Anatomy of the Cone and Fertile Stem of Equisetum . 233 
cation of that whorl. What I want to emphasize is that the difference 
between a pseudo-whorl and a local duplication of a whorl is one of degree ; 
consequently, in the classification of a group of traces, the drawing of 
a dividing line between a pseudo-whorl and a duplicated whorl is a matter 
of individual interpretation, and the classification is inevitably in some 
cases an arbitrary one. It is obvious that by regarding a group of traces 
as a duplication of an existing whorl or as an independent whorl (pseudo¬ 
whorl), a given mesh may be regarded as of the order x, or of the order 
x+i. Even when this factor does not come into play, parenchymatous 
meshes of the same actual length may, in the same region of the cone, 
belong to different orders, and this for three reasons. Firstly, because the 
sporangiophores and traces are inserted so irregularly on the axis and axial 
stele that different portions of the same internode may vary considerably 
in height. Secondly, because parenchymatous meshes may arise and be 
closed at very different distances above the departure of a trace; occa¬ 
sionally a mesh makes its appearance at such a height above a trace that 
it originates but little below the level of departure of the traces of the next 
whorl above, and, if it is closed half-way up this second internode, this 
mesh of the second order is actually shorter than a mesh of the first order, 
extending throughout nearly the whole of one internode. Lastly, because the 
tendency of meshes to become decurrent downwards and to one side of traces 
above which they may be considered to have originated in the phylogeny— 
a tendency noted in the study of the cones of E. pctlustre and E. limosum 
(Browne, p. 673) is a well-marked character of cones of E. maximum —and 
often increases the length of a mesh without bringing it under the category 
of a mesh of a higher order. These difficulties in the estimation of the 
height of the meshes, particularly in the classification of those of Cone B, 
must be borne in mind in studying the following analysis of the nature and 
number of these meshes. 
Meshes originating within the Cones themselves. 
(i. e. above the lowest whorl of sporangiophores.) 
Orders :i 2 3 4 5 6 7 8 9101117 Total. 
Cone A 29 30 12 7 5 2 3 o o o o o 88 
Cone B 33 46 32 23 10 5 5 4 3 2 3 1 167 
This table does not give the full number of meshes extending into the 
cone, for some of the meshes arising above the annulus, and some of those 
arising above the uppermost whorl of leaves, persist into the cone for varying 
distances. The number and nature of these two sets of meshes arising 
respectively above the annulus and above the uppermost whorl of leaves are 
given in the two following tables. Those, in the first of these two tables, of 
orders higher than the first order, and in the second table of ord ers higher than 
