APEX or THE ROOT OF OSMTJNDIA AND TODEA. 91 
again shows the group of four initial cells, as seen in a trans- 
verse section through a young lateral root of about the same 
age as that in fig. 31. It is thus shown that the meristem of 
the lateral root of Angiopteris evecta assumes its definite 
character from the very first, and the four initial cells make 
themselves at once apparent. It is thus impossible to obtain 
evidence as to the origin of the mode of growth with a group 
of initial cells from the mode of their origin in the individual 
lateral root, since the character of the meristem is defined from 
the very first. 
It is worthy of note that cells of the cortex lying around 
and outside the rhizogenic cell also undergo occasional and 
irregular division, but do not appear to play an important part. 
Other cells of the endodermis and cells of the pericambium 
also divide freely (fig. 30). 
General Consideration of the Results. 
The foregoing description of the structure of the meristem 
of the roots in Osmunda and Todea shows, in the first place, 
that there is no such strict uniformity in these plants as is 
found in the roots of Equisetum and the Polypodiaceae, 
on the one hand, and, according to Schwendener, in the 
Marattiacese on the other. Secondly, the structure of the 
meristem, as above described in the Osmundacese, fluctuates 
in its characters between those two well-marked types, and 
affords numerous intermediate examples between them. 1 In 
order that those intermediate examples may be duly appre- 
ciated, it will be necessary to enter briefly upon the considera- 
tion of those two typical systems of construction between which 
the Osmundacese oscillate. The first, that typical of the 
Polypodiaceae, is shown diagrammatically in fig. 33, which is 
quoted from Sachs ( f Arbeiten/ Bd. ii, Taf. iv, fig. 12). Here 
the periclinal walls in the body of the root constitute an inter- 
1 It may be objected that the roots observed may have been in a resting 
state, and so the apical cell may have been segmented as described by Sachs 
Arbeiten,’ Bd. ii, p. 90). This, however, was not the case, as at least the 
large majority of the roots were in a state of active growth. 
