440 MR. G@. BREBNER ON THE ORIGIN OF THE 
give rise to the loose internal tissue of the filiform thallus. The 
peripheral rows of discoid cells, and such portions of the more 
central ones as reach the surface, divide up, in the first instance 
as seen in fig. 4 C, and by subsequent repeated (sub-dichotomous?) 
branching give rise to the well-known candelabra (Armleuchter) 
arrangement of the periphery of the mature filiform thallus, as 
seen in transverse or longitudinal section. There is some reason 
for believing, from certain appearances in the growth of fig. 5, that 
the apical cells which carry on the growth of the annual thallus 
and its branches are simply developed from specially favoured 
apical cells of the specialized filaments. This club-shaped struc- 
ture acquires additional interest from v. Zeller’s remarks on the 
exotic Dumontia fastigiata *. This alga it seems is found in three 
different forms which are so distinct that, if their specific identity 
had not been proved, they would not be recognizable as the same 
plant. At the meeting reported, v. Zeller exhibited three speci- 
mens: the 1st was a simple club-shaped utricle, indented at the 
top; the 2nd was a dichotomously-branched, hollow, slippery, 
friable mass, adhering firmly to paper on drying; the 3rd was 
firm, cylindrical, and copiously branched. It is the first of these 
which is the most interesting from the point of view of the 
present investigation, and is probably comparable to the club- 
shaped mass of fig. 5, only much longer persistent. 
That the stages described above are truly young conditions of 
the D. filiformis of algologists is shown by the low-power fig. 6 
and by fig.7. The former shows an easily recognizable Dumontia 
annual thallus, a, growing from the perennial creeping thallus, 
er.th. The latter illustrates a more highly magnified base of a 
still older filiform thallus showing the relation to the attaching 
“dise.” The medullary portion of the former is seen to be 
composed of greatly elongated articulations which have arisen 
from the more central filaments of the original group. Likewise, 
as far as the magnification would allow, the branching of the 
peripheral filaments is shown, as also of more internal ones which, 
by their more rapid growth, have become superficial during part 
of their course. 
It may be worth while to note here that the origin of the 
upright thallus from its attaching base is totally at variance with 
* “Ueber vielgestaltige Algen.” Wiirttemb. naturwissen. Jahr esberichte, 
1876, p. 90. 
