the Anatomy of the Cone and Fertile Stem of Equisehtm. 451 
itself. In Cones A, C, and E much of the parenchyma found in the steles 
consists of meshes that arose above the last whorl of leaves and persist well 
into the cone. In Cone E, indeed, only two fresh meshes originate within 
the cone proper — i. e. definitely above the basal whorl of sporangiophores. 1 
But even in the cones of E. debile , where the vascular system is so reduced, 
never less and sometimes more than half the parenchymatous meshes 
originating within the cone itself arise vertically above traces of the node 
below. 2 Taking into consideration the reconstructions of the steles in all 
the nine species studied, the proportion of meshes that arise vertically above 
traces is very large. Moreover, the fact that this relationship is much more 
prevalent in species with more metaxylem and shorter meshes is surely an 
argument in favour of the primitiveness of that position for those who, like 
Barratt and myself, look upon the primitive type of vascular structure 
in the cone as a siphonostele. Though it is suggested that the primitive 
form of mesh was probably one of the first order, originating at a point 
superposed to and some way above a trace of the whorl below, it is not, of 
course, claimed that in any one cone on the line of descent of Equisetum 
a parenchymatous mesh of the first order was found above every trace. 
Indeed, this was almost certainly not so, since some of the existing species 
of Equisetum retain in places wide internodal sweeps of xylem, apparently 
a vestige of the former siphonostele, although all the existing species of the 
genus possess some meshes of orders higher than the first. 
The only argument given in support of the suggestion that the factor 
governing the distribution of the meshes of parenchyma may be a mechanical 
one is that ‘ the species with large and heavy cones have more abundant 
xylem and more regularly developed network ’ (Barratt. p. 227). It has 
already been pointed out that in the cones of E. maximum , the largest of 
the genus, there is a very irregularly developed network: and relatively little 
xylem. Of course, in a cone of this species which, like Cone A of my paper 
on E. maximum , consists of over six hundred sporangiophores there are 
actually a large number of tracheides. although the xylem is poorly 
developed relatively to the parenchyma. 3 I have previously suggested that 
while the length of the cones seems to have little effect on the height of the 
meshes the width of the stele has a bearing on the closure or persistence of 
the meshes through the node. I wrote : ‘ But if the reduction in width 
of the stele did not keep pace with the reduction of the xylem at the 
1 In estimating the nature and incidence of the parenchymatous meshes of this cone and of 
Cone B of the same species it must be remembered that the broken line enclosing a dotted surface 
marks the extent to which the differentiation of metaxylem would presumably have attained had the 
cone been mature. 
2 An indirect confirmation of the view that the cones of this species are amongst the most 
reduced of those studied is afforded by the fact that in the branches bearing them the ring of xylem 
at the last node may be incomplete, a condition not hitherto recorded for other normal nodes. 
3 This cone was not exceptionally large. It was about three inches long, and I have seen cones 
fully four and a half inches in length. 
