OF THE FOSSIL PLANTS OF THE COAL-MEASURES. 
207 
nearly parallel to the last, but a quarter of an inch nearer the free extremities of the 
leaves. Each petiole is now seen to be deeply indented by a sharply defined groove 
running along the centre of its upper surface, which appears to have become generally- 
depressed, the lower surface having undergone little change of form ; but the two margins 
of each petiole have become winged or prolonged laterally into broad membranous 
expansions (18, l'), explaining corresponding appearances seen in the longitudinal section 
(16, l’). These membranous expansions are mutually disposed with the utmost regula- 
rity, the one proceeding to the right, from each petiole, always overlying the margin of 
that approaching it from the opposite direction. The position of the vascular bundle of 
each petiole is now very distinctly marked by a small semilunar opening (m), from which 
the vessels have disappeared. A corresponding section of the specimen represented by 
fig. 20 displays this sharply defined semilunar orifice yet more strikingly. 
The section Plate XXVI. fig. 19 is a transverse one, made in the line x, x of fig. 16, 
so as to intersect some of the petioles near their extremities, and yet at right angles to 
their longer axes, as is done at l". They are here seen to have become yet thinner and 
more flattened at their central portions, though retaining the central groove on each 
upper surface. The same regularity in the superposition of the thin margins is found 
here as in fig. 18. Fig. 19, k represents the layer of tubular prosenchyma, and l the 
turgid bases of three more petioles. That the latter only exhibit the marginal membra- 
nous expansion on one side (19, l') is due to the fact that the section has passed obliquely 
through the petioles, and only crossed in the plane of those expansions on one side. 
Plate XXVI. fig. 20 is a longitudinal section of a fragment of the same species of Lepi- 
dodendron as the last, supplied to me by Mr. Whittaker, of Oldham, and which I have 
figured because it displays very clearly the somewhat elongated form of parenchyma of 
which these petioles consist. At their free extremities the latter have become yet more 
membranous than is the case with those of the corresponding, but less highly magnified, 
figure 16. This difference indicates that the shrivelling process incident upon the decay 
of the leaves has been, as might be expected, a gradual one ; and that by the time it 
reached the portions of the petioles represented in the tangential section fig. 17, a 
natural cicatrix would have been formed at which the decay would be arrested, and which, 
when the shrivelled portion had fallen away, would exhibit the ordinary lozenge-shaped 
scars seen in the common examples of Lepidodendron. The generic distinctions that 
have been drawn between types that retain and others that do not persistently retain 
their petioles, appear to me to be of more than doubtful value, since, in all probability, 
they represent temporary rather than permanent conditions. That some species have 
retained their petioles longer than others is sufficiently probable, but I believe this to be 
all that is implied by such differences. 
Resuming the task of tracing the development of the exogenous type amongst these 
Lepidodendroid stems, we come to a series of specimens of which that represented in 
Plate XXVIII. fig. 21 & Plate XXVI. fig. 22 is a marked example. These plants 
correspond very closely with the Anabathra of Witiiam and with the Diploxylon of 
2 e 2 
