98 
F. 0. BOWER. 
It lias now been amply shown that the structure of the 
meristem in the roots of the Osmundaceae possesses characters 
intermediate between the two types of construction above 
described, and that these characters are not constant even in 
roots from a single plant. The question remains whether 
there is variation in the individual root from time to time. 
Beyond some few cases of irregular structure, I have no evi- 
dence of any change of character of the meristem in the 
individual root; moreover, the observations on the origin ot 
lateral roots show that in both Osmunda and Todea dif- 
ferences in the mode of segmentation of the rhizogenic cells 
may be found, which are quite as great as those observed in 
mature roots. Thus it is possible that the differences of struc- 
ture observed in different roots may have had their origin in 
differences of the very first segmentation of the rhizogenic 
cells. I have, on the other hand, no reason for thinking that, 
within limits, variation of structure may not occur in the in- 
dividual root. 
It cannot escape observation that those roots in which the 
coaxial type of structure is most prominent the roots them- 
selves are habitually more bulky than those in which the type 
of structure is confocal. Are these characters dependent one 
upon another? My observations lead me, as above stated, 
to the conclusion that as regards the roots of the species 
Osmunda regalis, differences of bulk of the roots cannot be 
correlated with corresponding differences of meristematic struc- 
ture. Still, if we consider such a series of forms as P ter is, 
Osmunda, Todea, and Angiopteris, it certainly would 
appear that with an increase of bulk of the root there is also 
an advance in complexity of the meristematic structure, and a 
transition from the confocal to the coaxial type. Thus, a cor- 
relation which does not apply in detail for the individual species 
appears to hold for the roots of the above series of genera. 
As far as may be judged from observations at hand a similar 
progression may be traced roughly, but not in detail, for the 
meristem of the stem in different species of S elaginella . 1 
1 Strasburger, ‘Bot. Zeitg.,’ 1873, p. 79, &c. ; Sadebeck, ‘ Handbucb der 
