we not. know the history of the particular 
colony under observation. Sometimes the cells form a linear 
series, with the cells all in one plane (Fig. 25), or in different 
planes (Fig. 22). 
Harper (21) has given an explanation for this variation in 
