300 
protoplasm filling out the cell and not covered by chromatophores. In the other 
cells the nuclei are smaller and less conspicuous in a living state. In the thinner 
branches which sometimes occur in f. Turneri the number of nuclei may decrease 
to 3 or 4 (fig. 203) and the same may be the case with the sterile cells supporting 
the reproductive organs. The branches arise as small lenticular cells cut off by a 
concave wall at the upper end of the cell and containing from 
the first a number of nuclei (fig. 202). 
The hapters are numerous; they are frequently produced 
from a number of consecutive cells of the creeping filaments 
(comp. Dirrwyn |. c., LyxGByE Plate 40 C). They spring normally 
from the basiscopic end of these filaments and are as a rule 
unicellular. When meeting the substratum the cell is expanded 
in an adhesive disc composed of densely joined dichotomous ra- 
mifications of the cell (fig. 204). The hapters may exceptionally 
be two-celled (fig. 204), and it may also occur that a hapter is 
produced at the end of a shorter or longer filament (fig. 205). 
According to PRINGSHEIM (1862, p. 17, Taf. IV fig. 1 and Taf. VI 
figs. 1—2) the erect filaments may terminate in a feebly coloured 
hair and the same may be the case with the involucral branches 
of the cystocarps. The occurrence of these hairs is however, ac- 
cording to my ohservations, not normal but a comparatively rare 
phenomenon which I have observed only in a few -specimens. 
They arise from terminal cells of shorter or longer filaments, these 
cells becoming much longer, upwards a little thinner. The nuclei 
persist and become distincter while the chromatophores vanish 
and appear only as small feebly coloured grains (Comp. KoLDE- 
RUP ROSENVINGE, Hyaline hairs. Biol. Arb. tilegn. E. WARMING. 
1911, p. 210). Transitional forms between hairs and hapters may 
occur (comp. PRINGSHEIM and K. ROSENVINGE Il. cc.). 
The tetrasporangia are in the simplest case solitary and borne 
on a stalk-cell. This occurs particularly in the f. roseola; but a 
Fig. 206. second sporangium is here frequently present, terminal on a late-. 
Spermothannion repens ra] stalk-cell given off under the first. In the specimens from 
nuclei and numerous the North Sea, the Skagerak and the Northern Kattegat, which 
mal Cr PS are in great part referable to f. Turneri, the sporangia are 
placed in cymoid clusters, the ramification continuing in various 
degree. These clusters may be opposite or secund. In the f. Turneri verticillate 
clusters, in ternate whorls, may occur beside the opposite ones. The sporangia con- 
tain from the first one nucleus only, while the supporting cell contains several 
nuclei (fig. 207 B). This comes in existence, as far as I have observed, in the way 
that the sporangium only receives one nucleus by the division by which it is se- 
parated from the stalk-cell. Any disorganization of supernumerary nuclei as in 
