ORGANIZATION OF THE FOSSIL PLANTS OF THE COAL-MEASURES. 735 
exactly at its junction with the wood of the stem. It may be mentioned that root 4 
is giving off a rootlet. Fig. 184 from the next section shows the roots 3 and 4 in 
obliquely longitudinal section, the plane of which passes through their central 
cylinders. The vascular tissue of root 4 can here be traced through the cortex of the 
stem, The great interest of this series lies in the fact that it shows at one and the 
same time the characteristic root-structure of these organs and their continuity as 
endogenous appendages with the stem of Lyginodendron. 
In order to bring out the point in question more clearly, we have figured one of the 
roots on a larger scale (Plate 24, fig. 17). This is root 3 of figs. 18 and 18a. The 
more magnified figure is taken from the third section of the series, in which this root 
has become free from the stem, and is seen in complete transverse section. The root- 
like anatomy is here perfectly evident. The primary wood of the central cylinder has 
seven angles, at which the smallest trachese are situated, and is thus of heptarch 
structure. A certain amount of conjunctive parenchyma is present among the 
primary tracheze. There is a zone of secondary wood, which is so distributed that a 
principal ray corresponds in position to each of the protoxylem angles of the primary 
wood. The inner cortex is somewhat lacunar, and contains a great number of the 
dark-coloured elements, which we regard as secretory sacs. The cortex is limited on 
the exterior by a well-marked zone of large, clear elements. 
This brings us to our second point. Not only do these appendages of the stem of 
Lyginodendron present all the characters of roots, they are further identical with 
Kaloxylon Hookeri. This becomes evident at once on comparing our fig. 17 with 
Memoir VII., Plate 5, fig. 23, or with other transverse sections represented ir. 
Memoirs VII. and XIII., or again with our own, Plate 20, photograph 13, alf of which 
represent transverse sections of typical Kaloxylon. In every respect the identity is 
perfect: in the structure of the primary wood, the character and arrangement of the 
secondary tissues, the details of the cortex, and especially in the characteristic 
external zone, termed the “epidermal layer” in previous memoirs.* In all the 
points, in fact, by which Kaloxylon Hookeri was characterized, the roots of Lygino- 
dendron show an exact agreement with that fossil. 
If, after the consideration of this specially clear case, we refer back to photographs 
8, 9, and 10, we shall see that the roots there figured are of the same nature, though 
their Kaloxylon structure is less obvious, owing to slight modifications in that part 
of the root, which passes through the cortex of the stem. 
As regards the distribution of the adventitious roots on the stem, we have little to 
add to what has already been said. The bases of roots are very commonly found, 
especially on the larger stems. On the other hand, they were not present on all 
parts of the stem. Thus the series 1191-1198, which extends through the whole 
length of an internode, with the adjacent nodes, nowhere shows any sign of an 
adventitious root. 
. * Witttamson, “ Organization,” Part VII., 1875, p. 16; Part XIIL, 1887, p. 295. 
