Nr. 1] SUBAERIAL ALGAE FROM SOUTH AFRICA dl 
up to 1,5 mm. in diameter, and are closely appressed to ihe 
substratum. At times they grow so densely that the dises to the 
naked eye melt together to a cohesive inerustation of several em.” 
extent. I have often observed that where two thalli meet in this 
way, they do not grow into each other. PI. XI fig. 304 shows that 
then they join closely at the edges, and further growth in this 
direction ceases. Under these circumstances older dises may, at 
times, get a somewhat irregular circumference, as the growth con- 
tinues only in those directions where there are open spaces for 
a continuous growth. The covering of the stems thus gets thin and 
monostromatic all over, being composed of numerous single dises. 
From the regular series of cells in each thallus, however, the lines 
between the single dises may be easily discernable (vide fig. 298, 
pl. XI). Subjected to åa microscopic examination the species is found 
to be very characteristic by its regular cells, arranged in radial 
rows. Already in young and small dises the regular arrangement 
of the cells in radial rows is to be seen. Towards the circumference 
the cell rows divide dichotomouslvy, and in this way is formed å 
cohesive disc of a single layer, without holes or rents. The edge of the 
thallus is even and entire, always destitute of lobes, and never with 
a tendency to dilapidate into single threads. The individual cells 
are — at least in somewhat older specimens — distinctly rectangu- 
lar, on an average 1%%—2 times as long as broad. The size of the 
cells is, for the rest, somewhat varying. They are comparatively 
smallest on young and small thalli, and measure in full-grown 
specimens 7—9 u. The dise-sporangia are somewhat larger and more 
roundish than the vegetative cells, from nearly globose to ovoid. 
They occur without order, singly or several together, and any 
vegetative cell seems to be able to grow out into åa sporangium. Other 
reproductive organs do not occur in my material. 
When the cell-dises crack, for instance in being torn from the 
substratum in order to be put under the microscope, they are mostly 
split up radially along the cell-series. The connection among 
the cells in the disc is evidently stronger between the tangential 
walls, that is, between the individual cells in one and the same 
row, than between the cell-rows mutually. 
In the course of time quite a number of species of Phycopeltis 
have been described, partly under other names, as Phyllactidium, 
Ghromopeltis, Hansgirgia, and others. Several of them, however. 
seem to have been described on primitive and, in systematical 
regard, dubious characters only. It is probable that with the great 
range of variation which the single species of this family empiri- 
cally evince, several of these, described as species, in the future, 
when one day more closely examined and monographically treated 
according to modern prineiples, will turn out only to be modifi- 
cations due to habitats, and that the number of true, systematically 
