470 PROFESSOR W. C. WILLIAMSON ON THE ORGANIZATION 
of the vascular cylinder, and observed respecting them, “ This orientation of these 
small bundles from the innermost layers of the wood is suggestive of their primary 
relations to leaves rather than to branches;” and further pointed out that they 
differed only from similar leaf-bundles in living Conifers only in being in pairs instead 
of being single. From this I inferred that “ Either two bundles went to one leaf with 
a double midrib, or the leaves were arranged in pairs.” 
Since these remarks were penned I have examined various living Conifers in hope of 
finding in some of them a similar organization : I have succeeded in the case of Salis- 
buria Adiantifolia, but in it alone. Fig. 28 represents a transverse section of a twig 
of that plant of the first year’s growth, made immediately below the terminal leaf-bud. 
At a we have the medulla and the xylem zone at b ; c is the cambium layer with 
newly forming xylem and phloem zones.'"' At d is an inner and at e an outer cortical 
zone. Many of the cells of the medulla and of the inner bark contain large sphaero- 
raphides, these two tissues thus furnished being portions of the primitive fundamental 
tissue; the outermost bark, e, being an incipient cork periderm. At ff we find the 
xylem ring interrupted by extensions of the medulla—which connect the latter with 
the cortical fundamental layer. Two vascular bundles, g, are given off from the vas¬ 
cular zone, one on each side of this extension of the medulla, and proceed outwards, 
diverging but slightly as they ascend towards the periphery, which they reach in pairs, 
as at g', g'. In fig. 29 we have a transverse section of the petiolar base of one of the 
outermost leaf-scales of the leaf-bud, in which we again find a pair of these bundles, g, 
entering the leaf-scale. The prosenchymatous tracheides of each of these bundles are 
arranged in six or seven parallel and somewhat fan-shaped, radiating rows. In the 
matured leaf these two bundles obviously subdivide to form the well-known venation 
of the Salisburia. 
Seeing that this twin development of the foliar bundles appears to be limited, 
amongst living Conifers, to the Salisburia, in which plant it seems to hold a definite 
relationship to the peculiar multinerved structure and Adiantiform contour of the leaf, 
may we not recognize the probability that our British carboniferous Dadoxylons bore 
some remote genetic relations to the living Gingko ? No such duplex leaf-bundle 
appears to have been detected in any of the fine examples of Cordaites and their allies 
discovered by M. Grand-’Eury in the St. Etienne deposits. This Salisburian form 
seems to be confined to our British Dadoxylons . That this similarity of contour in 
the arrangement of the leaf-bundles is not merely accidental is rendered the more 
probable by the fact that transverse sections of young twigs of the ancient and modern 
plants display corresponding identities. Thus a section of the bark of a young 
Dadoxyloni corresponds closely with the similar section of a Salisburia, fig. 28, in 
both of which the bark is separated into an outer coarsely cellular periderm and an 
inner one composed of much more delicate elements. This observation applies to the 
* The section was made in April. 
f See Memoir VIII., fig. 34. 
