7 
Evolution of Branching in the Filicales. 
to activity when the growth of the apex of the main rhizome is 
arrested. The attachment of the branch stele is directly to the 
main stele, a short but variable distance in front of the related leaf- 
trace. There is some variation also in the behaviour of the vascular 
system of the branch, but in general it may be said that the branch 
supply originates as a single strand which rapidly becomes curved 
and then closes up into a xylem tube enclosing a “ mixed” central 
xylem. The parenchyma sooner or later replaces the central 
tracheides altogether, forming a definite pith. 
It is when we come to the Ferns with reticulate steles, however 
that we find the greatest diversity in the mode of origin of the 
branch vascular supply. In a large number of cases the vascular 
system of the branch takes its origin as a single protostelic strand 
although the main axis in that region is dictyostelic (several spp. of 
Aspidium, Alsophila excelsa , Fig. 1, F, Stvuthiopteris germanicci and 
Nephrolepis). The protostele arises from one of the cauline strands, 
and persists for a varying distance outwards before it expands in a 
funnel-like manner into a solenostele, this giving place sooner or 
later to a dictyostele. In Nephrolepis the protostelic stage is 
enormously long, being met with throughout the length of the stolon » 
in this genus, moreover, the main protostele does not itself expand 
into a solenostele, but gives rise to secondary protosteles which 
enter the bases of the lateral plants. There each expands first into 
an extremely short funnel-like portion immediately succeeded by a 
dictyostele (Fig. 1, G). The branch supply in Struthiopteris ger- 
manica originates, according to Stenzel, as an abaxially grooved 
strand which, traced distally, is seen to close up into a complete 
tube. This tube further on becomes perforated by the leaf-gaps, 
so that from this point onwards the branch has a stele exactly like 
that of the main axis. Secondary branches are recorded by Stenzel 
in this species. 
In Aspidium and Alsophila excelsa the protostelic stage is very 
short, being in the latter species succeeded by the solenostele near 
the point of exit of the branch-stele from the cortex of the main stem. 
Stenzel describes in the same plant one case in which the protostelic 
region was altogether unrepresented, the funnel-like expansion being 
sessile on the large cauline strand. 
A more interesting condition is recorded by Professor Bower 
( ioc. cit., 1913, p. 454 and pi. 33, fig. 13) in the stolon of Cibotium 
Barometz , the stele of which at its base contracted to the Lindsay a- 
stage, so that its phloem was continuous, through a gap in the 
