A Critical Study of certain Unicellular Cyanophyceae 99 
specialised character of the envelope, the pectic nature of the latter, 
the variability in the differentiation of the layers of the protoplasm, 
including the plasma-membrane, so that sometimes the cell may be 
homogeneous. The lack of a definite chloroplast and a nucleus is 
also of importance in connection with the primitive nature of the 
group. Incidentally, it seems altogether gratuitous to compare the 
central region with a nucleus, as suggested by many writers, especi¬ 
ally as the existence of chromosomes and even of chromatin is more 
than doubtful. It might better be compared with the endoplasm of 
many Flagellates and Rhizopods. In connection with the lack of 
nuclear structure it must be pointed out that this by no means pre¬ 
cludes a precise mechanism of heredity; even in the higher plants 
non-nuclear inheritance is held to be possible (e.g. plastid inheritance). 
The complete absence of flagella in the Cyanophyceae is a remark¬ 
able feature, which may or may not be a secondary character. Whilst 
it is true that the majority of Algae and Protozoa have been shown 
by recent work to be connected by all transitions with Flagellate 
forms, yet most of these organisms show a flagellate stage at some 
period in their life-history. In many ways the Chroococcaceae are 
as primitive or more so than the majority of Flagellata, although their 
structure is of quite a different type. In their mode of reproduction 
by binary fission rather than by multiple division the Chroococcaceae 
are less primitive than the allied Chamaesiphonaceae, although the 
latter are more advanced with respect to the thallus, in which apical 
and basal portions are differentiated. 
The general principle, that multiplicity of similar parts or struc¬ 
tures is more primitive than the condition in which such similar 
parts are few, may be applied to the axes of cell-division. The lowest 
members of the Chroococcaceae, then, are those showing division in 
three planes at right angles to one another ( Chroococcus , Glceocapsa). 
Reduction of the axes to two ( Merismopedia , Holopedium) or one 
(,Synechococcus ) may take place. Following the polarisation of growth 
especially developed in connection with the restriction of the division 
axes, the cell may become modified in form (Aphanothece, Tetrapedia). 
Further modification may take place by early constriction of the cell 
in the planes of division (Tetrapedia), this being paralleled in the cell- 
form of the majority of Desmids. Filamentous forms are made 
possible by the restriction of the planes of division; the method of 
origin of the Oscillatorian filament has been indicated. Amorphous 
types of colony may arise in the Chroococcaceae from definite types 
by the mucilage becoming diffluent (. Aphanocapsa , Aphanothece). 
7—2 
