Holden . — Some Fossil Plants from Eastern Canada . 249 
these illustrations the division is approximately in a horizontal direction, 
but more frequently it is oblique, so that in a given transverse section the 
trace appears single all the way out. That the foliar traces are not 
persistent, as in both the living members of the Araucarineae, may be 
inferred from their abundance in the twigs and complete absence in the 
mature wood. The pith itself never contains diaphragms like those of 
Cordaites , Tylodendron , and certain members of the living Abietineae. On 
the contrary, it is homogeneously parenchymatous, as shown in Fig. 20, 
a longitudinal section of one of the small lignitic twigs. 
The wood itself is composed of tracheides and rays. The former are 
isodiametric, with no trace of annual rings ; they are usually empty, though 
sometimes filled with a dark, pitch-like substance. The entire length of 
the radial wall is covered with bordered pits, which are always uniseriate 
and usually scattered. Fig. 22 represents a typical condition, though 
frequently they are further apart than is shown here, and rarely they are so 
closely compressed as to be flattened and angular. While they are as 
distant as the pits of the Abietineae and Taxodineae, they are never, as is 
the rule in these groups, 1 separated by the so-called bars of Sanio (cellulose 
thickenings embedded in the substance of the cell-wall). In this respect 
they resemble the wood of such types as Brachyoxylon 2 and Paracedroxylon? 
As opposed to the latter, however, the mouths of the pits are sometimes 
circular and sometimes elliptical. 
There is no vertical wood parenchyma, but ray parenchyma is 
extremely abundant. As is shown in Fig. 21, the rays are highly resinous 
and invariably uniseriate ; their tangential and horizontal walls are thin and 
unpitted, but the radial walls have from 2 to 4 small piciform pits to each 
cross-field. The characteristics of the rays are thus seen to be exactly like 
those of the living Araucarineae. 
Having described the salient points in the anatomy of our fossil, we 
may now consider its affinities. It seems probable that this wood 
represents the Dadoxylon Edvardianum described by Dawson 4 as charac- 
teristic of the Trias of Prince Edward Island, Quaco, and Martin’s Head, 
for although his specimens differ in having the radial walls of the tracheides 
covered with 1-2 rows of contiguous, hexagonal pits, they agree with ours 
in such characters as simple rays, and absence of either annual rings 
or transverse lamellae in the pith. The discrepancy in the matter of pitting 
is probably due to imperfect preservation of the tissue in question in the 
case of the material investigated by Sir William Dawson. 
1 Gerry, Eloise : The Distribution of ‘ Bars of Sanio ’ in the Coniferales. Annals of Botany, 
xxiv, pp. 119-24, 1910. 
2 Hollick, A., and Jeffrey, E. C. : Studies of Cretaceous Coniferous Remains from Kreischer- 
ville, N.Y. Mem. N. Y. Bot. Garden, 3, pp. 1-138, 1909. 
3 Sinnott, E. W. : Paracedroxylon. Rhodora, II, pp. 165-73, 1909. 
4 Dawson, J. W. : Acadian Geology, 4th edition, Notes and Addenda, p. 99. 
