420 Prof. H. Karsten on the Formation, 
& 39). It is only at the spot where it formed an annular fold, 
in consequence of its depression, that it is any longer visible, 
very strongly thickened, often im layers. 
In the course of the further development and outgrowth of 
the adjoining living cell, this membrane of the inverted cell, 
which thus serves the living cell as an external envelope com- 
parable to a cuticle, is absorbed in the same way as the inter- 
mediate substance and cortical tissue which is dissipated before 
an adventitious root. (See my account of the Organs of Vegeta- 
tion of the Palms, pl. 4. fig. 8.) 
This absorption of the inverted septum (figs. 37, 38, 39, & 
42 a, b) proceeds from below upwards (from the centre towards 
the periphery), in such a manner that the rather thick fold is, at 
a certain phase of development, the sole indication of the dis- 
vlaced septal wall; it then surrounds and somewhat constricts 
the daughter cell, in the form of an annular fibre, within the 
enveloping membrane (fig.42 a; the immediately preceding stage 
of this fold is represented at 0). 
This structural condition occurs not only in the course of 
artificial preparations, but also very frequently as a consequence 
of injuries in the course of the natural growth of the plant. 
These annular folds, when their origin is not recognized, may 
easily be mistaken for those others which have been regarded 
as incipient and suppressed septal structures. 
The remaining cylindrical dead wall of the cut joint-cell 
sometimes presents a delicate, though very definite, longitudinal 
striation (figs. 31, 33, 35). 
Not unfrequently, moreover, a cell is prolonged, in the manner 
described, through two or more adjoining cells, on account of — 
circumstances operating prejudicially on their vitality ; and we 
then find on the surface of these long joint-cells several annular 
folds situated at distances determined by the length of the 
original cells. . 
It is the general, but not universal rule, that the development 
of isolated joint-cells proceeds in the manner illustrated in figs. 
33, 34, & 385, the adjoining dead cell-wall being regularly and 
equally invaded by the growing cell, and apparently expanded 
passively. At times one side or one portion of the septum ap- 
a 
pears to offer greater resistance than the rest, or else the vegeta- _ 
tive energy of the extending cell is more considerable on one 
side than in the median line; and the consequence is the pro- 
duction of irregular forms of the annular rings, as exhibited in 
figs. 87 & 39a. 
It also now and then happens that the growing cell, when it 
has just advanced into the cavity of the cut cell, does not thrust 
the whole area of the septum downwards before it, but, as shown” 
