THE BOTANY OF A COAL MINE. 
293 
markings, which are the only characters by which Sigillaria 
and Lepiclodendron can be separated, show that the fossils so de- 
scribed were nearer to, if not genuine species of, the latter genus. 
In the absence, however, of stems showing structure, several 
specimens of Stigmaria have been found in which the various 
tissues have been preserved. In a beautiful series forming part 
of the collection of the late Eobert Brown, I have traced the 
various parts. The axis was composed of elongated scalariform 
cells. Around this was a compact cylinder of scalariform vessels, 
perforated by bundles, which passed to the rootlets, but without 
any trace of medullary rays. This was surrounded by a con- 
siderable mass of cellular tissue, forming the great bulk of the 
root, through which passed the bundles to the rootlets. 
The scalariform tissue and the absence of medullary rays are 
of importance in estimating the systematic position of the fossil, 
and these both clearly point to its being a true cryptogam. As 
the structure of Stigmaria agrees with what is known of the 
stems of species described as Sigillaria elegans , S. vascularis , 
Lepiclodendron Harcourtii , &c., there can be no doubt that it 
exhibits the true structure of Sigillaria itself. It is true that 
Dawson refers to this genus a stem composed chiefly of disc- 
bearing woody fibre, but it is not evident from either the 
drawing or description, that this remarkable stem really belonged 
to Sigillaria ; and it differs so much from the known structure 
of the roots as well as of the allied forms named that we hesitate 
to place it there. The medullary rays described and figured by 
Brongniart have not been seen in any other specimen, and an 
examination of his beautiful drawings convinces me that the 
interspaces between the vascular tissue, which he has described 
as medullar}^ rays, are accidentally produced by the splitting up 
of the tissues. He found no indications of the walls of the 
medullary cells, which would certainly have been apparent in a 
fossil so perfectly preserved as his Sigillaria elegans . 
All that is certainly known of Sigillaria , tends to establish 
the opinion advanced by Hooker, that it is allied to Lepido- 
dendron , and consequently belongs like that fossil, as we shall 
presently see, to the order Lycopodiacece. 
The extent of our acquaintance with Lepiclodendron is in 
singular contrast with what we know of Sigillaria. The only 
part of the plant about which there is the least uncertainty is 
the root. The stem, branches, foliage and fruit are all well 
known. 
Lepiclodendron was a branching tree of considerable size. 
It is distinguished from the other genera of coal plants by the 
lozenge-shaped leaf scars being arranged spirally on the stem. 
The axis of the stem (pi. xvi. fig. 12) is composed of elongated 
utricles of various sizes, irregularly arranged, and having thin 
