McNicoL — On Cavity Parenchyma and Tyloses in Ferns . 41 1 
closing membrane. Such, he also showed, was the case in fossil Ferns, and 
as there would be no actual resisting membrane over the pits to break 
through, this renders it somewhat more likely that the cells of the xylem 
sheath could grow out from one tracheid to another, and so throughout the 
bundle, just as in the xylem-sheath cells grow out from one spiral vessel to 
another in the formation of cavity parenchyma. Another argument in 
favour of the theory that these soft cells in fossil Ferns may be due to the 
xylem-sheath cells is that a petiole of Rachiopteris ( Zygopteris ) corrugata 
was recorded by Weiss (’06) as showing two small lignified tyloses in one 
of the tracheids, the others being filled by cells having the usual unthickened 
walls. This, of course, indicates that the tyloses were formed by the 
ingrowth of a soft wood cell and not by a fungoid organism, for the latter 
would not be capable of such lignification. Such an example is a strong 
argument in favour of the view that the soft cells in the tracheids are all due 
to the ingrowth of soft wood cells. I cannot explain the presence of these 
lignified tyloses except as being due to an isolated soft cell enclosed within 
the xylem, such as one occasionally sees in ferns which have a compact 
mass of xylem. So far as I know, no reference has been made to specimens 
of Rachiopteris in which tyloses occur only in the tracheids at the periphery 
of the bundle, as one might reasonably expect if the soft tissue were formed 
from the xylem-sheath cells, for the tyloses in this case should be formed 
first in the tracheids immediately adjoining the sheath cells. I have seen 
slides of Rachiopteris in which some cells, especially the smaller cells, to the 
outside are empty and others are filled with * tyloses ’ : some of the tracheids 
in the interior of the bundle had the soft tissue, others had not : the arrange- 
ment was quite irregular. I incline to the opinion that the occurrence 
in some fossil Ferns of soft pseudo-parenchyma sometimes stopping up 
the entire mass of tracheids, sometimes stopping up only a proportion, 
is due to the presence of fungi which had entered the plant. The phe- 
nomenon of the expansion of the mycelium of a fungus in the cavities 
of wood vessels is well known. Referring to the mycelium of the Agarici- 
neae ramifying in the wood of a host plant, Tubeuf and Smith (’97) write : 
‘The mycelium gradually spreads . . . into the vascular elements of the 
wood. . . . While previously it was simply filiform and furnished with 
numerous lateral hyphae, it now develops large bladder-like swellings, and 
at the same time the hyphae change into a kind of large-meshed paren- 
chyma, which like the tyloses of many dicotyledonous trees completely fills 
up the lumina of the tracheids.’ If the pseudo-parenchyma in fossil wood 
is due to the xylem sheath, one would expect to see in longitudinal sections 
evidences of the horizontal extension of these cells from one tracheid to 
another, such as one sees in modern Ferns, whilst if the tissue is due to fungi 
there would be less likelihood of noticing the passage horizontally of a fungal 
filament ; the direction of extension of a fungal filament would be generally 
F f 
