122 
Gerry .— The Distribution of the 
lowed by staining with haematoxylin, although the pit membranes are 
stained blue, there is not the slightest indication of bars in any of the 
above-mentioned genera. 
On the other hand, in Prepinus statenensis , where they were observed 
by Jeffrey, 1 they appear, as shown in Fig. 9, as prominent light bands. 
In an undescribed Pityoxylon from Martha’s Vineyard the characteristic 
double bars are well illustrated by Fig. 10. In Pityoxylon scituatense 
they appear unmistakably as shown in Fig. 11. Again we find them in 
a fossil Picea , dug up with a mammoth in Alaska, where they stand out 
almost as well, after the usual staining, as they do in the living material. 
In Sequoia Penhallowii , Jeffrey, of the California Gold Gravels the bars 
are strikingly present as appears in Fig. 12. The fine state of preservation 
of all these fossil woods makes the presence of the bars perfectly obvious 
even in unstained sections such as those of Prepinus statenensis , Fig. 9, 
and of Pityoxylon scituatense , Fig. 11, or even in spite of the fact that, 
being cellulose, the bars have sometimes become decomposed and left 
transparent spaces such as appear in Sequoia Penhallowii , Fig. 12, and 
in Prepinus statenensis, Fig. 9. 
w 
Conclusions. 
The distribution of the bars of Sanio as above described establishes 
a constant and useful diagnostic character in the determination of fossil 
woods. In woods with Abietineous affinities we always find bars of Sanio 
even though at the same time we may find more or less Araucarian-like 
pitting. But in the Araucarineae we never find bars, although in fossil 
forms such as the Araucariopityoideae and the Brachyphylloideae, we find 
Abietineous as well as Araucarian pitting. 2 
The presence of bars of Sanio in the Podocarpineae points with other 
evidence to the probability that they are more closely related to the 
Abietineae than to the Araucarineae. At least they may have sprung from 
a common branch, for the apparent relationship is strengthened by the 
presence of the tendency towards recapitulation and preservation in the root 
which indicates the primitive, rather than the recently acquired, character 
of the bars. The ancient character of the bars of Sanio is further empha¬ 
sized by the fact that they occur in Prepinus statenensis , which lays an 
undoubted claim to primitiveness on the ground of its double foliar trans¬ 
fusion sheaths and centripetal wood. Therefore, judging from the fossil 
evidence and from the well-developed distribution at the present time, the 
bars of Sanio appear to be a definite and constant anatomical characteristic 
of all the Coniferales except the Araucarians. 
1 On the Structure of the Leaf in Cretaceous Pines. Ann. Bot., vol. xxii, April, 190S. 
2 Hollick and Jeffrey: Studies of Cretaceous Coniferous Remains from Kreischersville, N.Y., 
p. 75. Mem. N. Y. Bot. Garden, No. Ill, May, 1909. 
