Osmundacece. 
263 
in this connexion, hut though they may conceivably have been derived 
from “ amphiphloic ” forms there is no actual evidence that they 
ever possessed internal phloem. 
If we do not accept FaulTs view of Osmund a cinnamomea 
as a degenerate type we seem bound to raise the question of why 
the internal phloem has arisen in the neighbourhood of the branching 
of the stem. It may have “ crept round ” the edges of the branch- 
gaps, or have “ subsided ” into them, but the use of such a 
phenomenon must remain a mystery. We may remember however 
that we found a somewhat similar difficulty in explaining the 
phloem-pockets of Gleicheniacea?, and there is little reason to 
suppose that these are reduction-phenomena. The solution of such 
problems may not improbably be found when we know more of 
the actual function of phloem. 
One of the strongest arguments against Faull’s view is 
undoubtedly that which has been advanced by Boodle, Seward and 
Ford, Chandler, and Kidston and Gwynne-Vaughan, namely the 
entire absence of any evidence of the existence of internal phloem 
in any of the young plants that have been examined. 
Fig. 91. Diagram of transverse section of an Osnnindaceous petiolar 
strand. 
Another significant fact is the absence of any sign of reduction 
in the leaf-trace. We have seen good reason to believe that in the 
main Leptosporangiate series the stele passed through the Lindsaya- 
stage on its way from the protostelic to the solenostelic condition, 
and that the continuous internal phloem of the stele is of value as 
giving a more adequate attachment to the phloem of the concavity 
of the leaf-trace. In the Osmundaceous leaf-trace we have phloem 
completely lining both the convexity and the concavity (Fig. 91) and 
yet the internal phloem of the trace is in connexion only with the 
external phloem of the stem. On the general principle of basipetal 
