96 
W. B. Crow 
The variation of the pigmentation of the Chroococcacece, together 
with the saprophytic tendencies of the group, would lead one to 
expect colourless representatives to be not uncommon. These are 
probably found in such bacterial genera as Pediococcns and Sarcina. 
It is of interest to note the extensive lipochrome production in certain 
species of the latter. 
Planes of Cell-Division 
Amongst the characters first adopted by the older algologists in 
distinguishing the various forms of unicellular blue-green Algae the 
spatial relation of the successive planes of division to one another 
played the most important part. This was perhaps due to the very 
obvious differences which may arise in colonial forms, as a result of 
differences in relation of the division planes, so that the distinction, 
for instance, between a Merismopedia and an Aphanocapsa is in this 
respect a very striking one. It is noteworthy that the various systems 
of classification of the bacteria have also been based largely on the 
character of the division planes, but these systems have met with 
adverse criticism since certain filamentous forms show several pos¬ 
sible types at different stages. The same objection might be raised in 
the case of the Chroococcacese. For at first sight it might seem that 
examples are not wanting in which cell-division in definite planes at 
right angles may give place to division in almost any plane. The fact 
that Gloeocapsa can pass over into an Aphanocapsa condition has 
already been mentioned. In all such cases, however, a critical 
examination of the actual phenomenon does not indicate that there 
has been any change in the actual relation of the division planes to 
one another ( i.e . in the polarity of the cell). Aphanocapsa-iorms are 
all characterised by the diffluent character of the mucilage. Within 
such mucilage rotation of the cells would become much easier than in 
the more compact types ( Gloeocapsa , Merismopedia). It is not easy to 
prove that such rotation occurs in Aphanocapsa because the cells are 
isodiametric, but in the similar diffluent Aphanothece- types with their 
elongated cells it is quite easy to observe. In Fig. i the shifting of 
two cells with reference to one another is shown diagrammatically. 
In all but one of the families of the filamentous Cyanophyceae 
there is absolute constancy of cell-division in the plane at right angles 
to the axis of the filament. Even when branching takes place this 
plane is not departed from, the well-known “ false-branches ” thus 
being formed. In the branching of the Stigonemacese, however, we 
get division planes at right angles to the normal direction. It is as 
