258 Browne.—A Second Contribiition to our Knowledge of the 
course, disappeared too. Otherwise the small bundle was perfectly normal, 
and as we pass upwards in the internode it moves gradually further and 
further away from its sister bundle, increasing a little in size. Though the 
series of sections did not extend very far up into the internode it seems clear 
that the bundle eventually became an independent trace-bearing strand. 
Phylog-eny of the Cone of E . maximum and its bearing 
on the Fossil Cones of the Equisetales. 
It has already been pointed out that the xylem in the axis of the cone 
of E. maximum is less well developed relatively to its size than that in the 
axes of the cones of E . arvense and E . palustre . If we were to construct 
a series showing progressive reduction of the vascular system of the axes of 
the cones of the four species studied, the order of the species would be: 
E. arvense , E. palustre, E. maximum , and E. limosum . The difference in 
the degree of reduction of the xylem relatively to the size of the stele 
is very slight between the last two species. The reduction of the xylem 
system seems to express itself in the phylogeny by the persistence of 
parenchymatous meshes upwards through more than one internode, by their 
extension laterally and for a little distance downwards, by the presence 
of a considerable number of unlignified parenchymatous cells mingled with 
the tracheides, and by the poor lignification of the latter. On the other 
hand, the occasional widening and consequent fusion of two or more bundles 
considerably below the level of departure of the traces gives rise to 
a local increase of xylem. This appears not to be a palingenetic character, 
since it is not found in the cones of the species that have, relatively to 
their size, a better developed vascular system. 
In the case of existing Equisetaceae there is every reason to believe 
that they are forms derived by reduction from plants resembling the larger 
mesozoic Equisetites. It would seem, however, from a recent paper of 
Compter’s on impressions from the Keuper that some specimens of Equise¬ 
tites had cones resembling those of the Calamariae rather than those of the 
Equisetaceae, since the fertile whorls seem to have been separated by 
sterile ones (Compter). Dr. Scott states that on the whole the dimensions 
of the mesozoic Equisetaceae decrease as we approach the later horizons 
(Scott, p. 83). It would therefore seem, a priori , natural to look upon the 
larger species as likely to show the more primitive structure. Of all the 
cones hitherto studied those of E. maximum are by far the largest. But 
we have seen that the axial steles of these cones are among the more 
irregular and, on the view advanced in these papers, the more reduced. 
The possibility at once suggests itself that this irregularity of the vascular 
system of the cone may, since it occurs in the larger cones, be a primitive 
character and not due, as I have maintained, to the extension of parenchy¬ 
matous meshes upwards, downwards, and laterally. I still hold that 
