Slopes .— The Internal Anatomy of ‘Nilssonia orientalise 391 
From most of the sections of the fossil one might conclude that resin- 
canals were absent, as they are in many Cycad genera. 
Two sections, however, clearly show canals which have every appear¬ 
ance of being resin-ducts. This is seen in Figs. 2 and 3, PI. XXVI, r. In both the 
leaf-sections which show the phenomenon, there is only one canal in the 
tissues. It lies near to the edge of the leaf, as is seen in Fig. 2, PI. XXVI. The 
appearance of the structure is exactly that of a resin-canal, with its epithelial 
lining partly preserved (see Fig. 3, PI. XXVI). From this, it seems very reason¬ 
able to conclude that the canal is really a resin-duct. It lies between two 
bundles, and is large compared to them. 
SC . 1 
Text-figure. Slightly diagrammatic drawing to show the anatomy of the leaf of Nilssonia 
orientalis, Heer. e. 1 , upper, and e . 2 , lower epidermis, st., stoma, only on the lower side. J ^. 1 and 
sc. 2 , groups of sclerenchyma above and below the bundle, s., bundle-sheath, px , protoxylem. x., 
xylem, which appears to be entirely centripetal, ph , space, in which are fragments of badly petrified 
phloem. 
As is seen in Fig. 2, PI. XXVI, the canal lies near the edge of the leaf, 
which must have been towards the apex in this section, and it is not impos¬ 
sible that an early type, such as the plant we are considering, may have 
had a large resin-canal near the border of the leaf, without having others 
regularly between each pair of bundles. 
The vascular bundles. The bundles lie approximately equidistantly 
in the leaf, about 0-5 mm. apart, and are simply collateral in structure. 
The bundle-sheath, though not highly specialized, is clearly recognizable 
round several of the better-preserved bundles. It consists of large, roundish 
cells (s., Text-fig.), on which I have not been able to observe any pitting. 
