700 Browne. — Contributions to our Knowledge of the 
(Hickling (1), p. 375), the non-alternation of the bundles in the cone of 
Equisetum seems clearly, from the comparative anatomy of the cone ; not to 
be a primitive feature. Moreover, Renault says of his Volkmannia gracilis, 
which is clearly a Pataeostachya (cf. Jongmanns (l),p. 332 ), that the vascular 
bundles alternate from one internode to the next. It seems shown by the 
full discussion above that this non-alternation of the bundles at the nodes 
of the cone of Equisetum is due to the persistence of parenchymatous 
meshes on either side of a strand of xylem, and that this persistence is the 
direct result of poor development of xylem. More generally, when there 
is no marked reduction of xylem, and when the position of traces is not dis- 
turbed by changes in the number or size of. members of successive whorls, 
the traces of successive whorls alternate with one another. When owing to 
reduction of xylem in the axis of the cone the traces are superposed, the 
sporangiophores which they supply still alternate regularly. This is, no 
doubt, due partly to exigencies of space in the young unelongated cone, but 
it may possibly be in part due to the retention of an ancestral feature. In 
Jeffrey’s photograph of a cone of E. arvense cut in two, some of the 
strands may not have alternated at the nodes owing to persistence of paren- 
chymatous meshes, while other strands may appear to be continued upwards 
above the node because the meshes of parenchyma originating above traces 
are generally narrow in this species, so that a longitudinal cut would not 
pass through many of them. 
Summary of Results. 
1. The xylem at the node of a cone of Equisetum consists of a woody 
ring or of woody bands of varying width ; in the internode the xylem 
usually breaks up into definite strands separated by parenchymatous 
meshes ; these generally arise vertically above the points of departure 
of the traces, often close above them, but never, except perhaps in the 
reduced apical region, immediately above them. 
2 . When the xylem is relatively well developed these meshes are 
closed again in the neighbourhood of the next node ; when less xylem 
is present some of the meshes tend to persist upwards through two or more 
internodes. When still less xylem is developed meshes of parenchyma tend 
to arise a little below and to one side of a trace ; they also frequently 
become widened laterally by fusion with a mesh that has originated 
independently or by congenital fusion with a mesh above a trace departing 
from the edge of a strand of xylem. 
3 . The cones of E. arvense, E. palustre , and E. limosum form a series 
showing increased reduction of the xylem ; pari passu with this decrease 
of the xylem the stele comes to form a more and more irregular network 
of strands. 
