Filicales. 
245 
assumes a central position in the internode; but at the node it 
connects again with the stele from which it originated. As this 
strand increases in size, phloem, endodermis and parenchyma 
appear in it, and, after it has thus attained to solenostely, another 
stele arises in the same way as a local thickening of the gap 
formed in this second solenostele by the departure of the strand 
that fuses with the original external stele when the latter gives 
off a leaf-trace. There can be very little doubt that polycyclic 
solenostely originated in the phylogeny in the same way as in the 
ontogeny, and that one of the ancestors of Matonia must have 
contained a ^single solenostele. Unfortunately, although we are 
acquainted with a certain number of Mesozoic fossils, such as 
Laccopteris and Matonidium, which are referred with some certainty 
to the Matonineae on account of the branching of their fronds, they 
are preserved only as impressions and we know nothing of their 
anatomy (27). 
There is little in common between the Matonineae and either 
Botryopterideae or Schizaeaceae. Such similarities as the origin 
of the sporangia and the modified dichotomy of the frond, in which 
the Matonineae resembles the two other orders, appear to indicate 
a remote common origin rather than a close affinity. 
It is to the Gleicheniaceae that the Matonineae show the 
closest affinity. It has been shown that the ancestors of Matonia 
must have passed through a solenostelic stage in which their 
stele resembled that of Gleichenia pectinata. Dr. Scott has 
shown that the acquisition of centrifugal xylem, advantageous 
because its development is unlimited, frequently causes the 
reduction and eventually the suppression of the centripetal 
wood or its conversion into pith (25). The centrifugal wood 
was presumably acquired early in the Gleicheniaceous cycle 
of affinity, since the protostelic species of Gleichenia are 
mesarch. It is not, however, contended that Matonia arose 
from a solenostelic Gleichenia, for solenostely, being only known 
in one exceptionally large species, appears to have arisen 
comparatively recently in the Gleicheniaceae. What is suggested 
is that the polycyclic solenostely of Matonia was evolved from 
simple or monocyclic solenostely; that this single presumably 
endarch solenostele may well have arisen from a mesarch soleno¬ 
stele or protostele by the abortion of the centripetal-xylem or its 
replacement by pith. Such a solid mesarch protostele would be 
distinctly Gleicheniaceous in its affinities. Such an origin would 
not be incompatible with the development of the frond as explained 
