559 
whereas the radial walls show no thickenings. This is undoubtedly due to the fact 
that the drawing has been made from a section which was not sufficiently thin. In 
examining a section only one cell thick it is 
evident that the thickenings in particular con- 
cern the radial walls. In the cells situated at the 
limit between two cortical layers the radial walls 
are much thickened at the base but the thickening 
diminishes upwards, with the consequence that 
the lumen of the cell is conical downwards and 
the area of the basal wall is very small. Thus 
no continuous thick transverse wall arises by Fig. 549. 
theminekeningmotethestransversesswall, but ithe, Afufelltaiplicatas Crossisection of older trond 
2 x showing several irregular layers in the cortex. 
thickened lower parts of the radial walls of 83:1. 
the limiting layer of cells appear in thick sec- 
tions as a thick refractive wall traversing the cortical tissue. The presence of a 
continuous thick transverse wall would not be easy to reconcile with the explana- 
tion given by Jönsson, and it would be very unfavourable for the conduction of 
matter in a radial direction. 
In older fronds, Jönsson usually found from 4 to 6 cortical layers, once up 
to 12. The layers may be regular, but irregularities occur, the layers being incom- 
plete, encompassing only a part of the stem, or secondary layers appear within 
the ‘primary ones, dividing them, over part of their extent, into two or three 
subdivisions. Printz (1926, p. 63), who often found 6—8 layers, thinks 
that they indicate the number of years, while Jonsson speaks with 
reservation of this question. I found the stratification so irregular 
in some cases that it was impossible to state the number of the 
layers. On one side of the frond the number of layers was high, 
e.g. 9, while on the other side it was much smaller owing to the fact 
that some boundary lines were vanishing, others confluent (fig. 549). 
Moreover, it was not always possible to distinguish the secondary 
layers from the primary ones. If the layers were regular annual pro- 
ductions like those of the ligneous Dicotyledons, one might expect 
to find the number of layers increasing from the top downwards; 
but that is, at any rate, not always the case. For in a large specimen 
I found about 3 layers near the base, at a higher level about 7. The 
diameter of the stem and the thickness of the cortex was also smaller 
than at a level one cm higher up, namely 530 w and 130 « against 
Fig. 550. 
Ahnfeltia plicata. 2 : 
End of frond 670 w and 248 w. In the same frond the thickness of the cortex 
with nemathecia. 
i arly ‘ as 057125; 
December. 2/,:1 increased from the top to nearly one cm above the base (100; 125 
158; 178; 200; 248 u). In this case the cortex has thus continued 
growing except in the lowermost part of the stem. A regular increase of the number 
of cortical layers could not however be ascertained in this case. 
