Filicales. 
24 1 
genera. Mr. Tansley and Miss Lulham have traced the evolution 
of solenostely from it by the decurrency into the internode of the 
phloem-strand (in which, as it increases in size, endodermis and pith 
are developed) and by its junction with the similar strand developed 
in connection with the leaf-trace of the node below (36). All the 
phases of this process have been found in the Cyatheaceae or 
Polypodiaceae. It appears that the first tissue to appear within the 
stele of these orders when protostely was departed from was in 
every case phloem. The great majority of the Cyatheaceae and 
Polypodiaceae, even the complex polycyclic dictyostelic forms, pass 
through an ontogenetic phase in which their xylem, like that of 
Lindsaya, contains only a strand of phloem. Mr. Chandler examined 
a large number of young ferns of these orders and came to the 
conclusion that those cases in which other tissues appeared within 
the xylem before internal phloem were due to “ short-cuts ” of 
evolution and that the parenchymatous cells in the centre of the 
stele were best regarded as potential phloem (13). On the other 
hand certain species of Vittaria and Antrophyum possess little or 
no internal phloem even when mature (17), (20). This is clearly 
due, as Dr. Jeffrey first asserted, to reduction (20). 
A considerable number of ferns have retained a solenostelic 
structure. Mr. Gwynne-Vaughan has shown that two types of 
dictyostely have been evolved from it by the overlapping of leaf- 
gaps. In the dorsiventral forms the leaves are inserted in two rows, 
and when the gaps left in the stele by the departure of their traces 
overlap, the dorsal part of the solenostele becomes a small dorsal 
strand, while the ventral part from which no leaf-traces are given 
off remains unreduced. In transverse section this type usually 
shows a large gutter-shaped stele open dorsally, and in the dorsal 
gap a much smaller stele ; but if the leaves are inserted nearly 
opposite one another on the sides of the rhizome the dorsal 
and ventral steles are more or less of the same size. In the radial 
dictyostelic forms the leaves are inserted all round the stem so that 
the gaps left by them are constantly changing, and the vascular 
strands therefore continually anastomosing. Mr. Gwynne-Vaughan’s 
work proves that dictyostely (so-called polystely) arises not by 
branching, but, as Dr. Jeffrey first contended (20), by the overlapping 
of the leaf-gaps. Miss Ford, however, describes the stele of 
Ceratopteris thnlictroides as dividing directly into two (15), and Mr. 
Chandler describes how the stele of Polypodiuiu aureum appears to 
break directly into two and later into more numerous strands (13). 
