194 Atkinson. — A Monograph of the 
It is in this way, by the development of the prostrate form 
from the rhizoids of the Chantransia- form, and new elements 
of the Chantransia- form in succession from this, with the pro- 
cess repeated successively for several weeks, that the extensive 
mats are produced covering the rocks where the social 
Lemanea-^\dsv\.^ are found. 
B. Chantransia-form. The origin of the Chantransia-i orm 
has already been noted ; its growth and special characteristics 
remain to be noticed. In the subgenus Sacheria it is much 
smaller than in the subgenus Lemanea , in length ranging 
about 2 mm. or less, while the diameter of the cells is less, 
ranging from 15 j x to 30 jx. The cells are of nearly the 
same diameter throughout the length of the filament. In the 
growing cells the endochrome is first flocculent, then granular, 
and finally becomes arranged in plates along the inner 
periphery of the cells. In many cases the plates of endo- 
chrome are in close contact and some of them fused. Fig. 8 
represents the arrangement of the endochrome. The endo- 
chromic-plates break down very easily, especially in the sub- 
genus Sacheria , and it is necessary to have fresh material, and 
examine it soon after collection, in pure water. 
In Lemanea australis the Chantransia- form at its origin is 
about 30 jut in diameter. As it increases in length it also 
increases in diameter. See Fig. 21. At a distance of four to 
five cells from the point of origin it is nearly of the normal 
diameter. The diameter of well-developed filaments ranges 
from 50 fx to 120 fx. The terminal, growing, cell is elongate 
oval, gradually becoming cylindrical or oblong, or in some 
cases broadly oval. The endochromic contents of the grow- 
ing cell are very flocculent, more dense at the distal end and 
sides, usually with an irregular hyaline space near the centre. 
It is of a bluish-green color. As the cell becomes older the 
of the tufts. It is quite likely that some of the rhizoids from lower primary groups 
of cells, where the sexual shoot is sterile and no cortex is developed, take part in 
multiplying the protonema, for they arise from cells of the sexual shoot which 
correspond to the generative filaments ; but that rhizoids develop from cortical 
cells and then produce protonema is highly improbable. 
