3 1 B Holden . — On the Anatomy of 
Figs. 13 and 14, PL X IX 5 represent two of these groups in transverse section, 
and Figs. 15 and 16 in longitudinal. The most striking feature about them 
is the large amount of xylem parenchyma occupying the centre of each 
group. As far as the writer is aware, such a condition is unique among 
stems of the Cordaitalean type, though it is characteristic of certain 
Palaeozoic ferns belonging to the Botryopterideae and allied forms. At the 
same time, it is worth noting that Mesoxylon shows a tendency in this 
direction (see especially 1 Annals of Botany vol. xxvi, PI. XC, Fig. 23). It 
was stated above that during the long internodes there is no distinction 
between cauline and foliar bundles, but at the nodes a difference may be 
observed in the relative size of these primary groups, those associated with 
the leaf-traces being distinctly smaller. This may be made out in Fig. 7, 
PI. XVIII. 
Secondary Wood. 
The secondary wood conforms in general with the cosmopolitan 
Palaeozoic genus Dadoxylon . The tracheides are small and unpitted tan- 
gentially, but the radial walls have crowded, one or two serial, bordered 
pits (Fig. 18, PI. XIX). These are flattened by mutual contact until often 
they become roughly hexagonal in outline ; the pore is elliptical. The 
rays are low (2 to 7 cells in vertical series), and throughout they are uni- 
seriate (Fig. 11, PI. XVIII). This feature is shared by Cor daites and Meso- 
xylon, though it is in marked contrast with the broad-rayed Megaloxylon , 
Poroxylon , or Pitys. Like them too, the horizontal and end walls are thin 
and unpitted, but the lateral pits are unique. In all previously described 
woods of this age, the rays communicate with the tracheides by means 
of several small so-called ‘ piciform ’ pits ; these are half bordered, and are 
in groups of 4 to 8 in each cross-field. In our Indian specimen, how- 
ever, the pits are much larger, and have lost their borders, forming the 
typical ‘ Eiporen ’ of Gothan. 1 They are in groups of 1 to 4, the general 
tendency being for several small ones to fuse and form a single pit, as has 
been described by Bailey 2 in various species of Pinus . Stages of this process 
are shown in Fig. 18, PI. XIX. The importance of the pitting of medu- 
llary rays has been emphasized, perhaps unduly, by Gothan ( 1 . c.). Since 
‘ Eiporen * such as are found here are characteristic of the genus Phyllocladus 
and its allies, he proposes to include all woods showing this feature in the 
form genus Phyllocladoxylon. The genus Xenoxylon is distinguished by 
having pits which are even larger, one such embracing the entire cross-field. 
Some of the pits of our fossil would bespeak its inclusion as Phyllocladoxylon, 
others as Xenoxylon. However convenient such a classification may be 
systematically, its value from a phylogenetic standpoint may be questioned. 
Among living Conifers, these large ‘ Eiporen ’ occur not only in the 
2 Bailey (1910). 
1 Gothan (1905). 
