STRUCTURE OF LEPIDOSTROBUS. 365 



and showing occasional transverse or oblique cross- walls. At j9 is seen a zone of cells 

 occupying a position between the xyleni and phloem — a position corresponding with that 

 occupied by the cells shown at p in the transverse section (PI. XXXVI. fig. 12) and 

 doubtless representing modified cells of the inner sheath. Here the cells appear to be 

 fairly thick -walled and exhibit great elongation in the vertical direction. 



When the bundles enter the outer cortex various points of interest arise. PI. XXXVI. 

 ■fig. 6 is from a photograph of a portion of a tangential section passing through the outer 

 cortex and showing some of the emerging leaf-trace bundles cut more or less transversely. 

 It will be seen that below each bundle there is an empty space, Pa., of considerable 

 size. The same space is also seen in PI. XXXVI. fig. 7, which is a photograph of a 

 nearly radial section passing through the outer cortex and proximal end of a sporophyll. 

 In the transverse section (PI. XXXVI. fig. 1) a similar space, P«., is seen heliind the 

 vascular bundle, as already mentioned in the description of Williamson's figures. In 

 the slides from which Williamson's figures were made the origin of this space cannot be 

 determined, but in PL XXXVI. fig. 10, which represents part of a radial section, its 

 continuity with the middle cortical space is clearly seen. In this figure, i.e. is the inner 

 cortex, o.c. the outer cortex, and l.t. a leaf-trace bundle passing outward. The bundle 

 is broken ; but the continuity of the space Pa., below the bundle, with the middle 

 cortex space, m.c, is quite clear. Unfortunately the tissue formerly occupying this space, 

 as also the middle cortex itself, has entirely disappeared. Further reference to this 

 space, which undoubtedly represents the tissue named the parichnos by Bertrand *, in the 

 vegetative region, will be made in the description of Wild's Cone {L. oldhamius /3). 



Reverting again to PI. XXXVI. fig. 7, a leaf-trace bundle is seen intersected longi- 

 tudinally as it is passing into the pedicel of a sporophyll. The xylem is shown at x. and 

 the phloem at p!, while between the two is a rather thick-walled tissue p. Surmounting 

 the bundle is a parenchymatous tissue sh., and below the phloem the remains of a dark- 

 looking tissue shJ \ doubtless both sh. and sh! represent the inner cortical sheath. On 

 PL XXXVII. fig. 13 these structures are shown on a larger scale, the reference letters being 

 the same as before. The remarkable development of the tissue between tbe xylem and 

 phloem in this region is shown on PL XXXVII. fig. 14. This is from a radial section of 

 a leaf- trace bundle in the outer cortex just as it begins to bend more sharply outward 

 to the sporophyll. Here the tissue p., between the xylem x. and phloem p. , is very 

 strongly developed, resulting in the w^ide separation of the two latter tissues. The 

 drawing represents the bundle in its natural position, and it will be seen that the cells ^. 

 are so arranged that their long axes are parallel, not with the length of the leaf-trace, 

 but with the main axis of the cone. 



PL XXXVII. fig. 15 is an enlarged drawing of one of the bundles shown in the 

 tangential section through the outer cortex (PL XXXVI. fig. 6). The bundle is passing 

 obliquely outward, and so is not cut quite transversely. At x. the xylem is shown ; at ^'. 

 the phloem elements with quite thin walls ; while between the two the fairly thiek-wallcd 

 tissue p. is seen. At sh., sh. are some clear-looking cells quite distinct from those of the 

 outer cortex o.c, and doubtless representing the inner cortical sheath. 



* " Remarques aur le Lepidodendron Harcourtii" Trav. et Mem, Facult Lille, vol. ii. (18D1), Mem. 6, p. 84. 

 SECOND SERIES. — BOTANY, VOL. V. 3 K 



