STRUCTURE AND AFFINITIES OF A LEPIDODENDROID STEM. 925 



In considering the affinities of the Dalmeny stem reference must be made to a 

 specimen recently described by Renault and Roche * from the Culm of Esnost under 

 the name Syringodendron esnostense. The petrified block shows several spirally 

 disposed elliptical areas on the weathered surface, which represents the ends of 

 parenchymatous cell-groups, called by the authors organes aeriferes. Internally there 

 is a segment of a band of secondary wood 20 mm. in diameter, succeeded by portions of 

 an annular zone of primary wood and a large pith. Beyond the secondary wood, and 

 separated from it by a space filled with siliceous material and plant debris, there is a 

 broad band of secondary cortex described as consisting of concentric zones of " suber." 

 In a tangential section of the secondary wood some broad medullary rays are repre- 

 sented containing diploxyloid f vascular bundles passing horizontally outwards, practically 

 identical with our PI. I. fig. 7. One of the vascular strands is also shown in transverse 

 section immediately beyond the secondary wood of the stem, agreeing closely with the 

 leaf-trace bundles of the Dalmeny stem (e.g,, PI. III. figs. 16 and 19). These bundles 

 are described as passing out to branches, and the true leaf-traces are considered to have 

 been obliterated as the stem had most probably lost its leaves ; one small and indistinct 

 bundle is shown in a tangential section of the wood passing through one of the broad 

 rays traversed by a " branch-bundle," which Renault and Roche speak of as a foliar 

 vascular strand. It would seem, however, much more probable that the diploxyloid 

 xylem strands are the true leaf-traces. 



There is, in fact, the closest resemblance between the French and Scotch stems, and 

 if not specifically identical, the two must be very nearly allied. The photograph of the 

 "concentric bands " of secondary cortex suggests a comparison of the fairly regular lines, 

 which break the regularity of the radially disposed phelloderm cells, with the rows of 

 secretory strands in the Dalmeny plant. The elliptical groups of cells visible externally 

 as slightly projecting areas in the French stem agree in structure, except in their some- 

 what larger dimensions, with the parichnos strands met with in tangential sections of 

 the Arran and Dalmen}' stems. 



The old name Syringodendron, usually regarded as standing for a partially decorti- 

 cated and old Sigillarian trunk, is used by Renault and Roche as an autonomous genus 

 on the grounds that the parichnos strands occur singly and not in pairs, and that no 

 evidence has been found suggesting the formation of the single groups by the fusion of 

 two originally separate arms. These reasons seem hardly sufficient for raising Syringo- 

 dendron to generic rank. In Syringodendron esnostense we have in all probability an 

 old stem — very closely allied to the Dalmeny species — which was most probably either 

 a Sigillaria, a Lepidodendron, or a Lepidophloios. To discuss this question would 

 involve us in an examination of the anatomical and other characteristics which dis- 

 tinguish these three genera. The structure of Lepidodendron and Lepidophloios is 



* Renault and Roche (97). 



t The term " diploxyle " is used by Renault to describe vascular tissue|consisting of primary (generally centripetal) 

 xylem and secondary centrifugal xylem ; he includes under this head the leaf-traces of Cycads, which are now termed 

 Mesarch by many writers ; they are not truly diploxyloid in that they consist of primary xylem only. 



