Lycopodiales . 183 
principal steles, other accessory ones (11). Mr. Tansley, in a recent 
course of lectures, suggested that the multiplication of the steles 
of S. incequalifolia might be due to an increase of branching, for 
while the leaf-traces are still mainly given off from the original 
central cylinder, the accessory steles contribute to the formation of 
the branch traces. 5. laevigata var. Lyalin has a dicyclic soleno- 
stelic rhizome and a highly polystelic erect stem. If we suppose 
(as seems reasonable, from an analogy with the Ferns) that the 
internal strand of the rhizome arose, in the phylogeny, as a 
thickening of the edges of the gap in the woody ring, formed in 
this case by the departure of a branch-trace, and that this 
thickening became detached as an independent cord fusing at 
intervals with the outer stele, it is clear that there must have been 
among the ancestors of this species a form in which the rhizome 
possessed a single solenostele without an internal strand. It seems 
equally reasonable to suppose that this solenostele was derived 
from a protostele, for the evolution of solenostely from protostely, 
has often been traced, though not in this phylum, where indeed 
solenostely appears to be very rare. When traced upwards the 
outer ripg-stele of the rhizome passes into the four principal leaf- 
trace-bearing steles of the erect stem ; these therefore apparently 
represent the original hypothecated protostele; the numerous 
accessory steles are presumably of secondary origin. 
Another point of considerable interest is that the base of the 
aerial stem of Selaginella spinulosa shows secondary growth in 
thickness (9). This character would seem to be best regarded as a 
recently acquired one, for it is not known in any other species and 
the genus as a whole does not appear to be reduced, since we found 
reason to conclude that there had been within the genus an evolution 
from a cylindrical protostele to a dorsiventral stele, to a dicyclic 
solenostele and to polystely. The genus thus appears to be on the 
up-grade of evolution and in Selaginella spinulosa the increasing 
complexity seems to have shown itself in the formation of a small 
amount of secondary xylem. 
In the recent genus the fructifications are cones, but M. Halle 
has lately shown that in the Carboniferous Selaginellites (Lycopodites) 
elongatus the sporangia were borne in the axils of ordinary leaves, 
as they are in some species of Lycopodium (12). This may well be 
a primitive condition, but in that case heterospory must have pre¬ 
ceded the evolution of a cone ; if so the strobiloid condition has 
originated independently in Lycopodium and Selaginella. The fact 
