396 PROCEEDINGS OF THE GEOLOGICAL SOCIETY. [Apr. 22, 



into bright coal, is very thin ; it is marked with longitudinal furrows 

 and ridges ; the latter are about half an inch in width : the furrows 

 are deep, but not always parallel and continuous, frequently waving 

 and running into each other; they are also occasionally wrinkled, 

 the effect probably of the squeezing which flattened the stem. As 

 the furrows approach the base of the tree they become less distinct, 

 but nevertheless do not wholly disappear ; they may even be traced 

 some six or eight inches down the upper surface of the roots. I 

 could not observe any appearance of leaf-scars. 



The stem and roots are filled with a fine-grained greyish white 

 sandstone. In the latter I observed some slight remains of a central 

 pith, but there are no traces whatever of organic structure in the 

 sandstone of the stem. 



The roots, which are true Stigmarige, with rootlets or (as they 

 generally have been called) leaves spreading out in every direction, 

 are about three inches in diameter at their junction with the stem; 

 at their extremities, or as far as they could be traced, they are flat- 

 tened to a depth of about one inch : two of the roots, it will be ob- 

 served by fig. 3, have been followed to their terminations, where 

 they gradually thinned out to a mere line in one direction, being 

 about three-fourths of an inch in width. They are thickly studded 

 with tubercles, presenting an imperfect spiral arrangement, and co- 

 vered with a thin bark or coating of carbonaceous matter. 



The rootlets, varying in length from three to twelve inches, as 

 shown in fig. 2, are compressed or flattened, being much broader 

 near their junction with the roots than at any other point : they 

 must have been composed of some soft substance, as they are in all 

 cases flattened, whether lying in horizontal or vertical positions. 



All the other stems represented in the section resemble so nearly 

 the one I have attempted to describe, that it is not necessary' to refer 

 to them further, enough having been said, I hope, to prove that 

 Stigmariae and their leaves are in reality the roots and rootlets of a 

 class of trees allied probably to Sigillariae. If this be admitted, we 

 must conclude that all beds in the coal-measures containing Stig- 

 mariae, with their delicate rootlets united to them, have at successive 

 periods been surface-soils, supporting forests of trees ; that in the 

 section before us, which represents only a perpendicular thickness 

 of forty-four feet, there have been at least seven such surfaces, cor- 

 responding with an equal number of subsidences ; and that the in- 

 tervals of rest between each successive subsidence were compara- 

 tively small, until the bed became the surface-soil, which continued 

 in a state of repose of suflicient duration to produce the immense 

 mass of vegetable matter required for the formation of the main 

 coal. 



