622 
be interrupted. I have also met with smaller groups of inflated cells consisting of 4 
or 5 cells, resembling those figured by Batrers but containing flat chromatophores 
lining the outer walls and one to three dense bodies, perhaps nuclei (fig. D). These 
cell-groups seemed not fully normal. — Fig. H shows a row of inflated cells the 
content of which is contracted and more or less globular, undoubtedly monospores; 
the callus buttons of the separating walls are very distinct. 
Fig. 618, A—H show details from Conchocelis growing in a tube of Pomatocerus 
triqueter gathered in the same locality near Frederikshavn in July, while figs. J—L 
are from similar specimens gathered in a neighbouring locality three days later. 
The tube of Pomatocerus was coloured finely rose for a long way by the perforating 
Alga. The youngest part of the red shell contained principally very long straight filaments. 
A group of fertile cells in connection with thin filaments is shown in fig. K. Numerous 
large complexes of fertile cells were met with in the older part of the shell (fig. G). 
The cells of these cell-rows were cylindrical, scarcely swollen, usually a little longer 
than broad, and the diameter smaller than in the specimens from Spirorbis, viz. 
9—11 u. The structure of the fertile cells was not obvious, although the specimens had 
been treated with picric acetic acid or with Nawashin’s mixture. An intensely stained 
nucleus could, however, be observed, but the shape and the number of the chromato- 
phores were indistinct and a pit in the transverse walls was not usually discernible. 
In some cases dense bodies, intensely stained with hematoxylin, were observed in 
the chromatophores (fig. H); as these bodies were in some cases angular, they were 
perhaps crystalloids. In a transverse section of a shell treated with Nawashin’s 
mixture and stained with hzematoxylin after Heidenhain the fertile cell-rows were 
seen growing inwards in the shell from the vegetative layer. With high power of 
enlargement a central callus button proving the presence of a pit-connection was 
ascertained (fig. L). A number of strings recalling the arms of a stellate chromato- 
phore were seen radiating from the central intensely stained nucleus; they seemed 
to be strings of protoplasm, whereas the presumed chromatophores were comparatively 
small and feebly stained. 
The specimens drawn in fig. 619 grew in Spirorbis but agreed as to the fertile 
cell-rows with those just described from Pomatocerus. Fig. C shows a two-celled 
group of fertile cells given off from a spindle-shaped cell. The nuclei were intensely 
stained, whereas the chromatophores were very indistinct. The fertile cells were 
cylindric, which is also apparent in fig. E, where four globular monospores are to 
be seen. 
The investigations here communicated have shown that there is a distinct 
difference between 1) a vegetative stage consisting of branched filaments composed 
of long, thin, cylindric cells and inflated, more or less spindle-shaped cells, both 
with narrow transverse walls, and 2) fertile branched cell-rows consisting of uni- 
form, broad cells separated by broad cross-walls. The presence of a central pit in 
the cross-walls has been ascertained, and it has been shown that the cells contain 
a nucleus and probably always more than one chromatophore. The formation of a 
