386 Farmer and HilL — Arrangement and Structure 
strands apparently replace the cylindrical strand below. 
Furthermore, the peripheral phloem, instead of remaining as 
a strictly external ring surrounding the cylindrical xylem, is 
seen to differentiate in an intravasating manner through a gap, 
and thus to ‘ dip down,’ so to speak, into the pith. Then 
as the gaps close up above, this phloem, which has thus 
become ‘ invaginated ’ into the centre is found to have dis- 
posed itself as a mantle lining the internal surface of the wood 
(Fig. 23). The endodermis does not necessarily take any share 
in the process at least in these early stages, and thus the 
new phloem internally borders directly on the pith-paren- 
chyma. It is only in rather exceptionally favourable instances 
that this pseudo-solenostelic condition can be clearly made 
out in Angiopteris , owing to the fractionation of the vascular 
cylinder in connexion with the increasingly crowded state 
of the foliar gaps. In Marattia , where the gaps are not so 
numerous, the character is more easily observed. The phloem 
on the inner and outer sides of the xylem is constantly 
continuous through the foliar gaps, and hence the appearance, 
as presented in sections, is given of a change from a ‘ mono- 
stelic 1 to a repeatedly branching ‘ polystelic 5 or 6 dialystelic * 
condition. And owing to the reparation of the foliar gaps 
above the exit of the traces, ‘ gamostely 5 (or solenostely) is 
the rule, save for the absence of the internal endodermis. 
As the fractionation of the vascular cylinder progresses, how- 
ever, the inner phloem also may come to be lined by an 
endodermis, and the arrangement thus brought about coin- 
cides with that stelar arrangement termed by Jeffrey 1 
amphiphloic siphonostely, if the pith be regarded as excluded 
from the stele. We shall, however, discuss this point of view 
in detail further on. 
It may be especially remarked here that the external 
endodermis is always present in all the plants of Angiopteris , 
Marattia , and Kaulfussia that we have examined. It is easy 
to demonstrate its presence by treating sections with sulphuric 
1 E. C. Jeffrey, The development, structure, and affinities of the genus Equisetum . 
Mem. of the Bost. Soc. of Nat. Hist., v. 
