DEVONIAN PLANTS. 27 



My first examination of sections of this plant, showed that I had to deal with a 

 structure of much finer quality than found in any of the species of Nematophyton, and I 

 was at first led to consider that it might be a plant of very different type. Upon closer 

 inspection, however, this view reqtiired important modification and, as will be seen from 

 the following description, it now seems most probable that we have to deal with an 

 organism which, if not generically related, is at least allied to Nematophyton through the 

 general character of its structure. 



The specimens examined hj me are all somewhat highly crystalline and the structure 

 has, in consequence, suffered important modifications which, although the general form 

 and disposition of the cells can still be made out, have tended to obscure the structure 

 and render details difficult of determination. The material is also, in consequence of its 

 crystalline character, very friable, and this has seriously interfered with securing sections 

 of sufficient thinness to make a close examination possible. 



In transverse section the structure is found to consist of small and closely compacted 

 cells — the tissue being so dense as to make the contiguous cells appear in most intimate 

 contact. This relation is probably the result of alteration and not that which existed in 

 the growing plant, since cells are found at intervals which are as distinct as in Nemato- 

 phyton, while it is also evident from the oblique section of many, that they were not all 

 parallel to the axis of growth. 



The cell walls are thin, and it is also a fact of some interest, in this connection, that 

 the tissiie presents almost the exact appearance of the intercelltilar structure seen in cross 

 sections of N. Logani, with reference to size, form and general disposition of the cells and 

 thickness of the walls. Occasionally, somewhat linear tracts are found, within which the 

 cells follow a direction transverse to the axis of growth. 



Scattered through the otherwise homogeneous tissue at rather wide intervals, are 

 relatively large, rounded or oval openings. These, so far as can be determined at present, 

 appear to be structureless, i.e. there is no sejjarate wall, nor are there any surrounding 

 cells such as form the resin passages of modern plants. They measure 10-13.5 ji in 

 diameter and their proper significance is at present a matter of doubt, although the in- 

 variable presence in them of highly crystalline silica siiggests that they may have been 

 caused by mechanical separation of the surrounding cells. Small and irregular open 

 tracts are also to be seen in the tissue, similar, although much smaller, to those in N. 

 crassum. 



In longitudinal section the cells are found to be distinctly vermicular, and although 

 the more compact nature of the stnicture tends to render them less sinuous than in either 

 N. Logani or N. crassum, yet the peculiar way in which the cells interlace and sometimes 

 cross one another very abruptly, leaves no room for doubt upon this point. Furthermore, 

 the cells are non-septate and in diameter are tolerably uniform, measuring 5 - 8 /<. In 

 this latter we again notice a curious resemblance to the intercellular structure of both 

 N. Logani and N. crassum, and also of iV. htxiim. In N. Logani the filaments show a variation 

 in size ranging 3.7-8.9//. In N. luxum the corresponding structures range 3 - 9 /( 

 and in N. crassum 2-10 //, so that the cells of the medulla in this plant may be fairly 

 regarded as essentially of the same size. Moreover, in N. laxum, it frequently happens 

 that the secondary filaments constitute^ the only structure within fairly large areas, 

 and in such cases these cells are generally found to run parallel to the axis of growth 



