STRUCTURE AND AFFINITIES OF A LEPIDODENDROID STEM. 913 



afforded by the photograph in PL III. fig. 23 ; in the upper part of the section the more 

 interna] secondary elements are shown, with their reticulate, spiral or scalariform bands, 

 and below we have the more delicate recently formed peripheral tracheids. This section 

 shows very clearly the meristematic tissue which has been pointed out in fig. 19 ; the 

 cells have extremely thin walls and square ends ; they are slightly longer than broad, as 

 seen in tangential section (PL III. fig. 23 m, m). In PL III. fig. 17 the transverse 

 section has just missed a leaf-trace, but the accompanying thin- walled meristem is 

 clearly shown, with the cells disposed in a fan-shaped manner. 



The presence of undoubted secondary xylem in the leaf-trace bundles is a fact of 

 considerable importance from the point of view of the affinities of the plant, and as 

 proving that the distinction insisted upon by some writers between the Sigillarise of the 

 section Leiodermaria and Lepidodendroid leaf-bundles cannot be maintained .* 



Owing to the absence of the greater portion of the cortex in the Dalmeny stem, the 

 course of the leaf-traces cannot be followed through the cortex, but the partially 

 disorganised tissues of two traces have been met with in the outer cortical tissues. 

 Before passing to the consideration of the outer cortex, the tissues which immediately 

 succeed the secondary xylem, and the question of secondary growth must be dealt 

 with. 



v. Tissues in contact with the xylem ; and secondary growth of the central cylinder. 

 — The first point to notice in connection with the growth of the secondary wood is the 

 juxtaposition of small tracheids and the most external tracheids of normal size, as in 

 the local patch shown in fig. 11, PL II. Instead of finding the large tracheids at the 

 limits of the secondary wood succeeded by thinner-walled elements of the same breadth, 

 which gradually pass into obvious cambium cells, we have a sudden transition from the 

 wide elements (t', PL III. fig. 19) to the much narrower elements t't". It is physically 

 impossible that the latter should ever have been able to alter their form so as to fit on 

 to the older and larger tracheids. It would seem that the constant association of these 

 dissimilar tracheids at the periphery of the wood points to the existence of unfavourable 

 conditions of growth ; had the stem continued to grow, the period during which such 

 conditions occurred would be shown by the occurrence of a ring of smaller tracheids ; 

 an extension round the whole circumference of such inequalities as are found here and 

 there in the secondary wood {e.g., PL II. fig. 11). It may not be unreasonable to 

 suggest that this irregularity in the growth of the wood was due to the harmful effects 

 of neighbouring volcanic activity. Prof. Williamson has expressed the opinion that the 

 large Arran trees (" L&pidodendron Wunschiannni"} perished probably "in conse- 

 quence of the mephitic vapours which filled the atmosphere"; the forests grew in a 

 district " as volcanic as Auvergne." t 



* In a tangential longitudinal section through the secondary wood of a specimen of Lepidodendron vasculare (No. 

 495) in the Williamson Collection, an outgoing leaf-trace seen in a wide medullary ray shows a tendency of some of 

 the elements to arrange themselves in a fan-shaped manner like that of the secondary tracheids in PI. I. fig. 7 and 

 PI. IV. fig. 30. 



t Williamson (96), p. 175. 



