STRUCTURE AND AFFINITIES OF A LEPIDODENDROTD STEM. 911 



thickening began there were some at least of these isodiametric elements present on the 

 surface of the corona, and additional elements of this type were formed during the 

 process of secondary growth ; but this question must be considered in more detail in 

 connection with the leaf-traces.* 



The rows of secondary tracheids are separated at intervals by bands of one, two or 

 several rows of radially elongated elements characterised by spiral or partially reticu- 

 late thickening bands. In the stem before us there appears to be an almost complete 

 absence of any simple unthickened parenchymatous elements in the bands of medullary- 

 ray tissue. The nature of the medullary-ray elements is shown in PI. I. fig. 6,1" which 

 represents the appearance of a broad " ray " as seen in a transverse section of the 

 secondary wood, also in PI. IV. fig. 24 : in figs. 7, 30 and 32, the medullary-ray tissue 

 is seen in longitudinal sections of the wood. In a tangential section of the secondary 

 xylem, the medullary rays are seen to vary considerably in size, consisting of a single 

 row of a few cells deep, or of bands of considerable length and varying breadth 

 (PI. IV. fig. 30) ; a leaf- trace may be met with in the broadest rays, as seen in trans- 

 verse section, on its way through the secondary wood (PL I. fig. 7 It, and PI. IV. fig. 

 30 It). In the broad medullary ray shown in the slightly oblique tangential section 

 of the wood in PI. I. fig. 7, it is possible by careful focussing to examine the walls of 

 the tissue, and in every case the elements appear to possess the characteristic thickening 

 bands. In radial section the medullary -ray elements are clearly seen in PI. IV. fig. 32 ; 

 this section shows the radially elongated form and other characters of the tissue. The 

 simple nature of the pits in the secondary tracheids which abut on medullary-ray 

 elements is demonstrated in PI. IV. fig. 29. 



There is the same absence of any concentric lines of growth in the secondary 

 tracheids as in Palaeozoic stems generally. Here and there, however, the uniformity 

 in the size of the constituents of the wood is interrupted by the occurrence of groups of 

 smaller elements suddenly abutting on the walls of the larger tracheids. This is well 

 shown in PI. II. fig. 11. Some of the tracheids immediately internal to the patch of 

 smaller elements are seen to have their cavities occupied by thin-walled parenchymatous 

 cells t ; a similar instance of tiillen in the tracheids has also been observed at one place 

 in a longitudinal section of the wood. This local irregularity in the size of the tracheids 

 may, no doubt, be regarded as the expression of some interference with the normal 

 progress of secondary growth. 



iv. Leaf -trace bundles. — The prominent ridges of the corona, as seen in PI. IV. fig. 

 24, gradually detach themselves from the primary wood and constitute the leaf-traces in 

 such manner as Bertrand and other writers have described in various Lepidodendroid 

 plants. For a short distance a leaf-trace pursues an almost vertical course approxi- 



* Williamson and Scott [(94), p. 926], in their account of the secondary thickening in Sjjhenophyllum plurifoliatum , 

 speak of part of the secondary xylem as separated from the primary xylem by a layer of thin-walled tissue. 



t Cf. the medullary rays of Stigmaria ficoides, Lepidodendron vasculare (specimens 362 and 1609, Williamson Col].), 

 also L. brevifolium (No. 489), etc. 



