the Anatomy of the Cone and Fertile Stem of Equisetum . 445 
well be looked upon as supra-annular in origin or as having become, speak- 
ing phylogeneticaily, decurrent slightly below the trace of the sporangio- 
phore, with the formation of which it may be associated. Such decurrence 
of meshes below traces is not uncommon in the cones of certain species 
(cf. Browne (2), p. 235). 
It is in Cone D of this species that the anastomoses characteristic of the 
strands of the axis at the level of the annulus are most clearly shown. In 
Cone D four out of six meshes are closed in this region, and three fresh ones 
arise, one of the first, one of the third order, and one that persists unclosed 
throughout the whole cone. 
In the matter of supra-annular anastomoses Cone E represents a much 
reduced condition. No fresh meshes arose in this region, and in the imma- 
ture phase of development represented in Text-fig. 7 none was closed ; but 
the distribution of the phloem indicated that had the metaxylem reached 
its full development the mesh between the second and third strands would 
have been closed in the neighbourhood of the insertion of the annulus. 
It is characteristic of the cones of E. debile, other than Cone D, that 
some of the meshes originating below the cone, and persisting in it, extend 
upwards throughout the greater part of the length of the cone. Thus, 
in Cone A one of the meshes arising above the uppermost vegetative node 
is only closed at the level of the uppermost whorl of sporangiophores. 
Cone C affords the most extreme example of unclosed meshes. Only three 
meshes arise within the cone, two of them unclosed, and one above the 
annulus, also unclosed. Four of the five parenchymatous meshes found 
in the internode below the annulus enter the cone, and three of them 
persist right through it and become confluent round the vascular strand 
of the acumen. One of these meshes has persisted unclosed not only through 
the whole cone, but also through the uppermost vegetative node (cf. p. 435). 
In the cones of E . variegatum , of which serial transverse sections were 
prepared, no fresh parenchymatous meshes arose in the stele between the 
level of insertion of the annulus and of the sporangiophores. In Cones A 
and B none of the parenchymatous meshes originating above the leaf-traces 
of the last vegetative node was closed in this region. It is possible that the 
junction of the fourth and fifth strands near the bottom of the reconstruction 
of the stele of Cone C represents a supra-annular fusion, for the anastomoses 
of the strands associated with the insertion of the annulus frequently occur 
somewhat above the attachment of the latter. 
It should be noted that in Cones A and C, though there are seven 
sporangiophores in the basal whorl of the cone, and seven strands below the 
annulus (these strands having arisen above and between seven leaf-traces), 
the seven traces entering the sporangiophores are not given off each by 
a single strand of the internode below, as would have seemed likely in the 
absence of any clearly supra-annular branchings of the strands. On the 
Gg 
