Anatomy of the Cone and Fertile Stem of Equisetum. 235 
The Course of the Vascular Strands in the Cone. 
In an analysis of the structure of the cones of E. arvense f E. palustre , 
and E. liwiosum , it was pointed out that the irregular network of the strands 
in the pone of the last species had probably arisen in the phylogeny from 
the more regular type of stele found in E. arvense . Reasons were given 
for believing that the more primitive type of mesh was one of the first 
order; that is to say a parenchymatous tract of tissue arising above a trace 
that has departed and one confined to a single internode. In the phylogeny 
meshes of a higher order seem to have arisen when the development of 
additional xylem at the fertile node was insufficient to close a pre-existing 
mesh by the union of two neighbouring strands. Any increase of xylem 
at the node tended, when insufficient to close a mesh, to narrow it. It was 
found that in cases in which a wide sweep of xylem extended uninterruptedly 
upwards, above a trace that had departed, throughout a whole internode, 
the absence of a parenchymatous mesh was due to unusual development of 
the woody tissues. Such cases were uncommon even in E. arvense ; they 
occurred more rarely still in E. palustre , and none were observed in E. li- 
mosum . In all other cases the fact that no fresh mesh originated above 
a trace seemed to be due to poor development of xylem. Thus, if owing 
to the development of little additional xylem in the trace-bearing region 
a trace was given off from the edge of a vascular strand, the dying out of 
the xylem vertically above this trace involved, not the formation of a fresh 
mesh, but the sudden widening of one arising at a lower level. In cases 
where even less xylem was produced at the node, the parenchymatous 
meshes on either side of a strand persisted unnarrowed through the trace¬ 
bearing region, and no fresh mesh was formed, though the trace departed 
from the middle of the strand. A tendency was noticed in cones showing 
a reduction of the vascular system for a parenchymatous mesh to become 
decurrent for a little distance below and to one side of the trace above which it 
was considered, phylogenetically, to have arisen. Lastly, in cones showing 
very considerable reduction, some of the narrower strands occasionally passed 
through a node without giving off a trace. For a fuller discussion of the 
ways in which the reduction of the xylem made itself felt in A. arvense , 
E. palustre , and E. limosum , the reader is referred to pp. 668-73 and 
pp. 675-8 of my earlier paper on Equisetum . 
In the cone of Equisetum maximum , though there are certain characters 
not found in the cones of the other species, the structure of which was 
investigated, reduction of the xylem of the cone has proceeded along the 
same general lines as in those species. That is to say, parenchymatous 
meshes tend to persist through numerous internodes, to become extended 
laterally, and for a little distance downwards ; narrow strands also occa¬ 
sionally fail to give off traces at a node (cf. especially the reconstruction of 
