meristems of Ferns as a Phylogenetic Study . 3 1 5 
Ferns. If, on the other hand, a comparison be drawn 
between the structure in the Osmundaceae, and that de- 
scribed by Schwendener for the Marattiaceae 1 it will be seen 
that in the character of the meristem the family of the 
Osmundaceae fluctuates between the two types, and affords 
intermediate examples between them. In order to make this 
point clear a comparison may be drawn between Sachs’ 
diagram of the type of the Polypodiaceae (Fig. 1) and a 
diagram which I have constructed with the intention of repre- 
senting the Marattiaceous type (Fig. 17). In the former the 
periclinal walls in the body of the root constitute an inter- 
rupted series of confocal paraboloid surfaces, their common 
focus or centre of construction being situated in the apical cell 
itself : the periclinal walls in the root-cap constitute a series 
of similar curves which are however not confocal, though they 
are coaxial. Since the anticlinal walls cut the periclinals at 
right angles, those in the body of the root present a concave 
surface to the axis of growth, while those in the root-cap, cut- 
ting the coaxial curves, present a convex surface. 
In the Marattiaceous type, according to the description 
given by Schwendener (and my own observations on the roots 
of Marattia fraxinea and Angiopteris evecta bear out his con- 
clusions rather than those of Russow 2 ), the arrangement differs 
from the above type in certain important points, and a 
diagram may be drawn to show the scheme of construction as 
in Fig. 1 7. In this type of construction there are in the first 
place walls in two radial planes, which cut one another at 
right angles, and their line of intersection 3 is the organic axis 
of the root. The periclinal walls are none of them confocal, 
neither those which lie in the body of the root, nor even those 
in the procambial cylinder ; they are, however, all coaxial, and 
their common axis is the line of intersection of the radial 
walls, that is the organic axis. As a necessary consequence 
1 l.c. 
2 Vergl. Unters. p. 107, Fig. 158. Compare also De Vriese and Harting, Mono- 
graphic des Marattiacees, Fig. 27. 
3 As a matter of fact the two planes do not cut one another accurately along one 
line ; this has been noted by Schwendener. 
