4 Prof. F. Schmitz on the 
up the individual filament cells, rather cut portions of the 
margins off*, which then become developed into longer or 
shorter lateral branches. 
These marginal cells are formed on the individual joint-cells, 
sometimes singly, sometimes in plurality, sometimes simulta- 
neously, sometimes one after the other, and, according to their 
number and their earlier or later development, they produce a 
very heterogeneous habit of the cell-division and of the ramifica- 
tion of the individual filaments. A very widely spread mode 
is that at the upper end of a newly produced joint-cell a 
branch-cell is at once separated off which extends itself by the 
side of the terminal cell of the filament almost as rapidly and 
strongly as the latter itself, and thus readily produces the 
appearance of a regular dichotomous branching; in what 
follows this mode of ramification is characterized as subdicho- 
tomous as by Bornet. 
At each division of a Floridean cell a peculiar pit is formed 
in the organic central point of the dissepiment formed ; this 
maintains the two sister cells in communication so long as 
they remain alivef. 
overlooked. In literature therefore there are many contradictory state- 
ments (by Nageli, Kny, Reinke, &c.). In all cases, however, that I have 
hitherto been able to investigate I have been unable to confirm these 
statements ; and now, after very numerous investigations, I believe I am 
justified in establishing the above propositions as generally applicable to 
the Florideee. Should exceptions really occur here and there, they cer- 
tainly take place extremely seldom. 
* Upon the circumstance that in the Floridez a joint-cell is never 
divided by a transverse wall, or by an (organically) median longitudinal 
wall, but that rather lateral marginal cells only are cut off from such 
joint-cells, depends also the fact that the cellular tissue of the Floridean 
thallus is always to be referred back to a system of branched cellular fila- 
ments, even when the thallus is solid and the individual cells are held 
together without gaps. Every cell-body constructed under this condi- 
tion must show the same behaviour; and it is due only to the cessation 
of this condition that the roots, stems, leaves, &c. of the Archegoniata 
and Phanerogamia also cannot be referred back to a system of simple 
branched cell-filaments. 
The Ascomycetes, however, show an exactly similar behaviour to the 
Floridese ; in the great majority of them a transverse division of the indi- 
vidual joint-cells of vegetative cell-filaments takes place but rarely. In 
consequence of this the construction of the thallus out of branched cell- 
filaments usually appears just as distinctly in these Ascomycetes also as in 
the Floridez ; and the occasional occurrence of parenchymatous cell-bodies 
(e. g. in the foundation of the perithecia of Pleospora herbarum (cf. Bauke 
in feet ee maa NG pp. 315 et segg.]) is probably due, 
ust as in the Floridez, only to a close firm conglobatio 7 - 
Tyne branched eelleainiienia Be Ts ee 
+ Such a pit consequently occurs both in the lower and in the upper 
septum of each joint-cell. Between the two pits of these two transverse 
