PRIMARY STRUCTURE OF CERTAIN PALAEOZOIC STEMS. 359 



into the radial series of the secondary wood. I have not been able to observe a 

 longitudinal section showing these xylem- strands, but in one case a displaced tracheide, 

 belonging to the more external part of the group, showed a spiral thickening. 



The xylem-strands, being both small and few in number, form only an inconspicuous 

 feature in the general structure, but none the less they are perfectly distinct, espe- 

 cially where approaching their exit from the pith, and here represent mesarch primary 

 wood-strands, quite comparable to those of Lyginodendron, though on a smaller scale. 



Although there is no series of successive sections in which the course of the bundles 

 could be followed, there is little doubt that the distinctly double strands are those near 

 their exit, preparing to furnish the twin-bundles of a leaf-trace, while the fused or 

 single strands are cut lower down in their course. The arrangement of the bundles 

 indicates a 2/5 phyllotaxis as probable. The stem differs from Lyginodendron in the 

 fact that the leaf-trace divides into two before leaving the pith, and that it passes out 

 almost horizontally. In the former point our fossil resembles Poroxylon, in the latter 

 it agrees with some of the other Dadoxylons figured by Williamson.* The elements of 

 the outgoing leaf-traces are seen in one of the transverse sections, and have spiral or 

 reticulated sculpturing. The leaf-traces are also seen in pairs in the tangential 

 sections ; the small strand of transversely cut tracheides, constituting the trace itself, 

 is accompanied by contorted elements belonging to the secondary wood, such as are 

 usually associated with outgoing bundles. In some cases the irregular secondary 

 elements have insinuated themselves among those belonging to the leaf-trace. 



A Dadoxylon from Brazil, probably of Permian age, described by ZEiLLERf under the 

 name of D. Pedroi, bears some slight resemblance to our fossil. The pith, as shown in 

 transverse section, has three prominent angles, recalling the five similar projections in 

 our specimen, and, like them, marking the lines along which lateral appendages were 

 inserted. The medullary rays are usually uniseriate, and the secondary wood agrees 

 fairly well with that of D. Spenceri. The pith, however, is enormously larger than in 

 our fossil, having a diameter of 37 or 38 mm. as against 5 or 6 mm. The structure 

 appears to have been purely endarch, as the author shows the spiral elements on the 

 inner edge of the wood. There is thus little to connect D. Pedroi with our fossil, 

 except a certain general similarity in the transverse section. 



The stem of Metacordaites Rigolloti of Renault \ is at once distinguished from that 

 of our plant by the fact that each leaf-trace, in passing through the wood, consisted of 

 live bundles, not to mention other differences. 



I have named the Hebden Bridge stem Dadoxylon Spenceri, in commemoration of 

 the discoverer, to whom we owe so many valuable contributions to Carboniferous 

 Palseobotany. I have not thought it necessary to found a new genus for its reception, 

 as the primary xylem-strands are much reduced, and the stem in other respects agrees 



* L.c, Part viii., Plate 9, figs. 44 and 46. 



t " Note sur la Flore fossile des Gisements houillers de Rio Grande do Sul.," Bull. Soc. Geol. de France, Ser. 3, 

 t. 23, 1895, p. 619. Text-figs. 8-19 ; pi. ix., fig. 4. 



I B. Renault, "Note sur le Genre Metacordaite," Soc. d'Hist. Nat. d'Autun, 1896. 



