88 
F. O. BOWER. 
into consideration, it seems probable, though not proved, that 
there is variation in meristematic structure, within certain 
limits, in the individual root during its development just as in 
Selaginella Martensii, as demonstrated by Treub, the 
form of the apical cell is liable in the individual stem to 
certain variations of form. 
The study of the anomalous structure of the meristem of 
the root of Osmund a affords us information as to another 
question, which has an important bearing upon views as to the 
connection between external form and internal cellular struc- 
ture. The question is this — whether any constant relation is 
to be found between the bulk of the individual root and the 
type of structure of its meristem ? In order to answer this 
question for the roots of Osmunda regalis the sections 
from which the drawings were made were measured, so as 
to express in y^L-g-ths of an inch the transverse diameter of 
each root at the level of the apical cell or apical group. The 
result of the measurement is attached to each figure. Though 
this method of measurement is open to objection, in that it 
does not take account of the actual curve of surface of the 
apex of the root, still the results thus obtained give at least 
an approximate measure of the bulk of the roots. Further, 
as median longitudinal sections do not show any considerable 
variation of the actual curve of surface, this may be at least 
provisionally neglected. A comparison of these measurements, 
and of the meristematic structure of the apices in question, 
will show clearly that there is no strict and constant relation 
between them. Thus, of the three observed examples of a 
four-sided cell, one was o Q ths of an inch in diameter, another 
Y^l^ths, and a third T ^ - 0 -ths. Conversely, figs. 3, 4, 5, 7, 8, 
11, 16, and 18 are all from roots giving within narrow limits 
the same measurement. On the other hand, it is worthy of 
note that fig. 1 represents almost the smallest, and fig. 14 the 
most bulky of the roots investigated. Thus, my observations 
do not bear out in detail the idea which suggests itself from a 
comparison of the roots of the Polypodi acese with those of 
