98 STUDIES IN FOSSIL BOTANY 



various veins of the leaf (Fig. 36, C). Further division 

 of the bundles took place within the leaf itself, as it 

 widened out, or divided into segments. Where the 

 leaves were linear, the course of the bundles was the 

 same, but in this case each leaf received a single strand 

 only, as in the British species, 5. plurifoliatum. 

 Numerous vascular bundles, apparently equal in 

 number to the leaves in a whorl, have been found in 

 this species in the outer cortex of the node. They 

 no doubt arose, here also, from the subdivision of a 

 smaller number of strands, given off from the angles of 

 the stele. The branching of the leaf-traces within the 

 cortex is very characteristic of Sphenophyllum. 



■ The anatomical structure of the leaf was fully 

 worked out by Renault. The vascular bundles are 

 small and simple, apparently of the concentric type. 

 The tissue of the leaf was strengthened by bands of 

 sclerenchyma, lying next the epidermis of the upper 

 and lower surfaces. The parenchyma is uniform through- 

 out. There is thus nothing remarkable in the anatomy 

 of the leaf, unless it be the strong mechanical construc- 

 tion, a point of some interest, for it seems to show that 

 these leaves could not have been floating, as some 

 botanists have assumed, though with but little reason. 

 The narrow, unifascicular leaves or segments of J>. 

 plurifoliatum had a rather simpler structure. 



The roots have also been investigated by Renault. 

 The specimens which he examined were of diarch 

 structure, with secondary wood, resembling that of the 

 stem. Somewhat similar structures are occasionally 

 found in association with the Burntisland Sphenophyllum 

 (S. insigne), and with the Coal-measure species, 5". pluri- 



