FILAMENTOUS THALLUS OF DUMONTIA FILIFORMIS. 439 
becomes pointed and converted into an apical cell (fig. 4, ap.c’). 
The distal non-active part of the row loses connexion with the 
parent filament after the differentiation of the discoid cells in 
the latter (fig. 1, 6). What exactly happens to these dissociated 
portions of filaments was not made clear by the preparations. 
The growing endogenous group of filaments certainly makes its 
way to the surface; but whether this is simply due to a mecha- 
nical forcing, or is effected by the aid of a digestive process, 
analogous to that by means of which lateral roots in higher plants 
penetrate the cortex, will have to be left for further investigations 
to decide. These fragments probably disappear, and are not 
simply displaced, as at a later stage the protruding group is seen 
to be cleanly and sharply marked off from the adjacent somewhat 
compressed filaments of the creeping thallus. The walls of these 
fragments evidently become very mucilaginous. This might very 
well be due to a enzyme secreted by the specialized filaments. 
That enzymes, under certain conditions, play a very important 
part in the solution of cell-walls has been demonstrated by Prof. 
Marshall Ward* in the case of fungal hyphe. Prof. Reynolds 
Green also has added considerably to our knowledge of the 
solvent action of enzymes in his work on the germination of the 
pollen-grain f. ‘ 
The central filaments of the group, whether immersed or 
superficial, first begin to elongate. Indeed the development of 
the whole group, be it endogenous or exogenous, is centrifugal. 
This can be realized to a certain extent by inspection of fig. 2, 
where the peripheral filaments are seen to be in a less advanced 
stage of septation than the more central ones. The elongation 
of the specialized filaments is due to the growth in length of the 
discoid cells, which at the same time increase very much in 
diameter. The conical appearance seen in fig. 1 a may become, 
in consequence of this, very much accentuated. A section 
showing the earliest stage of protrusion has not been figured, 
but the appearance of the annual thallus, in surface-view, soon 
after it has made its way out, is shown at a in fig. 5. Sections, 
optical or other, of this club-shaped mass show that the articula- 
tions of the central filaments have become very long, and thus 
* Marshall Ward, “A Lily Disease.” Ann. Bot. 1888, p. 337 et seq. 
+ Reynolds Green, “ Researches on the Germination of the Pollen-grain and 
the Nutrition of the Pollen-tube.” Phil. Trans. vol. 185 B, 1894. 
