86 
W. B. Crow 
regions the activities of the cell are largely directed towards the 
formation of this internal membrane and less active mucilage secre¬ 
tion would occur externally in these zones. Similar constrictions 
occur between the coenocytes of many Siphonocladiales. 
Returning to the modifications of cell-form in the Chroococcaceae 
themselves, it may be noted that the elongated type may sometimes 
be best explained as due to mutual pressure of the cells of a colony. 
Thus an examination of the tabular colonies of Merismopedia (e.g. 
M. elegans A. Br.) from their edge, shows that the cells are elongated 
at right angles to the surface of the whole group. Cell-division does 
not occur in this direction and it is easy to relate the elongated form 
to pressure rather than to polarised growth. Cell-elongation of the 
kind indicated here reaches its maximum development in the genus 
Holopedium. I 
The pear-shaped or heart-shaped cells of Gomphosphceria repre¬ 
sent another distinctive type. In these cells we can speak of a dis¬ 
tinct base and apex and hence the genus must be relegated to the 
Chamsesiphonaceae. The fact that the stalks at the base of the cells 
are joined to one another instead of to a substratum, thus giving the 
plant a free floating habit, has obscured the true affinities of the 
genus. The common occurrence of “gonidia” in Gomphosphceria 
verifies this view and thus places it outside the scope of the present 
investigation. 
Cell-Size 
When material of a single species from a single habitat is dealt 
with, no great variability in the character of cell-form is encountered. 
The character of cell-size, however, appears to be somewhat more 
plastic. Table I gives some data obtained by actual measurement of 
collections of various species for the character in question. It was 
not thought necessary to take more than one hundred individuals, 
since the number of classes into which the variates could be arranged 
was limited by the means of measurement. Table II gives the re¬ 
corded diameter of the protoplast for a number of well-defined 
species and such as are easily recognisable apart from the size of cell. 
These tables show that there is an appreciable variability. In the 
case of some of the species of Glceocapsa and of Chroococcus in Table I 
such variability may indicate the existence of several distinct races 
morphologically indistinguishable. But variation is also appreciable 
in the cell-size of Ccelosphcerium , Microcystis and Merismopedia in 
the table, where the protoplasts measured in each collection were 
members of one and the same colony, so that here we are dealing with 
